Search

In recent years, the construction and innovation of distribution network have been accelerating, and the overall investment in distribution network has been increasing year by year. Therefore, the investment efficiency of distribution network has become a key concern. At present, the evaluation system of distribution network often focuses on the technical aspects such as planning effect, safety and reliability, while lacks the evaluation of investment efficiency. In terms of investment decision, it usually makes a simple proportional allocation according to historical data. To solve the problems of the investment decision in distribution network construction, this paper puts forward an multi-factorial investment decision model, which is guided by the investment efficiency evaluation while satisfying the demand of distribution network. Firstly, the index system of input-output assessment, scoring function and investment benefit score evaluation model of regional distribution network are proposed. Secondly, the investment decision-making model based on genetic algorithm is proposed, which coordinate the balance between economy and safety of distribution network. Finally, a regional distribution network is taken as an example to test and verify the scientific usefulness of the model. The practical application results show that the multi-factorial investment decision model can promote the development and improvement of the distribution network more reasonably and economical efficiently.

With the wide spread of power quality monitors in smart grid, information-oriented use of power quality disturbances has become a concerned research area. In this paper, a two-step strategy classification system for various voltage sags is developed. A typical classification system is composed of one selector of discriminative feature and one classifier which the former is fundamental and critical. Further research on classification ability of physical features with respect to a given classification task is necessary. By studying the intrinsic relations between features and classification task, an automatic feature selection based classification system is presented. The main advantages of the proposed method are its ability to select features highly related to the given classification task, and its ability to reduce the dimensionality of the feature set simultaneously. In order to validate the proposed method, three common voltage sags causes, namely grid fault, large motor starting and transformer energizing, are considered. Simulating and practical recorded data are tested. Experimental results have shown that the proposed method is accurate, time-saving, and robust.

The electronic structure of point defects at 4H-SiC/SiO 2 interface was investigated by density functional theory. The results demonstrate characteristic features of the point defects at the SiC/SiO 2 interface, including carbon vacancy, silicon vacancy, divacancy, and antisite pair. The density of states changes after the defects are introduced, and defects energy level appear in the band gap. The defects energy level leads to Fermi energy changes, while the density of states of defect energy level is greatly related to electronic state of several atoms in the vicinity of the defect. Formation energy is particularly high for Si vacancy, and is strongly related to the different chemical potential, which differs from C-rich to Si-rich interface.

Fluorescence technique serves as a soft sensor with the ability to estimate the shape of emission and excitation spectrum and the information of concentrations of each fluorophore in multi-component fluorescent substances. Noise exist in each measurement inevitably. The impact of smoothing on parallel factor model of fluorescence emission excitation matrix is studied by compare nine methods. Smoothing can obtain more smooth resolved spectra, while the advantage of predication concentrations is not so obviously with the methods and processes used in this paper. Other more smoothing methods need to attempt to discover the obvious advantage of predication concentrations.

The ability to accurately and efficiently detect the contents and constituents of petroleum pollutants is important for environment of our ecological system. The fluorescence spectrum technology is applied to petroleum pollutants with high precision. A new type of portable petroleum pollutants fluorescence spectrum detection system based on multichannel sample cells and its optical structure is proposed. It is able to amplify the fluorescence signal without changing the volume of sample cell by increasing the optical path of the detection system using the parallel excitation of multiple microchannel sample cells. Through the experiments, it verifies the effectiveness of the microchannel fluorescence spectrum detection system and shows that the multichannel fluorescence spectroscopy has a stronger signal compared with general fluorescence spectrometer in the same experiment samples under the condition of the spectrum detection performance. It takes us one step closer to the realization of the real-time detection using fluorescence spectroscopy.

