There have been many studies for automatic failure restoration problem of power system. This paper proposes a new method of determining restorative operation process using dynamic programming. In this paper, a branch on-off operation is corresponding to a multi-stage decision of DP method, and the operation is carried out on every stage from initial failure system to final restorative system. Moreover, a power system configuration is formularized by laws of Kirch-Hoffs.
For a practical use of this method, it is necessary to accelerate calculation time. Therefore, two measures are taken as follows, 1) priority branch method for selecting an operation branch. 2) state reduction method for reducing system states with lower estimation evaluated by the objective function on every stage. By using this method on a model system, the effectiveness of this algorithm has been demonstrated.

View full abstract

In this paper, we study on the cutting plane methods for Lagrangian relaxation based unit commitment algorithm. In the algorithm, non-differentiable optimization methods can be applied to optimize the dual function, and a subgradient method which needs parameter tuning and has some drawbacks such as computational inefficiency and oscillating behavior is commonly used. The cutting plane method and the central cutting plane method are applied to the algorithm and implemented using re-optimization techniques. Numerical example shows that both methods are accelerated by the re-optimization techniques and have good convergence property without parameter tuning.

View full abstract

Assuming that photovoltaic (PV) systems are prevailing in residential houses under carbon tax regime, an economic performance of PV systems is investigated from standpoint of an electric utility. The economic performance is estimated with a buying price of PV electricity and a PV economic index which is defined by ratio of the buying price to the cost of generation of the electric utility. Because these values depends on electric power development and operation, the best mix and the operation of power plants are obtained by the linear programming under restrictions on power generation. Then, the buying price of PV electricity is calculated from the total cost of the electric utility. The buying price means the upper limit under which the electric utility never suffers a loss. Moreover, the buying price is compared with the power generation cost. Parameters are the prevalence attainment ratio of PV system (0_??_00%), the upper limit of newly-developed nuclear power plant (0_??_4 GW/10y) and the generated energy ratio of coal thermal plant (0_??_15%). Chubu Electric Power Company, Inc. is adopted as the electric utility. The calculation results show that the buying price of PV electricity increases linearly with increasing of the carbon tax rate and its values are 9 and 11.5 yen/kWh when the carbon tax rate are 0 and 25 thousand yen /t-C, respectively, which does not affected with the prevalence attainment ratio of PV system and the upper limit of newly-developed nuclear power plant. It is not the carbon tax rate but the newly-developed nuclear power plant that influences on the PV economic index. The values of the PV economic index are 1.35_??_1.45 and 1.50_??_1.60 when the newly-developed nuclear power plant are 0 and 4 GW/10y, respectively. These results show that the economic performance of PV system is increased by developing the nuclear power plant at a certain rate and introducing the carbon tax.

View full abstract

Thyristor-controlled series capacitor (TCSC) is being developed to improve transmission capability of power systems. TCSC is thought to compensate transmission line reactance without causing subsynchronous resonance (SSR). However, in order to evaluate its effect quantitatively, we must calculate frequency responses of generator damping torque. Simulations need long computing time, and it is hard to choose frequency freely. In this paper, we propose a method of analytically calculating the damping torque. First, when a generator rotor oscillates at a frequency, two voltage components appear. We analytically calculate its damping torque from small current variations due to the voltage components. The damping characteristic changes depending on a method of firing thyristors. The best characteristic is obtained when we trigger referring to the fundamental wave of TCSC voltage or current. By choosing an appropriate firing angle, we can drastically reduce negative damping by TCSC. The damping characteristic is closely related with the system impedance. The fact that TCSC has large resistance in 0_??_60Hz, substantially contributes to improve the characteristic. Lastly, we execute numerical simulations of SSR, and examine the validity of our investigation.

