Condition Based Monitoring (CBM) Using the AVEVA PI System

March 4, 2025 lamrconsultinginc@gmail.com No Comment 7 Views Share on

Recently, we successfully implemented a condition-based maintenance solution for a client using the AVEVA PI System. This innovative solution has transformed their asset management practices, improving visibility into generating assets and reducing the need for plant outages. Let’s dive into the details of their implementation and the benefits they’ve realized.

Business Challenge

The client faced several challenges in managing its renewable energy assets:

• Lack of visibility into generating assets
• Reliance on plant outages and routine maintenance to investigate asset degradation
• Decreasing outage flexibility due to market demand constraints

To address these issues, we set out to optimize resource usage through condition-based maintenance, leveraging the AVEVA PI System ecosystem.

Solution Implementation

Data Pipeline

We implemented a comprehensive data pipeline using AVEVA PI System components:

1. AVEVA PI System: Collects data points from assets and stores them in the PI Data Archive
2. AVEVA PI Asset Framework: Builds analytical models to format and contextualize data
3. AVEVA PI Vision: Constructs a platform to collate and present relevant information from various data sources

Solution Components:

1. Hydro Unit Stopping Sequence Analysis

Problem:

• Hydro generation units not stopping as expected, leading to forced outages
• Loss of potential generation during outage periods
• Uncertainty of root causes (e.g., wicket gate failure, degrading brake pads)

Approach:

• Utilized time-series data collected via AVEVA’s PI system with PI Asset Framework and PI Vision
• Created event frames for each unit’s stopping sequence
• Performed analytics on raw data to generate contextualized information
• Presented information using PI Vision

Outcome:

• Provided data points and visuals of all prior stopping sequences, offering insights into component-level conditions
• Created trends of unit stopping times to monitor the degradation of generating asset components over time

2. Hydro Unit Fatigue Monitoring

Problem:

• No indication of unit fatigue correlating to raw, real-time data
• Lack of visibility into hydro operating metrics
• Insufficient data to drive condition-based maintenance decisions

Approach:

• Utilized time-series data from AVEVA’s PI system with PI Asset Framework and PI Vision
• Performed expression analysis to gain insights into operating metrics such as:
  • Unit starts/stops
  • Tailwater Depression (TWD) operations
  • Station loading
  • Time within various generation ranges

Outcome:

• Highlighted the way Meridian Energy operates their generating assets
• Provided a foundation for future, complex analytics

Results and Benefits

The implementation of the AVEVA PI System ecosystem has delivered significant benefits to the client:

1. Increased Asset Transparency: Reduced the need for plant outages to perform analytics by improving visibility into generating assets.
2. Data Contextualization: Created processes to collect and contextualize information, delivering it to various business units in relevant formats.
3. Centralized Platform: Developed a centralized platform to present data at different depths, accommodating all levels of interest within the organization.
4. Improved Decision-Making: Provided insights into component-level conditions and asset degradation trends, enabling more informed maintenance decisions.
5. Operational Efficiency: Enhanced the ability to monitor and analyze unit performance, leading to optimized resource usage and improved asset health.

Future Work

We will not stop at the aforementioned achievements. We have identified several areas for future development:

1. Overload Analysis: Investigating the impact of overload conditions on revenue, maintenance frequency, and asset health.
2. Operational Vibration Analysis: Implementing advanced monitoring and alarming systems for vibration-related issues.
3. Component Condition Heat Map: Developing a visual representation of asset health across various components.
4. Fatigue Metric: Creating a single metric to describe unit fatigue, enabling easier comparison and decision-making.
5. Integration Enhancements:
   • Establishing a bi-directional channel between the PI System and work management system
   • Managing alarms and notifications through Asset Framework
   • Integrating PI Web API into their data pipeline

Conclusion

Our implementation of condition-based maintenance using the AVEVA PI System demonstrates the power of data-driven asset management in the renewable energy sector. By leveraging advanced analytics and visualization tools, they have significantly improved their ability to monitor, analyze, and maintain their generating assets.

This project serves as an excellent example for other energy companies looking to optimize their maintenance practices and improve operational efficiency. As we continue to build on this foundation with more complex analytics and integrations, our client is well-positioned to maintain their leadership among its peers.