Articles

Conceptualization of Markov Processes in Cyber-Physical Systems: Modelling, Prediction, and Control

/in /by

Cyber-Physical Systems (CPS) bring together physical processes with computing, communication, and control. They often operate in environments full of uncertainty, noise, and constant change, which makes traditional deterministic models struggle to capture how these systems really behave. This work introduces a more flexible framework based on Markov processes that helps model, predict, and control CPS in a more realistic way. By viewing system behaviour as probabilistic transitions between states, it becomes easier to analyze uncertainty and understand how the system evolves over time. The study looks at discrete-time Markov chains and expands the discussion to Hidden Markov Models (HMMs) and Markov Decision Processes (MDPs), allowing both visible and hidden aspects of system dynamics to be represented. It outlines a well-defined process for defining states, calculating transition probabilities, and making forecasts. The paper explores, in addition to that, the use of control techniques based on the use of probability theory and shows that these methods have a greater level of robustness compared to traditional control techniques. An example is given to show how this model improves performance and flexibility. All in all, Markov modelling is a good starting point for dealing with the challenges in CPSs, paving the way for integration with other tools.

Analysis of Technical Electrical Issues and Energy Losses in a 20/0.4 kV Power Distribution Network: A Case Study of Sheberghan City

/in /by

Power distribution systems in developing countries often face significant technical and non-technical losses, which reduce efficiency and reliability. This study focuses on the 20/0.4 kV distribution network of Sheberghan City, Afghanistan, aiming to identify key sources of energy loss and propose practical solutions for improving system performance. A mixed-methods approach was adopted, combining quantitative field measurements with qualitative insights from utility staff. Data were collected through direct site visits, operational records from the Sheberghan Electricity Department, and structured interviews with engineers. Energy losses were calculated for each feeder and transformer by considering conductor types, network topology, load patterns, voltage drops, and seasonal variations. The total annual energy loss was found to be 3,012,381.5 kWh, accounting for approximately 10.19% of the system’s total energy input, equivalent to a financial loss of over 18 million AFN annually. Voltage drop analysis revealed an average loss of 1.2643 kV in the 20 kV lines and up to 16 V in the 0.4 kV branches. Technical losses were mainly caused by conductor resistance, transformer overloading, lack of reactive power compensation, and outdated infrastructure. Non-technical losses included billing inaccuracies, illegal connections, and human errors during operations. This study highlights the urgent need for modernization in the Sheberghan distribution system. Implementing smart grid technologies, upgrading transformers, installing capacitor banks and voltage regulators, and adopting distributed generation strategies can significantly reduce losses and improve reliability. Additionally, training programs for technical staff and stricter enforcement of operational standards are essential to mitigate human-induced inefficiencies. The findings provide actionable insights for other developing regions seeking to enhance the efficiency and resilience of their medium- and low-voltage distribution networks.