Adaptive neuro-fuzzy modeling of battery residual capacity forelectric vehicles

This paper proposes and implements a new method for the estimation of the battery residual capacity (BRC) for electric vehicles (EVs). The key of the proposed method is to model the EV battery by using the adaptive neuro-fuzzy inference system. Different operating profiles of the EV battery are investigated including the constant current discharge and the random current discharge as well as the standard EV driving cycles in Europe, the US, and Japan. The estimated BRCs are directly compared with the actual BRCs, verifying the accuracy and effectiveness of the proposed modeling method. Moreover, this method can be easily implemented by a low-cost microcontroller and can readily be extended to the estimation of the BRC for other types of EV batteries     

Neural network-based receiver for wireless communications

A novel approach, based on neural networks, for the design of a receiver for TDMA wireless communications is described. The receiver utilises three functional blocks: time-frequency analysis, feature extraction, and pattern classification. Computer simulation results are presented comparing the performance of the new receiver against a conventional MSK receiver for a Rayleigh fading multipath channel. The results show that the new receiver is capable of achieving a performance comparable to that of the MSK receiver without the regular transmission of a training sequence     

Neural network-based systems for handprint OCR applications

Over the last five years or so, neural network (NN)-based approaches have been steadily gaining performance and popularity for a wide range of optical character recognition (OCR) problems, from isolated digit recognition to handprint recognition. We present an NN classification scheme based on an enhanced multilayer perceptron (MLP) and describe an end-to-end system for form-based handprint OCR applications designed by the National Institute of Standards and Technology (NIST) Visual Image Processing Group. The enhancements to the MLP are based on (i) neuron activations functions that reduce the occurrences of singular Jacobians; (ii) successive regularization to constrain the volume of the weight space; and (iii) Boltzmann pruning to constrain the dimension of the weight space. Performance characterization studies of NN systems evaluated at the first OCR systems conference and the NIST form-based handprint recognition system are also summarized.     

Neural network-based F0 text-to-speech synthesiser for Mandarin

A neural-network-based approach to synthesising F0 information for Mandarin text-to-speech is discussed. The basic idea is to use neural networks to model the relationship between linguistic features. Extracted from input text and parameters representing the pitch contour of syllables. Two MLPs are used to separately synthesise the mean and shape of pitch contour, using different linguistic features. A large set of utterances is employed to train these MLPs using the well known back-propagation algorithm. Pronunciation rules for generating F0 information are automatically learned and implicitly memorised by the MLPs. In the synthesis, parameters representing the mean and shape of the pitch contour of each syllable are generated using linguistic features extracted from the given input text. Simulation results confirmed that this is a promising approach for F0 synthesis. The resulting synthesised pitch contours of syllables match well with their original counterparts. Average root mean square errors of 0.94 ms/frame and 1.00 ms/frame were achieved     

Advanced motor developments for electric vehicles

Advances in motors and drives can make a significant improvement in the performance and cost effectiveness of electric vehicles (EV's)--increasing the vehicle range by as much as 20 percent, while reducing costs as much as 15 percent below present levels. Preliminary studies suggest that improvements in conventional dc motors and drives can only achieve modest improvements (about 3 percent in range and 10 percent in costs). Two advanced motor concepts-- electronically-commutated motors and permanent-magnet motors-- appear more promising.     

Zinc-bromine battery design for electric vehicles

Design projections for zinc-bromine batteries are attractive for electric vehicle applications in terms of low manufacturing costs ($28/kWh) and good performance characteristics. Zinc-bromine battery projections (60-80 Wh/kg, 130-200 W/kg) compare favorably to both current lead acid batteries and proposed advanced battery candidates. The performance of recently developed battery components with 1200 cm²electrodes in a 120V, 10 kWh module is described. Similarly constructed smaller scale (600 cm²) components have shown lifetimes exceeding 400 cycles and the ability to follow both regenerative braking (J227aD) and random cycling regimes. Initial dynamometer evaluations of full scale 20 kWh batteries is expected in early 1984.     

Neural network-based adaptive beamforming for one- andtwo-dimensional antenna arrays

We present a neural network approach to the problem of finding the weights of one- (1-D) and two-dimensional (2-D) adaptive arrays. In modern cellular satellite mobile communications systems and in global positioning systems (GPSs), both desired and interfering signals change their directions continuously. Therefore, a fast tracking system is needed to constantly track the users and then adapt the radiation pattern of the antenna to direct multiple narrow beams to desired users and nulls interfering sources. In the approach suggested in this paper, the computation of the optimum weights is accomplished using three-layer radial basis function neural networks (RBFNN). The results obtained from this network are in excellent agreement with the Wiener solution     

Neural network-based CAD model for the design of square-patchantennas

An artificial neural network (ANN) has been developed and tested for square-patch antenna design. It transforms the data containing the dielectric constant (ϵ_r), thickness of the substrate (h), and antenna's dominant-mode resonant frequency (f_r) to the patch length (l)     

Neural network-based dynamic channel assignment for cellular mobilecommunication systems

