Search

In this study, the simulation parameters obtained from experiment were used to verify the heat transfer parameters of graphene thermal interface materials (TIMs) on a single chip. The optical and power parameters of the unfilled high-power commercial LED lighting are measured by using integrating sphere and power analyzer in order to establish a fast and accurate simulation mode. The heat dissipation effects of composite and multilayer graphene-made TIMs in a closed and compact space are investigated using the heat-radiating materials with different thermal emissivity. By applying the established simulation mode, the simulation error in the main heating source of the light-emitting diode is between 3% and 5%. The use of composite graphene-made TIMs in the LED device can effectively reduce the operating temperature of the PCBA in a closed and compact space. The experimental result of temperature drop of the chip is up to 6.8 ° C and the simulation result shows the temperature drop is up to 3.9°C.

This paper aims to design an electrical holding device using NdFeB N35 permanent magnet to achieve saving energy as well as increase holding power. The NdFeB N35 permanent magnet is placed inside the iron yoke so that it retains the magnetic force at all times, and the electrical coils energizes to increase or decrease the surface magnetic force to achieve stronger adsorption or easy removal of the adsorbate. The design is achieved through the analysis of the magnetic model in ANSYS Maxwell. The analysis errors between the ANSYS Maxwell model and the prototype device are under 5%. Then, under the condition when the external shape is unchanged, the ANSYS Maxwell model is applied to obtain the optimal suction power with respect to the variable internal dimensions. The yoke size can change the relative position and size of the coils to form a new magnetic circuit. When the coil is less than 1mm from the surface of the suction cup, the surface magnetic flux density will be largely increased. At the same power, increasing the winding coils does not improve the performance of the suction power because of increasing the coil resistance. The enlarged voltage input can raise the coil current such that the surface magnetic flux density can be enhanced to produce the more suction power. In the future, the proposed approach can be applied the electromagnetic chuck design in robotics application.

Traditional IC packaging requires chips to be assembled at the same level, while recently thrived 2.5D/3D IC packaging utilizes skyscraper approach to stack various types of chips with diverse functions occupying similarfootprint, and this approach not only can reduce overall package size, but also can improve electrical interconnection performance. The primary difference between 2.5D/3D IC lies in the implementation of medium wafer with Through Silicon Vias (TSV), called Through Silicon Interposer (TSI). Although 2.5D IC packaging possesses the advantages of low power consumption, high data bandwidth, great electrical performance, and miniaturized form factor, the challenges of huge investment required on technology development and process manufacturing comparing to traditional 2D IC can't be ignored. Therefore, numerous innovative concepts have been proposed to reduce the cost on TSI fabrication, including but not limited to, different method of electrical interconnection, cheaper interposer utilization from material point of view, or even reconsideration of the necessity of interposer. One of the latest approaches is to utilize Fan-Out (FO) configuration to fully remove the requirement of interposer and use Epoxy Molding Compound (EMC) notonly as encapsulation of functional dies but also as structure supporter for further assembly processes. The above approach utilizes multi-layer structure composed of passivaton (PSV) layers and Re-Distribution Layers (RDL) on a carrier as frontside process supporter, and subsequent compound molding process is implemented to supportbackside process after the removal of carrier mentioned herein. For multiple processes and materials involved in fabricating diverse kinds of packages, one of the most important technical difficulties is warpage control. Since warpage behavior of the multi-layer structure directly influences the workability of both bumping and assembly processes, it's extremely crucial to control warpage in advance to ensure smooth subsequent wafer handling and processing afterwards. In order to have desirable warpage at final stage, the material properties and mechanicalbehaviors of individual materials within this multi-layer structure should be further investigated and characterized. There were much efforts dedicated on the warpage characterization of different types of composite structures in the past few years, and we have also demonstrated our research progress previously regarding warpage control from numerous diverse factors at bumping process stage, including carrier types, PSV layer material types, PSV layer thickness, PSV layer quantities, and RDL pattern density. These crucial factors have been discovered and proven to affect warpage to a different extent at bumping process stage. In our latest studies, the scope of warpagecharacterization is further expanded and widened and specifically focused on post-bumping assembly processes, including compound molding and grinding process. The parameters of EMC, including but not limited to, material type, Coefficients of Thermal Expansion (CTE), thickness, shrinkage ratio, and EMC ratio within the whole package. The correlation between warpage and these factors are elaborated and discussed in this paper. With these findings, wafer warpage of the entire composite structure composed of PSV films, RDLs, and EMC are characterized andcomprehensive understanding about the effects of different factors can be obtained and thus assist us in warpage modeling for different types of newly-developed innovative packages in the years to come.

In this study, Nd 2 O 3 was used as the dopant and the commercial Sr 2 SiO 4 :Nd phosphors was produced by using solid-state reaction method. The Sr 2 SiO 4 :Nd phosphors was sintered at the temperature range of 1000°C∼1300°C for 2 h. From the results of X-ray diffraction patterns, even sintered at 1000°C only Sr 2 SiO 4 phase was observed, no secondary and unknown phases were residual in the calcined Sr 2 SiO 4 :Nd powders. We had found that the sintering temperature had large effect on the luminescence characteristics of the Nd 2 O 3 -doped Sr 2 SiO 4 phosphors. The CIE colorimetric data of Sr 2 SiO 4 :Nd phosphors as a function of sintering temperature was also investigated in this study. We found that the Sr 2 SiO 4 :Nd phosphors revealed a dark blue or a ultraviolet light.

