As power generation plants age, having flexible and efficient operating systems becomes even more critical to meet changing electrical power demands.
To help meet these demands, power generation plants require continuous production with minimal risk of interruption.
As a result, power generation stakeholders are developing online operational status maintenance strategies and procedures to prevent failures, ensure an optimized maintenance schedule, and help avoid economic and environmental impact.
The most effective plans combine two elements: they include online health monitoring strategies, such as trending factors, for example, major trips and alarms that are designed to address degradation, along with best practices. traditional to condition monitoring.
A power plant uses a large number of electric motors and rotating equipment ranging from hundreds to several thousand for critical and non-critical assets in a given power plant. Most motor and rotating equipment failures will manifest as vibration or excessive temperature.
Unforeseen equipment failures can cause damage, affecting plant uptime.
These critical and non-critical assets are vital to efficient and reliable operation and act as the leading indicators of a plant's effectiveness. Fortunately, rotating equipment and engine failures can be reduced through online condition monitoring predictive maintenance programs.
What is health monitoring?
In today's increasingly competitive global economy, a predictive maintenance strategy is needed to ensure reliable production and customer satisfaction, two of the most important aspects of running a successful operation.
Health monitoring is one aspect of predictive maintenance that provides all the information needed to make maintenance planning decisions. This involves comparing key metrics, such as vibration and power consumption, with baseline values for normal behavior to determine if there is any degradation of equipment health (See Figure 1).
It consists of collecting data, processing and analyzing signals to provide a complete picture of the machine's operating status.
The Electrical Power Research Institute has calculated comparative maintenance costs in dollars per horsepower (HP) for different maintenance strategies. According to research, a scheduled maintenance strategy is the most expensive, $24.00 per HP. A reactive maintenance strategy is the second most expensive, $17.00 per HP, but it can also be dangerous. A predictive maintenance strategy is the most cost-effective, at just $9.00 per HP, and reduces the risk of damage from catastrophic failure to secondary equipment and people.
Comparison of manual and automatic monitoring of the operating status
Traditionally, health monitoring is applied through routine rounds of manual diagnostics. However, trends such as the use of lower cost sensors and automated control systems and the emergence of Big Data analytics are driving the adoption of automated solutions. Applying online health monitoring to both critical (think turbines) and non-critical (such as compressors, pumps and fans, see Figure 2) assets in a given production environment provides the greatest visibility into the overall reliability of the asset fleet or plant, which helps companies to fully understand their operations and make business decisions.
For large and expensive equipment and rotating machinery, the cost of implementing an online health monitoring solution is easily justified.
• The most important benefit is increased revenue, which is due to maximum uptime and optimal efficiency of production machinery.
The correct operation of the machines guarantees maximum performance and by monitoring the production machines end users can also detect defects in the production of products based on the behavior of the machines, which reduces waste and the use of raw materials, while increasing the quality of the product.
• End users can also get cost reduction through such a system. Thanks to strategic repairs, the operation and maintenance costs of machines equipped with a health monitoring system can be significantly reduced.
The system can also identify failure developments early enough to properly schedule maintenance in case of planned shutdowns, avoiding costly plant shutdowns.
• By monitoring various performance parameters, the health monitoring system can actually help warn of imminent risk of failure and help prevent serious injury. Online monitoring systems also eliminate the need for workers to enter hazardous environments to take action.
• Workforce optimization – Manual diagnostic rounds can be extremely time-consuming and require significant travel and setup time, leaving less time for specialists to actually analyze the data and assess required maintenance. Additionally, many industries are reporting that qualified vibration and predictive maintenance experts are nearing retirement. Online health monitoring helps ensure dedicated personnel are spending maximum time on the highest value tasks.
• Fewer data gaps: By making manual rounds to collect data, company line operators can typically only collect a few measurements per month for each piece of machinery, at most. A typical power generation utility takes more than 60.000 measurements per month. In certain cases, line operators, manually taking note of data values, can make mistakes or even copy previous results. Online monitoring eliminates these errors and helps ensure continuous data collection.
• Improved Diagnostics: By using a single database, more baseline and historical trend data is available to predict failures with greater statistical significance. Furthermore, due to manual diagnosis, the interpretation of a failure is often based on the experience and knowledge of a specialist, and this experience can vary significantly from one specialist to another.
Considerations for the choice of a system for monitoring the state of operation
Before choosing a health monitoring system, it is important to understand what types of machines and faults need to be monitored. The extent and number of machines (eg monitoring of bearings, gears, motors and transformers with respect to simply critical turbines) and the types of measures needed to detect failures will form a basis for this decision. For example, in many cases, the use of fusion measurements from different types of sensors will provide a more accurate diagnosis.
Having identified these criteria, it is important to consider the following when choosing a vendor for a health monitoring solution (see Figure 3):
a) The flexibility of the solution to be scaled according to evolving needs, such as support for new types of algorithms, support for a wide variety of I/O and emerging sensors, and the ability to scale to a large number of systems.
b) Opening up the platform to allow access to raw engineering measurements and extend the solution to meet maintenance program requirements.
c) Interoperability with third-party hardware and software packages that allow integration with existing CMMS and ERP systems; as well as with any historical database or business process management software used.
d) The breadth and quality of the company's product offering, including the robustness of the hardware and the number of algorithms available.
e) The price of a monitoring hardware and software solution, including, if possible, the scaling of an online health monitoring solution to cover, for example, most rotating machine assets.
f) Services are offered to help facilitate an end-to-end solution from asset to IT infrastructure, either directly or through a network of partners.
When implementing a large-scale health monitoring system, there are three main technology considerations that come into play.
The first is data management, which involves the use of a proper data structure and reflection on the database for easy data mining, alarm capability, and application of an aging strategy. The second is data analytics, which includes application-specific algorithms and higher-level predictive forecasting or analytics.
It involves real-time decisions and embedded intelligence close to the sensor source, as well as performing data-at-rest analytics on servers using aggregated data from multiple machines.
As the number of data acquisition or monitoring systems increases, data management and analysis become increasingly complex and a third consideration becomes critical, systems management.
Remote administration of a large number of monitoring systems helps increase the reliability, usability and availability of the overall solution. For example, thanks to the NI Insight CMTM Enterprise software suite, end users can more efficiently perform tasks such as viewing the health of all systems, connecting to the network and acquiring accurate data, as well as remote configuration. of channels and uploading firmware application images to systems.
Ultimately, this software solution allows users to view and manage data and results, making remote management of a large number of monitoring systems simple.
