The electrical degradation of zinc oxide varistors (ZnO; Bi2O3 and CoO or Bi2O3 and MnO2) was evaluated by ICTS (Isothermal Capacitance Transient Spectroscopy) method from -180°C to 80°C to investigate the role of each additive. Bias dependence of ICTS signal indicates that those interface states mainly contribute to ICTS signal at temperatures near room temperature and shallower bulk traps mainly contribute to that at lower temperatures. An analytical function for transient capacitance considering both capture and emission of electrons by bulk traps and distributed interface state is newly proposed. Interface states and bulk traps are separately evaluated. Density of number and level for bulk traps for both varistors did not change after degradation. It is confirmed that the degradation of varistors with the additive Bi2O3 and CoO is due to decrease of the density of interface state, and that of varistors with the additive Bi2O3 and MnO2 is due to the shallowing of the interface state levels.
In this paper we introduce the newly developed measuring system of partial discharges (PD) in after-laying tests on site for a long distance gas insulated transmission line (GIL) in a tunnel. We evaluated the length of harmful particles for the overvoltages to be considered for the Shinmeika-Tokai line. The detectable length of the particles by the PD sensors was compared with the harmful particles and the high sensitive PD measuring system which can locate PD signals was developed, considering noise levels in the tunnel. We applied the system to the after-laying test on site and confirmed that the developed system was effective to verify the required performance for a long distance GIL in a tunnel.
Performance and robustness evaluation of an improved H∞-PSS is presented in this paper. The proposed H∞-PSS is compared with the conventional and the optimized PSS which parameters are optimized by minimizing an eigenvalue performance index, taking into account several operating conditions of the power system. Simulation results show that for the nominal operating condition, both the proposed H∞-PSS and the optimized PSS give similar and better performances than the conventional PSS. However, when the robustness is considered over a wide range of operating conditions, the proposed H∞-PSS gives a better performance than the optimized PSS.
This paper presents an application of the interior point methods to decoupled optimal power flow (OPF) problems. The decoupling approach is exploited by decomposing the OPF formulation into a P-problem (P-θ active power model) and a Q-problem (Q-V reactive power model); which simplifies the formulation, improves computation time. We have compared the basic primal-dual interior point method and the predictor-corrector method which is a variant of the primal-dual method on the computational behavior. It was confirmed by numerical studies that the predictor-corrector method has a numerical robustness.
A two-phase flow analysis with a particle model is carried out for a pulsed MHD channel of which working body consists of the combustion gas of a solid propellant and liquid particles of Al2O3. The gas phase and the liquid phase interact with each other by the transfer of momentum and heat. The behavior of liquid particles is described as the statistical summation of each particle's behavior by using the particle model, where the collision-coalescence and the fragmentation of particles are taken into consideration. The average diameter of particles grows from 4μm at the nozzle inlet to 20 μm at the nozzle throat due to the collision-coalescence, increasing gradually to 25μm through the downstream part of the generator duct. The fragmentation of particles occurs near the nozzle throat. The velocity and temperature distributions show that the behavior of the liquid phase along the duct lags behind the gas phase in velocity and temperature and that the growth of particle diameter increases the velocity, lag and the thermal lag.
Two-dimensional analyses with the κ-ω turbulence model are carried out in order to reveal the performance characteristics of pulsed MHD channel under strong MHD interaction. When the Faraday current becomes large, a large boundary-layer separation is induced in the anode side, leading to a shock wave and a large electrode voltage drop. The separation narrows down the width of core flow, which induces the deceleration and acceleration of core flow. When the applied magnetic flux is varied proportionally to the channel current, the channel voltage increases with the channel current because of the increase of magnetic flux density if the channel current is small. On the contrary, when the channel current becomes large, the increase of channel current decreases the channel voltage because of the deceleration of flow. The saturation of channel current is reconstructed by the analysis as was found in a hot-fire test.