Conventional dynamic channel assignment schemes are both time-consuming and algorithmically complex. An alternative approach, based on cascaded multilayered feedforward neural networks, is proposed and examined on two cellular systems with different configurations. Simulation results showed that the blocking performance of our multistage neural network approach can match that of an example conventional scheme with less complexity and higher computational efficiency. The example scheme considered here is the ordered channel search, which can achieve a reasonably high spectral efficiency as compared to that of an ideal dynamic channel allocation algorithm. We conclude that our neural network approach is well-suited to the dynamic channel allocation problem of future cellular or microcellular systems with decentralized control     

A torque controller suitable for electric vehicles

A torque controller suitable for electric vehicles is studied. The controller ensures that an induction motor generates motor torque efficiently, stably and accurately. The torque control system feeds back an assumed motor torque calculated using the secondary magnetic flux and the torque current detected from current sensors of the primary currents. The motor torque is controlled by using the torque current reference determined from the generated secondary magnetic flux and the magnetizing current reference. The magnetizing current reference is determined on the basis of the torque current reference so that motor torque generation efficiency is always optimal. The magnetizing current regulator is operated according to the magnetizing current reference. This ensures the motor generates the motor torque stably even in transient states. Fundamental performance characteristics, such as response, accuracy and efficiency of the motor torque are verified by simulation and experiments. The torque controller is judged suitable for the drive system of electric vehicles     

Advanced electronic control systems for electric vehicles

The characteristics and relative advantages of several types of electric vehicle control systems are discussed. An advanced electronic control system for a separately excited dc motor is described. Performance characteristics of an electric vehicle using this control are discussed and recommendations for future systems are given.     

A battery state-of-charge indicator for electric wheelchairs

Deep-discharge-type lead-acid batteries used in most electric wheelchairs require acurrate battery state-of-charge indication to prevent stranding and to provide economical operation of the wheelchair. A monitoring technique combining the open-circuit voltage and the coulometric measurements had been previously implemented on a microcomputer-based circuit. This adaptive monitoring technique enables the monitor to adjust to different battery sizes as well as the aging process. Several improvements are reported. A technique has been developed to enhance the acurracy and reduce the required rest period of the open-circuit voltage measurement. The open-circuit voltage recovery curve is approximated by two asymptotes on a semilog scale. The open-circuit voltage is then extrapolated from the slope of the first asymptote before it fully stabilizes. The accuracy of the monitor has been verified in field tests, and comparison with a commercial battery monitor shows it to be superior in several respects     

Analysis of anti-directional-twin-rotary motor drivecharacteristics for electric vehicles

A new power train for electric vehicles is proposed using an anti-directional twin rotary (ADTR) motor. A stator in a conventional motor was reformed to be movable, and the stator (outer rotor) rotates in the opposite direction to the inner rotor. In this paper, several characteristics of an induction motor-type ADTR motor are reported. When an ADTR motor is used in electric vehicles, the direction of one of the rotors should be reversed and both rotors rotate in the same direction, propelling the two wheels of the electric vehicle. The torque of the wheels can be balanced without a differential gear. The fundamental torque-balancing characteristics of an ADTR motor are clarified, namely, the torque balance theory, the torque-speed characteristics, the rotor-speed transient characteristics and the transient torque response under speed sensorless torque control     

电动汽车交流充电桩系统设计

简要介绍了电动汽车交流充电桩的内容和研究现状.针对需要设计了基于STM32F107VCT6微控制器的控制系统解决方案,并详细介绍了控制系统的软硬件设计和桩体的电气部分设计.该系统提供包括人机交互、充电控制、电能计量、IC卡付费、票据打印、运行状态监测、充电保护和充电信息存储和上传等多种完善的功能,满足充电过程的要求.     

Novel permanent magnet motor drives for electric vehicles

Novel permanent magnet (PM) motor drives have been successfully developed to fulfil the special requirements for electric vehicles such as high power density, high efficiency, high starting torque, and high cruising speed. These PM motors are all brushless and consist of various types, namely rectangular-fed, sinusoidal-fed, surface-magnet, buried-magnet, and hybrid. The advent of novel motor configurations lies on the unique electromagnetic topology, including the concept of multipole magnetic circuit and full slot-pitch coil span arrangements, leading to a reduction in both magnetic yoke and copper, decoupling of each phase flux path, and hence an increase in both power density and efficiency. Moreover, with the use of fractional number of slots per pole per phase, the cogging torque can be eliminated. On the other hand, by employing the claw-type rotor structure and fixing an additional field winding as the inner stator, these PM hybrid motors can further provide excellent controllability and improve efficiency map. In the PM motors, by purposely making use of the transformer EMF to prevent the current regulator from saturation, a novel control approach is developed to allow for attaining high-speed constant-power operation which is particularly essential for electric vehicles during cruising. Their design philosophy, control strategy, theoretical analysis, computer simulation, experimental tests and application to electric vehicles are described     

电动汽车非车载式充电机系统研制

针对我国电动汽车成为国家十二五重点发展战略性新兴产业的实际情况,为了实现基于IGBT器件的非车载充电机,采用了ADUC813为控制板核心、模块化设计的方案,主回路电路采用国产IGBT器件作为逆变控制器件,控制板电路包括信号采集、反馈及UCC3895移相控制电路,采用ZVSZCS软开关技术降低开通损耗,解决了IGBT器件的电流拖尾,提高电路效率,通过试验采集ZVZCS波形结果,验证了该系统方案满足研制要求。