Search

Saliency-based position sensorless drive with the pulse width modulation (PWM) switching frequency injection can achieve the high position estimation bandwidth. However, in order to obtain the rotor position, multiple currents should be sampled every PWM cycle to calculate the saliency reflected current difference. Considering the low-cost inverter using shunt resistors for phase current sensing, it is not possible to obtain this saliency current difference because machine current can only be measured once the low-side switch is ON. To implement the PWM injection sensorless drive on shunt-based sensing inverters, this paper proposes two modified PWM voltage controls for the measurement of saliency reflected current difference. The first modified PWM is based on the asymmetric PWM for both the field-oriented control (FOC) and position estimation. This proposed PWM requires only a single interrupt service routine (ISR); however, the current total harmonic distortion (THD) is higher. By contrast, the second modified PWM consists of both discontinuous PWM (DPWM) for FOC and asymmetric PWM for the position estimation. Although two ISR’s are required for this proposed PWM pattern, the current THD is reduced through the separated PWM control. All the proposed PWM patterns are experimentally verified on an interior PM machine where the inverter with shunt resistors is used for current sensing.

Conventionally, dc-dc converters need a large input/output side electrolytic capacitor to suppress the voltage ripple which is caused by the PWM switching and achieve the desired voltage stable performance for the dc microgrid application. However, the large capacitor will increase the dc-bus short-circuit current dramatically when the short-circuit occurs. In this paper, a coupled-inductor-based filter, which ensures the lower voltage ripple and lower short-circuit current of the dc microgrid, is proposed to realize the electrolytic capacitor. Moreover, the long lifetime film capacitor can be used to replace the electrolytic capacitor by reducing the required capacitance. The computer simulations and hardware experimental results are shown to verify the performance of the proposed coupled-inductor filter and short-circuit current reduction technique.

This paper analyzes the saliency-based position estimation for the permanent magnet (PM) machine with non-sinusoidal inductance. Considering the saliency-based sensorless drive relying on the spatial information of machine inductance, an ideally sinusoidal inductance is assumed to estimate the rotor position. However in actual PM machines, non-sinusoidal inductances are typically resultant due to the stator slotting effect, rotor topology, and most importantly, the saturation caused by magnet flux and coil flux. In this paper, a systematic analysis of saliency-based position estimation errors caused by the non-sinusoidal inductance is proposed. It is shown that the saturation decreases d-q self-inductances while increases d-q mutual-inductances. In addition, secondary inductance harmonics increase as saturation increases. The saliency-based drive ultimately fails because of considerable position errors. This paper clearly defines the feasible drive operating area for highly saturated PM machines. An optimal current control strategy is developed to maintain the drive stability for full load operation. Finite element analysis (FEA) is used to simulate the inductance saturation at various load conditions. A 5kW interior PM (IPM) machine with highly non-sinusoidal inductance is analyzed.

In this paper, a step-up battery charger for the stand-alone wind power system is proposed. In general, maximum power point tracking (MPPT) is investigated to harvest the maximum wind energy. However, most of standalone type wind turbine systems adopt batteries and it is hard to apply a MPPT strategy on a battery-charging type wind turbine system. The proposed power converter can harvest the maximum power from the wind turbine while generating pulsating current for the battery bank. Based on the circuit analysis, the maximum power point tracking function is realized by the circuit parameter design of the power converter and the characteristics of wind turbine. Also, with the interleaved operation between the step-up power converters, the current ripple of the input can be reduced. Advantages of the proposed converter include: maximum power tracking, control circuit simplicity, low voltage stress, no current sensor, and high reliability. Hardware experimental results from a 100 W prototype circuit are presented to verify the performance of the proposed power converter.

This paper proposes a modified sinusoidal pulsewidth modulation (SPWM)control single-phase inverters for the tidal current generation system. In general, a two-stage configuration is used to convert the generating AC power from the tidal current generator to DC power by a rectifier and transfer the DC power into AC power by an inverter to provide a stable voltage source. It is noticed that a small size capacitor of the rectifier is possible and the discontinuous current can be improved. However, a small size capacitor implies the higher voltage ripple and results in a lots of low-order voltage harmonics of the inverter with traditional SPWM control. Therefore, a modified modulation control strategy is developed in this paper. The proposed inverter has lower total harmonic distortion (THD)of the output voltage with high input voltage ripple. The low-order harmonics of output voltage can be reduced effectively and the THD is less than 2.5% in the experiments.

