Technology Update: Process control simulations adapt to growing complexities
Hardware-in-the-loop simulations for process control speed verification, startup, and system optimization. Learn from other industries like automotive and aerospace where simulation is used across a variety of validation and testing tasks.
Although simulation technology has been used in the process control industry for a number of years, the computer processing power available in today’s engineering desktop is helping engineers build better systems while reducing their overall design cost. Learning from other industries like automotive and aerospace where simulation is used across a variety of validation and testing tasks, process automation engineers are beginning to use the same hardware-in-the-loop (HIL) simulation techniques for their applications to address the growing complexities of their next generation systems. Examples show how HIL simulations can be used in process control and explain what engineers should consider when implementing process control simulation.
In recent years, process control engineers have felt the competitive pressure typical of other engineering fields. The need to implement controllers with more advanced functionality, but with fewer resources and in less time is bringing innovative approaches to this application space. Process control industries like oil and gas have typically been reluctant to adopt new technologies in favor of more proven solutions, but increasing market pressures and advances in computing technologies are pushing engineers to look for innovative ways to make process control more efficient.
For instance, National Oilwell Varco (NOV), a leading provider for the worldwide oil and gas industry, was researching a new algorithm to avoid stick-slip. Stick-slip is a dysfunction of rotary drilling machines, characterized by large cyclic variations of the drive torque and the rotational bit speed. These variations are a major source of problems, and can cause excessive bit wear, premature tool failure, and poor drilling rate.
When looking for a way to test the new algorithm on a programmable logic controller (PLC), NOV had two major concerns. First, the control algorithm could not be fully tested during commissioning because not all the physical components of the drill are present. Secondly, field testing is time consuming and expensive because stick-slip conditions occur occasionally, resulting in an undue amount of waiting at the test site.
Although there isn’t a simple solution to these challenges, examining how other industries have optimized processes has proven to be an excellent source of new ideas. One of the most promising is the extensive use of simulation to provide more efficient validation and testing capabilities. NOV connected a real-time computer simulation of the pumping system to the top drive control PLC communicated through the Profibus industrial network. This setup allowed simulation of different operating conditions so that the algorithm can be tested over a wide range of conditions safely and on-demand in the lab. In addition, they were able to easily repeat the test to compare and fine-tune the PLC code without leaving the laboratory.
Hardware in the loop
You may ask, “Why is simulation in general (and HIL simulation in particular) applicable to the process control industry?” The answer to this question is that the application requirements are the same, such as: determinism, real-time execution, hardware I/O communication, datalogging services, stimulus generation, model execution, etc.
HIL has proven to be a valuable technique capable of solving different challenges. It is also an established technology in industries such as automotive and aerospace, which have been using it successfully for decades. HIL testing is a technique in which a real-time simulation of the plant system being controlled is used to allow more robust, more repeatable, and more efficient testing of the implemented control system. It allows engineers to simulate different operating conditions, as well as sensor and network failure before field installation at a much lower cost and with no risk to equipment or personnel. This provides a valuable tool for plant operations in industrial environments where field testing can be very costly and may have significant limitations in regard to the ability to create test conditions and even more to repeat those conditions to verify a fix to the control system.
Solix Biofuels was started with the goal of delivering a scalable and cost-effective way to produce biofuels from algae. The goal of its Algal Growth System (AGS) is to acquire data and provide the controls to manage the growth process. It needed a technology platform that would support both R&D experimentation and industrial operation, thereby accelerating the transition of this new technology from the laboratory to the demonstration plant. Solix started by building a simulated system of the growth process so that it could develop the control system, even before the system was completed. Once the control algorithm was thoroughly tested and validated using the simulation, it deployed it to an NI programmable automation controller (PAC). The ability to create real-time simulations and deploy to real-time hardware without code modification allowed Solix to complete its project in record time.
There are different types of HIL systems, but most of them have the following components in common: a real-time processor, I/O interfaces, and an operator interface. The real-time processor is the core of the HIL test system. It provides deterministic execution of most of the HIL test system tasks, such as the hardware I/O communication, data logging, stimulus generation, and model execution. Although the rate-of-change in process control systems in often slow, a real-time system is still recommended for the HIL simulation to provide an accurate representation of the simulated components. The analog, digital, and bus protocol signals are used to connect the plant simulation to the deployed control system.
You can use them to process stimulus signals, acquire data for logging and analysis, and provide the sensor/actuator interaction between the PLC being tested and the virtual environment being simulated by the model. The operator interface communicates with the real-time processor to provide test commands and visualization. Often, these components also prove configuration management, test automation, analysis, and reporting tasks.
Challenges: HIL for process control
One of the challenges of implementing an HIL testing system in the process industry is modeling the plant that the PLC or PAC will be controlling. There are a couple of approaches, each with its own list of pros and cons. The first approach is to use first principle math to analyze the different equations that govern the dynamics of the process under control. This approach requires detailed knowledge of the process and takes time to validate the accuracy of the model with real-world data. The second option is to measure the data as it enters the system and the results, or response that it provides. Using techniques similar to curve fitting, a mathematical model can be derived based on the stimulus and response data that is recorded from the process. It’s important to note that these two approaches are not mutually exclusive, as they can be combined for faster model development and fidelity.
Another challenge is the number of I/O channels involved in the system. Implementing and maintaining wiring for high-channel-count systems can pose costly and time-consuming challenges. These systems can require hundreds to thousands of signals to be connected between the controller and the HIL test system, often spanning many meters to compensate for space requirements.
Fortunately, deterministic distributed I/O technologies can help address these wiring complexities and provide a modular connectivity to the control system, which allows for efficient system configuration modification. Instead of routing all connections back to a single rack containing one or more real-time simulators and I/O interfaces, you can use deterministic, distributed I/O to provide I/O interfaces located in close proximity to each control system without sacrificing the high-speed determinism necessary for accurate simulation of the virtual parts of the system.
The application of HIL in process control can help save money by offering a sandbox where engineers can debug and test the code that runs of the PLC before the installation. It can also allow you to evaluate different control strategies and provide a starting point for the tuning process. Now the question to ask yourself is, “Does HIL simulation make sense for you?”
Javier Gutierrez is senior product manager for LabVIEW Simulation and Control Design Tools at National Instruments. www.ni.com/simulation
Simulation Software in Motion Control
Dynamic system simulation