Blend vegetable oils supply our daily nutrition. A judge of quality guides us to choose appropriate blend vegetable oils. The blend oil market has the problem of mislabeling and ambiguity of ingredients proportion. A novel classification and evaluation algorithm for blend vegetable oils is proposed to standardize the blend oil markets. After double times of fluorescence intensity values of blend vegetable oils, the algorithm gets the summation of an excitation-emission-matrix along the direction of excitation wavelength (or the direction of emission wavelength) over the region of the characteristic spectra which means an area with higher fluorescence intensity values or an area different from others and then gets the envelope and calculates its statistical characteristic values. Take the statistical characteristic values to do partial least squares (PLS) regression to the values of classification and evaluation. At last symbolic results are presented. The algorithm offers a way of standard of blend vegetable oils using judge of classification and evaluation of blend vegetable oils. It is useful to build a health blend vegetable oils market.

The development of optical fiber temperature sensor in recent years, promoting the development of temperature measurement on line. For example, temperature monitoring in the high-voltage power system equipment to monitor the operational status of their security operation and it plays a decisive role in life too. The system uses semiconductor absorption optical fiber temperature sensor to measure temperature, the range of the temperature measurement at 20°C to 175°C.

Researched on a kind of photoelectric current transformer based on the Rogowski coil, it solved the electrical insulating problem between the high and low voltage sides using the optical fiber to transmit the pulse signal from the high-pressure side. The laser energy can meet the demand of electricity in high voltage side steadily. The V/F conversion of digital signal processing method makes the transmission signal has a strong anti-interference ability, and experimental results show that the maximum error is 0.2% in the range of 0~50A.

A temperature measuring driving scheme with high resolution is proposed which is based on the optical fiber coupling closed-loop oscillatory quartz crystal. Using the optical feedback stabilized 5mW LD as illuminant, the photic driving distance is 50m. The power of designed micro-consumption temperature-sensitive sensor probe is less than 0.5mW. The design idea of this new sensor and the key technology of realizing photic driving are introduced in this paper. Experimental results show that the temperature sensor can resist high-voltage and electromagnetic interference, and in the range of 0°∼+200°C, sensitivity can reach 1000Hz/°C.

A novel optical fiber gas sensor based on cascade of hollow-core photonic crystal fibers (HC-PCF) is proposed. Using the comb filter characteristic of Fabry-Perot (F-P) interferometer, Near-infrared evenly-spaced multi methane absorption lines (R3, R4 and R5 lines in 2v3 overtone band around 1.66 µm) were detected simultaneously by tandem F-P interferometer and broadband SLED light source. Difference absorption detection based on equivalent absorption coefficient was used as signal processing method. Chirped fiber grating with its central wavelength around R4 was adopted as optical band-pass filter. The light power corresponding to three absorption lines and three reference wavelength was detected simultaneously by PIN PD. The diffusion of gas inside of the hollow core is analyzed and the optimal design of sensing head is completed. The feasibility of the system for methane detection is verified by experiments.

Non-touch online thickness measurement using CCD vision system is present. Measurement principle, composition of the detection system and the operation process are expatiated. Mathematical morphology theory is successfully used in thickness image analysis and image processing. Distance of two light beams represents the measured thickness. Center orientation is a key aspect of improving precision of thickness measurement. Morphology thinning algorithm is proposed to locate the central lines of line-structured light image. The algorithm can safely thin entities to their central points. Experimental results show that the measurement error is less than 0.8%, and the morphological algorithm presented here is adaptable to real time image processing.

The singular value eigenvectors are often used to recognise the different kinds of mineral oil. The eigenvectors are obtained by the Excitation-Emission Matrix (EEM) factorization from the three-dimensional fluorescence spectroscopy. They are complicated and not easy to be recognised by the simple formula. A new type neural network-wavelet neural network (WNN) was introduced. The singular value eigenvectors were used to be the input of the WNN. The mapping relation was obtained by the WNN between the singular value eigenvector and the species of the mineral oil. The WNN realized the recognition of the different kinds of mineral oil. The experiment result indicates that the right of the distinguish rate is 95%. The WNN has much higher resolution and less training times than BP networks.