Time filter

Source Type

Liu K.,Tsinghua University | He G.,Shanghai JiaoTong University | Huang L.,Hainan Power Grid Corporation | Gu Z.,Hainan Power Grid Corporation
Dianwang Jishu/Power System Technology | Year: 2016

Quantum particle swarm optimization (QPSO) may fall into local optimal solution when solving problems with variables with limited definition domain. In order to solve this problem, quantum particle swarm optimization based on asymmetric quantum potential (AQPSO) is proposed to improve QPSO quantum model. In this method, particles are located in asymmetric potential well. Potential well parameters are determined by current optimal location and definition domain. When solving wave-function indicating particle distribution, a method of reducing parameter number is proposed. This parameter-reducing method simplifies algorithm flow by specifying only one parameter: out-of-limit probability. Simulation results prove that global search performance of the new algorithm is improved significantly with advantage in solving complex problems with high-dimensional variables and strong interference. © 2016, Power System Technology Press. All right reserved.

Yu J.,Hainan Power Grid Company | Guo Z.,Harbin Institute of Technology | Bai X.,Harbin Institute of Technology | Liu R.,Harbin Institute of Technology
Diangong Jishu Xuebao/Transactions of China Electrotechnical Society | Year: 2010

The traditional static security analysis based on the data of one time section can not adapt to the continuous state changing of power system appropriately. To improve the deficiency, a time process-oriented static security analysis method is presented in the paper. Firstly, the basic idea of time process-oriented method is explained. Secondly, with the abundant measurements and load forecasting information, the characteristic vector of each time section is extracted. Then the load curve is partitioned into several time sub-processes by the improved K-means clustering algorithm considering the successive time series of power load. Thirdly, the indices extracting the maximum and minimum characteristic sections of every time sub-process are defined. Then the representativeness of the characteristic sections is verified. While ensuring the static security, the time process-oriented method can improve the utilization rate of transmission lines to some extent. Finally, the calculation and analysis of IEEE 30-bus system shows the effectiveness of the method proposed in the paper.

Sun C.,Xi'an Jiaotong University | Bie Z.,Xi'an Jiaotong University | Xie M.,City University of Hong Kong | Ning G.,Hainan Power Grid Corporation
IET Generation, Transmission and Distribution | Year: 2015

A random fuzzy model is proposed to express the probabilistic and possibilistic uncertainties of wind speed simultaneously. In this model, wind speed is represented by a random variable following Weibull distribution, indicating the probabilistic uncertainty. The Weibull distribution parameters of wind speed are fuzzy numbers, meaning the possibilistic uncertainty of wind speed. For estimating distribution parameters, a multi-objective optimisation problem is developed based on cumulative probability and probability distributions of wind speed. The proposed model is then combined with traditional generation system adequacy (GSA) evaluation method to investigate the effect of wind speed uncertainties on GSA. To overcome the difficulty in calculating fuzzy GSA indices, sparse grid is utilised to select collocation points and single-index regression is employed to fit the relationship between adequacy indices and wind speed parameters. This study illustrates the effectiveness of the model from its application to IEEE Modified Reliability Test System. Compared with previous researches, the proposed model is suitable for the case of incomplete data or containing some outliers. It provides more helpful interval information on wind speed and adequacy indices.

Zhang B.,Xi'an Jiaotong University | Guo D.,Xi'an Jiaotong University | Wang J.,Xi'an Jiaotong University | Huang R.,Hainan Power Grid Corporation | And 2 more authors.
Dianli Zidonghua Shebei/Electric Power Automation Equipment | Year: 2013

The analysis of fault simulation with electro-magnetic transient model for three kinds of wind power generator shows that, due to the current limitation of its controller, the direct-driven permanent magnetic synchronous generator will not supply big fault current when it is connected to the system with short circuit and its impact on over-current protection can be neglected; the induction generator or doublyfed induction wind power generator, when it is connected to the system, may cause the misoperation of over-current protection downstream to the grid-connection point and the refuse-to-trip of over-current protection upstream because of the shortened II-zone. The impact of wind power in-feed current or shunt current on short circuit current or the settings of different protection zones is analyzed by changing the grid-connection point, fault location, transmission line length and wind turbine capacity. The curve of short circuit current vs. the short circuit capacity ratio of grid-connection point of wind power shows that, in order to prevent the improper operation of protection, the short circuit capacity ratio of grid-connection point should be less than 10%.

Qiu Y.,Chongqing University | Qiu Y.,Hainan Power Grid Company | Tang J.,Chongqing University | Fan M.,Chongqing University | And 3 more authors.
Gaodianya Jishu/High Voltage Engineering | Year: 2013

SOF2 is a kind of important characteristic component of SF6 decomposed gases which are engendered by partial discharge. Through detecting SOF2 gas concentration and its varying pattern, we can judge the early insulation fault of SF6 gas insulated equipment. Considering the shortages of traditional SOF2 detection methods, we introduce a photoacoustic method, which has the advantages such as high sensitivity, no gas loss, online monitoring and so on. A broadband infrared light source is utilized to form a photoacoustic detecting device for experiments of SOF2 detection. The characteristic parameters of SOF2 are analyzed on the basis of experimental results, and the relationship between the signal of photoacoustic spectroscopy and the concentration of SOF2 is obtained. The lowest detectable concentration of SOF2 is about 4.6×10-6. Compared with the method used gas chromatography, the average error of photoacoustic method is about 5.9%, which indicates that the proposed photoacoustic spectroscopy device is effective in SOF2 detection. The theoretical and experimental results provide a key technical support for the development of photoacoustic online SOF2-monitoring system.

