PID (Proportional-Integral-Derivative)-based control has been commonly used for decades. And while it still is adequate for controlling many process parameters, the time may be right to take a step back and re-evaluate PID-based control in power generation and other industries.
The answer lies in an increasingly diverse energy mix that includes more renewable sources such as solar and wind, which is causing fossil-fueled plants to function differently than designed. For instance, combined-cycle and conventional fossil plants need to ramp and cycle more frequently in response to fluctuations in renewable energy generation.
During the recent Emerson Exchange Immerse event, Jim Nyenhuis explained how advanced process controls (APC) are helping plants remain competitive. In particular, he addressed how advanced control technologies using various model predictive control (MPC) strategies can augment or replace PID-based control and in doing so, help optimize steam temperature, drum level, coordinated ramp rates allowing for more automated overall control of processes.
Perhaps you’re intrigued by advanced process controls for power plant optimization, but unsure about whether this approach lives up to the hype. Comparing how APC based techniques and PID-based control respond to steam temperature variations during unit load-following operations can illustrate the advantages of model predictive control.
Why is this an important variable to optimize? Improving steam temperature control helps maximize unit efficiency while reducing stress on boiler tubes and turbine blades which, in turn, can significantly decrease maintenance costs and outage requirements. Optimal steam temperature management also improves ramp rates, which further contributes to increased revenue and unit flexibility.
Characteristics of PID-based and APC-based Control
As a first step, let’s review the characteristics of PID-based and APC-based control.
Simply put, PID-based control acts on the current error between a desired setpoint and the actual process variable. Conversely, model-based control involves creating a mathematical model that represents the system being controlled. This model is then used to predict the future behavior of the system and proactively generate control actions to maintain the desired process behavior.
In the real-world steam temperature example, the model-based application understands the transient impact of energy changes (changes in fuel firing, CT exhaust energy, duct burners, etc.) on steam temperature variations through the development of steam temperature process models. More specifically, application-specific algorithms are used to execute a forward looking predictive horizon model that accurately reflects the coupled relationships of controlled variables (a process variable that is to be driven to a particular setpoint, such as final steam temperature), manipulated variables (a signal or device that may be moved to achieve setpoint, such as a spray valve) and disturbance variables (a signal that has an impact on the process such as fuel quality, load or sootblowing) to quickly and accurately achieve the temperature setpoint. (See example on right, below.)
In contrast, a conventional PID-based control strategy (see example on left, below) offers a less accurate, slower response to reaching setpoint as the PID loops in areas where APC techniques have a good fit must be de-tuned due to long process time response characteristics to maintain an acceptable level of stability to support real world operations.
In one example site with the MPC models running, the plant saw immediate improvements in steady-state operation. More specifically, it achieved a 55% improvement in standard deviation of temperature versus setpoint as well as a nearly 53% improvement in standard deviation of spray flow and valve movement.
MPC models show similar operational improvements when used to optimize drum level and coordinated ramp rates , resulting in improved unit operation and performance.
Interested in learning more? You can read more about Jim’s session in Power Engineering. Learn more about Emerson Exchange Immerse and Jim’s presentation here. You can also find a wealth of information by visiting Emerson’s Ovation Advanced Power Applications website.