Advanced Materials Research Vols. 1065-1069

Abstract: After investigating the residual current protection and its major deficiencies, it is found out that the touch/Leakage current could be separated from the total residual current by the current separation technology , which is based on the phase relationship between normal residual current and touch/leakage current. A novel leakage current protection principle without protection dead-zone is proposed .The proposed method, which is put into practice on DSP , proves to be accurate and feasible in digital simulations. Compared to the existing techniques based on the residual current amplitude or its variation,the proposed method can eliminate protection dead-zone effectively and improve the reliability of protection devices drastically.

Abstract: In this paper the feasibility of natural excitation method which uses cross-correlation function instead of impulse response function of the response to identify the modal parameter of 500kV SF₆ current transformer was discussed .Four different algorithms were used to extract the modal parameter of 500kV SF₆ current transformer with the measured cross-correlation function obtained by natural excitation method. The results of modal parameter identification using natural excitation method and experimental modal analysis were compared in the experimental way.

Abstract: On the basis of monitoring surface deformation in some area subject to large transmission line tower base hazard, monitoring method description, monitoring data sorting and calculation of characteristic parameters for surface deformation are conducted to obtain relevant conclusions of the transmission line tower base deformation by sorting and analysis of monitoring results in this paper. Calculation, analysis methods and conclusions of this paper lay a foundation for analysis on stability of the transmission line tower base and provide a basis for design of the transmission line.

Abstract: With the implementation of energy-conservation power generation dispatching, power grid will lose some peak regulation capacity, and changes of load demand characteristic will further increase the difference between peak and valley load of power grid, especially in the Shanghai grid, primarily thermal power units. Its peak regulation capacity will be obviously insufficient, and the problem will become increasingly prominent. This presents a new topic for Shanghai power grid in the power generation market: Under the premise of meeting the customers’ need, how to provide a fair competitive environment that encourage power plants to participate in power peak regulation actively, and then lead the grid to safe, high quality and fairly competitive virtuous cycle.

Abstract: With the increase capacity of wind turbines, grounding device's influence on the equipment and personal safety is becoming more and more significant. In order to calculate the grounding resistance of a wind turbine more accurately, in this paper, based on the Neumann integral formula and the average potential method is applied to derive a formula which can calculate the value of grounding resistance. A general MATLAB program corresponding to the calculation formula is compiled. By comparing the numerical value with the empirical analysis results and building a grounding resistance measurement test platform the correctness of the formulas and program is verified.

Abstract: Harmony search (HS) algorithm is a good meta-heuristic intelligent optimization method, which has been paid much attention recently. However, intelligent optimization methods are easily trapped into local optima, HS is no exception. In order to improve the performance of HS, a new variant of harmony search algorithm with random mutation strategy (HSRM) is proposed in this paper. The HSRM uses a random mutation strategy to replace the pitch adjusting operation, and dynamically adjust the key parameter pitch adjusting rate (PAR). Experiment results demonstrated that the proposed method is superior to the HS and recently developed variants (IHS, and GHS) and other meta-heuristic algorithm.

Abstract: With the rapid development of rail transit, storage of mass data has become an urgent problem. In view of the problem of insufficient storage capacity and low-efficiency for traditional storage technology, maintenance monitoring data collected by MSS (Maintenance Support System) is taken as the background in this paper [1], an open source framework Hadoop is introduced to build MSS data storage architecture. Through building the virtual cluster environment, the Hadoop technology used in signal system of rail transit is verified feasible.

Abstract: A harmony search algorithm with opposition-based learning techniques (HS-OBL) to solve power system economic load dispatch has been presented. The proposed algorithm integrates the opposition-based learning operation with the improvisation process to prevent the HS-OBL algorithm from being trapped into the local optimum effectively. Besides, a new adjusting strategy is designed to dynamic adjust pitch adjusting rate (PAR) and harmony memory consideration rate (HMCR), which is to further improve the performance of algorithm. The HS-OBL is employed to solve 6 units and 13 units power system, the numerical results indicate that the HS-OBL has perform much better than harmony search(HS) algorithm and other improved algorithms that reported in recent literature.

Abstract: Harmony search (HS) algorithm is a new population based algorithm, which imitates the phenomenon of musical improvisation process. Its own potential and shortage, one shortage is that it easily trapped into local optima. In this paper, a hybrid harmony search algorithm (HHS) is proposed based on the conception of swarm intelligence. HHS employed a local search method to replace the pitch adjusting operation, and designed an elitist preservation strategy to modify the selection operation. Experiment results demonstrated that the proposed method performs much better than the HS and its improved algorithms (IHS, GHS and NGHS).

Abstract: The mathematic model of a two-bar truss is built in MATLAB and the analysis is carried out by the genetic algorithm toolbox. The parametric model of the planar truss is established by the ANSYS Parametric Design Language. Solutions are obtained using the first-order method native. Genetic algorithms don’t always display better properties than others. Finally, a joint optimization method is proposed, which combines MATLAB genetic algorithm toolbox and the numerical algorithm based on the quasi-Newton method. The method is identified through the numerical example of the two-bar truss. The results indicate the joint optimization method can always converge to the global optimal solution.