Titlebook: Control Performance Assessment: Theoretical Analyses and Industrial Practice; Pawe? D. Domański Book 2020 Springer Nature Switzerland AG 2

河流

Simulation Setuphe most representative measures are selected and validated with different SISO and MIMO simulated processes. These experiments address two main control algorithms used in process industry, i.e. PID and Model Predictive Control (MPC).

Dappled

向下

,Geistesst?rung und Verbrechen 1888,sly limiting initial investment and contractual risks. This chapter describes practical aspects how to conduct the study and how use in practice the knowledge and approaches described in the previous chapters.

decipher

Does Control Quality Matter? for control systems. The relationship is straightforward. Better control causes higher performance. Despite this clear relation and common understanding of the fact, the majority of the industrial loops is neither well tuned nor properly designed. Control engineers require tools and indexes that wo

FLAT

Step Response Indexesutput, manipulating variable, setpoint and control error. Analysis of this data in time domain is often the first shot in the CPA analysis. Once the time trends are properly acquired from the industrial system and delivered to the analysis, one can start playing with them. Step response and the asso

沖突

Basic Data-Based Measuresclassical example, being used by almost everybody. Actually, hardly anybody things about integral measure formulation—MSE is just selected. Its story started in the beginning of the 19th century and continues. However, there are much more integral indexes that might be used, like Integral Absolute E

他很靈活

jungle

Model-Based Measuresabout the loop or background process. They are fully data-driven. They all share the similar shortcut. None of them offers any distance from the measured index value to the optimal one. Thus, apart from the actual measured index value . one would require to estimate the lowest (the best achievable)

后來

Frequency Based Methodsfact, we need to identify process frequency characteristics, i.e. its model. Morover the process should be linear. The number of reports on CPA in frequency domain is very limited. These methods are hardly used in practice. Finally, the measures in the frequency domain are applicable to the linear p

解脫