As the area of glass panel becomes larger and the thickness becomes thinner (from 0.5mm to about 0.3mm), the method for the load of glass panel under the mass-production will be changed to another method, for example by platform with auto load function. In this study, we investigated a platform for generation six (G6, 1850 × 1500 mm) to generation eight (G8, 2460 × 2160 mm). The investigated auto panel-cutting platforms had the human-machine interface which could be designed to have the function of choosing the different cutting area. The investigated platform has two special functions, which can be controlled by the human-machine interface. The first is the diagnosis model, which is used to choose the cutting area for the analysis of the fail problem. The second is the accepted goods and unaccepted goods (OK/NG) model. The area of cutting glass panels could be controlled by the computer-integrated manufacturing (CIM) program.

By using Lyapunov equation approach associated with some linear algebraic techniques, new simple criteria for the robust stability and decay rate test problem of the time-delay systems are proposed. Comparing to well-known results, the presented criteria are sharper. The presented schemes then are applied to solve the mentioned problems for uncertain time-delay systems. Delay-dependent and delay-independent conditions for the robust stability and decay rate test are also developed. An interesting consequence of these results is that it is not necessary to solve any Lyapunov equation although the Lyapunov equation approach is used. Numerical examples are also given to compare the tightness of the presented results and some appeared in the literature.

With the computation power of the state-of-the-art processors and digital electronics, fast microprocessors and combination-logic devices, such as CPLD and FPGA, have become dominant in control engineering. Engineers usually expect that designed controllers can be realized accurately with fast processing speed. However, though the speed of digital electronics has been increased dramatically in the past decade, the resolution has not been improved significantly. As the demand of precision control increases with faster response, calculated outputs of the synthesized controller with fast sample rate but insufficient resolution will not only be distorted, but it can also lead to undesired results or become unstable. This paper provides a guideline to estimate the required wordlength for digital controllers in state-space realization.

This paper investigate on motion synchronization of multiple axes systems. Two different control strategies, a cross-coupling controller in feedback loop and an adaptive controller, were used to synthesize the synchronization compensator with the cross-coupling dynamics among the axes. By using these strategies, the asymptotic convergence of both tracking and synchronization errors are achieved. The comparison of the two controllers of the nominal plant is discussed. Experimental results of a three-axis motion system that include system uncertainties are also illustrated to verify the effectiveness of the proposed approach.

This paper investigates on motion synchronization of a multiple axes system. Two different control strategies, a cross-coupling controller in feedback loop and a linear quadratic optimal controller, were used to synthesize the synchronization compensator with the cross-coupling dynamics among the axes. These two methods are corresponding to SISO and MIMO approaches. With the strategies, the asymptotic convergence of both tracking and synchronization errors can be achieved. Simulation and experimental results of a three-axis motion system illustrate the effectiveness of the proposed approach.

This paper proposes a cascade control strategy, combined of a modified sliding mode controller and a quantized current regulator, for a Brushless DC Motor (BLDCM) to achieve high-performance speed control. In the current control loop, the quantized control law is derived by minimizing the instantaneous square value of a performance index signal using the theory of variable structure control to perform a fast current control. Then, a modified sliding-mode controller (SMC) with time-delay estimation is presented in its outer control loop to enhance the speed control without the need for the identification of system parameters or the estimation of the bounds of disturbances. Simulation results demonstrate the effectiveness of the proposed scheme under the influence of system uncertainties and disturbances.

This paper presents a direct torque control (DTC) scheme using space vector modulation (SVM) to achieve high performance speed control of a permanent magnet synchronous motor. Based on the concept of a field-oriented control principle, a Proportional-integral (PI) flux controller is first established to achieve fast direct flux control and then cascade schemes of PI torque and speed controllers are presented to enhance the objective of speed control by the direct torque method. The effectiveness of the proposed control scheme under the load disturbance and parameter uncertainties is verified by simulation results.

A new control scheme for current distribution in parallel connected DC-DC converters is presented. It uses an optimal multi-input multi-output (OMIMO) discrete integral variable structure controller (DIVSC) to achieve both output current equalization and output voltage regulation. The controller improves the dynamic performance of a parallel buck converter; the output voltage is robust against load disturbances and the output currents are very well equalized. The analysis and design of the controller is developed. Automatic master slave method is adopted in this work. DSP based experimental results show good dynamic performance of the proposed control scheme.

This paper proposes a sliding mode controller (SMC) design for a Brushless DC Motor (BLDCM) to achieve high-performance speed control. Based on Pulse-Width-Modulation (PWM) technique, a PWM-based controller in the inner current loop is first developed to perform a fast current control. A sliding mode speed controller is then designed in its outer control loop to enhance the speed control. Finally, the effectiveness of the proposed control scheme under the load disturbances and parameter uncertainties is verified by simulation results and comparison with proportional-integral control (PIC) approach.

