In engineering fields, one of the most important applications of artificial neural networks is a non-linear parametric model that can approximate any continuous input-output relation. Especially, in next day peak load forecating, the ability of approximation of artificial neural networks is very useful. However, an important but difficult problem exits in applying a neural network to next day peak load forecasting. This problem is to estimate the quality of approximation of a neural network. This quality depends on the architecture of the neural network applied, as well as on the complexity of the target relation. In other words, the pmoblem is to determine the optimal number of parameters in the neural network.
This paper proposes a new method for the selection of the optimal neural network architecture for next day peak load forecasting. The proposed method is based on a new information criterion. This paper also applies the proposed method to next day peak load forecasting constructed by real data and shows the comparison of error estimation of some models for peak load forecasting.

抄録全体を表示

By ZEV regulation of the California government in 1990, not only the United States but also European countries and Japan are largely interested in replacing gasoline car by electric vehicle. Electric utilities expect that the electric vehicle improves the difference between the daytime and nighttime of a recent demand as a new market of electric power demand.
There are some reports that indicates on the danger which generates the new peaks as the electric vehicles spread through the market in charging start hour. As the countermeasure for avoiding this problem, the authors propose that 1) inverse load flow by discharging of the contract private use EV is carried out as an energy shift in daytime, 2) the regional charging time shift method is introduced in the midnight charging time zone, in this paper. Therefore, the energy actual consumption of the automobile was investigated, and future demand was predicted by the mathematical consumption modeling.
Especially, it is described a strategy of optimum introduction on scheme of effective electric vehicle for energy shift in section 2.2. And authors point out the necessity of market regulation for EV.

抄録全体を表示

Photovoltaics(PV) system is expected to play an important part in electricity supply as a clean and renewable power source. But, because of the unstability and uncontrollability of the output of PV system, its economic value is significally decreased as a power source. One way to improve the economic value of PV system is to combine the system with a storage facility. We evaluated the economic value of the storage facility which is optimally managed in conjunction with PV systems.The probabilistic relationship between electricity demand and the output of PV system is also considered.
The results quantitaitively show that the main factor to determine the economic value of storage facility is its contribution in cutting the peak of electricity demand. The dependance of the economic value on the load profile and the capacity of PV systems is also examined.

抄録全体を表示

Most of the approaches aiming at the reduction of the power flow computation time approximates the Jacobian matrix. Thus itt degrades the convergence compared to the conventional Newton method. This paper proposes a new approach for reducing the processing time by considering the fact that half of nodes in real power systems arc floating nodes to be able to remove. In the conventional reduced matrix approach where the floating nodes are removed, the sparsity is being lost. The method in this paper does nott remove all the floating nodes but. keeps sortie nodes by using a optimal criteria for keeping the sparsity. The criteria is to indicate the minimum number of elements in the reduced matrix. This method has been applied to a 1000-node test. system. It. was verified the number of elements of the Jacobian has been reduced to about a half against the conventional matrix. And computation time has been remarkably improved without sacrificing the convergence characteristics for the power flow computation.

抄録全体を表示

This paper studies the power flow computation for advanced distribution systems where multiple PV-buses appear. The background for this study is that with the introduction of various dispersed generation sources into modern and future distribution systems, it may be necessary to treat multiple PV-buses in distribution power flow calculations. Meanwhile, in analysing and planning such complex systems, hundreds or thousands of power flow computations are required. Therefore efficiency of the power flow method is of vital importance for reducing the computing burden. The method presented in this paper is a modification to conventional implicit Z matrix method by which only one voltage specified bus can be dealt with efficiently. The modification is based on direct mathematical derivation and therefore overcomes drawbacks of heuristics based techniques. The method tries to solve for the total unknown variables simultaneously without use of trial and error iterations. Computational experience shows that the method is especially suitable for power flow study of distribution systems. Comparisons with Newton power flow method are provided to illustrate the efficiency of the method.

