Identification and classification of voltage and current disturbances in power systems are important tasks in the monitoring and protection of power system. Most power quality disturbances are non-stationary and transitory and the detection and classification have proved to be very demanding. The concept of discrete wavelet transform for feature extraction of power disturbance signal combined with artificial neural network and fuzzy logic incorporated as a powerful tool for detecting and classifying power quality problems. This paper employes a different type of univariate randomly optimized neural network combined with discrete wavelet transform and fuzzy logic to have a better power quality disturbance classification accuracy. The disturbances of interest include sag, swell, transient, fluctuation, and interruption. The system is modeled using VHSIC Hardware Description Language (VHDL), a hardware description language, followed by extensive testing and simulation to verify the functionality of the system that allows efficient hardware implementation of the same. This proposed method classifies, and achieves 98.19% classification accuracy for the application of this system on software-generated signals and utility sampled disturbance events.

No abstract

The spline interpolation using the sampling bases has a feature of real‐time computation which is not observed in other methods since the result of interpolation is derived directly from the sampled value sequence. Then it is necessary that the sampling function should be truncated in a finite interval. This paper aims to clarify the relation in this method between the error in the result of interpolation due to the truncation and the length of the truncation interval. The result is summarizes in a correspondence table for the truncation interval length and the error. In the proposed interpolation method, the truncation interval length to ensure the given tolerable range of error can be determined using the correspondence table.