1 Practical plantwide process control Part 1 Sigurd Skogestad, NTNU Thailand, April 2014 2 Part 1 (3h): Plantwide control Introduction to plantwide control (what should we really control?) Part 1.1 Introduction. Objective: Put controllers on flow sheet (make P&ID) Two main objectives for control: Longer-term economics (CV1) and shorter-term stability (CV2) Regulatory (basic) and supervisory (advanced) control layer Part 1.2 Optimal operation (economics) Active constraints Selection of economic controlled variables (CV1). Self-optimizing variables. Part 1.3 -Inventory (level) control structure Location of throughput manipulator Consistency and radiating rule Part 1.4 Structure of regulatory control layer (PID) Selection of controlled variables (CV2) and pairing with manipulated variables (MV2) Main rule: Control drifting variables and "pair close" Summary: Sigurds rules for plantwide control 3 Objective: Put controllers on flow sheet (make P&ID) Two main objectives for control: Longer-term economics (CV1) and shorter-term stability (CV2) Regulatory (basic) and supervisory (advanced) control layer Part 1.1 Introduction 4 Why control? Operation time Actual value(dynamic) Steady-state (average) In practice never steady-state: Feed changes Startup Operator changes Failures .. - Control is needed to reduce the effect of disturbances - 30% of investment costs are typically for instrumentation and control Disturbances (ds) 5 Countermeasures to disturbances (I) I.Eliminate/Reduce the disturbance (a)Design process so it is insensitive to disturbances Example: Use buffertank to dampen disturbances (b)Detect and remove source of disturbances Statistical process control Example: Detect and eliminate variations in feed composition inflowoutflow T in T out 6 Countermeasures to disturbances (II) II. Process control Do something (usually manipulate valve) to counteract the effect of the disturbances (a) Manual control: Need operator (b) Automatic control: Need measurement + automatic valve + computer Goals automatic control: Smaller variations more consistent quality More optimal Smaller losses (environment) Lower costs More production Industry: Still large potential for improvements! 7 Classification of variables Process (shower) u input (MV) y output (CV) d Independent variables (the cause): (a)Inputs (MV, u): Variables we can adjust (valves) (b)Disturbances (DV, d): Variables outside our control Dependent (output) variables (the effect or result): (c)Primary outputs (CVs, y 1 ): Variables we want to keep at a given setpoint (d)Secondary outputs (y 2 ): Extra measurements that we may use to improve control 8 Inputs for control (MVs) Usually: Inputs (MVs) are valves. Physical input is valve position (z), but we often simplify and say that flow (q) is input 9 Idealized view of control (PhD control) Control: Use inputs (MVs, u) to counteract the effect of the disturbances (DVs, d) such that the outputs (CV=y) are kept close to their setpoints (y s ) d, y s = y-y s 10 Practice: Tennessee Eastman challenge problem (Downs, 1991) (PID control) TCPCLCCCxSRC Where place?? 11 Notation feedback controllers (P&ID) TCTC 2nd letter: C: controller I: indicator (measurement) T: transmitter (measurement) 1st letter: Controlled variable (CV) = What we are trying to control (keep constant) T: temperature F: flow L: level P: pressure DP: differential pressure (p) A: Analyzer (composition) C: composition X: quality (composition) H: enthalpy/energy T s (setpoint CV) T (measured CV) MV (could be valve) 12 LC H HsHs Inflow (d) Outflow (u) INPUT (u): OUTFLOW (Input for control!) OUTPUT (y): LEVEL DISTURBANCE (d): INFLOW Example: Level control CLASSIFICATION OF VARIABLES: 13 How we design a control system for a complete chemical plant? How do we get from PID control to PhD control? Where do we start? What should we control? and why? etc. 14 Plantwide control = Control structure design Not the tuning and behavior of each control loop, But rather the control philosophy of the overall plant with emphasis on the structural decisions: Selection of controlled variables (outputs) Selection of manipulated variables (inputs) Selection of (extra) measurements Selection of control configuration (structure of overall controller that interconnects the controlled, manipulated and measured variables) Selection of controller type (LQG, H-infinity, PID, decoupler, MPC etc.). That is: Control structure design includes all the decisions we need make to get from ``PID control to PhD control 15 Page Buckley (1964) - Chapter on Overall process control (still industrial practice) Greg Shinskey (1967) process control systems Alan Foss (1973) - control system structure Bill Luyben et al. (1975- ) case studies ; snowball effect George Stephanopoulos and Manfred Morari (1980) synthesis of control structures for chemical processes Ruel Shinnar (1981- ) - dominant variables Jim Downs (1991) - Tennessee Eastman challenge problem Larsson and Skogestad (2000): Review of plantwide control Previous work on plantwide control: 16 Main objectives control system 1.Implementation of acceptable (near-optimal) operation 2.Stable operation ARE THESE OBJECTIVES CONFLICTING? Usually NOT Different time scales Stabilization fast time scale Stabilization doesnt use up any degrees of freedom Reference value (setpoint) available for layer above But it uses up part of the time window (frequency range) 17 Example of systems we want to operate optimally Process plant minimize J=economic cost Runner minimize J=time