نشریه مهندسی برق و مهندسی کامپیوتر ایران
سال نهم شماره 2 (پیاپی 26، تابستان 1390)

In this paper a new method is proposed for separation and geolocation of moving emitters using their radiated signal which has no assumption about their numbers, positions and signal types. In the first step of the proposed method, all available signals in a scene are received using several sensors. In the second step a vector of time differences of arrivals between the signal received by each sensor and signal received by the reference sensor is extracted and two spaces of TDOA vectors are constructed for successive time slots. Finally a combination of multiple hypothesis tracking and data association algorithms are applied to extract and confirm meaningful strings of vectors from successive TDOA vector spaces that each string indicates an emitter. Obtained results from evaluation of the proposed method and comparing them with results obtained from existing methods, show that it can separate and track several emitters with linear, nonlinear, constant velocity and variable velocity motions. Also the proposed method shows an acceptable ability to separate and track emitters with parallel and intersecting trajectories and maneuvering emitters with greater performance than existing methods and without losing processing speed.

In this work, in order to increase the capacity of a recurrent neural network, we present a model for extracting common features and sharing them across data. As a result of using this model, extracted principle components of data will be invariant to unwanted variations. The recurrent connection of the network removes the noise using a continuous attractor formed during the training phase. The defined speaker codes will be transformed to the information need for switching the continuous attractor in the input space. As a result, speaker variations can be compensated and the recognition will performed when a clean signal is available. We compared the performance of this method with a reference network described in the paper. The results show that the proposed model is more useful in removing noise and unwanted variations.We compared the performance of this method with the reference network. The results show that the proposed model performs better in removing noise and unwanted variations, it increased the phoneme recognition accuracy about 5% when the signal to noise ratio is 0 dB.

In this paper an approach to improve the quality of synthetic speech in formant and concatenative synthesis techniques is described. To deal with this problem we focused on using vocoders. In concatenative speech synthesis the idea is based on post processing the generated speech to reduce discontinuities. The post processing is consists of integrating Straight method to synthesis system in order to smooth the boundary between units. On the other hand, in formant synthesis we used multi excitation linear predictive method to replace simple excitation signal in Klatt method with multiband excitation. Our synthesis techniques were evaluated with respect to naturalness, fluidity and intelligibility based on subjective methods. These experiments clarified that the naturalness of synthetic speech can be improved by using our smoothing methods and multiband excitation signal.

Printing technology is known as one of the most suitable methods for adding electrical functionalities to textiles and inkjet method because of advantages such as low cost, availability, flexibility, … is a special method amongst all available printing techniques. This is the objective of this research to employ the novel method of using reactive inks in order to react with each other after being jetted onto the substrate for fabrication of simple electric circuit components. In this method, dilute solution of silver salt and a reducing solution are subsequently printed on each substrate. Oxidation-Reduction reaction between two inks deposits metallic silver nanoparticles by in situ reduction of silver salt forming an electrically conductive surface. The best reducing agent for inkjet deposition of silver was found to be ascorbic acid at normal pH. Conductive lines and patterns were fabricated on paper, plastic films and textile fabrics using the above technique and the effect of different parameters on their final conductivity were investigated and tried to gain the highest possible conductivities on each substrate. Based on our observations and results; inkjet technology posses very high potential for fabrication of silver nanoparticles containing patterns with conductivities up to 5x105 S/m for use as circuitry components.

In order to participate steam power plant in power system frequency regulating, in addition to producing the base load, the boiler and turbine should be controlled coordinately. Lack of coordinated control may lead to instability, cause oscillation in producing power and boiler parameters, and reducing the reliability and creating thermodynamic tension to devices. This paper proposes a sliding mode based controller to control two main boiler-turbine parameters; i.e., the turbine revolution and superheated steam pressure of the boiler output. For this purpose, complete and exact model of the subsystems including turbo-generator, turbine and related control systems are derived and the ability of the method is shown using this comprehensive model. The proposed method is simulated on the 320 MW unit of Islam-Abad power plant in Isfahan/Iran and its performance is compared with the related real PI controllers which have been used in this unit. The simulation results show the capability of the proposed controller system in controlling local network frequency and superheated steam pressure in the presence of load variations and disturbances.

As intermittent wind power generation becomes more significant in power generation, it becomes increasingly important to assess its impact on the generation reliability of power systems. Therefore, it is the objective of this paper to evaluate the impact of wind power on the power system reliability. In this paper, different approaches of wind power modeling are explained. Markov chain Monte Carlo (MCMC) and ARMA method are used to model of wind power output. Then Fuzzy-Markov method for wind power modeling is proposed. The proposed method is capable of modeling wind farms that have insufficient wind speed data. Finally, capacity credit of wind power is calculated.

In this paper, induction motor speed control drive is designed with application two multilayer feed-forward neural networks. That those are used one for generate PWM pulse and other for estimation of required torque and flux information. For trained of the PWM wave generate neural network is used from compound information two voltage and current classic model. Also, against general classic models for generate of the switching pulses is used as compound from reference voltage and current two motor phases. With these ideas are eliminated problems of the voltage and current classic models (flux saturation in current model for high speeds and voltage drop in voltage model for low speeds). As voltage profile is improved in this paper. The required feedback signals estimation (including: rotor flux, torque, etc.) is estimated by multilayer feed-forward neural network. That for robustness of the above estimator against rotor and stator resistances variations in time work of motor is used from compound trained data of the voltage and current classic models, because the voltage and current of the general classic models to sequence are independent of rotor and stator resistances. The simulation results by MATLAB-Simulink verify the proposed drive in improvement of the speed profile in transient and steady-state operating modes. Also, it verify clearly robust of the proposed drive against rotor and stator resistances variations in time work.