When a fault causes de-energization of several load sections, all the de-energized sections cannot always be restored by transferring them to other sound feeders, due to such constraint violations as line/transformer capacity or voltage drop. In such a case, it is desirable to restore such “priority load” as a traffic signal, hospital, Metro etc. prior to other “non-priority load”. Therefore, this paper proposes a new reconfiguration algorithm for distribution systems in which a restoration priority of the de-energized load is considered. In this algorithm, firstly, the priority load is restored as much as possible by applying the multi-stage support. Secondly, the remaining priority load is supported by asking to save non-urgent (unnecessary) electric power of the contracted customer which is most effective for restoring priority load. Thirdly, remaining nonpriority load is restored if enough power is available. Although the reconfiguration algorithm for distribution system must be formulated as a combinatorial optimization problem, the concept of the effective gradient method is successfully introduced in this paper, so as to avoid combinatorial calculations. Numerical results confirmed that the algorithm is valid and effective for a practical on-line use.
Recently, sectionalizing switches have been coming to be operated by remote control through the distribution SCADA system. However, the problem of determining the optimal switching sequence is a combinatorial optimization problem, and is quite difficult to solve. Hence, it is imperative to develop practically applicable solution algorithms for this problem. Several efficient algorithms have been developed for finding approximate solutions to such problems. These algorithms create a new arbitral distribution system configuration from an initial configuration, and some of these algorithms do not show a load transfer sequence to reach the objective system. Especially, there exists no systematic algorithm which shows the load transfer sequence from an emergency temporary system configuration to get back to the normal operation system. Therefore, another algorithm is necessary to find the load transfer sequence when the initial and the objective system configurations are known. Such an algorithm is proposed in this paper. The proposed algorithm finds the sub-optimal load transfer sequences based on a primal and dual effective gradient method. Several numerical examples show that the proposed algorithm is valid and effective for real scale distribution systems.
The digital simulation method has been utilized to analyze phenomena in power system. Since different algorithms can be applied, depending on the phenomena to be analyzed, digital simulation allows for high-precision analysis. However, it also has a disadvantages: it produces continuous phenomena which occur in actual systems only fragmentarily. Thus, when discussing important projects at research centers such as IREQ in Canada, use an analogue simulator to continuously analyze the phenomena from the moment the fault occurs until steady state. These analogue simulators however, consists of only a few generator models. They are only effective for analyzing phenomena in small-scale systems and do not allow for analysis of phenomena in large-scale systems over a long period of time. For this reason, the Kansai Electric Power Company (KEPCO) in coopration with Hitachi, Ltd. and Fuji Electric Co., Ltd., has developed the world's largest power system simulator (APSA: Advanced Power System Analyzer). The simulator will be used to analyze the evolution of accidents in actual systems and to analyze continuous system phenomena over long period. This paper describes an outline of the simulator.
This paper presents theoretical and experimental studies on voltage distribution of phosphoric acid fuel cell stack having a local short circuit cell. An improved performance analysis of phosphoric acid fuel cells which included non linear cell performance near zero current density is presented. According this theory, not only low cell voltage and unusual high temperature rise of the local short circuit cell, but also unusual high cell voltage of the adjoining cell are expected. These theoretical values are ascertained by experiments using a 3-cells stack of 60cm square which has an artificially short circuited cell. Unusual high temperature rises of the short circuited portion even on light load conditions were observed. Cell voltages of the adjoining cell in the portion close to the short circuit position were ascertained to increase in order to compensate for the decreased cell voltages of the short circuit position.
For the purpose of maintaining the power system security, the authors are developing an integrated security monitoring and control (ISMAC) system, in which both of the preventive and the emergency control basically have important roles. Static security control strategies, which deal with the steady state after the contingency is cleared, have been studied by various researchers including the authors. However, the basic research on dynamic security control, which deals with the transient state immediately after the contingency has occurred, is seldom found. This paper presents a method of dynamic preventive control for the transient stability considering plural contingencies collectively and coordinating the requirements from both security and economy flexibly. In the preventive control for the transident stability, it is necessary to judge the transient behaviors properly and also evaluate the security quantitatively for a postulated contingency. The authors define a security index which represents absorbing capacity of post-fault kinetic energy based on the energy function method. The index has several advantages of being possible to calculate the values fast, being easy to understand its physical meanings, etc. The effectiveness of the proposed method is ascertained through numerical examples for model power systems.
A pressboard providing both low permittivity and good mechanical properties is required as for insulating material in oil-immersed large power transformers insulation. Using a newly developed pressboard, insulation characteristics of inter-coil models are investigated under lightning impulse and AC voltage conditions. It is demonstrated that a spacer fabricated from the new pressboard material (permittivity of 3.5) increases the partial discharge inception and breakdown voltages up to 30%, compared with those of conventional pressboard spacers (permittivity of 4.7)
In high voltage measurement, digitizers are usually used for their convenience of automatic data acquisition. However, the dynamic characteristics of digitizers are not known in detail, especially in the measurement of single-shot waveforms. In this paper, a method for evaluating the dynamic characteristics of digitizers for the single-shot waveform is presented and the accuracy is examined by a numerical simulation. This method is also applied to the digitized data in an experiment and it is pointed out that the dynamic characteristics of digitizers should be evaluated not only for the steepness of the input waveform but also for the individual digital code which digitizers allocates to a small range of the analog input voltage. From these results, the method for estimating the dynamic characteristics of digitizers in the measurement of single-shot waveforms is improved. The dynamic characteristics evaluated by this method show very good agreement with those evaluated for the lightning impulse waveform.
SF6 gas has an excellent dielectric strength for the uniform electric field, however, flashover voltages are reduced for non-uniforrn electric field. To clarify the flashover mechanism, this paper deals with the effects of applied voltage waveforms and gas pressure in the presence of strong inhomogeneous field. For discharge detection, high-speed streak photographs and current measurements were taken. Experimental results proved that voltage-time characteristics have a minimum value in the time range between several micro seconds and 10 micro seconds. Flashover voltages increase for long wave front duration by corona stabilization effects, however, under fast-rising impulse and fast-oscillating impulse voltages leader type flashover appears immediately after the voltage application. Under fast-oscillating impulse voltages, flashover voltages are nearly equal to those of nonoscillating impulse voltages in the time range of several micro seconds, however, bipolar currents flow before flashover occurrs. The corresponding high-speed streak photograph reveals that luminous events and eliminations appear alternately. This indicates that the back discharges occur in the injected space charges. It also clarifies that discharge propagations occur when positive leader currents flow, and they do not occur when negative leader currents flow. These results show that the flashover mechanism for oscillating impulse voltages is a little different from that for non-oscillating impulse voltages, however, flashover voltage is determined by positive leader velocity.