Implementing a leak detection program is a complex task. However, regardless of approach, the first step toward improving leak detection in your operation is understanding your data. Not all data in your SCADA system is created equally. Plus, while the volume, quantity, and timeliness of data gathered is important, ultimately, it’s about the quality of your data.
The temptation is to build your system to gather as much data as possible. However, without focusing on data quality, it is difficult — if not impossible — for both the automated systems and controllers to identify and respond to abnormal operating conditions.
Pipeline operators need to leverage available technology to gather quality data at regular intervals and create real-time views of the data. Appropriate real-time views ensure that everyone in your operation understands the meaning and context of the data.
So, instead of asking how much data is enough to support your operation, ask these three questions to improve your approach to data acquisition for leak detection purposes:
- What are you trying to accomplish?
- What are the limitations of your SCADA infrastructure?
- How much data do you need to accomplish each critical function?
What Are You Trying to Accomplish with Your SCADA System?
When thinking about leak detection, your SCADA system needs to be set up with the capability to collect a lot of data, but the conversation is more than the amount of data to collect.
You need to look at your program and instruments to determine the accuracy and repeatability of those instruments, including how many bits of data you get back when the instruments are polled. Also, depending on the leak detection approach, accurate timestamping of the data is a key consideration.
This might mean that it’s better to grab more detailed, accurate data from the instruments versus collecting as much data as possible. There is a trade-off between those two elements, which points to the question, “What you are trying to accomplish?”
– The most common use of the data is in the control room to monitor pressures and flows to understand the operating condition. Additional normal data requirements include collecting safety-related alarm information, such as H2S, and measurement audit trail data used by the measurement group to perform their accounting and verification processes.
– You might also be passing this data to a computational pipeline model that’s taking the data, performing calculations, and using calculations to assess whether or not there is a leak in the pipeline.
– Other stakeholders could be the engineering department looking to historize the data, the financial group needing data for forecasting, and the facilities management system that needs to determine the level of ongoing support and service for the equipment.
Obviously, many people in your operation need to use the data. All of this goes into the conversation of how much data is needed and what is required to contextualize the data appropriately for each stakeholder’s specific use.
What Are the Limitations of Your SCADA Infrastructure?
Your operation does not have an unlimited budget for data collection. So, you need to be smart about having the right-size SCADA infrastructure to support your goals for collecting the data.
This step also requires looking at the system constraints and making sure that whatever you are doing to try to get the right amount of data is achieved in a comprehensive manner.
One central issue to consider is the actual communications capability that exists within your SCADA system. Why is this important? It will impact how often you collect data and what you can do with the data.
– Satellite communication: there is a natural delay between the time requesting and receiving the data. If you need to see the data change more quickly than a several second delay, then satellite might not be the right fit for your operation.
– Poll-Response Protocols: most legacy protocols are based on poll response where the host makes a request to a single remote and waits for a response from the end device before moving to the next device in the circuit. You do not get any data unless you initiate the request and limitations are created based on the number of devices on the circuit. The host might only receive data every 15 minutes because there is so much data being moved on the circuit.
– MQTT (publish/subscribe protocol): MQTT communication is driven by the remote, and data only moves to the host when there is a change. While MQTT has been around for a while, it is now a popular protocol in the Industrial Internet of Things (IIoT) Era because you can move a lot more data and move it a lot more quickly without bogging down the system.
Overall, your communications capability, budget, and use for the data will impact what you’re actually able to accomplish in your system.
How Much Data Do You Need to Accomplish Each Critical Function?
The last major consideration for the data used in leak detection is being able to support critical functions.
When you aim for very high accuracy and very high sensitivity in leak detection, the amount of data, the accuracy of the instruments, and the timestamping of all that data becomes very critical.
- When there is an alarm, you want the ability to analyze what’s happening at the relevant location.
- If there is an issue at a compressor station or around a pump, you want data from around the station — typically suction, discharge pressures, and machine operating condition.
In these scenarios, you want the data quickly, and you want the alarm to enunciate as soon as it moves outside of the desired operating range.
Building out these capabilities in your SCADA system will allow controllers and other personnel to achieve situational awareness by understanding what is causing the alarm and how to formulate the appropriate response.
EnerSys Provides Pipeline SCADA Consulting Support
Every pipeline operation faces unique challenges trying to collect the right amount of data in your SCADA system to improve leak detection.
Remember, instead of asking how much data is enough, ask what kind of data you need for what purpose. Then, ask how to supply that data within the configuration of your system to support critical functions.
You might need to tinker with a few elements of your system to get this correct. EnerSys is also equipped with pipeline SCADA consulting capabilities to support your operation.
Our company is rooted in telemetry, data communications, and SCADA support. We then applied our knowledge to develop industry-leading software and consultation services that help operators achieve excellence through high-performance pipeline control.