View full abstract

Eigenanalysis methodologies for large-scale power systems are considered to be already well-established and many methods have been proposed since the middle of the 1970s.
The features of these methods are, 1) to utilize the sparsity techniques, 2) to obtain specific set of eigenvalues, 3) to have numerical robustness, etc.
On the other hand, recently digital excitation controllers have been aggressively introduced in power systems and a method that can grasp the characteristics of power systems including them is strongly desired. Eigenvalue methods are suitable for identifying fundamental characteristics of power systems by using eigenvalues and eigenvectors obtained from a state matrix linearized around an equilibrium point.
A digital control system includes 0-th order hold or 0-th order delay function with sampling period T, these elements can be drscribed as difference equations whose characteristic is different from differential equations. Those elements make structure of the state matrix dense, therefore, it is difficult to apply conventional sparsity technique to the state equation including difference equations. An eigenvalue method for large scale-power systems including digital controllers has not been established yet due to such reasons.
In this paper, the authors propose an efficient eigenvalue method which can deal with large-scale power systems including digital controllers by using newly developed sparsity technique.
To confirm the effectiveness of the proposed method, test calculations were carried out by using a 16 generator and a 64 generator systems. The results show that the proposed method was able to obtain eigenvalues with sufficient accuracy and high speed.

View full abstract

R&D planning is important for promoting R&D effectively and efficiently. But it is examined and decided mainly by experts' judgements based on their intuition and experience. The authors consider that systems analysis should be introduced to make R&D planning more rational and persuasive and to make R&D more effective and efficient.
This paper has two purposes. One is to propose a method for analyzing cost-effectiveness of R&D, which is based on cost-efficiencies derived from technological progress models with variable of cumulative R&D expenditure. And the other is to validate plan of R&D on solar cells in the (New-) Sunshine Program of Japan by using the proposed method with demonstrating its effectiveness.
Based on the cost-effectiveness analysis, R&D goals or allocation of R&D expenditure for multicrystalline silicon (mc-Si) solar cells might not be appropriate during FYs 1997-2000. And during FYs 1989-1996, R&D on me-Si solar cells might have been planned without well forecasting increase of the mc-Si solar cell production by subsidization programs after that.
This analysis demonstrates that the cost-effectiveness analysis of R&D by using the proposed method is effective and helpful in planning R&D. The authors expect that this method will be used widely for decision making in R&D planning for energy technologies.

View full abstract

In this paper we consider a damping control of low-frequency oscillations in an electric power system. On the basis of the hypothesis that an auto-parametric resonance model can explain a power oscillation, we propose a new model for a robust damping control, by which the system maintains stability even if some auto-parametric resonance happens. With this model, we can express a parametric variation of a principal oscillation mode and a class of uncertainties which cover neglected dynamics. Since the model has a certain structure of uncertain parameters, we design a robust controller via μ-synthesis. The robust controller which can be obtained from the presented design strategy has the property that the control performances are more sophisticated in comparison to controllers designed with other existing methods based on the H_∞control.

View full abstract

Power system stabilizing control plays an important role to remain in synchronism during major disturbances resulting from sudden or sustained load changes, either the loss of generating or transmission facilities. In this paper, two methodologies to allocate TCSC (Thyristor-controlled Series Capacitor) controllers have been applied to stabilize disturbances in power systems. One takes steady-stability of power system into consideration. The other considers transient stability. Both methodologies figure out location points and the number of TCSC controllers satisfying the stability index. After allocating TCSC controllers, this research applies robust controller design for them. As the methodology of robust controller design, output feedback H_∞ control has been utilized. Nowadays, LMI (Linear Matrix Inequality) approach is very effective, since it solves the Lyapunov inequality without considering the weighting coefficients used in other control theories. This paper has designed the robust TCSC controllers for multi-machine systems based on the LMI approach. In the proposed approach, input information of TCSC controllers is only angular velocity deviations of generators as observable value. The proposed LMI-based approach is proved to be applicable for the design of robust TCSC controllers to enhance robustness and fast response by simulations on a test system.

