Search Results

With reference to present literature, most OEMs are working on reducing product development time by around ~20%, through seamless integration of digital ecosystem and focusing on dynamic customer needs. The Systems Engineering approach focuses on functions & systems rather than components. In this approach, designers (Computer Aided Design) / analysts (Computer Aided Engineering) need to understand program requirements early to enable seamless integration. This approach also reduces the number of iterative loops between cross functions thereby reducing the development cycle time. In this paper, we have attempted to tackle a common challenge faced by Closures (Liftgate) engineering: meeting slam durability fatigue life while replicating customer normal and abusive closing behavior.

Nowadays, the rapid growth of civil aviation transportation demand has led to more frequent flight delays. The major problem of flight delays is restricting the development of municipal airports. To further improve passenger satisfaction, and reduce economic losses caused by flight delays, environmental pollution and many other adverse consequences, three machine learning algorithms are constructed in current study: random forest (RF), gradient boosting decision tree (GBDT) and BP neural network (BPNN). The departure flight delay prediction model uses the actual data set of domestic flights in the United States to simulate and verify the performance and accuracy of the three models. This model combines the visual analysis system to show the density of departure flight delays between different airports. Firstly, the data set is reprocessed, and the main factors leading to flight delays are selected as sample attributes by principal component analysis.

With the popularization of electric vehicles, the safety performance of electric vehicles has drawn much attention. However, the gears of electric vehicles are more prone to failure at high speeds, which can affect the safety performance of the vehicle. This topic proposes a electromechanical coupling model, which is composed of a permanent magnet synchronous motor model, a vehicle longitudinal dynamics model and a transmission system model, and will be applied to gear fault diagnosis. First, the sensitivity of the gear fault to the stator current signal, the electromagnetic torque signal and the q-axis current signal is investigated based on the time-varying meshing stiffness obtained by the potential energy method. The discrete wavelet algorithm is used to decompose the stator current signal, and the d1 component with obvious fault information is obtained.

Time-domain and frequency domain methods are two common methods for fatigue damage and life assessment. The frequency domain fatigue assessment methods are becoming increasingly popular recently because of their unique advantages over the traditional time-domain methods. Recently, a series of moment of load path based multiaxial fatigue life assessment approaches have been developed. Among them, the most recently developed effective second moment of load path (ESMLP) approach demonstrates its potentials of conducting fatigue damage and life assessment accurately and efficiently. ESMLP can be used for fatigue analysis even without resorting to cycle counting because of its unique mathematical and physical properties, such as quadratic form in the kernel of the moment integral, rotationally invariant, and being proportional to damage. Developing a better parameter for frequency-domain analysis is the driving force behind the development of ESMLP as a new fatigue damage parameter.

Switching frequency is very important to the control of the inverter as well as driving motor system. This paper presents a design method of model predictive direct torque controller for permanent magnet synchronous motor considering switching frequency optimization. Firstly, the mathematical models of PMSM and inverter and their integrated models in discrete time are constructed. Secondly, the principle of the model predictive control method and its application in direct torque control are described, and the controller design is carried out using its integrated model. When constructing the cost function, the optimal switching frequency of the inverter is used as a condition to further limit it. Finally, the simulation models of the proposed method and the traditional direct torque control method are established to verify their superiority.

Vehicle failure prediction technology is an important part of PHM (Prognostic and Health Management) technology, which is of great significance to the safety of vehicles and to improve driving safety. Based on the vehicle operating data collected by the on-board terminal (T-box) of the telematics system, the research on the state of vehicle failure is conducted. First, this paper conducts statistical analysis on vehicle historical fault data. Preprocessing procedures such as cleaning, integration, and protocol are performed to group the data set. Then, three indexes including recency (R) frequency (F), and days (D) are selected to construct a vehicle security status subdivision system, and K -Means algorithm is utilized to divide different vehicle categories from the perspective of vehicle value. Labeled information of vehicles in different security status are further established.

At present, the remote monitoring cloud platform of many automobile companies only displays the collected data information, and it does not fully mine the deep-level information of the data. This paper uses data mining and machine learning methods to build a driver's driving style and driving condition prediction and recognition model based on the historical driving information generated by the vehicle, so as to improve the supervision and safety of the driver and the vehicle by automobile companies and other automobile-related industries. First, 36 standard driving cycles are utilized to construct an initial operating condition block data set. Second, we obtain the feature variables of driving style and driving conditions through feature engineering, and two recognition model data sets use the principal component analysis (PCA) and clustering algorithm for data dimensionality reduction and cluster analysis.

