In the golf “industry”, when you play a round, there’s inevitably a water hazard to contend with. Well, in the water industry, when implementing the PI System, we have a “golf hazard”. When starting out, one has a very poor handicap. Very few people are familiar with the system and a lot of mistakes can be made. However, if you engage an expert (golf pro), someone who knows how to play the game, how to hold the club, how to swing the software, you can setup the correct foundations – before you the swing the club. And for the more experienced pundits, you will know that the correct follow-through is just as important as the swing; so training, mentoring and continual improvement is essential to successful implementation of a PI system in your organisation.

 

This paper presents TasWater’s journey in implementing the PI System to their organisation. The trials, tribulations and learnings they’ve experienced in this rapid rollout; including the roll that analytics have played (and the lack thereof). This journey continues as TasWater integrates more data into their PI System, whilst simultaneously continuing with training, mentoring and continual improvement.

In this session we will cover what OSIsoft has been working on for the PI System with new releases and technology (including some live demos!). We aim with every release to make the PI System as robust and secure as possible to meet customers’ data infrastructure needs and also make sure users have easy access to the data they need to perform their jobs. We will also touch on what is currently in the product pipeline.

Integrating OT and IT data can present challenges however the insight and value it brings to operations and analytics are worth the effort. OSIsoft and the PI System, as an Enterprise Data Infrastructure solution, can offer ways of organising and contextualising operational data so it can be combined with business data for advanced analytics (e.g. machine learning and AI, pattern recognition, big data analytics). We will have a look at some of the technology and also an approach to reduce the uncertainty in this convergence.

Modern wind turbines are highly automated machines which generate enormous volumes of data on every aspect of their operation. The potential uses of this tide of data are endless; the challenge is in extracting actionable insight to drive improved turbine performance.   A great visualisation can summarise a vast amount of data from a wind farm and highlight patterns or insights which are otherwise hidden.  Two methods that AGL has used to identify outliers are:
1.    Training and tuning of advanced pattern recognition models
2.    Building custom visualisations to compare performance across similar assets

As an innovative business, SA Water are determined to lead the way, as such we are desperate to invest in advanced analytics, complex machine learning and progressive cloud computing. However, if our determination and desperation leads to incorrect methodology and poor techniques, we will fail. We must learn to walk before we can run, perhaps even crawl, in order to obtain the right business outcomes. 

Understanding the data we have and the relationships between different data sets will enable effective data consumption, trust in the data and governance into where the data belongs.  Through this we can develop solid fundamentals which will allow us to ‘walk’ and expand our techniques to finally allow us to run and lead the way.

TransGrid operates and manages the high voltage electricity transmission network in NSW and ACT, a system comprising of 99 substations and power station switchyards and over 12,000 kilometres of transmission lines operating at up to 500kV.  The Asset Monitoring team monitor the condition of all assets and manage the operation and maintenance of the assets which was difficult with data stored in a dozen different systems.

 

The Asset Management Information Portal (AMIP) brings together real-time data, asset specifications, maintenance records, manual readings and many other data sources into a single data model in the PI System; and provides a single point of reference offering tactical information to users through intuitive PI Vision and ArcGIS dashboards; as well as providing a solid foundation for the future development of analyses.  AMIP gets situational awareness where it’s needed fast, shortening issue resolution time and creating efficiency.