The rise of renewable energy sources forces hydropower plants to change often, wearing out important parts faster. To keep these plants safe and dependable, we need models that track and predict essential component health. Two primary approaches dominate the development of these degradation models: physics-based and traditional data-driven models. Physics models are grounded in domain knowledge of failure processes but lack a robust predictive element. Traditional data-driven models provide insights derived solely from sensor data and maintenance records. In response to the limitations, we propose a physics-informed data-driven approach to predict future degradation of essential electrical and mechanical components. It uses operational data sourced from the extensive Hydropower Research Institute (HRI) database, providing a well-informed level of confidence in prognostic predictions related to asset degradation and remaining operational life.