High power hydraulic units play a leading role in the safety of the power supply system; its safety, or the ability to stay in work is a high priority. Therefore, attention should be focused to the safety operation of hydraulic units, diagnostics and possible forecasting and determination of the technical condition. A novel approach offered in the article allows to extend sensor data application from the production cycle monitoring to the maintenance tasks. Legacy systems contain information regarding the whole production cycle and store working conditions information from all machines. The proposed methodology aims to bridge, with the power of data mining technics and machine learning. Within the framework of the developed methodology, the weighting coefficients of the parameters characterizing the technical condition of hydraulic units have been determined and their norms and evaluation criteria have been developed. A methodology for assessing the technical condition of high power, slow-rotating hydro units has been developed, which combines knowledge from legacy systems, and data analysis of an online sensor system. The proposed system extends the basic Condition Based Management - CBM functionalities with the integration of decision support systems technologies to enhance the interaction among humans and machines, improving the performance of the maintenance. A use case of Monitoring system for proactive maintenance of hydro-turbines is also discussed in this research.