Green process plant Minimize J=environmental impact (with given economic cost) General multiobjective: Min J (scalar cost, often $) Subject to satisfying constraints (environment, resources) Part 1.2 Optimal operation (economics) 18 Theory: Optimal operation Objectives Present state Model of system Theory: Model of overall system Estimate present state Optimize all degrees of freedom Problems: Model not available Optimization complex Not robust (difficult to handle uncertainty) Slow response time Process control: Excellent candidate for centralized control (Physical) Degrees of freedom CENTRALIZED OPTIMIZER 19 Practice: Engineering systems Most (all?) large-scale engineering systems are controlled using hierarchies of quite simple controllers Large-scale chemical plant (refinery) Commercial aircraft 100s of loops Simple components: on-off + PI-control + nonlinear fixes + some feedforward Same in biological systems 20 Process operation: Hierarchical structure Manager Process engineer Operator/RTO Operator/Advanced control/MPC PID-control u = valves 21 Translate optimal operation into simple control objectives: What should we control? CV 1 = c ? (economics) CV 2 = ? (stabilization) 22 Part 1.2 Optimal operation (economics) Goal: Select economic controlled variables (CV1) Active constraints Self-optimizing variables 23 Outline Skogestad procedure for control structure design I Top Down Step S1: Define operational objective (cost) and constraints Step S2: Identify degrees of freedom and optimize operation for disturbances Step S3: Implementation of optimal operation What to control ? (primary CVs) (self-optimizing control) Step S4: Where set the production rate? (Inventory control) II Bottom Up Step S5: Regulatory control: What more to control (secondary CVs) ? Step S6: Supervisory control Step S7: Real-time optimization 24 Step S1. Define optimal operation (economics) What are we going to use our degrees of freedom (u=MVs) for? Typical cost function*: *No need to include fixed costs (capital costs, operators, maintainance) at our time scale (hours) Note: J=-P where P= Operational profit J = cost feed + cost energy value products 25 Optimal operation distillation column Distillation at steady state with given p and F: N=2 DOFs, e.g. L and V (u) Cost to be minimized (economics) J = - P where P= p D D + p B B p F F p V V Constraints Purity D: For example x D, impurity max Purity B: For example, x B, impurity max Flow constraints: min D, B, L etc. max Column capacity (flooding): V V max, etc. Pressure: 1) p given (d)2) p free (u): p min p p max Feed: 1) F given (d) 2) F free (u): F F max Optimal operation: Minimize J with respect to steady-state DOFs (u) value products cost energy (heating+ cooling) cost feed 26 Step S2. Optimize (a) Identify degrees of freedom (b) Optimize for expected disturbances Need good steady-state model Time consuming! But it is offline Main goal: Identify ACTIVE CONSTRAINTSs A good engineer can often guess the active constraints 27 Step S2a: Degrees of freedom (DOFs) for operation NOT as simple as one may think! To find all operational (dynamic) degrees of freedom: Count valves! (N valves ) Valves also includes adjustable compressor power, etc. Anything we can manipulate! BUT: not all these have a (steady-state) effect on the economics Steady-state DOFs 28 Steady-state degrees of freedom (DOFs) IMPORTANT! DETERMINES THE NUMBER OF VARIABLES TO CONTROL! No. of primary CVs = No. of steady-state DOFs CV = controlled variable Methods to obtain no. of steady-state degrees of freedom (N ss ): 1.Equation-counting N ss = no. of variables no. of equations/specifications Very difficult in practice 2.Valve-counting (easier!) N ss = N valves N 0ss N specs N 0ss = variables with no steady-state effect Inputs/MVs with no steady-state effect (e.g. extra bypass) Outputs/CVs with no steady-state effect that need to be controlled (e.g., liquid levels) 3.Potential number for some units (useful for checking!) 4.Correct answer: Will eventually find it when we perform optimization Steady-state DOFs 29 N valves = 6, N 0y = 2*, N DOF,SS = 6 -2 = 4 (including feed and pressure as DOFs) Typical Distillation column *N 0y : no. controlled variables (liquid levels) with no steady-state effect 4 DOFs: With given feed and pressure: NEED TO IDENTIFY 2 more CVs - Typical: Top and btm composition Steady-state DOFs 30 Steady-state degrees of freedom (N ss ): 3. Potential number for some process units each external feedstream: 1 (feedrate) splitter: n-1 (split fractions) where n is the number of exit streams mixer: 0 compressor, turbine, pump: 1 (work/speed) adiabatic flash tank: 0 * liquid phase reactor: 1 (holdup reactant) gas phase reactor: 0 * heat exchanger: 1 (bypass or flow) column (e.g. distillation) excluding heat exchangers: 0 * + no. of sidestreams pressure * : add 1DOF at each extra place you set pressure (using an extra valve, compressor or pump), e.g. in adiabatic flash tank, gas phase reactor or absorption column *Pressure is normally assumed to be given by the surrounding process and is then not a degree of freedom Ref: Araujo, Govatsmark and Skogestad (2007) Extension to closed cycles: Jensen and Skogestad (2009) Real number may be less, for example, if there is no bypass valve Steady-state DOFs 31 Potential number, N ss = 0 (column distillation) + 1 (feed) + 2*1 (heat exchangers) + 1 (split) = 4 With given feed and pressure: N ss = 4 2 = 2 Distillation column (with feed and pressure as DOFs) split Steady-state DOFs 32 . BUT A GOOD ENGINEER CAN OFTEN GUESS THE SOLUTION (active constraints) What are the optimal values for our degrees of freedom u (MVs)? Minimize J with respect to u for given disturbance d (usually steady-state): min u J(u,x,d) subject to: Model equations (e,g, Hysys): f(u,x,d) = 0 Operational constraints: g(u,x,d) < 0 OFTEN VERY TIME CONSUMING Commercial simulators (Aspen, Unisim/Hysys) are set up in design mode and often work poorly in operation (rating) mode. Optimization methods in commercial simulators often poor We use Matlab or even Excel on top J = cost feed + cost energy value products Step S2b: Optimize for expected disturbances 33 Step S3: Implementation of optimal operation Have found the optimal way of operation. How should it be implemented? What to control ? (primary CVs). 