The objective of this paper is to propose an inductorless non-isolated step-down converter. The proposed converter is composed of multiple charging pump DC-DC converters with interleaving operation. Based on the characteristic of charging pump capacitor, the presented converter does not require an inductor and can obtain the step-down output voltage. Advantages of the proposed converter include: inductorless power conversion, control circuit simplicity, low voltage stress, low input current ripple, low output voltage ripple, and fault-tolerant operation for each DC-DC converter module. Based on the circuit analysis, the power loss can be evaluated with the capacitance of bridge capacitor and switching frequency. Also, with the interleaved operation between the charging pump converters, the current ripple of the input can be reduced without the need of input inductor. The computer simulations and hardware experimental results are shown to verify the performance of the proposed inductorless step-down converter. The peak efficiency from an input voltage 100 V, output voltage 50 V, and output power 75W prototype is around 97.2%.

The objective of this paper is to propose a reactive power control strategy of the grid-connected inverter for microgrid application. The proposed strategy can actively participate to regulate the voltage, realize the islanding detection, reduce the power quality impact of islanding detection and simplify the controller design. When the distributed energy resource (DR) of the microgrid system generates power to the grid, it may cause the voltage of the point of common couple (PCC) rising. The proposed strategy is to use the grid-connected inverter with the reactive current control to regulate the PCC voltage. Meanwhile, the reactive current disturbance is used to detect the variable frequency in islanding condition. Also, the experimental results are shown to confirm the validity of the design concept of the proposed strategy.

In this report, we use series of cobalt derivatives, conducting polymer and carbon materials as the counter electrodes to replace the conventional platinum in dye sensitized solar cells. The basic cobalt derivatives, Co12 and Co15, showed the capability as the counter electrode with proper short-circuit current (J SC ) and open-circuit voltage (V OC ), but the poor conductivity and redox ability result in bad fill factor of the device. After mixing Co12 and Co15 with carbon black, the device performances were enhanced obviously. The further improve was done by combine Co derivative and carbon black to form the new type material CoCB as the counter electrode. The great conductivity and redox ability enhanced both J SC and V OC , the optimized device show the power conversion efficiency of 7.13%, similar to the control device with conventional Pt electrode. The result showed the potential of cobalt derivatives as the counter electrode of DSSCs.

The objective of this paper is to propose a cascaded multi-level inverter for battery energy storage systems. This paper proposes the implementation of the multi-level inverter system having five H-bridge structure inverter modules connected in series with distinct linked batteries. Based on the developed controller, the proposed inverter system can uninterruptedly operate to overcome the disadvantage of power downtime while one of the battery modules is under fault condition or maintenance. By taking advantage of the hot swappable capability, not only the flexibility and fault tolerance of the proposed system can be improved, but the capacity expansion is possible for the battery energy storage systems in microgrids. The control strategy is implemented on a 32-bit floating point TMS320F28335 DSP and tested in standalone mode and grid-connected mode. Finally, the simulations and hardware experimental results from the prototype system are presented to verify the validity of the proposed inverter.

The main purpose of our research is to investigate the outcome and impact of the new e-platform program, the Taiwan Indigenous Culture Inclusive Health Science Education Program for the 5th grade elementary students. The materials of course were developed based on two core knowledge systems: the first one was from the "empirical science" which examines indigenous traditional knowledge and combines it with modern evidence-based scientific knowledge. We have analyzed the effectiveness of the new course with three tribal indigenous groups of Taiwanese, Tayal, Paiwan, Bunun. To probe into the indigenous students' learning attitude, learning knowledges and learning behavior. In our research, the quasi-experimental survey method, included a pre-, post- tests were used with one experimental group and a control one. The data from the survey was analyzed by SPSS for Windows 10.0 version and t-test analyses were performed. The study results are as follows: 1) There are significant differences between the two groups in knowledge learning and enhancing outcome in acquiring health science knowledge among three-tribe experimental groups 2) The results in acquiring healthy attitude and behavior were vary difference since the program only last for 8 weeks. A longer program might be more effective.

