Earticle

In this brief, a novel pathological continuous speech detection method based on time domain features weighted. First, different optimal threshold for time domain features, including zero crossing ratio, short-time energy and autocorrelation, are obtained from training speech data. Second, a difference evaluation technique is proposed, and with it, the difference of the same time domain feature selected from testing speech data and training speech data were obtained. Finally, to distinguish a given speech well, a novel weighting method based on difference evaluation for each kinds of time domain is employed, respectively. Experiments were conducted on the pathological speech database to prove the power and effectiveness of the proposed method. Results obtained shown that this method outperforms other early proposed time domain feature method, creating a more reliable technique for pathological continuous speech detection.

A kind of autonomous navigation strategy for an outdoor mobile robot is proposed in the real world environment. Encoder disk, electronic compass and GPS are used in the dead reckoning system of outdoor mobile robot. To reduce dead reckoning error, the extend Kalman filter method is adopted to realize data fusion from encoder disk and electronic compass. GPS receiver is used to correct the dead reckoning data to eliminate the accumulation error of the dead reckoning. Laser range finder is used to detect the obstacle and establish the environment map. Filed D* algorithm and DWA method are used to realize the obstacle avoidance and path planning. The experimental results show that the outdoor mobile robot can realize the autonomous navigation under the complex road condition.

The speed is higher, the kinetic energy is greater. In order to ensure the safety of a new generation train running in a high speed, it is necessary to research on its braking performance. In this paper, the braking force, running resistance, braking time, braking distance and the deceleration generated by the train with two kinds of braking wings were analyzed while the high-speed train was doing deceleration movement. And the results were compared and analyzed between the train with and without braking wing, and between the two kinds of braking wings. The results showed that the high speed train with braking wings made much contribution to the acceleration in the braking process, especially the train is in high speed.

In order to find new methods for preparing and improving the performance of optical attenuation slice. Vacuum magnetron sputtering method was used to prepare Ni/SiO2 composite film optical attenuation slices, with vacuum magnetron sputtering apparatus, at 5,10,15,20,25min sputtering time, 0.2,0.4,0.6,0.8,1.0Pa sputtering pressure and sputtering power 300~1200W. The XRD, SEM, EDS, AFM and 722 spectrophotometer also were used to study the effects of different sputtering time, sputtering pressure and sputtering power on the film structure, surface morphology, composition, three-dimensional structure, surface roughness and light attenuation rate of optical attenuation slice samples. The results indicated that: Ni/SiO2 composite films were formed, instead of simple physical adsorption between Ni film and SiO2 substrates with magnetron sputtering by XRD analysis and calculation, with uniform grain size of 25.96, 32.38, 32.29,26.95, 25.92nm, respectively. The main component element was Ni, but there were few impurities deposited on the substrates. Impurities were reduced gradually with the increase of sputtering time, mainly resulted from two sputtering of Ni atoms by EDS; The film surface was smooth and dense, flatness and organizational structure of Ni film were better, surface roughness was 1.267nm at 25min sputtering time, 0.4Pa pressure and 400W power with SEM and AFM. Light attenuation rate of optical attenuation slices was different in different process parameters, and the maximum reached 0.52, film flatness, impurities and defects were the main reason by the light attenuation performance analysis.

An efficient controller based on joint between original proportional-integral-derivative (PID) controller and fast genetic algorithm (FGA) is proposed to enhance the transient response of high order synchronous machine. PID controller has several advantages in practical design of synchronous machine compared with another controller。The transient response parameters with PID controller of synchronous machine are reasonable, but they susceptible to the local minima problem and not adequate. Therefore, FGA is used to overcome this problem and to enhance the transient response. Performance of the proposed controller is tested through a high order synchronous machine model and then compared with the original PID controller. In this paper the weak coupling relationship between the automatic voltage controller (AVR) and automatic generation control (AGC) in the synchronous machine has been proved to simplify the overall model. The results demonstrate that, the proposed controller is an efficiency and better than a conventional PID controller.

Modulation type is one of the most important characteristics used in radar signal recognition. This paper proposes a method to realize modulation identification. This algorithm applies wavelet transformation to the signal, and then uses manifold learning method to reduce the high dimension and extracts the recognition feature. The proper threshold value is set as the classifier to achieve the purpose of recognizing 5 kinds of signals （2ASK, 2FSK, 2PSK, LFM, CP）in Gauss white noise environment. The algorithm requires priori signal information no other than signal-to-noise rate. Simulation result indicates the algorithm achieves good performance.

The dry cutting experiment for nickel-based high temperature alloyGH4169 was conducted with Al2O3- SiCw whisker toughening ceramic blade WG300 and Si3N4-Al2O3 ceramics (Sialon) blade SX9. The main wear mechanism of ceramic cutting tools were analyzed by means of observing the wear morphology about cutting tools with scanning electron microscope, and measuring concentration of each element in wear tiny region lying in the ceramic tools with Energy Dispersive Spectrometer(EDS). The results show that the main wear mechanism of ceramic cutting tool was presented by cohesive wear, abrasive wear and oxidation wear. Cohesive wear occurs mainly on the rake face and were influenced by the contact stress state between the contact points of cutting tool and workpiece material.

Recently, the smart grid and microgrid based on information technology (IT) and their applications have being studied worldwide because of its high-efficiency energy consumption and variety of energy. Many universities are studying a green campus to operate the energy system efficiently based on the microgrid. Especially, it is important to meet the balance between energy supply and load with minimal cost in the green campus operation. Especially, the demand response (DR) is an important function of optimal operation on the smart grid environment. In this paper, a mathematical model for optimal operation based on DR of the energy system of the green campus is proposed. Through the simulation, its feasibility is tested and analyzed.

Intelligent control that can be used learning ability and human experiences, is widely used in industrial application such as motors speed control. This paper presents Proportional-Integral-Derivative Controller (PID controller) based on Adaptive Neuro-Fuzzy Inference System (ANFIS) for Permanent Magnet Direct Current (PMDC) motor. ANFIS provides combination of artificial neural network and fuzzy inference systems therefore ANFIS uses advantages of them simultaneously. The proposed PID controller Coefficients are determined by ANFIS. The proposed controller based system is compared with Internal Model Control (IMC) PID controller based system. The Comparison shows that proposed controller improves characteristics in different conditions such as no load, increasing reference speed, applied load and noisy load. Proposed controller based system can improve system performance by smaller fuzzy rule set.