1.Active constraints 2.Self-optimizing variables (for unconstrained degrees of freedom ) 34 Cost to be minimized, J=T One degree of freedom (u=power) What should we control? Optimal operation - Runner Optimal operation of runner 35 1. Optimal operation of Sprinter 100m. J=T Active constraint control: Maximum speed (no thinking required) CV = power (at max) Optimal operation - Runner 36 40 km. J=T What should we control? CV=? Unconstrained optimum Optimal operation - Runner 2. Optimal operation of Marathon runner u=power J=T u opt 37 Any self-optimizing variable (to control at constant setpoint)? c 1 = distance to leader of race c 2 = speed c 3 = heart rate c 4 = level of lactate in muscles Optimal operation - Runner Self-optimizing control: Marathon (40 km) 38 Conclusion Marathon runner c = heart rate select one measurement CV = heart rate is good self-optimizing variable Simple and robust implementation Disturbances are indirectly handled by keeping a constant heart rate May have infrequent adjustment of setpoint (c s ) Optimal operation - Runner c=heart rate J=T c opt 39 Expected active constraints distillation Both products (D,B) generally have purity specs Valuable product: Purity spec. always active Reason: Amount of valuable product (D or B) should always be maximized Avoid product give-away (Sell water as methanol) Also saves energy Control implications: 1.ALWAYS Control valuable product at spec. (active constraint). 2.May overpurify (not control) cheap product valuable product methanol + max. 0.5% water cheap product (byproduct) water + max. 2% methanol + water Selection of CV 1 40 Example with Quiz: Optimal operation of two distillation columns in series 41 Operation of Distillation columns in series With given F (disturbance): 4 steady-state DOFs (e.g., L and V in each column) DOF = Degree Of Freedom Ref.: M.G. Jacobsen and S. Skogestad (2011) Energy price: p V =0-0.2 $/mol (varies) Cost (J) = - Profit = p F F + p V (V 1 +V 2 ) p D1 D 1 p D2 D 2 p B2 B 2 > 95% B p D2 =2 $/mol F ~ 1.2mol/s p F =1 $/mol < 4 mol/s < 2.4 mol/s > 95% C p B2 =1 $/mol N=41 AB =1.33 N=41 BC =1. 5 > 95% A p D1 =1 $/mol QUIZ: What are the expected active constraints? 1. Always. 2. For low energy prices. QUIZ 1 42 DOF = Degree Of Freedom Ref.: M.G. Jacobsen and S. Skogestad (2011) Energy price: p V =0-0.2 $/mol (varies) Cost (J) = - Profit = p F F + p V (V 1 +V 2 ) p D1 D 1 p D2 D 2 p B2 B 2 > 95% B p D2 =2 $/mol F ~ 1.2mol/s p F =1 $/mol < 4 mol/s < 2.4 mol/s > 95% C p B2 =1 $/mol 1. x B = 95% B Spec. valuable product (B): Always active! Why? Avoid product give-away N=41 AB =1.33 N=41 BC =1. 5 > 95% A p D1 =1 $/mol 2. Cheap energy: V 1 =4 mol/s, V 2 =2.4 mol/s Max. column capacity constraints active! Why? Overpurify A & C to recover more B QUIZ: What are the expected active constraints? 1. Always. 2. For low energy prices. Hm.? Operation of Distillation columns in series With given F (disturbance): 4 steady-state DOFs (e.g., L and V in each column) SOLUTION QUIZ 1 43 Control of Distillation columns in series Given LC PC QUIZ. Assume low energy prices (pV=0.01 $/mol). How should we control the columns? HINT: CONTROL ACTIVE CONSTRAINTS Red: Basic regulatory loops QUIZ 2 44 Control of Distillation columns in series Given LC PC Red: Basic regulatory loops CC xBxB x BS =95% MAX V1 MAX V2 1 unconstrained DOF (L1): Use for what?? CV=? Not: CV= x A in D1! (why? x A should vary with F!) Maybe: constant L1? (CV=L1) Better: CV= x A in B1? Self-optimizing? General for remaining unconstrained DOFs: LOOK FOR SELF-OPTIMIZING CVs = Variables we can keep constant WILL GET BACK TO THIS! SOLUTION QUIZ 2 QUIZ. Assume low energy prices (pV=0.01 $/mol). How should we control the columns? HINT: CONTROL ACTIVE CONSTRAINTS Hm. HINT: CONTROL ACTIVE CONSTRAINTS! 45 Comment: Distillation column control in practice 1.Add stabilizing temperature loops In this case: use reflux (L) as MV because boilup (V) may saturate T 1s and T 2s then replace L 1 and L 2 as DOFs. 2.Replace V1=max and V2=max by dpmax-controllers (assuming max. load is limited by flooding) See next slide 46 Control of Distillation columns in series Given LC PC TC T1 s T2 s T1 T2 Comment: In practice MAX V2 MAX V1 CC xBxB x BS =95% 47 a)If constraint can be violated dynamically (only average matters) Required Back-off = measurement bias (steady-state measurement error for c) b)If constraint cannot be violated dynamically (hard constraint) Required Back-off = measurement bias + maximum dynamic control error J opt Back-off Loss c c constraint c J More on: Active output constraints Need back-off Want tight control of hard output constraints to reduce the back-off. Squeeze and shift-rule The backoff is the safety margin from the