Process Historians can sound confusing, but necessary for collecting and tracking historical data. There are three major approaches to Process Historians: Built-in Trending with an HMI, a Database-backed storage option, and a Full-Blown Process Historian. We’ll go through each of the options in depth. Plus you can check out the video below for the highlights. An example of one of Corso’s Process Historian included in the video.
Virtually all modern HMI platforms have some form of built-in trending. This will allow you to store values for various datapoints and make them available to operators on a built-in trending component. This option is extremely valuable to provide information to operators while they are on the plant floor. A common use case is showing pump speeds for the current shift, or tank levels for the last 24 hours.
Generally these interfaces are not well suited for navigating large datasets. They are more designed to show a static number of tags on a trend for a particular time period. If you need to collect a larger number of tags, or make data available to another interface, such as a third-party trending tool, a reporting back-end, or Excel the next step would be to move towards a database-backed system.
Using scripting you can generally get data from an HMI system into a database. Some HMI packages natively utilize SQL databases for storage. Otherwise you can use something like OPC-HDA or dataloggers to get data from your process into a database.
Databases provide easier access to data. This can include the ability to do some more advanced analytics than you can do in an HMI trending tool. Additionally Database-backed options allow the data to be made available for use in third party tools, or even web-based systems.
The example video is using a database-backed system Corso has implemented is exposing historical data for a golf-course pump station manufacturer. This allows users on mobile devices via a web-site with a powerful trending and reporting package not available in the standard HMI interface. This is an extremely popular feature that can be replicated for virtually every application. For more information on this please see our Remote Viewing Application.
The downside to database-backed systems is that they don’t utilize compression. As your dataset grows it can impact performance when accessing the data. This can be an extreme detractor if you are creating a huge data set, have a lot of tags, and/or are collecting the data in close intervals. A Database-backed option can also require working with someone who knows how to extract the data to turn it into useful information.
If you are going to be collecting more than a few thousand datapoints, or want to collect data on the level of milliseconds or even faster, you will likely be looking at a full-blown Process Historian.
Full Process Historian
Process Historians generally utilize data compression techniques to reduce the overall file sizes of your stored data. They can also limit the data stored to values that have changed outside of some deadband to prevent storing duplicate values over and over. Process Historians are also usually paired with some form of analysis tool. These tools are designed to work directly with the data to turn it into useful information. These tools can include basic trending, reporting backends, and various data mining tools you can use to understand operations in more detail.
Many process Historians also offer an SQL Query Interface. This brings together the power of the Historian’s compression, with the open-ended nature of a database. Thus making Process Historians easy to integrate with other systems. This concept can even be extended to the golf-course pump station example. Process Historians are usually an added expense beyond the SCADA system, and can be somewhat cumbersome to configure until you become familiar with the interfaces, but provide a robust and powerful solution for turning data into information.
Layering The Options
A common solution to the Process Historian conundrum in many facilities is layering options on top of each other. This type of solution is common in many facilities where staff requires multiple layers of information. Operators and plant floor level staff use the built-in trending tools for a more real-time view of the data. Process Engineers and design/optimization staff will use the data to have an “all-time” view of the data. Additionally this will provide the ability to compare various process conditions against each other to find optimal process conditions
Like most everything, it is about having the right tool for the job. Sometimes you will need more than one tool.
If you are looking for more information on Process Historians, you can check on our Remote Viewing Application. Stay tuned for our next installment on Process Historians, where we will compare the big four brands (Siemens, Inductive Automation, Rockwell Automation, Wonderware) and show you examples of each.
If you cannot wait, send us a request and we’ll send you a copy of our White Paper on comparing Process Historians.