抄録全体を表示

This paper presents a new method of simulation for unsymmetrical faults in power system. The new method is based on the phase coordinate method. The symmetrical component method has been used traditionally for unbalanced faults in power system. This is mainly because the synchronous generators can be easily treated by the method. However, when analyzing a system that has elements with unsymmetrical impedance, the symmetrical component method can not be used. In such cases, the simulations can be easily executed by using the phase coordinate method. However, there are few papers that present detailed procedures of simulation based on the phase coordinate method. This paper, at first, presents phase coordinate models of the elements in power system; i.e., synchronous generator, transmission line, transformer and load. Next it presents a method of power flow calculation of a power system including elements with unsymmetrical impedances. Finally it presents detailed procedures of simulations based on the phase coordinate models derived.

抄録全体を表示

Cogeneration system (CGS) is expected to be one of the most efficient energy supply systems. Many discussions concerning CGS have so far focused on the energy saving and/or cost benefit for the individual energy demand. On the other hand, this paper describes the potential of CGS evaluated from the viewpoint of the reduction of total energy consumption in the whole energy system. In our numerical simulation, it was revealed that the higher efficiency would be required for the introduction of CGS from the viewpoint of the energy saving of whole energy system than that of the individual demand. This is attributed to the following reasons. In order to derive the high efficiency of CGS. it should be operated to match with the changing heat demand without wasting the excess heat output, resulting in the low annual load factor of CGS. On the other hand, the conventional generator with high generating efficiency can be operated with the high load factor throughout the year. Thus, when the heat supply efficiency of CGS is not so high, it is rather efficient for the whole energy system to replace the output of old-fashioned low-efficiency generators with that of advanced ones. Consequently, the improvement of annual load factor of CGS is inevitable to facilitate the introduction of CGS.

抄録全体を表示

In IEC publication 60060-1, “high-voltage test techniques”, the time to peak Tp of the switching impulse voltage is defined as “the time interval between the actual origin and the instant when the voltage has reached its peak value”. In this case, it is very difficult to determine precisely the time to peak due to the flatness of the switching impulse around the peak. The situation becomes even more so when digitally recorded switching impulse data include noise around the peak.
Therefore, IEC 61083-2, “Digital recorders for measurements in high voltage impulse tests” introduced, as an appendix, an approximation formula which determines the value of Tp from the interval time between the instants when the switching impulse reaches 30% and 90% of the peak value in its front.
Authors checked the formula and propose more precise approximation formula and a graphical method. The details of the newly proposed methods are discussed in this paper.

抄録全体を表示

This paper describes the insulation characteristics under polluted conditions for a polyme composite insulator and a 154 kV insulation arm composed of the insulators.
The equivalent salt deposit density (ESDD) on the housing of the polymer composite insulator wa almost the same level as that of the long rod type porcelain insulator subjected to a coastal exposur test for about 8 years. However, the surface leakage currents of the housing were smaller tha, those of the porcelain insulator, due to the restoring of water repellency on the surface of th housing. The withstand voltage of the 154 kV insulation arm obtained by a clean fog method in an, ESDD range from 0.03 to 0.5 mg/cm² was about 15% higher than that of a long rod type porcelai:i insulator, but was about 15% lower than that of a polymer composite insulator, due to the structur with plural units connected in parallel.

抄録全体を表示

In this paper, the static electrification phenomena on the dielectric materials by SF₆ gas flow were investigated. When pure SF₆ gas flows on the dielectric materials at high speed up to 25m/sec, no electrification was observed on the dielectric material surfaces. In case of the flow contained dielectric microscopic particles colliding with dielectric materials, it is confirmed that the SF₆ gas flow of speeds above 5m/sec causes the electrification on the dielectric surfaces. The faster the SF₆ gas flow, the higher the surface potential of the dielectric material is. Such phenomena of static electrification of SF₆ gas contaminated with microscopic dielectric particles were explained to be caused by the friction between the dielectric materials and the dielectric microscopic particles in SF₆. The polarity and the magnitude of the surface potential of the material, after flowing the SF₆ gas contaminated with dielectric particles, are confirmed by the result of a general friction experiment of the two dielectric materials. It is considered that a well-controlled gas insulated transformer not contaminated with microscopic dielectric particles in SF₆ can not generate electrification phenomena.