For the cases where perfect interface is not assumed and crack propagation path is unknown, the fully embedded zero-thickness cohesive element model (FECEM) is an alternative simulation method to model crack growth. In our newly developed FECEM model, the common element is triangle 3-node element under two-dimensional condition and the interface element is a quadrilateral cohesive element with zero thickness. From the simulation by FECEM, it is found that the elastic modulus of the second phase particle in a ductile matrix has a significant effect on crack propagation behavior. When the particle encounters the propagating crack, if it is softer than the matrix, the crack will be attracted towards the particle and then will puncture into it. If the particle is harder than the matrix, it will slow down the propagation of the crack.

Simulating quasi-static crack to obtain crack tip stress field is a criterion to evaluate whether a numerical simulation model is good at modeling discontinuous problems such as cracks. In this study, a fully embedded zero-thickness cohesive element model (FECEM) was developed based without assuming a perfect interface. By using XFEM (extended finite element method) and FECEM, respectively, 2-D finite element models were constructed to simulate the unilateral I-type and II-type cracks and mixed I-II type crack on 2-D rectangular plate. In XFEM simulation, J-integral was used to calculate the stress intensity factor on the crack tip, which was compared with the theoretical solution. Comparison of the simulated results by XFEM and FECEM models confirms that our newly developed FECEM model has a high reliability in simulating quasi-static crack without assuming a perfect interface.

Lane keeping assist (LKA) is an autonomous driving technique that enables vehicles to travel along a desired line of lanes by adjusting the front steering angle. Reinforcement learning (RL) is one kind of machine learning. Agents or machines are not told how to act but instead learn from interaction with the environment. It also frees us from coding complex policies manually. But it has not yet been successfully applied to autonomous driving. Two control strategies using different deep reinforcement learning (DRL) algorithms have been proposed and used in the lane keeping assist scenario in this paper. Deep Q-network (DQN) algorithm with discrete action space and deep deterministic policy gradient (DDPG) algorithm with continuous action space have been implemented, respectively. Based on MATLAB/Simulink, deep neural networks representing the control policy are designed. The environment as well as the vehicle dynamics are also modelled in Simulink.

As a global optimization method, dynamic programming (DP) can be employed to seek the optimal velocity with minimum energy consumption for EV on given driving cycles. Due to its terrible computational burden, conventional DP is not suitable for real-time implementation especially with higher dimensions. In this paper, we propose an iterative dynamic programming (IDP) approach to reduce computing time firstly. The IDP can obtain the optimal control laws alike the conventional DP by converging the optimal control strategy iteratively and save considerable computing time. Second, the developed IDP and model predictive control (MPC) are combined to establish a real-time cruising controller called IDP-MPC for an EV with terrain preview. In the predictive controller, we use the IDP to solve a constrained finite horizon nonlinear optimization problem.

The initial presence and dynamic formation of internal voids in human body models have been subjects of discussion within the human body modeling community. The relevant physics of the human body are described and the importance of capturing this physics for modeling of internal organ interactions is demonstrated. Basic modeling concepts are discussed along with a proposal of simulation setups designed to verify model behavior in terms of volume and pressure between internal organs.

Prediction and reduction of noise/vibration at the early design stage is important for motor design. Rapid design iterations require a platform where electromagnetic, structural and acoustic solvers can communicate with each other without user scripting or interventions. Based on the platform, multiple designs in a given design space need to be analyzed by distributed high performance computers automatically. To demonstrate such a multiphysics multi-objective optimization workflow, four geometrical variables for an interior permanent magnet motor are selected for optimizing the electric and acoustic performance (Figure 1). Average torque and equivalent radiated power level (ERP) are calculated for multiple design points and response surfaces are then created for the sensitivity study and optimization.

The paper proposes an energy-oriented torque allocation strategy to reduce the energy consumption of four-wheel independent driving (4WID) electric vehicles (EVs). To compute its energy efficiency accurately, the measured efficiency map is fitted by the cubic spline interpolation method. The energy-oriented torque allocation strategy is designed by minimizing the energy consumption during vehicle driving and braking. According to the varying requirement of torque and speed, the torque of front and rear axle cab be allocated dynamically by adjusting the torque allocation coefficients. To ensure the high efficiency of motors and reduce the energy consumption caused by current shock, the torque allocation coefficients are obtained from two kinds of sub strategies. A fuzzy logic controller is designed to combine the derived torque allocation coefficient above, by adopting Mamdani structure with 2 inputs and 1 output.

