Lecture 4: Process Control Methods - ENGI 24495 PDF

Lecture 4: Process Control Methods Control Systems – ENGI 24495 Outline of Presentation Closed-loop control system Transducers Final control elements On-off control Continuous control Advanced control techniques [email protected] Lecture 4...

Lecture 4: Process Control Methods Control Systems – ENGI 24495 Outline of Presentation Closed-loop control system Transducers Final control elements On-off control Continuous control Advanced control techniques [email protected] Lecture 4 – Process Control Methods Page 1 of 27 Review Controller Process Comparator Actuator Controlled Set point + error CO MV variable (SP) (CV) - Closed loop Sensor Feedback signal (FB) Sensor’s output (measured variable) is conditioned by a transmitter into a standard signal aka feedback control Goal: CV = SP Controlled variable changes due to disturbance, load change, or SP change With the feedback loop, it becomes self regulating and more effective. One loop is designed for one controlled variable. [email protected] Lecture 4 – Process Control Methods Page 2 of 27 Response Time of Closed-Loop input 𝑘 output 𝜏𝑆 + 1 1st order system k=gain 𝜏=time constant Time delay (or response time) is inherent to control loop; When a step change takes place, there is not an immediate response by the loop i.e., correcting action takes time. CV takes one time constant (𝜏) to approach 63.2% of the step input and 5𝜏 to approach steady state. [email protected] Lecture 4 – Process Control Methods Page 3 of 27 Response Time and Dead Time System goes through dead-time and transient time Dead time is the elapsed time between the instant a step input is applied and the instant a corrective action begins. There are two ways to reduce time delay (response time) of a control loop: 1. Selecting a controller with proper operating features (Today) 2. Properly tuning a controller (Next week) [email protected] Lecture 4 – Process Control Methods Page 4 of 27 Let’s talk about transducers … I/P Transducer: (-- required for operation of I/P) Converts current signal (I) to pneumatic signal (P) Pneumatic signal (P) actuates a control valve to adjust fluid through a pipe [email protected] Lecture 4 – Process Control Methods Page 5 of 27  P/I Transducer: Bubbler sensor measures the level of flammable liquid. Converts pneumatic signal (P) to current signal (I) Back pressure from bubbler sensor is proportional to level of the liquid in the tank i.e., High level ➞ 15 psi and low level ➞ 3 psi [email protected] Lecture 4 – Process Control Methods Page 6 of 27  I/E Transducer: Displays information for operator Transmitter ⟶ Acquires data for future use. Generates corrections Converts current signal (4-20mA) to a proportional voltage signal (0-10v) Output DC voltage is applied to indicator, recorder, or controller. These devices are connected in parallel. [email protected] Lecture 4 – Process Control Methods Page 7 of 27 Other Transducers Square-Root Extractor: Located between differential-pressure flowmeter and controller, to linearize the output signal from the differential- pressure flowmeter P&ID symbol Analog-to-Digital (A/D): Converts analog signal from sensors to digital signal used by controller. PLCs use analog input modules, and PCs use analog I/O card Digital-to-Analog (D/A): Converts digital signal from controller to analog signal used by actuator. PLCs use analog output modules, and PCs use analog I/O card [email protected] Lecture 4 – Process Control Methods Page 8 of 27 The Final Control Element, FCE (Actuator) FCE is manipulation device and directly influence the process variable Types include solenoids, motors, and valves The Solenoid Valve Electromagnetic device converts on-off electrical signal from controller (e.g., PLC) to on-off flow control; When coil is energized, flow passes; otherwise, flow stops [email protected] Lecture 4 – Process Control Methods Page 9 of 27  AC Motor Workhorse of industry; powers most machines e.g., pumps, blowers, etc. Speed is controlled by varying frequency 1-phase, 3-phase [email protected] Lecture 4 – Process Control Methods Page 10 of 27  DC Motor Used in high torque applications e.g., augers, conveyors, hoists, etc. Speed is controlled by varying DC voltage Shunt, series, compound [email protected] Lecture 4 – Process Control Methods Page 11 of 27  3-15 psi air The Control Valve pressure signal Diaphragm Variable control valve is very common in process control Adjusts flow of materials to process to change Spring process variable Main components: Valve position – Body: housing valve parts, inlet and outlet indicator ports for pipe connections – Plug: restricts fluid passage and hence adjusts flow rate; could be ball or disk Stem – Actuator: converts pneumatic signal to mechanical displacement to position plug Actuator force = (pressure) (diaphragm area) F=pxA [email protected] Lecture 4 – Process Control Methods Page 12 of 27  Two valve actions: Air-to-Open (A-O) and Air-to-Close (A-C) Arrow in P&ID symbol indicates the position of the valve when control signal fails. Fail-open Fail-closed [email protected] Lecture 4 – Process Control Methods Page 13 of 27 Valve Positioner Works as pneumatic amplifier to provide enough force for positioning the valve Benefits: − Overcome frictions in the valve − Reduce actuator dead-band and hysteresis effects of spring-diaphragm − Provide linear and precise positioning − Improve frequency response of closed loop Figure: Sliding gate valve with positioner [email protected] Lecture 4 – Process Control Methods Page 14 of 27 Valve Characteristics Flow through valve 𝑞=𝐶 ∆𝑃/𝑆𝐺 Valve coefficient C is related to c/c’s Quick-open: small ∆x → large ∆q (suitable for on-off control) Linear: ∆𝑞/∆𝑥 = 1 Equal percentage: ∆𝑞/∆𝑥 = 𝑘𝑞 Plug shape [email protected] Lecture 4 – Process Control Methods Page 15 of 27 Controller Controller Process Comparator Actuator Controlled Set point + Error PID CO MV variable (SP) e(t) or On-Off (CV) - Closed loop Sensor Feedback signal (FB) Common control modes: On-Off control ➞ CO: on or off Proportional (P) Continuous control Integral (I) (CV always tracking SP) Derivative (D) 𝑑𝑒 𝑡 𝐶𝑂 = 𝐾! 