Moving from planned to predictive maintenance20 August 1998
The installation of on-line, high performance turbine condition monitoring has enabled several Italian power plants to switch from planned to predictive maintenance.
The power generation systems at the Centrale di Moncalieri facility comprise a 132 MWe steam turbine and a 35 MWe gas turbine. The steam turbine generates 388 GWh annually for Turin, making it vital to the performance of the local power utility, Azienda Energetica Municipale of Turin (AEM).
Ansaldo Energia is supplying AEM with two Monsor systems – based on Solartron's 1051 on-line condition monitoring system and augmented by Ansaldo – for turbine, pump and auxiliary equipment monitoring. These will contribute to the utility's programme to upgrade Moncalieri as part of a city-wide cogeneration and district heating project. AEM's decision to install Monsor systems was based on experiences at similar plants in Italy, such as Ponti sul Mincio.
Condition monitoring minimises forced outages, increasing the interval between consecutive overhauls, minimising time for an overhaul, and optimising plant performance according to design. Importantly, it provides a method to improve residual life assessment, plant rehabilitation, and ultimately, plant life extension.
Initially, Italian power plant managers were hesitant to accept some of the more advanced facilities that have become available in recent years, but the increasing potential profitability of these features was becoming too significant to ignore.
Co-operation between Ansaldo and Solartron started with the development of the first Monsor system for monitoring and supervising shaft vibrations of the 320 MWe unit 2 at the ENEL Torvaldaliga Sud power plant. The system was commissioned by Ansaldo in 1996. At the same time, a second Monsor system was installed on another 320 MWe generator unit at the Fusina power plant.
Ansaldo's Monsor is a modular condition assessment system based on Solartron's 1051. The system offers dynamic and static measurement combined with analysis facilities during turbine run-up, coast-down and on-load, and the capability to build a machine life-cycle database to simplify fault diagnosis. Signal and data processing power can be distributed throughout the system to maximise the speed of producing results, while the software offers a machine-state sensitive graphical interface to improve the productivity of plant operators and machinery analysts.
Following implementations at the Torvaldaliga Sud and Fusina power plants, the operators of the Ponti sul Mincio power plant decided to install a Monsor system as part of a programme to extend the operating life of its 80 and 160 MWe units. The combined oil- and gas-fuelled power station at Ponti sul Mincio, near Verona, is a build-own-operate plant for the municipal authorities of Brescia, Verona, Rovereto and Vicenza. These authorities also provide transmission and distribution through the local grid, which is connected to the national grid.
Ponti sul Mincio station was commissioned in 1966 with a 80 MWe turbogenerator. The capacity was increased in 1983 with the installation of the 160 MWe unit. Both units were originally supplied and installed by Ansaldo Energia.
Prior to installing the Monsor system, maintenance was based on planned periodic inspection. In the event of forced outages, the broken or failed parts were replaced without any deep investigation into the origin of the failure. Monsor replaced the plant's paper-based vibration monitoring system – typical of small-scale power producers – with an on-line condition monitoring system, improving maintenance and overhaul planning, and helping to maximise turbine availability.
At Ponti sul Mincio, each bearing is equipped with two orthogonal mounted probes. Each probe supplies two signals proportional to displacement: relative shaft support movement and absolute support movement. Distributed high-performance data acquisition front-ends collect probe data and provide local processing before transferring it to the control room computer.
To simplify operation two distinct presentation environments for results are provided: the operator environment and the specialist environment. The operator environment is designed to support the day-to-day tasks of the plant unit control room operator without overwhelming him with data. Information is presented in simple pre-configured formats with customised data content, continuously updated in real-time. The specialist environment supports the vibration specialist, allowing him to view current and historical data, perform comparisons of data acquired at different times, and correlate data variations to enable trends to be predicted and the causes of incidents to be investigated.
Installation of the Ponti sul Mincio Monsor system has been fully justified. By using the system, the power plant operators have moved from planned to predictive, condition-based, maintenance, extending the period between major shut-downs from two to three years. This alone has paid for the system. In addition, two incidents have been managed by the system, further justifying its installation.
When restarting the 160 MWe unit during a minor overhaul, and again following a forced outage, high vibration levels in excess of 200 micron peak-to-peak were detected in bearing number 1. Investigation identified an oil-instability phenomenon due to wear in the babbit. At the following major overhaul, the bearing was restored.
Following a major overhaul of the 80 MWe unit, high vibration levels were detected at speeds over 1500 rpm. Analysis using the Monsor system ruled out any possibility of sliding, imbalance or bearing wear. Investigation showed the cause to be rotational acceleration of over 100 rpm/minute when the machine was not thermally stabilised.
These benefits encouraged the operators of Centrale di Moncalieri to progress from planned maintenance by adopting predictive, condition-based, maintenance.
The generation systems at Moncalieri comprise a 132 MWe De Pretto steam turbine and a Fiat TG20 35 MWe gas turbine. The steam turbine was installed in 1966, and the Monsor system, replacing a paper-based equipment monitoring system, contributes to AEM's programme to extend this turbine's operating life.
AEM installed the 35 MWe gas turbine in 1975 to meet peak power demands; this turbine is capable of a fast run-up, from 0 to 5000 rpm in less than four minutes. Solartron recently extended the acquisition performance of its 1051 system's front-end measurement modules, enabling the Monsor system to maintain lock at very high accelerations and capture and process vibration data in real time for this type of application.
The Moncalieri generators are being integrated into a large-scale district heating project, called Torino Sud, which will provide heating for 200 000 Turin residents. Integral to the project is the construction of a thermal ring to link AEM's other district heating plants, and collect waste heat from large industrial complexes inside the city. To deliver the hot water, AEM is installing 900 kW pumps. Unlike conventional turbine condition monitoring systems, Monsor can monitor reciprocating and oscillatory equipment such as pumps, and allow vibration specialists to call up comprehensive multi-plots to enable turbine and auxiliary equipment activity to be compared quickly – a proven technique for revealing impending faults at an early stage.
The change among Italian power plant operators from planned maintenance to a predictive, condition-based, maintenance strategy is gaining momentum. For such a strategy to succeed, an extensive and accurate measuring system is required, along with experience and expertise in interpreting the data gathered.
To speed the adoption of predictive maintenance, Ansaldo provides a vibration analysis service to Monsor system users in the form of on-line data interpretation and advice. Real time and historical data files can be transferred by high-speed communications link from each Monsor system to Ansaldo's own system, installed at the company's Genoa headquarters. Working in parallel with the power plant's on-site vibration analysts, and using the same software, problems can be quickly identified and resolved.
|Moncalieri turbine data acquisition information|
Modern Power Systems is a product of Progressive Media International. Registered office: 40-42 Hatton Garden, London, England, EC1N 8EB. © 2023, All Rights Reserved.