**2. LabVIEW role**

frequency with permissible range(1%) [3, 4]. Based those three attributes, any deviation from the given permissible limit is known as power disturbances. The power disturbances consist of different type of disturbances namely, voltages sags, swell, interruptions, transient over voltages, harmonics, and voltage imbalance. The

Referring to **Table 1**, the type of disturbances has its own unique characteristics.

definitions of each type power disturbances can be described as follow [5]:

*LabVIEW - A Flexible Environment for Modeling and Daily Laboratory Use*

Therefore, there are a paramount important to classify the type disturbance at initial stage of the PQ solution step. The classification of power disturbance is occasionally incorporating with signal processing technique which functioning as feature extraction tools. The popular signal processing technique that often used in extracting the feature are Short time Fourier Transform (STFT), Wavelet Transform (WT), S-Transform (ST), Hilbert Transform (HT), Hilbert Huang Transform (HHT), and Ensemble Empirical mode Decomposition (EEMD) [6–11]. STFT had been widely used as a tool in recognizing disturbances. Nevertheless, STFT has a limitation where it's only has a fixed windowing technique which is not fit for non-

stationary signal [12]. The improved WT has been introduced to enable the nonstationary signal processing approach in continuous windows size by using multiresolution analysis (MRA) [13]. However, this technique suffers from high losses of information during MRA process [14]. On top of that, the WT has disad-

S-transform (ST) is proposed to carry out the solution which experienced by STFT and WT [15, 16]. ST is superior signal processing tool, where its technique is improved and outcome the limitation that exist in STFT and WT. ST manage to provide the distinct features which facilitate the decision making tools i.e., Artificial Intelligent (AI) to produce the accurate classification result accuracy. From the literature studies, there are several types of AI that frequently used as a decision making mechanism for classifying power disturbances, such as the Support Vector Machine (SVM), Levenberg Marquardt Neural Network (LMNN), Probability Neural Network (PNN), General Regression Neural Network (GRNN), and Radial Basic Function Neural Network (RBFNN) [17–21]. However, due to the less computational burden and resilience performance, the SVM has been chosen as a decision-making tool to be used in this work. Both ST and SVM are embedded to

Voltage sag Temporary voltage reduction in magnitude between 10% and 90% of nominal voltage in RMS for duration few cycles to one minute Interruptions Interruptions can be defined as 0.9 pu reductions in voltage magnitude for

Swell An increase of RMS voltage from 0.1 pu to 0.8 pu for duration 0.5 cycle to

Harmonics A Sinusoidal voltages or currents which having frequencies that are integer

Interruptions The disappearance of the supply voltage on one or more phases. It is usually

Voltage imbalance The ratio of the negative sequence component to the positive sequence

multiple of the frequency at which supply system is designed to operate.

qualified by an additional term indicating the duration of the interruption

Sudden, non-power frequency change in the steady-state condition of voltage,

vantage where it fails to extract time-frequency information.

**Characteristic**

1 minute.

current, or both,

*The definition of each type power disturbance.*

duration less than 1 minute.

component which exceed 0.5%

**Type of power disturbance**

Transient over voltages

**Table 1.**

**144**

This research is conducted in LabVIEW software with the aid of National Instrument (NI) data acquisition module, where the entire data acquisition module, analysis module, features extraction process and data representation are developed and executed using this tools [23]. Along with enormous flexibility in interfacing and working with various type of real time input signal from different type of measurement tools, LabVIEW is offering a superiority analysis with a variety choice of programming tools that is appropriate to be implemented in real time power quality monitoring system. The flexibility of LabVIEW software also grants an excellent output result of the projects with lesser computational time. This develops system from this research basically is able to extract features and classify the PQ disturbance as illustrated in **Figure 1**.

Based on **Figure 1**, the developed system starts with the accumulating of PQ disturbances by using the PQ recorder i.e. Fluke 1750. Then those PQ disturbances are transferred to Chroma 61701 for storing and regenerating purpose. Then a National instruments Compaq Data Acquisition Chassis (NI cDAQ-9184) with NI 9225 card is used for high-speed acquisition of PQ disturbances. Further, the types of PQ disturbances are analyzed in LabVIEW interfacing since the voltage signals

**Figure 1.**

*System architecture of PQ analysis using LabVIEW software.*

which acquired from the NI 9225 card. Further work of classification process which based on ST and SVM are developed at LabVIEW platform.