Differences in cultural and the convenience of obtaining information expose a significant difference in the effectiveness of learning between Indigenous and non-Indigenous students. Indigenous students frequently misunderstand the content and examples provided by the teachers, and cultural constraint is an important factor in learning. This study created a new learning system with information technology for Indigenous students. We established and integrated indigenous culture into a cloud-based Indigenous elementary school e-learning system. Students can read e- books online and study new vocabularies with teachers' explanations. Moreover, students can engage in exercises after completing very segment of a whole lesson. After finishing the course, students take an examination for evaluating the learning effects. All examples used in the e-books match the students' culture and lifestyle and can facilitate their understanding of the contents. We administered the test teaching course at 4 aboriginal elementary schools and found that the average score has improved significantly(p

The objective of this paper is to propose an energy storage converter (ESC) for microgrid applications. Microgrid integrates distributed generators, load and ESC through various interface power electronic converters to form a small power system. Based on the design of the converter, microgrid can operate normally in grid-connected mode or standalone mode. In grid-connected mode, the grid current controller of ESC is used to regulate demand power for microgrid. When the utility grid fault occurs, the static switch between the utility grid and microgrid is turned-off and the controller of ESC transfers the operation mode from grid-connected mode to standalone mode to supply its power to the load as a stable voltage source. Therefore, the ESC becomes an important role for microgrid stability. In addition, a method of detecting the utility grid fault is proposed to reduce the detection time. The computer simulations and hardware experimental results from a prototype system are presented to verify the validity of the operation principle of the proposed strategy.

In this report, we composite cobalt and carbon black (CB) as the counter electrodes to replace the conventional platinum in dye sensitized solar cells (DSSCs). The basic cobalt derivatives, Co12 and Co15, showed the capability with proper J SC and V OC , but poor FF of the device. After mixing with CB, the device performances were enhanced obviously. The further improve was done by combine Co derivative and CB to form the new type material CoCB as the counter electrode. The great conductivity and redox ability enhanced both J SC and V OC , the optimized device showed the efficiency of 7.38%, higher than the conventional Pt electrode of 7.00%. The result showed the potential of cobalt derivatives as the counter electrode of DSSCs.

State of the art algorithms for stochastic multi-armed bandit problem are based on either one of the two principles: optimism in face of uncertainty, and probability matching. In this paper, we provide a unified approach to combine these principles. The major result is a new upper confidence bound formula, UCB-max, by which the p-value of an arm's UCB value roughly matches the arm's probability of been optimal. Our numerical study comparing UCB-max with other competitors (UCB-tuned, UCB-V, MOSS, KL-UCB and Thompson sampling) shows that UCB-max is remarkably efficient and stable.

This paper proposes an efficient active noise cancelling (ANC) circuit design for the in-ear headphones. We develop a hardware oriented Least Mean Square (LMS) adaptive algorithm, and design a modified high-performance feed-forward ANC architecture which only uses one microphone to effectively cancel broadband noise for the in-ear headphones. The proposed ANC circuit design has been successfully implemented through adopting the standard cell-based design flow based on the TSMC 90nm CMOS technology. Besides, the proposed design has been verified under versatile noisy scenarios for its real-time ANC performance by using a Field Programmable Gate Array (FPGA) platform. Experimental results show that the proposed high-performance circuit design can reduce disturbing noise of various frequency bands very well, and outperforms the existing works. The proposed design can attenuate the broadband pink noise between 300 Hz to 900 Hz, with a maximum performance of 18 dB. Moreover, the proposed design can achieve 406.5 k samples/sec data throughput rate at operating frequency of 50 MHz, at the costs of 111.7 k gates and power consumption of 4.47 mW.

An effective method, adaptive fuzzy sliding-mode (SM) PI Control to design full-bridge inverter is proposed in this paper. For practical applications, sliding mode control (SMC) with chattering phenomenon is a large hindrance. The modified model of the PI controller is built by the plant extension scheme, in order to relieve the chattering. Also, the fuzzy logic is applied to find the appropriate parameter values of the PI controller, and can adapt to plant parameter variations and external load disturbances. The proposed method has been realized for the actual full-bridge inverters controlled by DSP and CPLD. Simulation and experimental results of the closed-loop system show that the proposed method yields system performance improvement, such as low total harmonic distortion (THD), fast dynamic response, reduced steady-state error, and reduced chattering.

This paper is aimed to provide the detailed design and operational investigations of synchronous reluctance motors (SynRM) that are commonly assembled by laminated electromagnetic steels. The magnetic flux paths in the SynRM rotors with various barrier and segment compositions will be analyzed, along with the generated motor torque and loss assessments. As the related mechanical cutting and/or punching processes on the rotor steels are required to achieve desired direct- and quadrature-axis reluctance ratios, their impacts to the flux paths due to saturations will also be thoroughly evaluated. Assisted by 3-D finite-element analyses, the results clearly indicated that the additional annealing process on the electromagnetic steel strips must be introduced for these SynRM designs with smaller sizes and higher efficiency requirements.

