Search

Aiming at the problem that traditional machine learning methods rely on manual feature extraction in DC-DC circuit soft faults, it is important to effectively obtain the fault characteristics of DC-DC circuit soft faults. In this study, combining the advantages of one-dimensional convolutional neural network (1DCNN) and gated logic unit (GRU), a deep learning model for fault identification of time series signals is proposed to realize soft fault diagnosis of DC-DC circuits. In this study, a 1DCNN-GRU network model is constructed, and 1DCNN can directly perform automatic feature extraction on the data, while the GRU makes up for the shortcomings of CNN in processing time series data, thereby ensuring the comprehensiveness of the extracted features. For the hyperparameter problem in the network model, the powerful optimization ability of the gray wolf optimization algorithm is used to automatically search for the best hyperparameters in the 1DCNN-GRU network, and then the optimized 1DCNN-GRU network model is used for comprehensive feature learning. In order to meet the needs of deep learning for large data samples, the overlapping sampling method is used to enrich the data sample set. Experimental results show that the proposed method achieves 99.62% accuracy in DC-DC circuit fault diagnosis, and still maintains good robustness in noisy environment.

Three degrees of freedom servo system (TDFSS) is one of the key equipments of inertial testing, such as evaluation of inertial navigation system and test of inertial components. It is a kind of servo system with some non-linearity and uncertainty. This thesis takes advantage of the characteristic of Neural network in approaching non-linear function, applies the Neural network on the three-axis simulator, provides a method for the TDFSS. Simulating experiment has been used to verify the advantage of the scheme and achieved completely decoupling control. The scheme gives good control precision, and it is simply structured and easily implemented.Introduction

Gas monitoring is considered as the most important mean in the safe production of coalmine. Time delay inevitably exists in the current gas monitoring technology widely used today. This research uses Support vector machine to forecast gas concentration. ZigBee wireless network protocol is adopted so that a wireless gas monitoring network with concentration predicting function is obtained. In this way the present time delay in Gas concentration monitoring will be improved and the scope of monitoring will be expanded at the same time

A supervisory control and data acquisition system is designed for wind farms, including the local controller and remote monitoring terminal. The data of the sensors collected by the lower-computer is transmitted to the host computer monitor terminal through the GPRS wireless network after preliminary treatment. In the host computer, the data is processed and displayed by the virtual instrument. The remote monitoring terminal can also send commands to control the actuators in the wind farm. Wavelet de-noising principle is used to process the data collected, improving the reliability and stability of the system.

Along with the tension of global energy and the improvement of environmental awareness, more and more Research institute begin examining how to change low-concentration gas into treasure. So we design the intelligent controller with ARM920T as its main control chip, set up an intelligent controller with two inputs, which will dilute the concentration from 4%-8% to 3.8% and the transmission can be carried out under constant pressure. We have made simulation toward the system by using Simulink in MATLAB and Fuzzy toolbox. The results show that the control effect turns out to be fine and has reached the performance indicators of power supply control system of low-concentration gas. This program will be proved to be an ideal intelligent control plan.

To reduce the frequent occurrence accidents of mining areas, a set of monitoring system of monitoring the operational status of mining equipment is designed. The information of electromechanical equipment are collected by the on-off sensors with anti-interference, which are converted to the standard RS485 interface signal though the MCU（Microprogrammed Control Unit）, and transmission to the monitoring platform by the bus transfer. The monitoring system is convenient and saves the cable, which can monitor the operational status of the field devices with accurate and real-time.

Due to a variety of electrical equipment using, the grid power quality severely polluted. For current harmonic signal is generated in hoist converter, the methods is proposed based on Hilbert-Huang Transform (HHT) to detect current harmonic signal and improved Active Power Filter (APF). The simulation mode is established by MATLAB/Simulink software about shunt APF+SPWM inverter hybrid active filter device for harmonic motion detection and suppression simulation. The results show that the harmonic of hoist converter can be detected by this design method real-time and the harmonic can also be suppressed validly.

Within the framework of ellipsoidal bag model, the state at finite temperature is investigated, and the dependence of the energy of the system on the interquark separation is obtained. We calculate the spectrums of and statesat different temperature for TM and TE mode respectively.

This paper reviews twenty-eight Chinese medicine books with complete prescriptions prior to the Qing Dynasty, and analyzes the characteristics of acupoint selection and needling manipulations from the perspective of apoplectic symptoms. It is concluded that, in ancient times, apoplexy is often treated on the basis of its symptoms and a great number of acupoints are employed; hemiplegia is mainly treated by the acupoints of the Large Intestine Meridian and Gallbladder Meridian, with two key acupoints; coma is mainly treated by first-aid acupoints and qi-supplementing acupoints, with seven key acupoints; wry mouth and convulsion are mainly treated by the local acupoints; as for needling manipulations, moxibustion with moxa cones is principally used, while needling is less used.

Abstract: The quark condensate in nuclear matter is studied. We relate it to the nuclear sigma commutator, and then treat it as sigma commutator for quasi- particles in a self-consistent way. We find that the deviation from the linear expression is large at high density.