Condition monitoring: Automated solutions

•  Workforce optimisation: Manual diagnostic rounds can be time-consuming and require significant travel and setup time, leaving less time for specialists to actually analyse data and assess required maintenance. In addition, many industries are reporting that predictive maintenance and vibration specialists are nearing retirement. Online condition monitoring helps ensure that specialised personnel are spending maximal time on the highest value tasks.

•  Fewer gaps in data: When performing manual rounds to collect data, companies typically can only collect a few measurements for any given piece of machinery a month, if at all. A typical power generation utility takes over 60,000 measurements per month. In certain cases, line operators manually noting data values can make mistakes or even copy previous results. Online monitoring removes these errors and helps ensure continuous data collection.

Improved diagnostics: By using a single database, more historical trend and baseline data is available for predicting faults with greater statistical significance. In addition, with manual diagnostics, the interpretation of a fault is often based on the experience and knowledge of a specialist. This can differ significantly from one specialist to another.

System selection

Before choosing a condition monitoring system, it is important to understand what types of machines and which failure modes need to be monitored. Once this is identified, consider the following when choosing a vendor:

•  The flexibility of the solution to scale with your evolving needs, such as support for new types of algorithms, support for a wide variety of I/O and emerging sensors, and ability to scale to large numbers of systems.

•  The openness of the platform to allow you to gain access to the raw engineering measurements and extend the solution to meet your maintenance program requirements.

•  Interoperability with third-party hardware and software packages so that you can integrate with existing CMMS and ERP systems and any database historians or process management enterprise software used.

•  The breadth and quality of the company’s product offering.

•  The price of the monitoring hardware and the software solution, and how well it allows you to scale your online condition monitoring solution to cover the bulk of your rotating machinery assets.

•  The services offered to help facilitate your end-to-end solution from your asset to your IT infrastructure, either directly or through a network of partners.

When implementing a large-scale condition monitoring system, there are three main technology considerations that come into play. The first is data management, which involves using an appropriate data structure, database considerations for easily mining data, alarming capabilities, and implementing an aging strategy.

The second is data analytics, which includes application-specific algorithms and higher-level predictive analytics or prognostics. It involves both real-time decisions and embedded intelligence closer to the sensor source and performing at-rest data analytics on servers using aggregated data from multiple machines.

As the number of data acquisition or monitoring systems increases, data management and data analytics become more complex and a third consideration becomes increasingly important: systems management. Remotely managing large numbers of monitoring systems helps to increase reliability, serviceability, and availability of the overall solution, so that you can perform tasks such as viewing the health of all of your systems, connecting to the network and acquiring accurate data, and remotely configuring channels and uploading firmware application images to your systems.

National Instruments' NI InsightCM Enterprise is a ready-to-run software solution with tightly integrated and flexible hardware options for online condition monitoring that allows companies to gain insight into operations and maintenance programs and understand the health of rotating machine assets.

Kamalina Srikant is responsible for product management and market development for condition monitoring solutions at NI.