Connected and automated vehicles (CAVs) are gaining momentum, especially in the potential to improve road safety and reducing energy consumption and emissions. Lane-change maneuver is one of the most important conventional parts of automated driving. We address the problem of optimally CAVs to accomplish an automated lane-change and eliminate potential collision during the lane-change process on a curved road. Drivers’ safety, comfort, convenience, and fuel economy are also engaged in trajectory planning. We assume that the centripetal motion displacement and the rotational angular displacement meet the requirement of odd-order polynomial constrains. Then, the polynomial coefficient of the trajectory can be reduced and the mathematical model of virtual trajectory for lane-change can be designed based on the models of centripetal displacement and angular displacement by applying the above constrains and boundary conditions.

This paper proposes a semi-active hybrid battery system (HBS), composed by lithium iron phosphate battery (LFP) and lithium titanate battery (LTO) for electric vehicle (EV) to reduce the life cycle cost of energy storage system. Firstly, the topology of this HBS is introduced. The high energy-density battery, LFP is adopted as the primary energy source, while the high power-density one, LTO is connected in parallel with a bidirectional DC-DC converter and used as secondary energy source to extend the lifetime of HBS by reducing the current stress of LFP. The dynamic model of this HBS is built, in which, the LFP and LTO are both modeled as second-order RC model. In addition, dynamic semi-empirical degradation model of the LFP battery is chosen to estimate the lifetime of HBS. Secondly, a fuzzy logic controller with 3 inputs and 1 output is proposed to decide the power split between the primary and secondary power sources.

The combustion efficiency of direct injection engines is largely dependent upon the mixing of fuel in air, thereby creating a combustible mixture. Such a process is highly dependent upon the motion of the charge in the cylinder. The shape of the intake runners and valves determines the charge motion generated within the engine. Swirl and tumble, generated along the vertical and horizontal axis respectively, govern the charge motion and hence distribution of combustible mixture. Unlike traditional parametric optimization where the parameter space has to be predetermined, adjoint optimization utilizes the gradient of objective functions obtained from a computational fluid dynamics solution to modify the shape of the original CAD geometry. During the optimization process, specific parts of the geometry can be morphed in any direction freely. The final design is a fluid volume generated as a result of such adjoint computations.

An actual trend in the automotive industry is to have global products in order to have economy of scale. This paper presents how a Belt Drive Rack EPS developed for the North American market had to be modified in order to be assembled in a Vehicle sold all around the world. Main technical challenges for achieving that goal were generated from different Architectures, whether electrical or mechanical, used in each vehicle, Packaging issues and Regional Requirements. Main features affected are Database Configuration, Electromagnetic Compatibility, Smooth Road Shake mitigation and Pull Compensation.

Trivial Principal Component method (TPC) was developed recently to model a system based on measured data. It is a statistical method that utilizes Eigen-pairs of covariance matrix obtained from the measured data. It determines linear coefficients of a model by using the trivial eigenvector corresponding to the least eigenvalue. In general, linear modeling accuracy depends on the strength of nonlinearity and interaction terms as well as measurement error. In this paper, the TPC method is extended to analyze residual (error) vector to identify significant higher order and interaction terms that contribute to the modeling error. Subsequently, these additional terms are included for constructing a robust system model. Also, an iterative TPC analysis is proposed for the first time to correct the model gradually till the least eigenvalue becomes minimum.

Driver out-of-position (OOP) tests were developed to evaluate the risk of inflation induced injury when the occupant is close to the airbag module during deployment. The Hybrid III 5th percentile female Anthropomorphic Test Device (ATD) measures both sternum displacement and chest acceleration through a potentiometer and accelerometers, which can be used to calculate sternum compression rate. This paper documents a study evaluating the chest accelerometers to assess punch-out loading of the chest during this test configuration. The study included ATD mechanical loading and instrumentation review. Finite element analysis was conducted using a Hybrid III - 5th percentile female ATD correlated to testing. The correlated restraint model was utilized with a Hybrid III - 50th percentile male ATD. A 50th percentile male Global Human Body Model (HBM) was then applied for enhanced anatomical review.