Jiang C.,North China Electrical Power University | Liu W.,North China Electrical Power University | Zhang J.,North China Electrical Power University | Yu Y.,Hainan Power Grid Company | And 2 more authors.
Diangong Jishu Xuebao/Transactions of China Electrotechnical Society | Year: 2014

The reliability model of the wind farm is established considering the randomness of the wind speed, the power characteristics of the wind turbine, wind turbine operation conditions, derating operating status, step-up transformer and high-voltage transmission line failure rate and other factors. Because the traditional assessment method has large sample size and low efficiency for large-scale wind power penetration, the new assessment method for generation and transmission systems with large-scale wind power based on scattered sampling monte-carlo technique is proposed in this paper. This method has [0, 1] interval divided into several sub-interval in which the system index can be calculated independently which indirectly increase the sampling frequency of the fault condition. So the scattered sampling Monte-Carlo method could reduce sample times and enhance sampling efficiency. According to the random and intermittent of wind power, the risk indexes which fully reflect the impact of power generation and transmission systems are given. The calculation and analysis of improved IEEE-RTS 79 reliability test systems show that the proposed algorithm is effective.

Liu W.-X.,North China Electrical Power University | Jiang C.,North China Electrical Power University | Zhang J.-H.,North China Electrical Power University | Wang X.-W.,North China Electrical Power University | And 2 more authors.
Dianli Xitong Baohu yu Kongzhi/Power System Protection and Control | Year: 2013

A multistage reliability model of wind turbine is built utilizing a systematic method based on Markov chain approach, considering the drawback of the traditional wind turbine reliability model in sequential Monte Carlo Simulation. The probability of occurrence and duration of each state can be obtained using the state transition rate between each output power state of wind turbine calculated out with the regional wind regime of wind farm and operation historical data of wind turbine. On this basis, the double sampling method for the sequential Monte Carlo simulation is proposed. The simulation program for multistage reliability model of wind turbine is compiled. Then it is compared with the commonly used two-state model based on a single sampling method. Simulation results verify the feasibility of the proposed model based on Markov method. It can reflect accurately the output power of the wind turbine of any duration under fault conditions, improve the accuracy and expand the application range of the simulation model.

Qiu Y.,China Electric Power Research Institute | Yuan J.,Hainan Power Grid Company | Chen X.,China Electric Power Research Institute
Gaodianya Jishu/High Voltage Engineering | Year: 2013

In order to provide references for quality control of new SF6 gas in electrical equipment, the influence of trace impurities in new SF6 gases on SF6-insulating electrical equipment was experimentally investigated. An internal default as metal point discharge on the center rod was simulated by using a straight-line isolator installment which shared gas chamber with a current transformer. Then experiments of using SF6 new gas with various qualities under 2 voltage modes, single 220 kV single voltage mode and 220 kV/3 150 A synchronous upward current-voltage mode, were performed for about 100 h. In the experiments, variations of volume concentration of each impurity in the SF6 gases were detected by a gas chromatography mass spectrometry analyzer and a DPD SF6 impurity analyzer ( made in Canada). The results show that, more by-products like SO2F2, SOF2, and SO2 will be generated when electrical equipment is filled with new SF6 gas which has mass trace impurities, including fluorinated alkane, fluorizating sulfonyl, and carbon sulfur fluoride; considering the corrosive effect of SO2 on the equipment, it is concluded that when SF6 is filled with new gas with high level impurities, the life of electrical equipment will be shortened, especially that of breaks will be shortened.

Chen H.,China Electric Power Research Institute | Yang H.,China Electric Power Research Institute | Xu A.,China Electric Power Research Institute | Yuan C.,Hainan Power Grid Corporation
Proceedings - 2014 9th International Conference on P2P, Parallel, Grid, Cloud and Internet Computing, 3PGCIC 2014 | Year: 2014

Today, big data is not only the data scenario with large volume, but also high-speed and changing all the time. Such data streams commonly exist in Smart Grid facilities. Decision tree as one of the most widely-used analysis methods, has been applied in the decision support system for smart grid. This paper proposes a two-level classifier combining cache-based classifier and incremental decision tree learning, instead of the tree inductions using Hoeffding bound. The simulation result shows that the proposed approach has better accuracy. The combined method can handle high-speed data streams collected from power grid units. © 2014 IEEE.

Yang H.,China Electric Power Research Institute | Chen H.,China Electric Power Research Institute | Yuan C.,Hainan Power Grid Corporation | Lianhang F.,Hainan Power Grid Corporation
Proceedings - 2014 IEEE International Congress on Big Data, BigData Congress 2014 | Year: 2014

In big data era, more and more people concern on what hidden knowledge can be found from data. Today, big data is not only the data scenario with large volume, but also high-speed and changing all the time. Such data streams commonly exist in Smart Grid facilities. As previous research, incremental learning method was proposed to discover the decision model from the continuous data streams. The decision model is able to interpret the findings to an easily understood format that can be used by humans and machines. In this paper, we investigate the previous theories of incremental learning, and apply them for constructing a streaming process engine in power grid system. The advanced learning method produces an efficient way to handle the high-speed data streams that are captured from power grid units, and establishes a decision support system to forecast the trend of power load in certain period. © 2014 IEEE.

Loading Hainan Power Grid Company collaborators
Loading Hainan Power Grid Company collaborators