Search

Safety is an eternal issue in the civil aviation transportation. Once a civil aviation accident occurs, it will cause great casualties and economic losses. In order to ensure the civil aviation safety, the hazard identification and prediction of civil aircraft should be effectively and accurately realized. The civil aircraft uses Aircraft Communications Addressing and Reporting System (ACARS) to interact with the ground during flight. The data generated by ACARS has a simple structure and strong timeliness. In view of the advantages of ACARS data, a hazard identification and prediction method based on support vector machine optimized by particle swarm optimization (PSO-SVM) and long short term memory (LSTM) neural networks which uses ACARS report as analysis data is proposed. First, in order to reduce the identification and prediction time cost, the SVM-based recursive feature elimination method with cross-validation algorithm (SVM-RFECV) is used to select the characteristic parameters. Then, the SVM optimized by PSO is used to identify hazard based on the selected parameters. According to the identification results, the LSTM is used to predict the trend of the selected parameters to realize hazard prediction. An A13 report of APU generated by ACARS is selected as analysis data for hazard identification and predication in this paper. The analysis results show that the proposed identification method based on PSO-SVM and SVM-RFECV has high identification speed and accuracy. The proposed prediction method based on LSTM has the best prediction performance. The proposed method can effectively identify hazards and accurately predict the trend of parameters to improve the safety of aircraft.

Sensitivity is an important indicator for evaluating the performance of RFID tag. Firstly, this paper gives the definition of group tags sensitivity. Furthermore, the formula of the group tags sensitivity is deduced. Secondly, Semi-physical verification platform is used to measure the reading distance of the tags. Thirdly, the sensitivity for each tag in the tags group is tested. Finally, the group tags sensitivity is obtained and calibrated. The experimental results show that the formula accords with the actual result. Therefore, this work is very useful for dynamical measurement and calibration for RFID tags group sensitivity.

In RFID multi-tag network, the image of the tag needs to be collected. Therefore, the research on the effect of LED array light source on the image capture of RFID tag is proposed. According to the optical properties of LED, the radiance illumination expression of LED array light source is deduced, and the illumination distribution of LED array is simulated. We choose the appropriate arrangement of array light sources and analyze the effect of the array on the collection of the tags. The research results show that the LED array design can effectively provide a uniform illumination of the light source. The design of LED array can obviously distinguish the tag from the background and improve the quality of the acquired RFID tag image.

The RLC circuit model, named the Lorentz model, is commonly used for resonator. But when the resonators are strongly coupled with each other, the principle of superposition does not work. Traditional Lorentz model can not describe the interaction between wave modes, and Fano's resonance model is used. The paper introduces the background for the two resonant models. And then the behaviors of a two-arm resonant model in a waveguide are simulated and measured, the results of them are basically accordance.

One of the important advantages of RFID technology is to identify multiple targets at the same time. However, in order to identify multi-object at the same time, it is necessary to solve the problem of improving the performance of tag reading. Among the factors affecting the performance of tag identification, the geometric distribution of multi-tag is the key one. With the advantage of GA-BP neural network in optimization analysis, we do some researches about the impacts of the multi-tag's geometric distribution to the performance of reader. By training a large number of dynamic test data under the gate entrance environment, optimal RFID tag geometric distribution can be predicted by GA-BP neural network under the maximum or minimum reading distance. Furthermore, the dynamic reading performance of multi-tag system could be effectively improved.

Three-phase rectifier voltage often use software phase-locked loop, input voltage is appropriate. However, the software phase-locked loop had previously most locks can only perform accurate voltage in order that agreement cannot be done accurately lock in reverse order (lost key). To overcome this drawback in this article combines the principle of phase locked loop voltage argued upon, examined the nature of change of voltage vector angle reverse order, analyze the causes loss of lock reverse order part of a program to design self-adaptive phase locked loop can be done in the correct key sequence voltage and voltage upon reverse order. Also in this article also offer solutions using fractional integrated controller PIX to improve software self-adaptive phase locked loop instead of using the traditional PI controller as the software was designed before. Experimental results have demonstrated the method using fractional controllers to improve software PI phase locked loop three-phase self-adaptive software makes the phase locked loop to work more flexible, done properly locked voltage law in order upon agreement or otherwise. Carry extra one further step, analysis and experiments demonstrated using software adapts phase locked loop can make PWM (pulses width modulation) rectifiers work without influenced by the phase sequence of input voltage.

