BKW Energie in Europe owns or partially owns 91 hydropower plants with a total capacity of 1,604 MW. Some of its plants in Switzerland are more than 100 years old and were refurbished several decades ago. Because of their age, some of these facilities are about to reach the end of their original design life.
By Moritz von Plate
BKW used the automated Prognostic solution from Cassantec AG to determine the remaining useful life of several of its hydropower plants and key components that include turbines, generators and transformers. The solution helped BKW appreciably extend the remaining useful life of its hydropower assets by minimizing maintenance costs and scope, and relaxing the maintenance schedule. As a result, the utility benefits from an informed and targeted asset procurement and replacement schedule.
BKW is located in the canton of Berne, Switzerland. According to the company, hydroelectric power generation accounts for about 40% of its business. The company owns shares in hydroelectric and other types of energy facilities located in France, Germany and Italy, but the majority of its holdings are in Switzerland. In Berne alone, BKW owns and operates seven run-of-river hydropower plants:
- 15.2 MW Aarberg;
- 28.5 MW Bannwil;
- 8.3 MW Kallnach;
- 19 MW Kandergrund;
- 44.6 MW Muhleberg;
- 15 MW Niederried; and
- 18.7 MW Spiez.
Based on their age, BKW decided it needed a means to determine the remaining useful life of these power plants and key components in all of these facilities (i.e., turbine-generator units, transformers and control mechanisms) in an effort to provide guidance regarding steps that could be taken to extend their operating life.
The Bannwill facility, on the River Aare, was commissioned in 1970 and underwent a major rehabilitation in the late 1990s. The powerhouse has three turbines, generators and transformers. In April 2014, BKW contracted Cassantec AG to determine the optimal renewal time for plant components by installing Cassantec’s Prognostic solution.
At the heart of Prognostics lies a stochastic model. The empirical data compiled through condition monitoring tools are used and applied to the automated stochastic model. Optimizing maintenance decisions results from prognostic reports that provide the following information:
- Explicit future time horizon;
- Consolidating and prioritizing condition data; and
- Condition forecasts and related malfunction prognoses.
Additionally, the solution offers a comprehensive remaining useful life-distribution report, as well as an automated periodic result update.
Prognostics complements the alarm and diagnostic information given by condition monitoring and predictive analytics. The resulting foresight can be used for both the mid- and long-term maintenance planning as well as the overall fleet operations strategy.
Configuring the solution necessary to generate prognoses is efficient and requires minimal effort. Once the critical pieces of equipment have been selected, relevant malfunctions have been defined and suitable condition parameters have been formulated, the configuration is complete and prognostic reports can be generated for equipment onsite.
By using existing data history from the database, accurate prognoses can be compiled the moment the configuration is completed. Additionally, a machine learning mechanism built into the technology increases the precision of prognoses over time.
|The 28.5 MW Bannwill facility, on the River Aare, was commissioned in 1970 and underwent a major rehabilitation in the late 1990s.|
In the first step, current and historical vibration, temperature, pressure, flow and electrical condition data from BKW’s hydropower stations are used to forecast asset conditions over an explicit future time horizon. In the second step, these asset conditions are correlated to critical asset malfunction modes, formulated in close collaboration with BKW’s engineers onsite. In the third and final step, future malfunction risk with respect to these malfunction modes is computed and converted to a prognostic report.
By knowing what limits remaining useful life, the limiting factors can be actively managed, thus extending remaining useful life. BKW required that the Cassantec system be able to provide data on when to schedule maintenance. This requirement was part of the preparation for BKW’s operational strategy. One of the desired outcomes included lowering Bannwil’s costs for operation and maintenance.
BKW provided Cassantec with data on temperature, lubricants, vibrations, rotation speed and electrical data such as current and voltage, that were recorded and stored during Bannwil plant operations.
The computerized stochastic model analyzes, aggregates and correlates these data and develops malfunction modes, or life limiting factors. The resulting output, called remaining useful life distributions, can be aggregated at plant level and updated in desired time intervals (i.e., daily, weekly or monthly).
Cassantec presents the prognostic reports for the plant and all of its listed individual components in a secure online portal. BKW exports the data updates from their database and transfers them to Cassantec through a SharePoint server. Cassantec then calculates the prognoses via a cloud-based server system and renders the reports through a browser-based portal that is easily accessible to the client.
In 2014, after configuration, the analysis indicated Bannwil’s remaining useful life was being dictated by Generator 3. Increase in vibration displacement indicated a progressing defect in the generator, thereby limiting the generator’s remaining useful life beyond acceptable levels. An analysis of the vibration parameters showed that the generator’s remaining useful life is dependent on the respective load scenario. Under full-load conditions, the generator deterioration is faster than under part-load conditions.
From this information, BKW arrived at an operating strategy that limits the operating capacity of Generator 3 to roughly 80% of its capacity and, if necessary, would increase the operating capacity of Generators 1 and 2. That way, the vibration gradients from Generator 3 sufficiently lessened. Thus, through making the appropriate adjustments at Bannwil, BKW’s operating strategy has extended the entire plant’s remaining useful life.
“The Prognostic Solutions reports have improved our daily plant management and long-term planning,” said David Rhyner, BKW asset manager. “We can see what affect our decisions today will have on the entire plant operation. Thereby we expect to reduce our costs and to improve, today, the reliability of tomorrow.”
An update of the prognostic reports from Bannwil is sent monthly. As a result of successfully implementing Cassantec’s system at Bannwil, the utility is rolling out the product to Kallnach.
Moritz von Plate is chief executive officer of Cassantec AG.