DEVELOPMENT OF HIGH AVAILABILITY ALGORITHM FOR OPTIMIZING SCADA COMMUNICATION IN SUBSTATION AUTOMATION SYSTEM BASED ON REAL-TIME DATA MINING ANALYSIS

In the era of energy digitalization, SCADA systems play a crucial role in monitoring and controlling modern electrical infrastructure. However, PLN UP2B Kaltimra faces challenges with inconsistent High availability data in substations, causing delays in measurements, status updates, and control o...

Full description

Saved in:
Bibliographic Details
Main Author: Yolandanu, Satrio
Format: Theses
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/86887
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:In the era of energy digitalization, SCADA systems play a crucial role in monitoring and controlling modern electrical infrastructure. However, PLN UP2B Kaltimra faces challenges with inconsistent High availability data in substations, causing delays in measurements, status updates, and control operations, which hinder optimal performance. To address these challenges, this study aims to develop a High availability algorithm based on data velocity and veracity analysis. The testing was conducted using a Personal Computer (PC) or Virtual Machine (VM) equipped with Python and communication libraries in the Front-End network. Measurement data and equipment statuses were collected through Real-Time Data Mining, and the CRISP-DM methodology was applied to analyze measurement data and equipment statuses, identify delay factors, and develop algorithmic solutions. The analysis revealed that data availability instability is often caused by repeated data retrieval from a single sensor, resulting in congestion, delays, or data disconnections. Further investigation into network topology is recommended. Based on these findings, a high-availability algorithm was designed and tested in the SCADA system. The test results showed that the developed autorestart algorithm successfully increased communication data availability by 16.5%, reducing initial failure rates from 24% at one substation to only 7.5%. The average communication latency was also reduced from 0.12 seconds to 0.05 seconds. In the implementation of the algorithm with Simulated Annealing, data collection processes that previously showed a 25.33% failure rate became more stable, with the error rate reduced to 1% on the tested devices. This research contributes to improving the efficiency and reliability of SCADA at PLN UP2B Kaltimra, offering a relevant solution to address communication network challenges in modern electrical systems.