- June 21, 2017
- Schneider Electric
By Scott A. Kortier, Schneider Electric Software
Automation applications are benefiting from HMIs that have moved well beyond panel device replacers and are now collecting, filtering and displaying data from multiple machines or processes.
By Scott A. Kortier, InduSoft Web Studio Senior Technical Sales, Schneider Electric Software
Starting in the 1980s, HMIs were created to replace pushbuttons, pilot devices and meters—which were often mounted on huge panelboards in control rooms. But modern HMIs do much more than replace these panel devices, as they also provide knowledge and insights into the operation of a machine or process.
In some cases, an HMI is not just monitoring a single machine and its performance, efficiency and quality—but many machines, maybe hundreds. The data from these machines can be analyzed to allow operators to understand and act upon slight differences in performance, cycle rate or a process variable such as temperature, tension or viscosity.
This article takes a brief look at the history of HMIs as panel device replacers, and its growth into data collection and analysis.
Back in 80s
The original HMIs were designed to replace the huge panelboards found in control rooms (Figure 1). These HMI display terminals were powered by minicomputers, and connected to a distributed control system (DCS) which was the source of their information. Just about every large process plant had a DCS and multiple HMIs.
Figure 1, Panelboard: HMIs were originally used in many industrial plants and facilities to replace large panelboards filled with meters, pushbuttons and other panel devices.
On the machine automation side, HMIs were proprietary hardware terminals used to replace operator input devices such as pushbutton, switches and potentiometers—and to replace operator indication devices such as lights, meters and charts.
In these older systems, part of the information was data, but it also included formatting of the screen, such as drawing a rectangle with a number in it to better represent the data to operators. Significant overhead was needed to add this formatting, and for any graphics.
To reduce this overhead in a DCS, memory was added to the terminal. A simple command from the mainframe, draw screen 1 for example, would trigger the terminal to draw the screen graphics. The mainframe could then populate the display with data.
In the late 1980s, an oil and gas firm based in Southern California used this configuration and handwrote Modbus in the terminal. It wasn’t fast enough, so the customer asked for the Modbus program to be added to the terminal firmware. Once completed, it became the first HMI.
In these early years, a communication driver for connectivity to a controller was a board that needed to be added to the display hardware. Modbus, Allen-Bradley, Mitsubishi, Siemens and other boards cost around $1500. Jump ahead 10 years and the drivers became software that had to be purchased on a floppy disk and cost about $150 dollars.
When a software driver wasn’t available, which was often because most controllers were used proprietary operating systems, a custom driver had to be written, tested and maintained. Jump ahead another ten years and HMI software, such as InduSoft Web Studio, has software drivers available for free, hundreds of them.
HMIs are still replacing panel devices, albeit in a much simpler fashion in terms of implementation, and a much more sophisticated fashion in terms of capabilities. But in many applications, they are also used to collect, analyze and compare data from multiple machines to create actionable information.
Configure and View Information
Users start with standard, off-the-shelf, PC-based HMI programming software and create custom runtime displays and applications. But for many users, it’s better to not create every runtime from scratch, but to instead start with a template.
For example, InduSoft’s Andon Template allows a user to go to configure their software for an application and deploy it (Figure 2). The template is modularized, allowing users to modify and configure it without custom coding or program development.
Figure 2 OEE: This HMI screen presenting production and OEE data was quickly created using InduSoft’s Andon Template without the need for custom coding or program development.
The Andon template starts with a blank slate. A tree view is accessed to set up the application, for example a screen containing ten machines in two zones across five areas. The software allows the machines to be set up depending on zone and area. It also allows usernames, passwords, and user levels or roles to be set up.
Downtime and quality reason codes are then defined, and there could be hundreds of each. The application created using the templates is then configured to match the products, plants, machines, quality, etc.—and aggregate that information and log it into a database.
To view and use the data shown on HMI screens, users navigate down through a tree view from a central point. A key point is the HMI’s ability to find issues with machines and processes, and to alert operators so they can take action.
