This paper presents the results of measurement of harmonics in utility power distribution systems and proposes a harmonics suppression method based on capacity modulation of capacitors. Three years of survey with newly developed measuring instruments in a commercial area found that the peak harmonic value as high as 10% can occur during the time zone between 19:00 and 23:00 hours of any day as the power-factor angle leads prominently due to excessive quantities of capacitor at night when the electrical loads is low. The primary cause was identified with the resonance of harmonics from many sources. To suppress harmonics in the distribution system, the authors suggest that resonance frequencies are controlled by modulating the capacities of high-voltage customers' capacitors. Simulation showed the proposed method can reduce the harmonic voltage by half.

View full abstract

This paper proposes the cooperative fuzzy controller of AVR and GOV to improve the transient stability of power systems. The fuzzy rules to stabilize the power systems are selected from general ideas based on the sliding mode controller. The both AVR and GOV input are determined to satisfy system's stability from the values of state variables, acceleration power and magnitude of their products. Moreover, these control laws reduce chattering effect due to discontinuity of the control laws. Therefore, this fuzzy controller acts to improve the transient stability of power systems. By introducing the ideas of sliding mode control, the fuzzy rules and composition of membership functions can be constructed easily. The validity and usefulness of proposed fuzzy controller are tested on a single-machine infinite bus system. It is shown that the proposed controller has robustness regarding parameter errors in the synchronous machine and that the transient stability of power systems is improved by cooperative AVR and GOV control.

View full abstract

Generator control has been studied using various methods. Generator control is an effective way to improve the stability and reliability of power systems which are continually expanding and becoming more compli-cated. However, the comparative estimation of these control methods is not made using one common criterion. In this paper, three kinds of fundamental generator control methods, optimal control, fuzzy control and PID cuntrol, are considered with some proposed alterations and a uniform evaluation of these control methods is made quantitatively with consideration for practical use. From the simulation results, a uniform evaluation is obtained by using a one-machine infinite-bus system which includes AVR and governor as the control object model.

View full abstract

A method for several-days-ahead prediction of inflow rate into a dam was investigated. We introduced a new type of variable sampling rate model which aggregated rainfall information with different time duration depending upon the characteristics of the inflow rate into the dam. As the rainfall information was appropriate-ly aggregated into the model, the proposed variable sampling rate model was able to express both short term characteristics and long term characteristics of the inflow rate into the dam. We constructed five different models depending upon the amount of the inflow rates and selectively used one model out of the five models in predicting the inflow rate into the dam. The proposed method was evaluated based on the actual inflow rates data and corresponding rainfall data during past several month at Midorikawa Dam. The proposed method gave satisfactory accuracy in the several-days-ahead prediction of the inflow rate into the dam. We further evaluated the proposed variable sampling rate model by comparing a conventional constant sampling rate model without data aggregation. The proposed model gave 20% reduction in the error of 8-days-ahead prediction of the inflow rates data. The variable sampling rate model was proved to be effective for the prediction problems in wide actual fields.

View full abstract

It is important to include the behavior of load in studies of loss of generation and consequent load shedding and also in studies of long-term dynamics of a power system. This paper describes field test results of load characteristics in power system by using normal operating data. The necessary resolution of data for estimate of load characteristics are investigated and measuring devices are made according to this conclusion. Field test clarified magnitude of fluctuations in system frequency, bus voltage and active and reactive power of load under normal operating condition. Load-frequency and load-voltage characteristics constants are estimated from field test data. Fluctuations of system frequency and voltage behind the equivalent impedance are clarified to be independent to fluctuations of active and reactive power of the major bus load under the state of nominal frequency and voltage. The presuppositions of the proposed method are verified to be sound.

View full abstract

A correct and rapid inference is required for practical use of expert system for fault section estimation in power systems. This paper proposes a new model based expert system for fault section estimation using colored and timed Petri nets. The Petri nets simplifies the inference procedure very much because it can represent the causalities among faults of power system components, protective actions of relays and circuit breakers and can also represent the operating states of these protective devices explicitly, or more precisely whether a relay has been actuated correctly or falsely etc. In addition, time sequence information on actuated relays and tripped circuit breakers can be handled by using the colored and timed Petri nets. Therefore, the operating states of relays and circuit breakers can be inferred more correctly. Petri nets are suited for parallel processing so well that there is a possibility of reducing the inference time considerably. In the proposed system, the inference is executed by setting the initial states in the Petri nets using the information on the actuated relays and the tripped circuit breakers and then operating the Petri nets. Several test cases have been tested using a prototype system on Macintosh computer using Object Common Lisp and good results have been obtained.