View full abstract

In order to verify the efficiency or availability and stability of photovoltaic generation systems, huge system apparatuses are needed, in general, in which an actual size of solar panel, a type of converter system and some amount of load facilities should be installed in a particular location. It is also hardly possible to compare a MPPT control scheme with others under the same weather and load conditions in an actual PV generation system. The only and a possible way to bring above-mentioned problem to be solved is to realize a transient simulation scheme for PV generation systems using real weather conditions such as insolation and surface temperature of solar cell. The authors, in this paper, introduce a novel simulation method, which is based on EMTP and EMTDC, for PV generation systems under the real weather conditions. Firstly, VI characteristic equation of a solar cell is simulated and implemented, and then the real data of weather conditions are interfaced to EMTP and EMTDC program using Fortran program interface method. The outcome of the simulation demonstrates the effectiveness of the proposed simulation scheme. The result shows that the cost effective verifying for the efficiency or availability and stability of PV generation systems and the comparison research of various control schemes like MPPT under the same weather conditions are possible.

View full abstract

The recent trend of deregulated power industry makes a request for more economical operation of distribution networks. In such situation, it is necessary for power utilities to cooperate with customers in maintaining proper voltage profile in an economical manner. The authors have been discussing a new DSM system named as Customer's Cooperation System (CCS) that can provide incentive for customers in order to make them participate in the network operation.
This paper proposes a simulation model for the evaluation of effectiveness of CCS from the viewpoints of voltage management and economy. A novel point of the proposed model is that the decision making process of utility and individual customers is embedded in power flow analysis. The simulation studies show that the proposed model is suitable for analyzing the effects of customer's behavior on voltage profile, and which can be used to determine the incentive price for drawing the cooperative actions from customers.

View full abstract

Due to the rapid increase of miniaturized and networked indoor equipment (electric appliances) in highly information-oriented society, the number of lightning damage has been increasing. Since the indoor equipment has AC power line and communication line, the potential difference by lightning surge occurs between grounds, and the equipment connected between grounds becomes the passage route. As a result, when the equipment becomes passage route of the lightning surges, it causes the flashover and it become the damage. Above mentioned relation is also similar on the equipment with the grounding conductor.
In order to investigate the lightning damage of indoor equipment, authors prepared the model house and then surge protective elements installed in the indoor equipment was arranged. Using this experiment circuit, the relationship between operating condition and grounding resistance value for lightning surge from the power side was investigated experimentally. In addition, the application of low voltage arrester to AC power service entrance was examined experimentally for the purpose of the protection of indoor equipment against the lightning.

View full abstract

The oxidative degradation characteristics of a tape lapped joint (following TJ) were examined. Permeability constant, diffusion constant and consumption speed of Ethylene-Propylene rubber tape (following EPR tape), semi-conducting tape and cross-linked polyethylene insulations were measured. A certain amount of oxygen existed in the aperture of a conductor and a wire shield of a cross-linked polyethylene insulated cable. The simulation of the diffusion equation of oxygen in the TJ, which added the oxygen consumption clause, was carried out. It was shown clearly that this oxygen might oxidize locally and cut the EPR tape, which was the insulator of the TJ, by the simulation. The part imitation models of the TJ, which could supply air or oxygen only from a conductor side, were manufactured. These models were heated under constant temperature. The EPR insulation layer fractured from the inner side, and resulted in the whole destruction. Moreover, when the temperature of 105 degrees and operation voltage were added to the used TJ for 2, 877 hours, the insulation of the TJ was destroyed. Only EPR tape fractured over the length of 40cm, and the internal semi-conducting tape was not fractured. We presumed from the simulation result and the destructive aspect, insulation destruction of the TJ was carried out by oxidative degradation of EPR tape. When a high temperature operation of the TJ was carried out, it was shown that there was a possibility of the insulation destruction by oxidative degradation.

