Predicting energy production with data analytics
SAS has the right people to understand these processes. Christophe Deleplanque VP Innovation SPG Dry Cooling
performance of installations predicted 24 h in advance
SPG Dry Cooling uses SAS solution on Air-Cooled Condensers to analyze performance of installations
An Air-Cooled Condenser (ACC) is an essential part of a thermal power plant. It condenses the steam at the end of a turbine and returns the condensate to the boiler, completing the steam cycle. The performance of this installation determines the energy production. SPG Dry Cooling, a major ACC manufacturer, collaborates with SAS to deploy advanced analytics in order to predict the outcome 24 hours in advance.
SPG Dry Cooling is a market-leading supplier of air-cooled condensers for power plants. It’s hard not to spot these enormous pipelined structures that have the size of a football field. In Belgium, for example, the Herdersbrug power station in Bruges operates with an SPG Dry Cooling installation. A typical power plant generates enough electricity to support 250.000 households.
SPG Dry Cooling has produced over a thousand dry cooling systems. Usually, huge volumes of water are required to condense steam, but this is not the case with dry cooling – as opposed to wet cooling. This method is particularly favored in areas that have a low water supply and high water costs or are subject to environmental restrictions.
SPG Dry Cooling employs about 800 people worldwide and has its headquarters in Brussels. The company collaborates with SAS to deploy advanced analytics and predict the performance of their installations 24 hours in advance. The location was certainly an asset, because SAS is also well-represented in the Belgian capital.
“During the development process, our SAS experts visited SPG Dry Cooling on multiple occasions”, says Kaat Tastenhoye, Customer Advisor at SAS. “SPG Dry Cooling needed a solution to predict and optimize the performance of their condensers and the associated steam cycle. The input of SPG Dry Cooling’s thermal engineers was essential. We also provided training during the first end-to-end implementations.”
Scalability is an important reason why we wanted an experienced partner. The advanced analytical competence of SAS is beyond dispute Frédéric Anthone Manager Aftermarket SPG Dry Cooling
An air-cooled condenser is part of a steam cycle. First, water is turned into steam inside a boiler. This steam powers the turbine that generates electricity. The steam then flows from the turbine exhaust into the ACC where condensation occurs – ventilators blow cooler air on the steam heat exchanger. Finally, the condensate returns to the boiler in a closed loop. As the steam at the end of the turbine is at a low pressure, the ACC works at a pressure close to a vacuum and non-condensable gases are removed continuously.
The performance of the air-cooled condenser determines the amount of energy produced, but the installation is subject to several factors such as ambient temperature and wind. When it’s cold outside, steam condenses much better and the energy production will increase. Strong lateral winds, however, act as an obstacle for the ventilators and thus affect the overall capacity.
For the end users of SPG Dry Cooling’s products, it would be very valuable to forecast the performance of the installation in all circumstances. Fortunately, lots of parameter data captured by sensors are available for analysis.
SPG Dry Cooling requested the assistance of SAS to build a digital twin of the installation. “Scalability is an important reason why we wanted an experienced partner. The advanced analytical competence of SAS is beyond dispute”, says Frédéric Anthone, Manager Aftermarket at SPG Dry Cooling. “The SPG Dry Cooling solution can be implemented on all types of dry cooling installations, which will lead to huge amounts of data and result in more precise predictions.”
In an ideal scenario, more than 4.000 ACCs worldwide could be connected and share data for analysis. The predictive power of this solution has multiple benefits. First, the operators of the power plant know the output of their installation 24 hours in advance. This way, power suppliers have a much better idea of the amount of electricity they can bring to the market. This also gives them the possibility to optimize the net plant heat rate.
Another great advantage is predicting when the system needs maintenance. Cleaning for example an ACC requires a lot of water and is very costly in areas with a low water supply. The energy output also increases after the cleaning process. As the model offers better insights about performance, operators can now explore the limits of the installations and delay maintenance as long as possible
SPG Dry Cooling – Facts & Figures
Valuable information available to improve future ACC designs
Reliability & Cost Saving
Thanks to optimized maintenance of installations
Power suppliers can better forecast amount of electricity they can bring to the market
SPG Dry Cooling also benefits from these forecasts. As they usually develop condensers for the constructor of power plants, and not directly for the operators themselves, feedback about the life-time performance of installations is very scarce. Now SPG Dry Cooling’s engineers receive valuable information to improve future ACC designs.
“The data analytics solutions provided by SAS open the door to a wide range of possibilities, including building on our large assets fleet to continuously optimize our ACC solutions even in the most extreme operational conditions”, says Christophe Deleplanque, VP Innovation at SPG Dry Cooling.
“The SAS solution has already shown its robustness”, says Christophe Deleplanque. “The biggest challenge was the communication between thermal engineers and data scientists. For the latter, the source of data doesn’t really make a difference. Data may come from banks, pharmaceutical companies, or in our case, air-cooled condensers. However, we wanted to offer our customers real added value. So we needed something that goes beyond pattern recognition. SAS has the right people to understand these processes.”
“On top of that, we haven’t reached the limits of advanced analytics yet. There is still much more potential in the SAS product. It’s a great long-term solution.”