active constraint and is defined as the difference between the constraint value and the chosen setpoint Backoff = |Constraint Setpoint| CV = Active constraint 48 Hard Constraints: SQUEEZE AND SHIFT Richalet SHIFT SQUEEZE CV = Active constraint Rule for control of hard output constraints: Squeeze and shift! Reduce variance (Squeeze) and shift setpoint c s to reduce backoff 49 Example. Optimal operation = max. throughput. Want tight bottleneck control to reduce backoff! Time Back-off = Lost production CV = Active constraint 50 Example back-off. x B = purity product > 95% (min.) D 1 directly to customer (hard constraint) Measurement error (bias): 1% Control error (variation due to poor control): 2% Backoff = 1% + 2% = 3% Setpoint x Bs = % = 98% (to be safe) Can reduce backoff with better control (squeeze and shift) D 1 to large mixing tank (soft constraint) Measurement error (bias): 1% Backoff = 1% Setpoint x Bs = % = 96% (to be safe) Do not need to include control error because it averages out in tank CV = Active constraint D1xBD1xB 8 xBxB x B,product 2% 51 More on: WHAT ARE GOOD SELF- OPTIMIZING VARIABLES? Intuition: Dominant variables (Shinnar) More precisely 1.Optimal value of CV is constant 2.CV is sensitive to MV (large gain) Unconstrained variables 52 1.Optimal value c opt is constant (independent of disturbance d): 2.c is sensitive to MV=u (to reduce effect of measurement noise) Equivalently: Optimum should be flat Unconstrained optimum BADGood GOOD SELF-OPTIMIZING CV=c 53 Control of Distillation columns. Cheap energy Given LC PC Overpurified: To avoid loss of valuable product B CC xBxB x BS =95% MAX V1 MAX V2 1 unconstrained DOF (L1): What is a good CV? Not: CV= x B in D1! (why? Overpurified, so x B,opt increases with F) Maybe: constant L1? (CV=L1) Better: CV= x A in B1? Self-optimizing? Example. Overpurified 54 More on: Optimal operation Mode 1. Given feedrate Mode 2. Maximum production minimize J = cost feed + cost energy value products Two main cases (modes) depending on marked conditions: 55 Amount of products is then usually indirectly given and Optimal operation is then usually unconstrained maximize efficiency (energy) Control: Operate at optimal trade-off NOT obvious what to control CV = Self-optimizing variable Mode 1. Given feedrate J = cost feed value products + cost energy c J = energy c opt Often constant 56 Assume feedrate is degree of freedom Assume products much more valuable than feed Optimal operation is then to maximize product rate Mode 2. Maximum production Control: Focus on tight control of bottleneck Obvious what to control CV = ACTIVE CONSTRAINT c J c max Infeasible region J = cost feed + cost energy value products 57 Control structure design using self-optimizing control for economically optimal CO 2 recovery * Step S1. Objective function= J = energy cost + cost (tax) of released CO 2 to air Step S3 (Identify CVs). 1. Control the 4 equality constraints 2. Identify 2 self-optimizing CVs (Use Exact Local method and select CV set with minimum loss) 4 equality and 2 inequality constraints: 1.stripper top pressure 2.condenser temperature 3.pump pressure of recycle amine 4.cooler temperature 5.CO 2 recovery 80% 6.Reboiler duty < 1393 kW (nominal +20%) 4 levels without steady state effect: absorber, stripper (2), make-up tank Step S2. (a)10 degrees of freedom: 8 valves + 2 pumps *M. Panahi and S. Skogestad, ``Economically efficient operation of CO2 capturing process, Part I: Self-optimizing procedure for selecting the best controlled variables'', Chemical Engineering and Processing, 50, (2011). Disturbances: flue gas flowrate, CO 2 composition in flue gas + active constraints (b) Optimization using Unisim steady-state simulator. Mode I (nominal feedrate): No inequality constraints active Unconstrained degrees of freedom = 10 4 4 = 2 Case study 58 Exact local method * for finding 2 self-optimizing CVs The set with the minimum worst case loss is the best * I.J. Halvorsen, S. Skogestad, J.C. Morud and V. Alstad, Optimal selection of controlled variables Ind. Eng. Chem. Res., 42 (14), (2003) J uu and F, the optimal sensitivity of the measurements with respect to disturbances, are obtained numerically 59 39 candidate CVs - 15 possible tray temperature in absorber - 20 possible tray temperature in stripper - CO 2 recovery in absorber and CO 2 content at the bottom of stripper - Recycle amine flowrate and reboiler duty Best self-optimizing CV set in region I: c 1 = CO 2 recovery (95.26%) c 2 = Temperature tray no. 16 in stripper These CVs are not necessarily the best if new constraints are met Use a bidirectional branch and bound algorithm * for finding the best CVs * V. Kariwala and Y. Cao. Bidirectional Branch and Bound for Controlled Variable Selection, Part II: Exact Local Method for Self-Optimizing Control, Computers & Chemical Engineering, 33(2009), Identify 2 self-optimizing CVs 60 Proposed control structure with given nominal flue gas flowrate (mode I) 61 Mode II: large feedrates of flue gas (+30%) Feedrate flue gas (kmol/hr ) Self-optimizing CVs in region IReboiler duty (kW) Cost (USD/ton) CO 2 recovery % Temperature tray no. 16 C Optimal nominal point % feedrate % feedrate % feedrate %, when reboiler duty saturates (+20%) % feedrate (reoptimized) Saturation of reboiler duty; one unconstrained degree of freedom left Use Maximum gain rule to find the best CV among 37 candidates : Temp. on tray no. 13 in the stripper: largest scaled gain, but tray 16 also OK region I region II max 62 Proposed control structure with large flue gas flowrate (mode II) max 63 Switching policies CO2 plant (supervisory control) Assume operating in region I (unconstrained) with CV=CO2-recovery=95.26% When reach maximum Q: Switch to Q=Qmax (Region II) (obvious) CO2-recovery will then drop below 95.26% When CO2-recovery exceeds 95.26%: Switch back to region I !!! 