
District Cooling Plant Optimization using Machine Learning
The Zabeel District Cooling Plant serves chilled water to the DIFC (Dubai International Financial Center) chilled water network. The same network is served by one more District Cooling Plant. The Zabeel Plant is equipped with 3 number of Mitsubishi GART-350PL Chiller Pairs, each chiller pair has a capacity of 6250 Refrigeration Tons. The upstream chiller has 3350 Tons and the downstream chiller has 2900 Tons. The Plant has 3 Cooling towers, 4 number of Condenser Pumps on 3 duty with 1 standby arrangement, 4 Primary pumps on 3 duty with 1 standby arrangement and 5 Secondary chilled water pumps. The plant has 2 numbers of Thermal Energy Storage Tanks (TES). The objective of this project is to optimize the plant performance by taking advantage of the part load conditions with the use of Machine Learning (ML) techniques to enhance the efficiency of the plant. The plant has been operational since December 2021, and hence, a large amount of process data was available from the data archives. The plan was to demonstrate the savings (June 2024 to May 2025) with the Energy consumption of the previous year (June 2023 to May 2024). This is yet another project under Empower's ML/AI based optimization suite EmSMART.