This paper proposes model reference adaptive control (MRAC) for an integrated design of a voice coil motor and flexure mechanism to achieve a high-precision position. Based on the continuous deformation property of the flexure mechanism, the positioning system possesses linear motional characteristics without influence from the nonlinear phenomena of backlash and friction. MRAC controller design is then applied to enhance the positioning performance under the condition of system parametric uncertainties. Finally, simulation results verify the effectiveness of the proposed integrated mechanism design and control scheme to achieve a positioning precision better than 1µm.

By using linear algebraic techniques, this paper proposes a simple approach to derive new matrix bounds of the solution of the continuous algebraic Lyapunov equation (CALE). Both upper and lower matrix bounds are presented. Compared to existing works on this topic, these newly obtained bounds are less restrictive and easier to calculate. A numerical example is also given on order to compare the tightness of the presented results with those appearing in the literature.

This paper describes a typical application of controlling the slope of a piezoelectric cantilever. An analytical model for the cantilever beam describing the dynamic relationship between tip slope and the applied voltage is introduced. With closed loop feedback control strategy, the trajectory tracking error of the cantilever deflection angle is evaluated. To control the slope of the cantilever precisely with discrete-time devices, a suitable processor with enough resolution is necessary. With the consideration of limited resolution, an estimation method of required wordlength for the digitalized controllers is also proposed. A discrete-time LQR controller realized with different resolutions was used to validate the proposed method.

In this paper, an integrated hybrid design, consisting of two independent sets of cylinder controllers (a feedforward controller and a fuzzy tracking controller) and one coordinate fuzzy controller, is proposed for a dual-cylinder electro-hydraulic lifting system. The system, which is driven by a hydraulic oil supply, is used to achieve a synchronized positioning objective with unbalanced loadings, uncertainties, and disturbances. With the proposed controller structure, motion synchronization commands are generated by a coordinate fuzzy controller. The commands are then dispatched to the cylinder controllers. The two independent sets of cylinder controllers adaptively enforce the position-tracking commands to the individual single-rod actuators. The experimental results verify that the proposed hybrid controller structure can achieve the synchronous positioning objective of two independent, linear electro-hydraulic actuators driven by a single hydraulic oil supply. Meanwhile, the maximum synchronization error of the experimental system can be maintained within plusmn4 mm.

In this paper, the algorithm of time-delay estimation is considered to estimate unknown dynamics and to fulfill the feedback linearization framework. The design of perfect tracking control rooting on the poles-zeros cancellation is adopted to push the system's output to follow the desired trajectory. Thus, the feedforward controller with the capability of perfect tracking control in cooperation with the feedback linearization realized by time-delay estimation can improve the robustness of high-accuracy positioning performance of transmission mechanisms under model variations and unknown disturbances. The experimental results show the proposed control strategy is promising for the application of positioning mechanism in the presence of unknown plant parameters and external disturbances.

Fast Fourier transform (FFT) is widely used for the signal processing because of its computational efficiency. Most power quality meters and digital relays adopt FFT-based algorithm to characterize harmonics of the measured signals. A typical FFT-based algorithm may lead to inaccurate results, if the system frequency varies. This paper proposes an improved FFT-based algorithm for measuring harmonics. The performance of the proposed algorithm is then validated by testing the simulated and actual measured signals. Results are compared with those obtained by the conventional FFT algorithm and by a power quality meter. It shows that the proposed algorithm is computational efficient and the solution accuracy is maintained.

A control strategy, which combines predictive control with zero phase error tracking controller (ZPETC) and time delay scheme, is developed to overcome positioning and tracking difficulties for transmission mechanism with friction. Time delay control is used to cancel disturbances and potential nonlinearities. Predictive control uses future and past information to maintain servo performance, regulator performance and robust stability under the influence of disturbance and measurement noise. ZPETC is included to improve the overall system bandwidth. A cycloidal velocity profile is used to generate smooth reference motion to avoid saturating the actuator. Realistic numerical simulations that considered the effect of sampling, quantization and friction were performed to investigate the effectiveness of the proposed control scheme. Encouraging transient response and steady-state control performance were observed in the results of positioning control simulation of a one-dimensional transmission mechanism.

In general the air conditioning system of a building accounts for almost 40% of the total energy consumption. The majority of the air conditioning usage occurs during the day when the building is occupied and the energy cost is higher. By shifting the portion of the peak hour usage to off-peak hours, one can potentially save up to 50% of the cost of energy for a typical office building. The paper presents a distributed building air conditioning control network that utilizes the off-peak hours to make and store ice for peak hour cooling use. One advantage of the proposed approach is that current building temperature control limits can be easily incorporated into the network, since the distributed control uses only the existing I/O information and open network communication protocol. The proposed network has the potential to include another environment control system into the existing temperature control system to form a complete building eco-control system to achieve more efficient use of available energy and to improve overall occupant comfort and efficiency. Experimental results demonstrated the effectiveness of the proposed distributed network control.