64 Conclusion optimal operation ALWAYS: 1. Control active constraints and control them tightly!! Good times: Maximize throughput -> tight control of bottleneck 2. Identify self-optimizing CVs for remaining unconstrained degrees of freedom Use offline analysis to find expected operating regions and prepare control system for this! One control policy when prices are low (nominal, unconstrained optimum) Another when prices are high (constrained optimum = bottleneck) ONLY if necessary: consider RTO on top of this 65 Part 1.3 -Inventory (level) control structure Location of throughput manipulator Consistency and radiating rule 66 Outline Skogestad procedure for control structure design I Top Down Step S1: Define operational objective (cost) and constraints Step S2: Identify degrees of freedom and optimize operation for disturbances Step S3: Implementation of optimal operation What to control ? (primary CVs) Control Active constraints + self-optimizing variables Step S4: Where set the production rate? (Inventory control) II Bottom Up Step S5: Regulatory control: What more to control (secondary CVs) ? Step S6: Supervisory control Step S7: Real-time optimization 67 Step S4. Where set production rate? Very important decision that determines the structure of the rest of the inventory control system! May also have important economic implications Link between Top-down (economics) and Bottom-up (stabilization) parts Inventory control is the most important part of stabilizing control Throughput manipulator (TPM) = MV for controlling throughput (production rate, network flow) Where set the production rate = Where locate the TPM? Traditionally: At the feed For maximum production (with small backoff): At the bottleneck 68 TPM (Throughput manipulator) Definition 1. TPM = MV used to control throughput (CV) Definition 2 (Aske and Skogestad, 2009). A TPM is a degree of freedom that affects the network ow and which is not directly or indirectly determined by the control of the individual units, including their inventory control. The TPM is the gas pedal of the process Value of TPM: Usually set by the operator (manual control) Operators are skeptical of giving up this MV to the control system (e.g. MPC) The TPM is usually a flow (or closely related to a flow), e.g. main feedrate, but not always. It can be a setpoint to another control loop Reactor temperature can be a TPM, because it changes the reactor conversion, Pressure of gas product can be a TPM, because it changes the gas product flowrate Fuel cell: Load (current I) is usually the TPM Usually, only one TPM for a plant, but there can be more if there are parallel units or splits into alternative processing routes multiple feeds that do not need to be set in a fixed ratio If we consider only part of the plant then the TPM may be outside our control. throughput is then a disturbance 69 TPM and link to inventory control Liquid inventory: Level control (LC) Sometimes pressure control (PC) Gas inventory: Pressure control (PC) Component inventory: Composition control (CC, XC, AC) 70 Production rate set at inlet : Inventory control in direction of flow* * Required to get local-consistent inventory control TPM 71 Production rate set at outlet: Inventory control opposite flow* TPM * Required to get local-consistent inventory control 72 Production rate set inside process* TPM * Required to get local-consistent inventory control 73 General: Need radiating inventory control around TPM (Georgakis) 74 Consistency of inventory control Consistency (required property): An inventory control system is said to be consistent if the steady- state mass balances (total, components and phases) are satisfied for any part of the process, including the individual units and the overall plant. 75 CONSISTENT? QUIZ 1 76 Local-consistency rule Rule 1. Local-consistency requires that 1. The total inventory (mass) of any part of the process must be locally regulated by its in- or outflows, which implies that at least one flow in or out of any part of the process must depend on the inventory inside that part of the process. 2. For systems with several components, the inventory of each component of any part of the process must be locally regulated by its in- or outflows or by chemical reaction. 3. For systems with several phases, the inventory of each phase of any part of the process must be locally regulated by its in- or outflows or by phase transition. Proof: Mass balances Note: Without the word local one gets the more general consistency rule 77 CONSISTENT? QUIZ 1 78 Local concistency requirement -> Radiation rule (Georgakis) 79 Flow split: May give extra DOF TPM Split: Extra DOF (FC)Flash: No extra DOF 80 Consistent? Local-consistent? Note: Local-consistent is more strict as it implies consistent QUIZ 2 81 Closed system: Must leave one inventory uncontrolled QUIZ 3 82 OK? (Where is production set? NO. Two TPMs (consider overall liquid balance). Solution: Interchange LC and FC on last tank QUIZ 4 TPM 1 TPM 2 83 Example: Separator control Alternative TPM locations Compressor could be replaced by valve if p 1 >p G 84 Alt.1 Alt.2 Alt.3 Alt.4 Similar to original but NOT CONSISTENT (PC not direction of flow) 85 Example: Solid oxide fuel cell Solid oxide electrolyte TPM = current I [A] = disturbance O 2- e-e- CH4 H2O Air PC TC CC x CH4,s (in ratio with CH4 feed to reduce C and CO formation) CH 4 + H 2 O = CO + 3H 2 CO + H 2 O = CO 2 + H 2 2H 2 + O 2- 2H 2 O + 2e - (excess O2) O 2 + 4e - 2O 2- T s = 1070 K (active constraint) x H2 ?? 