Naveed Ahmed

Developing countries such as Pakistan require uninterrupted supply of electricity. Owing to the high energy demand in nearly every area of economy, the power transmission lines must be highly reliable, scalable, and less vulnerable to faults, reports WealthPK. According to Dr. Khurram Kamal of the Pakistan Navy Engineering College (PNEC), National University of Science and Technology (NUST), power plant failures may occur due to multiple factors, such as lightning strikes, damage to equipment at power generation stations, natural factors (such as trees falling on transmission lines), and human factors. 

Faults in power plants need to be detected, located, and classified quickly to ensure smooth operation. Therefore, artificial neural networks are considered to be a valuable tool when it comes to related applications of power systems. He said transmission, generation, and distribution are the main components of a power engineering system. At the same time, transmission lines are the medium for linking generation and distribution. Transmission lines are prone to numerous faults when exposed to the environment. 

“There are two main types of failures: symmetrical and asymmetric. A symmetrical fault occurs when the lines are connected simultaneously. We can call it a short-circuit situation. In these faults, phases and currents are equal. These are very severe faults and rarely occur in power systems,” Kamal said. He said in case of asymmetrical failure, the lines are unbalanced. These faults occur when two conductors come into contact, mainly triggered by swinging lines caused by wind and other factors.

These occurrences cause unbalance in the system, and the resistance values differ in each phase, causing an unbalanced current to flow in the circuit. Dr. Khurram explained that the transmission lines are considered the primary component of a power plant. It is crucial to locate faults accurately to repair them as soon as possible. These faults are more challenging to analyse. The quality of power supply will be affected by the time required to discover the fault. 

Owing to the system's nonlinearity, the artificial neural network has been verified to be an effective tool in the related applications of the power system. Compared to conventional methods, artificial neural networks have proved to be more robust and immune to disturbances by changing operating conditions in power systems.  He said using different optimization algorithms along with Artificial Neural Network (ANN) makes fast and accurate results achievable.

ANN can solve non-linear problems quickly, which is a hectic task for humans to translate in real-time. Its flexibility, programming techniques, and ability to solve problems rapidly make it one of the most widely used artificial intelligence techniques. The data-driven method of ANN is used in many industrial systems that involve input and output magnitudes of the system for detecting and classifying faults. 

He suggested that larger datasets could be easily organized and utilized by engineers nowadays using machine learning algorithms quickly. Machine learning mathematical models enable high-precision techniques to identify and classify faults in the transmission line of the power grid. An artificial Neural Network (ANN) is an intelligent system to identify and resolve the defects that commonly occur in a power plant network.

Credit : Independent News Pakistan-WealthPk