With their consistent production line load increments, the possibilities of substation congestion problems in the integrated steel companies are inevitable. To meet these power demands, well organized system infrastructure expansion plans must be designed in advance. An alternative scheme to alleviate some short-term facility overload problems is to integrate distributed generator (DG) units into the system to defer possible congestions. Based on substation operational constraints, systematic guidance for evaluating the optimal location and reliability indices of DG units into the system will be provided for related operations. The proposed scheme can be adopted by steel companies and other industries in determining the installation and integration schedule of DG units, and thus defer the required expansions of substation infrastructures.

The objective of this paper is to propose a multi-string single-stage grid-connected inverter (MSG-inverter) for PV power generation system. The proposed MSG-inverter can improve the power conversion efficiency, increase the overall PV array output power, reduce the output inductor size, eliminate the input current sensor, and simplify the control circuit. The proposed MSG-inverter is composed of distributed buck-boost type dc-ac inverters (BBIs) and an ac-ac filter. The distributed BBIs can eliminate the shading effect while fulfill the functions of dc-ac conversing and the maximum power point tracking (MPPT). The ac-ac filter is realized by 4 active switches operated at low switching frequency and an output inductor. It can convert the high frequency pulsating ac current generated by the distributed BBIs into a sinusoidal one with utility line frequency. Based on the developed equations, the MPPT feature of the proposed BBI can be achieved without the need of input current sensor. Also, with the interleaved operation between distributed BBIs, the current ripple of the output inductor can be reduced too. The computer simulations and hardware measurements are shown to confirm the validity of the operation principle of the proposed MSG-inverter.

This paper developed a speech controlled interface with cloud computing technology for vehicle on-board diagnostic (OBD) system. The proposed vehicle OBD system is constructed by two parts. They are OBD embedded global position system (GPS-OBD) module and vehicle surveillance server. The speech recognition task is performed in vehicle surveillance server, instead of GPS-OBD module. The speech signal spoken by driver will be first transferred into AMR format and then submitted to the speech recognition system built in the vehicle surveillance server via 3.5G wireless network. The proposed system can directly extract the speech features from AMR bit stream and carries out accurate speech recognition results. Meanwhile, the vehicle equipped with GPS-OBD module will consistently report its OBD-II data and GPS coordinates to the vehicle surveillance server. Hence, one can use the speech controlled interface to access vehicle operation information, e.g., vehicle location, vehicle speed, engine rpm, coolant temperature, fault code number, and others from GSP and OBD. The main advantage of the proposed system is that users do not need to install any speech recognition software in their GPS-OBD module before using the proposed cloud speech controlled interface. This will also benefit the cost and size reduction of the GPS-OBD module. The derivers could operate the proposed system without removing their eyes from the road and increase driving safety.

This paper introduces a new game named Chilled Domineering, together with its mathematical analysis. Chilled Domineering is modified from the Domineering game by allowing a player placing a 1x1 domino while unable to place a 2x1 domino at a position. After modifying the rule, each position in the game becomes an infinitesimal. Many interesting infinitesimals are found in this new game. This paper calculates the game values of all sub-graphs of 3x3 squares and presents a rule to determine the outcome of any sum of these positions.

This study presents a fast speaker clustering method based on multidimensional scaling. Speech segments are trained as initial acoustic models. MDS is utilized to transform acoustic models to a space with the coordinate best preserve the distances or dissimilarity between models. Speaker clusters are clustered using vector quantization on the MDS coordinates and the acoustic speaker models are trained on MFCCs features for each cluster. Experimental results show the proposed method outperforms the baseline speaker clustering method in lower execution time.

Triangular Nim, one variant of the game Nim, is a common two-player game in Taiwan and China. In the past, Hsu strongly solved 7 layer Triangular Nim while some of the authors recently strongly solved 8 layer Triangular Nim. The latter required 8 gigabytes in memory and 8878 seconds. Using a retrograde method, this paper strongly solves 9 layer Triangular Nim. In our first version, the program requires four terabytes in memory and takes about 129.21 days aggregately. In our second version, improved by removing some rotated and mirrored positions, the program reduces the memory by a factor of 5.86 and the computation time by a factor of 4.38. Our experiment result also shows that the loss rate is only 5.0%. This is also used to help improve the performance.