View full abstract

EPR tape is employed as the insulator of tape lapped joint (TJ), which is used for 66kV class XLPE cable. It has been found that, in case a TJ is operated at higher than the short-time allowable temperature of XLPE cable (105°C), its EPR tape is oxidized, which may break the tape to lead to dielectric breakdown in the TJ. Therefore, when one operates a XLPE cable at higher than this temperature, it is necessary to catch the degree of oxidation of the EPR tape.
In this research, we have developed a technique to investigate the degree of the oxidation of EPR tape. Oxygen induction time (OIT) observation gives the volume of remaining antioxidant in EPR tape, from which the degree of the oxidation is derived. We investigated the oxidation of EPR tape of the TJ which had been operated at high temperature by this technique. The result shows that the oxidation concentrates to the surroundings of the inner conductor.

View full abstract

Impulsive electromagnetic noises, generated by electrical discharges on power distribution lines, are mainly responsible for wide-band disturbances to communication systems. In this paper, the impulsive noises emitted from various discharge sources are analyzed. We employed a non-contact measurement system using the VHF wide-band antenna, which permits not only the waveforms of electromagnetic pulses but also the timing of pulses. By using this system, we measured the pulses generated by applying high voltage upon the discharge sources. The characteristics of a single pulse waveform are analyzed by digital signal processing techniques, which are Fourier and wavelet transforms. Moreover, the time-sequential pulse-height patterns of pulses are investigated. The pulse-height and pulse-repetition of both positive and negative pulses are discussed associating with the mechanism of discharge. It is shown that the results are useful to classify the kinds of discharge sources and assess the condition of insulators in distribution systems.

View full abstract

The aim of this paper is to evaluate the current breaking performance of a circuit breaker through simulation of the dynamic characteristics of its arc discharge, whose parameters were measured beforehand. Here the general arc model is taken as the fundamental. Recently, the time constant of the arc model has been considered to be variable, and there are two theories that describe the behavior of the parameter variation. One theory asserts that the time constant should vary as the function of the steady state arc conductance for the current at that instant, but the other asserts the function of the instantaneous arc conductance. It is necessary to discuss which assertion is true or better. It has been inspected firstly through numerical calculation and also the method for identifying the time constant from the experimental result has been discussed and developed. It is revealed that the results from both theories were of less difference for predicting the performance of an arc discharge through simulation, and they were applied in parallel afterward. In the latter part of the paper, the experimental results of three reported cases have been analyzed through the identification of their arc parameters. It is revealed that the time constant of the arc discharge may be an increasing function or a decreasing function concerning the variation of the arc conductance, depending on the types of the arc discharge. The behavior of arc discharge can be described well in concern with its cooling mechanism. It is verified that both theories can be applied to achieve the aim of the paper, and the behavior of the arc time constant has been revealed for the first time.

View full abstract

High reliability of electric power supply is requested by the rapid evolution of information-oriented and computerized society. On the other hand, it is important to consider the cost down on power distribution lines in power companies. From this point of view, the authors developed an analytical method that can calculate flashover rate caused by lightning considering both the reliability and cost performance. As the results of calculations, it is clear that the flashover rate is significantly influenced by various parameters such as the lightning current waveform and nearby buildings. Moreover it is clear that the flashover rates of power distribution lines for various lightning protection designs are much different even if the constructing costs of power distribution lines are the same.

View full abstract

This paper analyzes the distribution of the corona discharge current distribution and the ambient electric field intensity on an EHV transmission tower during a thunderstorm. In the measurement of corona discharge current from the arm ends of a tower at Mt. Okushishiku in 1996 and again in 1998, unexpected phenomena were observed in the corona current and the charge distribution. The different polarities of the corona discharge current were intermittently measured at the different arm ends of a transmission tower. Through these measurements, the initiation field intensity of the corona discharge current and the field intensity on the top of the tower were observed for the first time.