Load data captured by disturbance monitors have been analyzed and the static and dynamic characteristics of the observed behavior of a residential load have been determined. This paper describes simple models of the load behaviors observed during and shortly after a system fault, including the load dropping a voltage dip caused by a system fault and simple dynamics of load admittance due to an aggregate response of small induction motors to a voltage dip. The system impacts of observed load behaviors have been evaluated.
This paper describes an optimal TCSC (Thyristor Controlled Series Capacitor) control method which can be applied to a large-scale power system to suppress system disturbances. TCSC is introduced from time to time for increasing the transmission capability. The main objective of introducing TCSC is that it can be used for improving the system stabilization along with the existing equipment, thus cost involvement is very less. However, the practical control method should meet some requirements, such as simple and unique design, it should be applicable to a multi-machine system. The proposed method is based on the optimal regulator theory with quadratic performance index. And sensitivity coefficient to relate TCSC states and swing equations of the generators are used to the formulation. So the design is very easy. The improved transient performances of frequency deviations are shown in the simulation tests with 9-machine 38-bus system model. Three different cases of the selection index of TCSC allocations, namely relative angle, line length, and sensitivity coefficient are compared. Thus, using sensitivity coefficient is best method to find out the best allocations. Another considerations, tuning of the weighting matrix and a response example for a heavy fault are also discussed.
This paper proposes a support system for power system planning tasks which can adequately assist planning engineers. The proposed system possesses the following features. It can efficiently manage a set of data necessary for each analysis program by making use of an object oriented database with a version management mechanism. The second feature lies in a way of registering a series of analysis programs into the support system. The last is that the system equips graphical user interfaces aiming to support use of analysis programs and the above functions for planners. A prototype system has been constructed on a Sun Work Station with the object-oriented language (Smalltalk-80) and the database management system (Versant). The effectiveness of the proposed system has been demonstrated.
The authors have so far proposed an approach for generating system configurations based on the optimal switching power flow (OSPF) algorithm proposed by H. Glavitch. The proposed approach takes account of day to day outage works continuity by augmenting the tableau to consecutive three days. The ratio of the power flow of a line to its nominal capacity (referred to as iload ratioi) is chosen as an objective function in the pivot operations. However, since the proposed approach minimizes the objective function with respect to a line with the maximum ratio, it is impossible to completely avoid from causing overload at N-1 contingencies. The main contribution of this paper is to present an efficient method for generating system configurations that meet the security constraint for N-1 contingencies. We adopt the sum of two different functions for pivot operations: maximum load ratio under normal conditions and N-1 contingencies. Owing to the adopted composite functions and the fundamental characteristic of OSPF, the amount of computation for searching an optimal system configuration in consideration of N-1 security can be reduced significantly. In addition, we propose a new method for selecting pivot operations based on the tabu search so as to prevent a solution from falling into a local optimum. In order to confirm the effectiveness of the proposed method, a part of the 110kV transmission systems of the Chugoku Electric Power Company is chosen as a test system. Results obtained have shown the clear advantages of the proposed method in both the reduction of switching operations and the sustenance of N-1 security.
Reactive power supply sources with excellent voltage and dynamic stability characteristics are required to maintain system stability in a large-scale heavily loaded power system. The authors carefully examined the effectiveness of synchronous condenser application to such a power system in improving the stability margin. In an analysis example of a large, actual power system which has a comparatively higher nose tip voltage, a large capacity synchronous condenser can lower the nose tip voltage by about 4kV on the same demand conditions compared with shunt capacitor case. A large capacity synchronous condenser applied to a large-scale heavily loaded power system achieved the expected results with good operating characteristics.
This paper presents a new method of a daily peak load forecasting by structured representation on genetic algorithm for function fitting (STROGANOFF). The STROGANOFF is a hierarchical technique of multiple regression analysis method and a GA-based search strategy. Then, we use the fitness of MDL-based for the STROGANOFF. The proposed method is demonstrated by using the data of Chubu district in Japan.