With the development of large scale multi-layer multi-domain optical network, it is becoming extremely difficult to calculate a strategy and constraint condition based path when multiple network failures present themselves all at once. Therefore, routing technology has become one of the core issues concerning the survivability of the optical network of the next generation. A genetic algorithm based on load-balancing was proposed in this paper and the optimal solution could therefore be derived with a higher probability. The simulation and related analysis verified the effectiveness of the algorithm.

Along with the development of science and technology are the strict requirements for control quality and the capacity of the system. In recent years the use of the pulses width modulation (PWM) rectifier in control systems becomes more and more popular. In conventional control system, a small inverter is generally used to control the rectifier or power of the system, it therefore cannot satisfy the requirement of high performance and good quality control. A solution to this issue is the PWM rectifier control system that couples two or more inverters in parallel connection. The significant advantages of parallel inverter systems used in automatic control system are the capacity, reliability, flexibility and quality improvement and the ability of backing up the system. However, conducting parallel pair of PWM rectifier also exists many problems, such as the emergence of the equilibrium current circulation among the rectifiers, the existence of high harmonic components leads to unbalanced grid. Hence, a new suitable approach that could minimize the risks by optimizing the system in the course of work must be investigated. The proposed control system coupling two parallel three-phase PWM rectifiers and the application of direct current controller (DCC) to control systems in this paper will be analyzed with simulation and experimental results to produce a reasonable solution of this issue.

Modern electrical devices prepare less space for antennas. Both the miniaturization of antenna elements and decoupling of antenna arrays have become significant problems to solve. This paper firstly discusses the usage of lossless passive network to decouple strongly coupled antennas, such as some previous works done by other authors. Then this paper mainly concentrates on the Eigen mode decoupling theory and finds that the radiation efficiency is affected by the decoupled network because it changes the Rayleigh quotient of the whole network. Furthermore, this paper uses microstrip branch-line directional coupler to decouple two strongly couples dipoles. It comes out an isolation below −25dB at a center frequency of 3.25GHz. The passive decouple network's effect to the two-element array's radiation efficiency is analyzed in detail. And it turns out to a conclusion that the function of the decoupled network can increase the radiation ability of the whole structure.

A novel focusing lens conical horn antenna loaded with dielectric is designed in this paper. The design method of the dielectric surface equation is given. The comparison between the proposed antenna and traditional spot focusing lens antenna is well discussed. Simulations show that the proposed antenna has advantages of small reflection, low cross-polarization, symmetrical beam, and small focal spot. The antenna can be obtained very symmetrical focal spot. What's more, the focal spot is approximately circular. The average electric field strength at the focal point of the antenna can be improved by about 1/dBVm than traditional spot focusing lens antenna. This article provides a new focusing lens antenna design ideas, with good reference value.

Yeast cells were utilized as templates to synthesize ZnWO 4 hollow microspheres photo-catalyst in this paper. Moreover, the main influencing factors in the photocatalytic degradation of low concentration RhB dye wastewater had been studied by using as-obtained hollow ZnWO 4 photo-catalyst. The results of experiments indicate that pH, concentration of RhB dye and dosage of catalyst had significant influence on the degradation efficiency of RhB dye wastewater. The degradation rate reached 87.5% under the condition when amount of ZnWO 4 is 2g/L, pH 6, and the concentration of RhB at 0.5×10−5 mol/L.

Affinity propagation (AP) is a new powerful clustering algorithm based message passing between data points. One of the major problems in clustering is the determination of the optimal number of clusters. In this paper, we propose a new approach called Affinity Propagation Clustering based Particle Swarm Optimization (PSO-AP), which using Particle Swarm Optimization (PSO) algorithm to determination of the optimal clustering number of AP. Our PSO-AP method is absolutely “automatic”. PSO-AP represents the issue of finding the optimal clustering number of AP as an optimization problem of searching optimal solution of the input “preferences” space. It evaluates the particles' fitness using clustering validation indexes. Bounded constraint strategy of PSO used supervisor-student model. Several experiments data sets are presented to illustrate the simplicity and effectiveness of PSO-AP.

The real estate industry is one of important pillar industries of our national economy. The real estate industry is also a fund intensity industry, and it is closely related to the financial system.. Fluctuations, as a property of the real estate economy, are bound to make an impact on the financial system, leading to great financial risks, and even to a financial crisis, if not under proper control or management. This paper probes into the cause and reasons for financial risks based on the analysis on loan risks involved in real estate development and mortgage risks related to housing purchase; this thesis also makes suggestions regarding policies whose purpose is to prevent financial risks.