College and university professors do an excellent job in covering theoretical aspects of system identification, primary process control design, and PID control loop tuning methods. However practical exposure to hands-on and industrial system identification and process control optimization exposure is difficult for new students and graduates. Most academic schools and colleges do not cover system identification and process control optimization at the undergraduate level as this is perceived to be an advanced and graduate-level topic. Academic courses are often research-oriented and do not cater to the needs of the industrial control room environment. Furthermore, there is an excessive focus on academic topics like Laplace transforms and frequency domain which are not useful in the industrial control room environment where people prefer and work entirely in the time domain.
So, it is no surprise that when graduated students – both engineers and technicians enter the industry their skills are more academic and less practical, they are afraid of making process control-related changes and improvements in the control room. Since the plant’s operating performance can be impacted significantly by the choice of process control strategy and their optimal parameters, new process control engineers and technicians need to be formally trained on practical aspects, e.g., industrial system identification and process control optimization catering to the industrial control room needs and environment. They should be on practical industrial system identification, PID control loop tuning and optimization in a very real plant-like environment. They should have the freedom and ability to fearlessly drive PID control loops unstable, study sluggish PID control loops, control valve problems, and the effect of external unmeasured disturbances on the PID control loop quality.
Since the mid-1960s, only a handful of system identification algorithms have been invented and put to practical use. The most widely known and still used BJ (Box-Jenkins), ARMAX (autoregressive moving average models with exogeneous inputs), and step-response vector coefficient methods. These methods require extensive skills and training in order to use them correctly for the design of control systems. Industrial data are often superimposed with random noise, drifts, complex unmeasured disturbances, and nonlinearities. These menacing components make the system identification process challenging and result in potentially uncertain or inaccurate results. The commonly used ARMAX-based system identification is rather sensitive to the large levels of noise and significant unmeasured disturbances and does not work well on completely oscillatory closed-loop data. Products like Matlab are excellent in some applications, but rather cumbersome and complex for industrial use in the control room environment. In sensitive processes, intrusive step-tests can upset the process or drive product quality to go off-spec and may be difficult to conduct the tests safely.
Engineers often use the old trial-and-error method or directionally guess the PID control loop tuning parameters, feedforward parameters, and other advanced process control (APC) parameters. Practical tools and methods catering to the control room environment are not available easily, and even if they are available, they are complex, expensive, bulky, and often produce uncertain results. Therefore, there is definitely a need for simpler and more practical tools.
To address this long-term gap between the academia and industry and to facilitate the training and certification process of process control engineers and technicians, PiControl has developed new software for training, focusing on the needs of the control room. These include SIMCET for real-time dynamic process control simulation software and PITOPS for multivariable closed-loop system identification, PID/APC (Advanced Process Control) tuning and optimization software.
SIMCET is a real-time, online simulator for PID control loop tuning and optimization practice and for testing of PID control loop tuning skills providing the hands-on experience necessary for the practical control room environment. Typical examples, under the PID control loop tuning practices which can be seen in SIMCET are related to chemical, air separation, compressor, turbine, polyethylene, laboratory and other plants and processes. Under the new simulation in SIMCET any other custom simulation can be also easily configured to mimic a specific plant or process by using typical scheme of desired process. The uniqueness of SIMCET is that it also provides testing and grading features for testing PID control loop tuning skills of engineers and technicians. SIMCET provides a lot of other features which can be seen on the real DCS or PLC system in the control room.
PITOPS is a comprehensive and unique industrial system identification and process control simulator and PID tuning optimizer and APC parameter calculator which provides the following functionalities:
- Operates completely in the time domain.
- System identification based on the open-loop data (step changes of PID OP), closed-loop data (step changes of PID SP), completely oscillatory and non-steady-state closed-loop data, and stepless closed-loop data.
- Multivariable system identification using closed-or-open loop data.
- System identification based on the ultra-short duration data (1/5th of data).
- Control valve stiction identification.
- Pattern identification for unmeasured disturbances.
- Identification of process or equipment nonlinearities.
- Does not need data preconditioning (resample, high noise, missing data, outliers).
- Possesses all commercially available PLC/DCS PID algorithms.
- Master and slave control loop optimization (based on complex SP trajectory).
- Mathematical calculation of feedforward parameters (gain, deadtime, lead/lag).
- Inferential control loop design.
- Deadtime compensator control loop design (Smith predictor).
- PID control loop optimization based on the SP change (step or ramp), different disturbances (step, pulse, ramp, sine), control valve stiction, PID OP rate of change, over-or-undersized control valve.
Such software can tremendously help to improve process control education of students and plant personnel. This will help to improve process stability, reduce unplanned plant shutdowns and increase plant profits. PiControl Solutions possesses over 50 years of industrial process control experience. PiControl goals are to reduce the current gap between education and skills acquired from universities and the needs in the control room. An increasing number of university professors have recognized and seen the benefits of this new PiControl approach. Many professors are now using the new PiControl software and technology for both teaching and research.
Some of the academic ventures include: Modernizingtheir existing system identification, primary and advanced process control (APC) semester-based courses and to improve their laboratories and homework. They have put the focus more on industrial control design using industrial practical real plant data for student homework and labs. Professors are also using PiControl software and technology for designing virtual laboratories and offer remote teaching, share materials, and practices that suit current COVID-19 restrictions, and stay abreast in the digital era. Professors are also using PiControl software and technology for generating new ideas for MS/PHD research projects in process control resulting in new technical papers on research topics. PiControl is working with various professors to develop a college-industry alliance group to provide the industry with consulting, technology, and control room assistance including process control project execution and education courses. This way the college could generate significant consulting revenue, research topics, and students would have direct value to control room environment and industrial job needs.
