Process historians can be an integral part of the IIoT
Process historians are the first step up the automation pyramid beyond process controls and are the foundation for turning data into information. Process historians provide the data necessary for powerful data analysis tools to do their job. Following the current technological trends, they can even be moved out of datacenters and into the cloud.
Process historian essentials
At its most basic level a process historian is a database used to store time-series data from an industrial process. Data is collected either on a cyclical basis of usually 1 to 10 sec, or as data points change. Actual values, the associated timestamp and other information such as data integrity are written to the historian each time data is collected.
Once the data is in the historian, it can be accessed in a handful of ways. Many vendors offer data analysis packages that can be used in conjunction with the historical data for trending, reporting, and various database query capabilities. Most historians also can be queried directly using existing database systems, such as structured query language (SQL) or stored procedures. Other tools (big data analysis or machine learning algorithms) can be applied to the data to find interesting insights not easily obtainable through other methods.
One option available in the past few years is to move a process historian out of the plant and into the cloud using a hosted software as a service (SaaS). Some vendors offer historian as a service plan where a historian is configured on a remote machine and data access into and out of the historian is included with a subscription to the service. Historians also can be configured on cloud-based computing services that are managed by information technology (IT) departments.
Both options generally include things like service level agreements (SLAs) to define overall uptime and automated backup/recovery options. The one caveat to cloud-based historians is the increase of the "distance" between the process and the historical data, which requires additional networking resources as well as adding a potential hurdle to accessing the data in the event of intermittent network failure.
It is possible, though, to have the best of both options. This is especially true for companies that have remote systems installed all over the world. To make it work, though, local historians need to collect data from the system and feed that data to a supervisory level historian.
This provides all of the benefits of the cloud while eliminating the risk of network outages leading to data loss with local data storage that can be synced as network connections come back online. This approach is very popular with solar companies that are managing thousands of facilities around the world and need to provide data to their customers and power utilities with a minimal amount of interruption.
A historian by itself is not a tremendously useful tool. As described in Oil & Gas Engineering’s 5-part Intelligent Automation Series, the real power of a historian happens when the data is combined in the historian with analysis tools that turn the data into information.
The first tool, usually supplied with historians, is trending. Trends display time series data and the overall trajectory of data points in relation to one another. Most trend tools provide basic statistical analysis capabilities including averaging, integration, and range configurations to show when values have exceeded desired limits.
In addition to trending tools, many historians include the ability to query historical data using SQL tools. This opens the door for more advanced statistical analysis and allows historical data to be distributed to other programs using application programming interfaces (APIs). This makes getting process data into reports or Microsoft Excel an easy task for users.
Integrating IIoT with process historians
Historians can also be used to collect and store information from IIoT devices accessible through open platform communication (OPC) and MQ telemetry transport (MQTT) protocols and, using the various integration capabilities described above, can provide information back to IIoT devices.
One example of this type of integration using the API interface approach is in use by a company that manufactures golf course irrigation pump stations. Each system in the field has a local PC running a historian and a webserver. The webserver is configured with a reporting and trending interface accessible via browsers and mobile devices. It also has an API exposing historical data to third-party SaaS system.
The API can be queried with different parameters such as describing date ranges, specific data points to transmit, and login and security tokens to prevent unauthorized access to the system. The SaaS application integrates with other data sources including weather information, irrigation sensors, and GPS units on landscaping equipment to provide a cohesive monitoring system for all of the activities required to maintain a golf course.
Building the information pyramid
Using a process historian can be the first step towards a fully connected enterprise. Process data can be exposed to business systems, made available for statistical analysis, and integrated with third-party systems using APIs.
Some questions users should ask include: What data would be useful to collect? How can this data be made available to and consumed by other systems? What tools will be most useful to analyze the data today and a year from now as the data becomes an indispensable tool? What experience and ideas can IT and system integration partners bring to the table to build systems to provide context for the data? And finally, if a company isn’t using a process historian, why not start now?
Alex Marcy, P.E., is the owner and president of Corso Systems, a system integration firm based in Chicago, Ill. Edited by Chris Vavra, production editor, CFE Media, Control Engineering, firstname.lastname@example.org.
Process historians provide the data necessary for powerful data analysis tools to do their job.
A process historian is most valuable when the data is combined with analysis tools.
Process data can be made available for statistical analysis and integrated with third-party systems using APIs.
What applications could process historians serve and what benefits could they provide?
See additional stories from Marcy linked below.