𝑒 𝑡 + 𝐾" 3 𝑒 𝑡 𝑑𝑡 + 𝐾# 𝑑𝑡 Each mode has its own characteristics and is selected based on the requirements of the process [email protected] Lecture 4 – Process Control Methods Page 16 of 27 On-Off Control on Controller output (CO) off Simple and applied to slow systems where CV changes slowly (e.g., temperature, pressure) Controller and actuator has two extremes: ‘on’ or ‘off’ CV cycles around setpoint Noisy FB signal makes actuator bouncing, causing premature wear in actuator and affecting its lifespan [email protected] Lecture 4 – Process Control Methods Page 17 of 27 Hysteresis or Deadband On-Off control is usually used with hysteresis to reduce the effect of high switching rate of actuator – Switching Off point: CV = SP + DB/2 – Switching On point: CV = SP - DB/2 [email protected] Lecture 4 – Process Control Methods Page 18 of 27 Proportional Mode Can be accomplished in two ways 1. Time Proportional ➞ CO is switched either on or off Ratio of on-time to off-time is varied based on value of CV relative to SP Average CO = D xV E.g., CV=SP If V=10, D=0.6 CO = 0.6 x 10V =6V CVSP [email protected] Lecture 4 – Process Control Methods Page 19 of 27  2. Amplitude Proportional: The most common technique and works as amplifier 𝐶𝑂 = 𝐾! × 𝑒𝑟𝑟𝑜𝑟 (Kp is proportional gain) Adjustments: Percentage output change 1: Proportional gain ➠ K! = Percentage input change Controlled variable % change 2: Proportional band ➠ PB = x 100 Final control element % change 1 PB = x100 K! Sensitivity: Describes how fast the controller responds to its input; the larger the gain (𝐾$ ) or the narrower the PB, the more sensitive the controller is to input changes. [email protected] Lecture 4 – Process Control Methods Page 20 of 27 Integral Mode Assume load change for 1 CV min. P output rises and stops after 1 min, but error (offset) SP still exists. To eliminate offset, Integral (I) mode is added with P mode. Integral output continues to increase until CV=SP Integral is also called reset Adjustments: Integral Gain (Ki), Reset Rate, Tr (repeats/min), Rest Time, Ti (min/repeat) For PI controller, large Ki is used with Rest Time = 1/Rest Rate small Kp (wider bandwidth) [email protected] Lecture 4 – Process Control Methods Page 21 of 27 Derivative Mode Works on rate of change (slope) of the error (D=Kd x de/dt) Used with P or PI mode to reduce overshoot and settling time Unsuitable for systems exposed to noisy signals as D mode amplifies the noise D mode exhibits a boost action when the error is increasing and a breaking action when the error is decreasing Adjustments: Derivative gain (Kd) or derivative time (Td) [email protected] Lecture 4 – Process Control Methods Page 22 of 27 Continuous Control Modes TABLE 1: Proportional, Integral, and Derivative Mode Summary Mode Function Applications Proportional (P) To provide gain For small setpoint or small load changes Proportional-Integral (PI) To eliminate offset For large and slow setpoint or load changes Proportional-Derivative (PD) To speed up response For sudden setpoint or and minimize quick load changes in a overshoot slow-response system Proportional-Integral-Derivative (PID) To speed up response, For large and sudden minimize overshoot, setpoint or load changes and eliminate offset in a slow-response system [email protected] Lecture 4 – Process Control Methods Page 23 of 27 Advanced Control Techniques Comparator Actuator Process Controlled Set point + Error + variable (SP) C1 C2 e(t) (CV) - - S2 S1 Cascade control: Two loops in series; primary loop (or outer loop) to monitor controlled variable and secondary loop (or inner loop) to monitor manipulated variable. Tunning Inner loop is tuned first procedure: Time constant of inner loop is 3 to 10 times faster than time constant outer loop. [email protected] Lecture 4 – Process Control Methods Page 24 of 27 Ratio Control Controls the flow rate of one ingredient based on the flow rate of another ingredient. Flow rate of wild flow is measured and used as a reference to set the flow rate of controlled flow. Wild flow is uncontrolled. Applications: soft drink production, sewage water treatment [email protected] Lecture 4 – Process Control Methods Page 25 of 27 Feed-forward Control Disturbance FF Controller Comparator Actuator Process Controlled Set point + + variable Error + e(t) Controller (SP) (CV) - Sensor FF generates corrections to reduce or eliminate the error. The feed- forward and feedback actions can be done in one controller. – Compensates for measurable disturbances – Should be used with feedback control [email protected] Lecture 4 – Process Control Methods Page 26 of 27 Adaptive Control Adaptive controller is a sort of controller, which modifies its parameters in response to changes in the dynamics of the system and process disturbances. It is also used to accommodate nonlinear process. Controller has additional routines to estimate PID parameters. Estimation of Controller Parameters SP CV 𝐾# + 𝐾" + + 𝑇$ 𝑠 Process - 𝑠 FB Sensor [email protected] Lecture 4 – Process Control Methods Page 27 of 27