86 PC TC 87 LOCATION OF SENSORS Location flow sensor (before or after valve or pump): Does not matter from consistency point of view Locate to get best flow measurement Before pump: Beware of cavitation After pump: Beware of noisy measurement Location of pressure sensor (before or after valve, pump or compressor): Important from consistency point of view 88 OK? (Where is production set? NO. Two TPMs (consider overall liquid balance). Solution: Interchange LC and FC on last tank QUIZ 4 TPM 1 TPM 2 89 Where should we place TPM? TPM = MV used to control throughput Traditionally: TPM = Main feed valve (or pump/compressor) Gives inventory control in direction of flow Consider moving TPM if: 1.There is an important CV that could otherwise not be well controlled Dynamic reasons Special case: Max. production important: Locate TPM at process bottleneck* ! TPM can then be used to achieve tight bottleneck control (= achieve max. production) Economics: Max. production is very favorable in sellers marked 2.If placing it at the feed may yield infeasible operation (overfeeding) If snowballing is a problem (accumulation in recycle loop), then consider placing TPM inside recycle loop BUT: Avoid a variable that may (optimally) saturate as TPM (unless it is at bottleneck) Reason: To keep controlling CV=throughput, we would need to reconfigure (move TPM)** **Sigurds general pairing rule (to reduce need for reassigning loops): Pair MV that may (optimally) saturate with CV that may be given up * Bottleneck: Last constraint to become active as we increase throughput -> TPM must be used for bottleneck control 90 QUIZ. Distillation. OK? LV-configuration TPM 91 DB-configuration OK??? TPM 92 cc *But DB-configuration is not recommended! DB-configuration: Level control NOT consistent by itself, but can still work* if we add one (or preferably two) composition/temperature loops TPM cc 93 QUIZ * Keep p p min ** Keep valve fully open * ** 94 95 QUIZ 96 LOCATION OF SENSORS Location flow sensor (before or after valve or pump): Does not matter from consistency point of view Locate to get best flow measurement Before pump: Beware of cavitation After pump: Beware of noise Etc. Location of pressure sensor (before or after valve, pump or compressor): Important from consistency point of view 97 Often optimal: Locate TPM at bottleneck! "A bottleneck is a unit where we reach a constraints which makes further increase in throughput infeasible" If feed is cheap and available: Located TPM at bottleneck (dynamic reasons) If the flow for some time is not at its maximum through the bottleneck, then this loss can never be recovered. 98 Single-loop alternatives for bottleneck control Bottleneck. Want max flow here Alt.1. Feedrate controls bottleneck flow (long loop): FC F max Alt. 2: Feedrate controls lost task (another long loop): F max Alt. 3: Reconfigure all upstream inventory loops: F max Traditional: Manual control of feed rate TPM 99 May move TPM to inside recycle loop to avoid snowballing Example: Eastman esterification process Alcohol recycle Alcohol + water + extractive agent (e) Ester product Reach max mass transfer rate: R increases sharply (snowballing) 10 0 First improvement: Located closer to bottleneck 10 1 Final improvement: Located at bottleneck + TPM is inside snowballing loop Follows Luybens law 1 to avoid snowballing(modified): Avoid having all streams in a recycle system on inventory control 10 2 Where should we place TPM? TPM = MV used to control throughput Traditionally: TPM = Main feed valve (or pump/compressor) Operators like it. Gives inventory control in direction of flow Consider moving TPM if: 1.There is an important CV that could otherwise not be well controlled Dynamic reasons Special case: Max. production important: Locate TPM at process bottleneck* ! Because max. production is very favorable in sellers marked TPM can then be used to achieve tight bottleneck control (= achieve max. flow) 2.If placing it at the feed may yield infeasible operation (overfeeding) If snowballing is a problem (accumulation in recycle loop), then consider placing TPM inside recycle loop BUT: Avoid a variable that may (optimally) saturate as TPM (unless it is at bottleneck) Reason: To keep controlling CV=throughput, we would need to reconfigure (move TPM)** **Sigurds general pairing rule (to reduce need for reassigning loops): Pair MV that may (optimally) saturate with CV that may be given up * Bottleneck: Last constraint to become active as we increase throughput -> TPM must be used for bottleneck control 10 3 A purely top-down approach: Start by controlling all active constaints at max. throughput (may give moving TPM) Economic Plantwide Control Over a Wide Throughput Range: A Systematic Design Procedure Rahul Jagtap, Nitin Kaistha * and Sigurd Skogestad * Step 0:Obtain active constraint regions for the wide throughput range Step 1:Pair loops for tight control of economic CVs at maximum throughput Most important point economically Most active constraints Step 2:Design the inventory (regulatory) control system Step 3:Design loops for taking up additional economic CV control at lower throughputs along with appropriate