View full abstract

In order to obtain high efficiency of energy conversion in a wind turbine generator system, it is desirable that the magnitude of the turbine torque driving the generator is set to be proportional to the square of the rotor shaft speed. In a WTG system with constant magnetic field generator, this can be achieved by making the generator's output current be proportional to the square of the shaft speed. However, detailed characteristics of the wind turbine are requied in advance to determine the value of the proportionality coefficient.
This paper proposes the method which controls the proportionality coefficient and makes it match to the specific wind turbine under operation in natural wind field, without the preliminary knowledge of the turbine characteristics. The method can be applied to the systems with a propeller type wind turbine and a constant magnetic field DC generator.

View full abstract

This paper presents an efficient computational algorithm for evaluating the risks of uncertain factors in generation planning. Uncertainty in a multi-objective risk assessment problem can be divided into (a) uncertainty in possible conditions to be evaluated, (b) uncertainty in multi-objective decision making, and (c) uncertainty in risk which is obtained using imprecise information. The proposed method basically consists of two phases: (1) the aggregation of risks with respect to all the objectives for each uncertain factor, and (2) the linguistic expression of the aggregated risk of each uncertain factor. In the first phase, uncertainty in possible conditions is treated as fuzziness in planning parameters, and uncertainty in multi-objective decision making is also treated as fuzziness in decision making. Furthermore, the statistical trend of risk with respect to an objective for each uncertain factor can be extracted from uncertain risks obtained using imprecise information. In the method, both classes of fuzziness and the statistical trend of risk for each uncertain factor can be integrated into the risk of each uncertain factor, and then the risks of each uncertain factor with respect to all the objectives can be aggregated by extension principle in fuzzy sets theory. In the second phase based on the linguistic approximation technique, the aggregated risk of each uncertain factor can be represented by natural language. The proposed method can realize an effective and flexible decision support for evaluating the risks of uncertain factors in generation planning. The effectiveness and feasibility of the proposed method are demonstrated on a typical power system model.

View full abstract

Safety margin of internal insulation of power transformer windings should be selected to be the minimum necessary level by considering dielectric stresses not only under acceptance test but also under surge overvoltages of transmission systems. There is no practical way to measure the internal stress distributions of a transformer in service. However, such transient distributions can be simulated with a surge transient waveform measured in a substation and a transfer indicial response based on a measurement on the transformer under an untanked condition. This process consists in deconvolution of a measured response waveform by an excitation input waveform to obtain the transfer indicial response and convolution of the response with the surge waveform. The resulting waveform represents an estimated transient voltage which appears across the winding section when the surge will be applied to the transformer. A computer program for a workstation is developed and it is validated for its practical applicability through numerical processing examples of actually measured data at IREQ in Canada.

View full abstract

Measurement method of Insulation diagnosis for 22 kV cable has not yet been fully established. We measured electrical properties of the cables with different degradation grade by various methods. As a result, we found some desirable methods of insulation deterioration for future study. For DC leakage current and tanδ measurements, it is possible to analyze their effective data using these insulation diagnosis methods. Measurements of residual charge, loss current and transient current under DC superimposition enable us to detect the insulation degradation. Therefore it is necessary to gather a lot of data using these method for an effective insulation diagnosis. In this study, we also discussed the generating mechanism of residual charge and loss current, and influence of outer semiconductor layer in these methods.

View full abstract

Volt-ampere (V-I) characteristics are the most fundamental in metal oxide surge arrester modelin Although V-I characteristics have time domain properties of hysteresis and frequency dependency and such modeling research has been conducted, the residual voltage of an arrester has generally still been modelled only by the magnitude of discharge current, neglecting time domain properties.
Through the analysis of experimental results, the authors have found that energy dissipated during dischar reduces residual voltage and propose a new analytical model of residual voltage expessed by dissipation energy as well as discharge current as parameters. This model can be calculated in EMTP (Electro Magnet Transients Program) generally used in circuit analysis and well accounts for time domain properties above hysteresis and responses to steep wavefront current. It not only complies with the theory of conduction zinc-oxide element, but also quantifies the theory in a sense.

View full abstract

For the Insulation diagnosis of live cables, the method applying DC component current has been developed. The mechanisms for the evolution has been qualitatively explained on the behavior of space charges in XLPE. However, it has not yet been fully understood. Recently, it has been presented that DC component current could be measured owing to the presence of water trees closely linking between the inner and outer semicon-ducting layer. So that, we assumed that DC component current could arise from the composite mechanisms and investigated the phenomenon on the interface between the water and conductor or shielding layer by modeling the water-copper interface. In the copper-water-copper electrodes system, DC component current and the electromotive force due to the difference in the electrode potential were observed when AC voltage was applied. Moreover, from the X-ray analysis it was confirmed that Cu₂O (type p semiconductor) was produced on the surface of the electrode. When DC voltage was applied, weak polar effect probably due to the contribution of semiconductor or redox reaction was observed. From the results, it is considered that in the copper-water-copper electrodes system the electromotive force and the polar effect participate in the evolution of DC component current.

View full abstract