View full abstract

A transient response of a cable with a two-layer conductor, such as a conductor including a semiconductive layer, is analyzed by means of a finite-difference time-domain (FDTD) method. Prior to this analysis, the accuracy of a transient analysis by this method, is investigated in comparison with an EMTP simulation for a single-core cable without a two-layer conductor. Current waveforms at the sending and receiving ends of the cable, calculated by FDTD method, agree well with those calculated by EMTP. In the case of a simulation for the cable including a semiconductive layer which is the second layer of a two-layer conductor, current waveforms at the both ends of the cable are distorted depending on the conductivity of the second layer. Also, the propagation velocity of a surge current is dependent on it. When the conductivity of the second layer ranges from 0.001 to 0.1 S/m, the shunt admittance of the cable dominates the above phenomena. On the other hand, they are ascribable to the series impedance when the conductivity is higher.

View full abstract

We have been investigating the feasibility on Superconducting Fault Current Limiting Transformer (SFCLT) with the functions of fault current suppression and system stability improvement. This paper discusses these functions with consideration of overvoltage and thermal characteristics of SFCLT, and their dependence on recovery time into superconducting state after the quench of SFCLT. Simulation results using EMTP suggested that SFCLT would be coordinated with the background power system under the appropriate superconducting recovery time. The operating parameters of SFCLT were also optimized in a simplified power transmission model system.

View full abstract

There are many cases where trees grow around power system equipments such as transmission lines and distribution lines in Japan. For developing better lightning protection measures, it is essentially important to clarify influence of trees upon lightning stroke attachment manners. This paper presents results of experiment of discharge characteristics for an open wire of power distribution lines and a tree, which were placed close to each other. In this experiment, lightning impulse voltages were applied under the condition that an open wire and a tree were placed close to each other, and the rate of sparkover to the open wire or the tree was examined. Moreover, pictures of sparkover were taken by the video camera and the 50% sparkover voltages were measured for the open wire and the tree that were placed independently.
As a result, sparkover to the open wire only occurs, even when the open wire is lower than the tree by about 1.5 meters. For some configuration, the rate of sparkover to the open wire tends to increase in comparison with that of sparkover to the open wire via the tree when the value of applied voltage has gradually raised. From this experimental results, it is clarified that the shielding effect of trees for power distribution lines is not expected under some conditions.

View full abstract

When GIS insulation specifications are improved, it is necessary to recognize the insulation characteristics under oscillatory overvoltage waveforms occurring in the field. Actual overvoltage waveforms, which have oscillation and steep rising, are different from standard lightning impulse waveform. We call these actual waveforms “non-standard lightning impulse waveforms”. In this paper, insulation characteristics under non-standard lightning impulse waveforms which are simulated the actual overvoltage waveforms on GIS were investigated. As the first step, we picked up the important actual overvoltage waveforms for insulation design which are double-frequency oscillatory waveform and single-frequency oscillatory waveform. The experimental circuit was designed for generating simulated waveforms by EMTP analyses. Insulation characteristics under single-frequency oscillatory waveforms and double-frequency oscillatory waveforms, which were generated on the experimental circuit, were investigated. As a result, it was obtained that the insulation characteristics under single-frequency oscillatory waveforms are about 15_??_24% higher than that of standard lightning impulse waveform and insulation characteristics under double-frequency oscillatory waveforms are about 6_??_13% higher than that of standard lightning impulse waveform.

View full abstract

Unit step response (USR) characteristics of a measuring system for the test of chopped-lightning impulse voltages, which were computed by the Numerical Electromagnetics Code (NEC-2), are compared with those of the USR measuring circuit recommended by IEC. When an impulse voltage generator of high impedance is used in the test of a chopped-lightning impulse voltage, its response is similar to that of the USR measuring circuit, while a low-impedance generator is used, its response differs.
Although the influence of sphere electrodes of the chopping gap is identified as a spike in the initial rising part of the USR, it little influences the USR parameters of a system including a shield ring and a high-impedance impulse voltage generator. In the case of a system having a 100Ω or 200Ω damping resistor and a low-impedance generator, however, the response times become longer owing to the presence of sphere electrodes.

View full abstract

This paper proposes an approach based on wavelet packet transform (WPT) for rootmean square (rms) and power measurements. The WPT can decompose a waveform into uniform frequency bands, which are important for identification of harmonic components and measurement of harmonic parameters.

View full abstract