throughput manipulation strategy Moving TPM Warning: May get complicated, but good economically because of tight control of active constraints 10 4 10 5 Conclusion TPM (production rate manipulator) Think carefully about where to place it! Difficult to undo later 10 6 Part 1.4 Structure of regulatory control layer (PID) Selection of controlled variables (CV2) and pairing with manipulated variables (MV2) Main rule: Control drifting variables and "pair close" Summary: Sigurds rules for plantwide control 10 7 Outline Skogestad procedure for control structure design I Top Down Step S1: Define operational objective (cost) and constraints Step S2: Identify degrees of freedom and optimize operation for disturbances Step S3: Implementation of optimal operation What to control ? (primary CVs) (self-optimizing control) Step S4: Where set the production rate? (Inventory control) II Bottom Up Step S5: Regulatory control: What more to control (secondary CVs) ? Distillation example Step S6: Supervisory control Step S7: Real-time optimization 10 8 II. Bottom-up Determine secondary controlled variables (CV2) and structure (configuration) of control system (pairing, CV2- MV2) A good control configuration is insensitive to parameter changes Regulatory layer 10 9 Step 5. Regulatory control layer Purpose: Stabilize the plant using a simple control configuration (usually: local SISO PID controllers + simple cascades) Enable manual operation (by operators) Main structural decisions: What more should we control? (secondary cvs, CV 2, use of extra measurements) Pairing with manipulated variables (mvs u 2 ) CV 1 CV 2 = ? Regulatory layer 11 0 Objectives regulatory control layer 1.Allow for manual operation 2.Simple decentralized (local) PID controllers that can be tuned on-line 3.Take care of fast control 4.Track setpoint changes from the layer above 5.Local disturbance rejection 6.Stabilization (mathematical sense) 7.Avoid drift (due to disturbances) so system stays in linear region stabilization (practical sense) 8.Allow for slow control in layer above (supervisory control) 9.Make control problem easy as seen from layer above 10.Use easy and robust measurements (pressure, temperature) 11.Simple structure 12.Contribute to overall economic objective (indirect control) 13.Should not need to be changed during operation Regulatory layer 11 1 Stabilizing control: Use inputs MV 2 =u 2 to control drifting variables CV 2 GK CV 2s u2u2 CV 2 CV 1 Key decision: Choice of CV 2 (controlled variable) Also important: Choice of MV2=u 2 (pairing) Primary CV Secondary CV (control for dynamic reasons) Process control: Typical drifting variables (CV2) are Liquid inventories (level) Vapor inventories (pressure) Some temperatures (reactor, distillation column profile) Regulatory layer 11 2 Degrees of freedom unchanged No degrees of freedom lost as setpoints y 2s replace inputs u 2 as new degrees of freedom for control of y 1 GK CV 2s u2u2 CV 2 CV 1 Original DOF New DOF Cascade control: Regulatory layer 11 3 Example: Exothermic reactor (unstable) u = cooling flow (q) CV 1 = composition (c) CV 2 = temperature (T) u TC CV 2 =T CV 2s CC CV 1 =c CV 1s feed product cooling LC L s =max Active constraints (economics): Product composition c + level (max) Regulatory layer 11 4 Details Step 5 (Structure regulatory control layer) (a) What to control (CV2)? 1.Control CV2 that stabilizes the plant (stops drifting) 2.Select CV2 which is easy to control (favorable dynamics) In addition, active constraints (CV1) that require tight control (small backoff) may be assigned to the regulatory layer.* *Comment: usually not necessary with tight control of unconstrained CVs because optimum is usually relatively flat Regulatory layer 11 5 Control CV2 that stabilizes the plant (stops drifting) In practice, control: 1.Levels (inventory liquid) 2.Pressures (inventory gas/vapor) (note: some pressures may be left floating) 3.Inventories of components that may accumulate/deplete inside plant E.g., amine/water depletes in recycle loop in CO2 capture plant E.g., butanol accumulates in methanol-water distillation column E.g., inert N2 accumulates in ammonia reactor recycle 4.Reactor temperature 5.Distillation column profile (one temperature inside column) Stripper/absorber profile does not generally need to be stabilized Regulatory layer 11 6 ( b) Identify pairings = Identify MVs (u 2 ) to control CV2, taking into account : Want local consistency for the inventory control Implies radiating inventory control around given flow Avoid selecting as MVs in the regulatory layer, variables that may optimally saturate at steady-state (active constraint on some region), because this would require either reassigning the regulatory loop (complication penalty), or requiring back-off for the MV variable (economic penalty) Want tight control of important active constraints (to avoid back-off) General rule: pair close (see next slide) Details Step 5b. Regulatory layer 11 7 Step 5b. Main rule: Pair close The response (from input to output) should be fast, large and in one direction. Avoid dead time and inverse responses! Regulatory layer 11 8 Sigurds pairing rule: Pair MV that may (optimally) saturate with CV that may be given up Main reason: Minimizes need for reassigning loops Important: Always feasible (and optimal) to give up a CV when optimal MV saturation occurs. Proof (DOF analysis): When one MV disappears (saturates), then we have one less optimal CV. Failing to follow this rule: Need some fix when MV saturates to