Controlling User Access
It doesn’t matter which machine or desk the HMI is attached to, it can connect to data from several machines, and present it to a variety of users such as operators, supervisors or managers. The user’s security rights determine which data can be accessed, and what level of remote control is permitted.
The operator may just need local machine data, the maintenance person may need downtime information, and the quality manager might just need statistical information. The user login can automatically filter the data as well. With a modern HMI, the data can be filtered at the machine level up to the central sever level, to view information from one, some or all machines.
Correspondingly, each users’ level of remote control can be set. Local operators would normally have the highest level of control. Some remote users, such as management personnel, might have no remote control permitted, with view-only capability.
IoT with HMI
Some modern HMI software platforms have the pieces in place for users to create their own IoT applications. This often requires moving and using data full circle, from edge devices to the cloud and back again to the plant floor. These HMIs can work at the edge device level, the machine level or the supervisor level. They can send the data to an online repository such as Wonderware online, a storage and reporting service.
In this context, an edge device is a sensor, transmitter, motor drive or other field device collecting data in an industrial application, and maybe performing some control, as with a motor drive. HMI software that can be licensed to run on Linux enables embedding an HMI in the edge device, providing it has the capability to host Linux. A modern HMI embedded directly in a motor drive, for example can directly log data to a database, independent of the system HMI, bypassing the system controller and making it a smart edge device.
There are many cloud depositories with analytics that an HMI can use. Once data from HMIs is in the cloud, users located anywhere in the world can then analyze and run reports on the data. These results can be pushed back down to any HMI in the plant to provide actionable information. These HMIs can be full-function PCs, or simple edge devices running Linux.
Over the years, the progression in many manufacturing applications is from gaining hours per day in production time by automating manual processes, to gaining minutes per day by making incremental improvements. Now, advanced HMIs and analysis are helping users gain precious seconds of production time each day. A little optimization to many machines adds up.
By evaluating the performance of many similar machines or devices and analyzing them, these modern HMIs are collecting and providing actionable information in many different formats—everything from a PDF file to data displays accessed via web browsers.
This information can be displayed on HMIs, many of which are equipped with web server functionality, allow any remote device capable of running a web browser to access the information. These remote devices include laptops, smartphones and tablets.
From One HMI to Many
The information gathered by HMIs from controllers and smart components such as motor drives can be part of an Andon system showing live values on multiple 50-inch big screen monitors mounted in the rafters. Display data often includes target, actual and difference for any KPI. The large text can be seen from hundreds of feet away (Figure 3).
Figure 3 Andon Display: Modern HMI software such as InduSoft Web Studio can display high level production information on a big screen, and push detailed data to dozens of thin clients.
In an application with InduSoft Web Studio, a main HMI can connect to many secure viewer thin clients, similar to the way mainframes use to feed data to terminals, but with many more capabilities and features. With a secure-viewer thin client loaded on a PC, or on any device capable of running a web browser, HMI screens can be displayed.
It’s not uncommon to have dozens of these thin clients driven from a single HMI, each showing different types of production information. One Web Studio user had 50 thin client monitors displaying information in webpages. On the main runtime HMI, a simple application was created, a single screen with one object and one script program. It fed all fifty of the thin clients, each with a different production information screen.
Each of the 50 monitors has their own unique computer name, such as PC 1, PC 2, etc. Based on the area, different production information web pages are displayed. The screen is a full edge-to-edge web browser object, and the script says if I’m PC 1, show web page x, if PC 2, show web page y, etc. All thin clients are fed from a single InduSoft runtime HMI. Each thin client can thus display unique or identical information.
HMI software has come a long way over the last 30 years, from replacing a single push button station to pushing production information to 50 or more thin client monitors throughout a manufacturing facility. For IoT applications, a modern HMI can gather the data, push it to the cloud for analysis and then send it back as actionable information for use by an operator, supervisor or manager.
Did you enjoy this great article?
Check out our free e-newsletters to read more great articles..Subscribe