抄録全体を表示

This paper evaluates thermodynamic characteristics and energy saving potentials of four kind of power generation plants which generate power by utilizing waste heat from refuse incineration plants. Investigated four plants are: conventional refuse incineration power plant, improved conventional refuse incineration power plant, repowering system which is constructed by incorporating a gas turbine into a refuse incineration power plant (super refuse incineration power plant), and H₂O turbine power generation system utilizing waste heat from refuse incineration power plant (hyper refuse incineration power plant). The structure and advantages of these systems are described. Power generation efficiency and exergetic efficiency have been estimated and compared, by using simulation models constructed. Energy saving effects have also been evaluated if each system utilizing all the disposed refuse in Japan were assumed to be constructed. It has been shown that the hyper refuse incineration power plant has the highest efficiency among 4 systems with 37.0% exergetic efficiency and 51.3% fuel-base efficiency, and that it has the potential for saving 13.0% of the amount of natural gas consumed in Japanese thermal plants in the year 1993.

抄録全体を表示

This paper describes a series of development tests carried out when developing a fault location system applied for the main bus of GIS (Gas Insulated Switchgear) by using shock pressure relays. In the main bus of this GIS, each gas section per bay is connected by gas pipings.
This paper also refers to the basic construction of the software of the fault location system actually applied to GIS with rated voltage less than 168kV.

抄録全体を表示

Three-dimensional behavior of disturbances in nonequilibrium cesium seeded argon plasma in a disk MHD generator is simulated in r-θ-z coordinate system. The results reveal three-dimensional structures of ‘streamer’ and ‘domain’, which are known as nonuniformity caused as the result of ionizaion instability. The fluctuation of low electron temperature which exceeds the critical amplitude grows and causes the streamer structure on the r-θ plane, whereas the structure becomes uniform along the z-axis immediately. On the other hand, a high electron temperature fluctuation causes the domain structure and it forms a completely three-dimensional nonuniformity, which can be ascribed to the long relaxation time for argon ions.

抄録全体を表示

This study is aimed to develop a method of reliable automatic decision making for washing the polluted insulators in substations near coastal areas based on the actual data of human decision making, pollution deposits and weather information. The automatic decision making by use of the weather information is derived based on Bayesian rule of probability theory. Effectiveness of the proposed method is assured by a new technique which uses the modified actual data of pollution deposits and weather information.

抄録全体を表示

We propose a new cascading thermal system of recovering low-temperature heat lower than 300°C by using chemical heat pump systems and thermoelectric devices, which we call heat regenerating and recycling system. We investigate a concept and an efficiency of its system. We perform calculations of entire efficiency of the system and confirm that the entire efficiency of our proposed system becomes larger than that of conventional cascading system numerically. We also examine effects of efficiencies of each chemical heat pump system and each thermoelectric device on entire efficiency of the system.

抄録全体を表示

The Swedish (Karolinska Inst.) investigation of epidemiology, which has been carried out on cancers in children residing near high-voltage power lines in and outside Stockholm county during 1960_??_1985, shows us the conclusion that there is a significant association between the historical magnetic fields calculated and childhood leukemia. Employing the results of the stratum-specific analyses, and assuming that the relative risk of childhood leukemia is significant whenever the lower limit of its 95% confidence interval is above unity, we can find out that the association is seen for children diagnosed 1975_??_1985 residing outside Stockholm county, while it is not seen for those diagnosed 1960_??_1974 residing in and outside Stockholm county. We can conclude that the association is not a causal relation, because consistency of association was not always seen throughout the areas and periods investigated. Therefore, we have no alternative but to consider that the conclusion which suggests that the same association is to be observed in every area and at every period is not valid.

抄録全体を表示