remain optimal, like reconfiguration (logic) backoff (loss of optimality) BUT: Rule may be in conflict with other criteria Dynamics (pair close rule) Interactions (avoid negative steady-state RGA rule) If conflict: Use reconfiguration (logic) or go for multivariable constraint control (MPC which may provide built-in logic) Regulatory layer LV TC T s. loop LV TC T s LV TC T s TSTS (a) Normal: Control T using V (b) If V may saturate: Use L 11 9 TC TSTS LV TC T s. loop LV TC T s LV TC T s Normal: Control T using V 12 0 Why simplified configurations? Why control layers? Why not one big multivariable controller? Fundamental: Save on modelling effort Other: easy to understand easy to tune and retune insensitive to model uncertainty possible to design for failure tolerance fewer links reduced computation load Regulatory layer 12 1 Hierarchical/cascade control: Time scale separation With a reasonable time scale separation between the layers (typically by a factor 5 or more in terms of closed-loop response time) we have the following advantages: 1.The stability and performance of the lower (faster) layer (involving y 2 ) is not much influenced by the presence of the upper (slow) layers (involving y 1 ) Reason: The frequency of the disturbance from the upper layer is well inside the bandwidth of the lower layers 2.With the lower (faster) layer in place, the stability and performance of the upper (slower) layers do not depend much on the specific controller settings used in the lower layers Reason: The lower layers only effect frequencies outside the bandwidth of the upper layers 12 2 XCXC TC FC ysys y LsLs TsTs L T z XCXC Cascade control distillation With flow loop + T-loop in top 12 3 QUIZ: What are the benefits of adding a flow controller (inner cascade)? q z qsqs 1.Counteracts nonlinearity in valve, f(z) With fast flow control we can assume q = q s 2.Eliminates effect of disturbances in p1 and p2 Extra measurement y 2 = q 12 4 Summary: Sigurds plantwide control rules Rules for CV-selection: 1. Control active constraints Purity constraint on expensive product is always active (overpurification gives loss): 2. Unconstrained degrees of freedom (if any): Control self-optimizing variables (c). The ideal variable is the gradient of J with respect to the inputs (Ju = dJ/du), which always should be zero, independent of disturbances d, but this variable is rarely available Exception (is available!): Parallel systems (stream split, multiple feed streams/manifold) with given throughput (or given total gas flow, etc.) Should have equal marginal costs Jiu = dJi/du, so Ju = J1u - J2u, etc. Heat exchanger splits: equal Jchke temperatures, JT1 = (T1 Th1)^2/(T1-T0) In practice, one prefers to control single variables, c=Hy (where y are all available measurements and H is a selection matrix), which are easy to measure and control, and which have the following properties: Optimal value for c is almost constant (independent of disturbances): Want small magnitude of dcopt(d)/dd. Variable c is sensitive to changes in input: Want large magnitude of gain=dc/du (this is to reduce effect of measurement error and noise). If the economic loss with single variables is too large, then one may use measurement combinations, c=Hy (where H is a full matrix). 3. Unconstrained degrees of freedom: NEVER try to control a variable that reaches max or min at the optimum (in particular, never control J) Surprisingly, this is a very common mistake, even (especially?) with control experts Rules for inventory control: 1. Use Radiation rule (PC, LC, FC ++) 2. Avoid having all flows in a recycle system on inventory control (this is a restatement of Luybens rule of fixing a flow inside a recycle system to avoid snowballing) A special case is a closed system Rules for pairing: 1. General: Pair close (large gain and small effective time delay) 2. CV1: Sigurds pairing rule: Pair MV that may (optimally) saturate with CV that may be given up 3. CV2 (stabilizing loop): Avoid MV that may saturate 12 5 Part 1 (3h): Plantwide control Introduction to plantwide control (what should we really control?) Part 1.1 Introduction. Objective: Put controllers on flow sheet (make P&ID) Two main objectives for control: Longer-term economics (CV1) and shorter-term stability (CV2) Regulatory (basic) and supervisory (advanced) control layer Part 1.2 Optimal operation (economics) Active constraints Selection of economic controlled variables (CV1). Self-optimizing variables. Part 1.3 -Inventory (level) control structure Location of throughput manipulator Consistency and radiating rule Part 1.4 Structure of regulatory control layer (PID) Selection of controlled variables (CV2) and pairing with manipulated variables (MV2) Main rule: Control drifting variables and "pair close" Summary: Sigurds rules for plantwide control 12 6 Plantwide control. Main references The following paper summarizes the procedure: S. Skogestad, ``Control structure design for complete chemical plants'', Computers and Chemical Engineering, 28 (1-2), (2004). There are many approaches to plantwide control as discussed in the following review paper: T. Larsson and S. Skogestad, ``Plantwide control: A review and a new design procedure'' Modeling, Identification and Control, 21, (2000). The following paper updates the procedure: S. Skogestad, ``Economic plantwide control, Book chapter in V. Kariwala and V.P. Rangaiah (Eds), Plant-Wide Control: Recent Developments and Applications, Wiley (2012). More information: All papers available at: