Wireless Measurement and Control - AIChE New Orleans

1,778 views

Published on

Wireless Measurement and Control - Opportunities for Diagnostics Process Metrics Inferential Measurements and Eliminating Oscillations

Presented by Greg McMillan on March 15, 2011.

Published in: Education
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,778
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
163
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Wireless Measurement and Control - AIChE New Orleans

  1. 1. Wireless Measurement and Control - Opportunities for Diagnostics, Process Metrics, Inferential Measurements, and Eliminating Oscillations <br />AIChE New Orleans Section Meeting March 15, 2011<br />
  2. 2. Welcome<br />Gregory K. McMillan <br />Greg is a retired Senior Fellow from Solutia/Monsanto and an ISA Fellow. Greg was an adjunct professor in the Washington University Saint Louis Chemical Engineering Department 2001-2004. Presently, Greg contracts as a consultant in DeltaV R&D via CDI Process & Industrial. Greg received the ISA “Kermit Fischer Environmental” Award for pH control in 1991, the Control Magazine “Engineer of the Year” Award for the Process Industry in 1994, was inducted into the Control “Process Automation Hall of Fame” in 2001, was honored by InTech Magazine in 2003 as one of the most influential innovators in automation, and received the ISA Life Achievement Award in 2010. Greg is the author of numerous books on process control, his most recent being Advanced Temperature Measurement and Control. Greg has been the monthly “Control Talk” columnist for Control magazine since 2002. Greg’s expertise is available on the web site: http://www.modelingandcontrol.com/<br />
  3. 3. Top Ten Things You Don’t Want to Hear on a Startup <br />(10) You need the owner to be a little more patient (supplier expert).<br />(9) Don’t bother with a checkout - just light it up! What is the worst that can happen?<br />(8) We didn’t do any simulation or testing. We decided that would spoil the adventure.<br />(7) I don’t understand. It fit fine on the drawing.<br />(6) Cool - This is my first time in a real plant (supplier expert).<br />(5) I tried to open the valve and nothing happened. Wait! The same valve on the other reactor just opened.<br />(4) Should the Variable Frequency Drive smoke like that?<br />(3) I don’t understand. I am sure I left all your tools and radios in a box right here.<br />(2) The CEO is holding on a phone for you.<br />(1) Boom!!! WHAT was that?!?!<br />Source: “Final Word on Instrument Upgrade Projects”, Control Talk, Control, Dec 2010<br />http://www.controlglobal.com/articles/2010/InstrumentProjects1012.html<br />
  4. 4. ISA Automation Week - Oct 17-20<br />Process Automation <br />Hall of Fame Speakers<br />Charlie Cutler<br />Bela Liptak<br />Russ Rhinehart<br />Greg McMillan<br />Terry Tolliver<br />
  5. 5. Advances in Smart Measurements<br />Technological advances in sensing element technology<br />Integration of multiple measurements<br />Compensation of application and installation effects<br />Online device diagnostics<br />Digital signals with embedded extensive user selected information<br />Wireless communication<br />The out of the box accuracy of modern industrial instrumentation has improved by an order <br />of magnitude. Consider the most common measurement device, the differential pressure <br />transmitter (DP). The 0.25% accuracy of an analog electronic DP has improved to 0.025% <br />accuracy for a smart microprocessor based DP. Furthermore, the analog DP accuracy often <br />deteriorated to 2% when it was moved from the nice bench top setting to service outdoors in <br />a nasty process with all its non-ideal effects of installation, process, and ambient effects [1][16]. <br />A smart DP with its integrated compensation for non-ideal effects will stay close to its inherent <br />0.025% accuracy. Additionally a smart DP takes 10 years to drift as much as the analog DP <br />did in 1 year.<br />
  6. 6. Smart Transmitter Auxiliary Variables<br />The availability of auxiliary process variables in a smart wireless pH transmitter, provide early indicators of performance problems. The use of these variables by online data analytics tools could detect abnormal conditions and predict sensor life.<br />
  7. 7. Smart Transmitter Diagnostic Messages<br />“Fix Now” and “Fix Soon” alerts are provided along with common causes and recommended actions<br />
  8. 8. Wireless Opportunities<br />Wireless temperatures and differential pressures for packed absorber and distillation column hot spot and flow distribution analysis and control<br />Wireless temperatures for finding the column control point with the largest and most symmetrical change in temperature with reflux/feed or steam/feed ratio<br />Wireless temperatures for heat transfer coefficient metrics (fouling and frosting)<br />Wireless temperatures and flows for measurement and control of reaction rate and crystallization rate from heat transfer (BTU/hr measurement and control)<br />Wireless temperatures and differential pressures for fluidized bed reactor hot spot and flow distribution analysis and control<br />Wireless temperatures and flows to debottleneck coolant systems<br />Wireless pressures to debottleneck piping systems, monitor process filter operation, and track down the direction and source of pressure disturbances<br />Wireless pressures to compute installed control valve characteristic (flow versus stroke) and variable speed drive installed characteristic (flow versus speed) <br />Wireless instrumentation to increase the mobility, flexibility, and maintainability of lab and pilot plant experiments. <br />Wireless pH and conductivity measurements for <br />(1) Selecting the best sensor technology for a wide range of process conditions(2) Eliminating measurement noise(3) Predicting sensor demise(4) Developing process temperature compensation(5) Developing inferential measurements of process concentrations(6) Finding the optimum sensor location <br />http://www.isa.org/InTechTemplate.cfm?template=/ContentManagement/ContentDisplay.cfm&ContentID=80886<br />
  9. 9. WirelessHART Network Topology<br /><ul><li>Wireless Field Devices
  10. 10. Relatively simple - Obeys Network Manager
  11. 11. All devices are full-function (e.g., must route)
  12. 12. Adapters
  13. 13. Provide access to existing HART-enabled Field Devices
  14. 14. Fully Documented, well defined requirements
  15. 15. Gateway and Access Points
  16. 16. Allows access to WirelessHART Network from the Process Automation Network
  17. 17. Gateways can offer multiple Access Points for increased Bandwidth and Reliability
  18. 18. Caches measurement and control values
  19. 19. Directly Supports WirelessHART Adapters
  20. 20. Seamless access from existing HART Applications
  21. 21. Network Manager
  22. 22. Manages communication bandwidth and routing
  23. 23. Redundant Network Managers supported
  24. 24. Often embedded in Gateway
  25. 25. Critical to performance of the network
  26. 26. Handheld
  27. 27. Supports direct communication to field device
  28. 28. For security, one hop only communication</li></li></ul><li>WirelessHART Features<br />Wireless transmitters provide nonintrusive replacement and diagnostics<br />Wireless transmitters automatically communicate alerts based on smart diagnostics without interrogation from an automated maintenance system<br />Wireless transmitters eliminate the questions of wiring integrity and termination<br />Wireless transmitters eliminate ground loops that are difficult to track down<br />Network manager optimizes routing to maximize reliability and performance<br />Network manager maximizes signal strength and battery life by minimizing the number of hops and preferably using routers and main (line) powered devices<br />Network manager minimizes interference by channel hopping and blacklisting<br />The standard WirelessHART capability of exception reporting via a resolution setting helps to increase battery life<br />WirelessHART control solution, keeps control execution times fast but a new value is communicated as scheduled only if the change in the measurement exceeds the resolution or the elapsed time exceeds the refresh time<br />PIDPLUS and new communication rules can reduce communications by 96%<br />
  29. 29. Broadley-James Corporation Bioreactor Setup<br /><ul><li>Hyclone 100 liter Single Use Bioreactor (SUB)
  30. 30. Rosemount WirelessHART gateway and transmitters for measurement and control of pH and temperature. (pressure monitored)
  31. 31. BioNet lab optimized control system based on DeltaV </li></li></ul><li> Elimination of Ground Noise by Wireless pH <br />Incredibly tight pH control via 0.001 pH wireless resolution <br />setting still reduced the number of communications by 60%<br />Temperature compensated wireless pH controlling at 6.9 pH set point<br />Wired pH ground noise spike <br />
  32. 32. Wireless Bioreactor Adaptive pH Loop Test<br />
  33. 33. University of Texas Pilot Plant for CO2 Research<br /><ul><li>The Separations Research Program was established at the J.J. Pickle Research Campus in 1984
  34. 34. This cooperative industry/university program performs fundamental research of interest to chemical, biotechnological, petroleum refining, gas processing, pharmaceutical, and food companies.
  35. 35. CO2 removal from stack gas is a focus project for which WirelessHART transmitters are being installed</li></li></ul><li> Wireless Conductivity and pH Lab Setup<br />In the UT lab that supports the pilot plant, solvent concentration and loading were varied and the conductivity and pH were wirelessly communicated to the DCS in the control room <br />
  36. 36. Effect of Ions on Conductivity<br />Conductivity measures the concentration and mobility of ions. Plots of conductivity versus ion concentration will increase from zero concentration to a maximum as the number of ions in solution increases. The conductivity then falls off to the right of the maximum as the ions get crowded and start to interact or associate (group) reducing the ion mobility. <br />
  37. 37. 40 oC<br />Conductivity (milliSiemens/cm)<br />30 oC<br />20 oC<br /> Effect of Solvent on Conductivity<br />Conductivity in the operating range of 25% to 30% by weight solvent is relatively unaffected by solvent concentration <br />
  38. 38. Conductivity (milliSiemens/cm)<br />40 oC<br />30 oC<br />20 oC<br /> Effect of CO2 Load on Conductivity<br />Conductivity shows good sensitivity to CO2 loading that can be fitted by a straight line whose slope depends upon temperature above 30 oC<br />
  39. 39. Effect of Solvent on pH<br />pH measures the activity of the hydrogen ion, which is the ion concentration multiplied by an activity coefficient. An increase in solvent concentration increases the pH by a decrease in the activity coefficient and a decrease in the ion concentration from a decrease per the water dissociation constant.<br />pH is also affected by CO2 weight percent since pH changes with the concentration of carbonic acid.<br />Density measurements by Micromotion meters provide an accurate inference of CO2 weight percent. <br />
  40. 40. Effect of MEA Solvent on pH<br />
  41. 41. Effect of PZ Solvent on pH<br />
  42. 42. +<br />+<br />Elapsed <br />Time<br />+<br />+<br />Elapsed <br />Time<br /> Enhanced PID Algorithm for Wireless (PIDPlus) <br /><ul><li>PID integral mode is restructured to provide integral action to match the process response in the elapsed time (reset time set equal to process time constant)
  43. 43. PID derivative mode is modified to compute a rate of change over the elapsed time from the last new measurement value
  44. 44. PID reset and rate action are only computed when there is a new value
  45. 45. If transmitter damping is set to make noise amplitude less than sensitivity limit, valve packing and battery life is dramatically improved
  46. 46. Enhancement compensates for measurement sample time suppressing oscillations and enabling a smooth recovery from a loss in communications further extending packing -battery life</li></ul>TD<br />Kc<br />TD<br />Kc<br />Link to PIDPlus White Paper<br />http://www2.emersonprocess.com/siteadmincenter/PM%20DeltaV%20Documents/<br />Whitepapers/WP_DeltaV%20PID%20Enhancements%20for%20Wireless.pdf<br />
  47. 47. Flow Setpoint Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  48. 48. Flow Load Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  49. 49. Flow Failure Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  50. 50. pH Setpoint Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  51. 51. pH Load Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  52. 52. pH Failure Response - PIDPlus vs. Traditional PID <br />Enhanced PID <br />Sensor PV<br />Traditional PID <br />Sensor PV<br />
  53. 53. PIDPlus Benefits Extend Far Beyond Wireless - 1 <br />The PID enhancement for wireless (PIDPlus) offers an improvement wherever there is an update time in the loop. In the broadest sense, an update time can range from seconds (wireless updates and valve or measurement sensitivity limits) to hours (failures in communication, valve, or measurement). Some of the sources of update time are:<br />Wireless measurement default update rate for periodic reporting (default update rate)<br />Wireless measurement trigger level for exception reporting (trigger level)<br />Wireless communication failure <br />Broken pH electrode glass or lead wires (failure point is about 7 pH)<br />Large valve operating on upper part of installed characteristic (low sensitivity)<br />Valve with backlash (deadband) and stick-slip (resolution and sensitivity limit)<br />Operating at split range point (discontinuity of no response to abrupt response)<br />Valve with solids, high temperature, or sticky fluid that causes plugging or seizing <br />Plugged impulse lines<br />Analyzer sample processing delay and analysis or multiplex cycle time<br />Analyzer resolution and sensitivity limit<br />
  54. 54. PIDPlus Benefits Extend Far Beyond Wireless - 2 <br />The PIDPlus executes when there a change in setpoint, feedforward, or remote output to provide an immediate reaction based on PID structure<br />The improvement in control by the PIDPlus is most noticeable as the update time becomes much larger than the 63% process response time (defined in the white paper as the sum of the process deadtime and time constant). When the update time becomes 4 times larger than this 63% process response time that roughly corresponds to the 98% response time frequently cited in the literature, the feedforward and controller gains can be set to provide a complete correction for changes in the measurement and setpoint.<br />Helps ignore inverse response and errors in feedforward timing<br />Helps ignore discontinuity (e.g. steam shock) at split range point<br />Helps extend packing life by reducing oscillations and hence valve travel<br />Since the PIDPlus can be set to execute only upon a significant change in user valve position, the PIDPlus as a valve position controller offers less interaction and cycling for optimization of unit operations by increasing reactor feed, column feed or increasing refrigeration unit temperature, or decreasing compressor pressure till feed, vent, coolant, and/or steam, valves are at maximum good throttle position.<br />Website entries on Enhanced PID Benefits<br />http://www.modelingandcontrol.com/2010/08/wireless_pid_benefits_extend_t.html<br />http://www.modelingandcontrol.com/2010/10/enhanced_pid_for_wireless_elim.html<br />http://www.modelingandcontrol.com/2010/11/a_delay_of_any_sorts.html<br />
  55. 55. Enhanced PID<br />Traditional PID<br />PID PV<br />PID Output<br />Limit Cycles from Valve Stick-Slip<br /> Enhanced PID Can Eliminate Valve Limit Cycles<br />
  56. 56. Enhanced PID Can Maximize Production Rate<br />ZC-3<br />TC-3<br />ZC-4<br /><<br />maximum <br />production<br />rate<br />CTW<br />vent<br />CAS<br />ZC-2<br />PC-1<br />condenser<br />CAS<br />FC-1<br /><<br />TT<br />feed A<br />FT<br />PT<br />CAS<br />FC-2<br />RC-1<br />ratio<br />CAS<br />feed B<br />TC-1<br />TT<br />FT<br />ZC-1<br />CAS<br />coolant<br />makeup<br />TC-2<br />Valve Position Controllers (VPC)<br />ZC-1,2,3,4 are enhanced PID with <br />directional output velocity limiting <br />and position noise band set to reduce<br />interactions and limit cycling<br />TT<br />reactor<br />product<br />
  57. 57. Self-Regulating Process Open Loop Response<br />Response to change in controller output with controller in manual<br />% Controlled Variable (CV) <br />or<br />% Controller Output (CO)<br /> CV<br />Kp = DCV / DCO <br />Self-regulating process gain (%/%)<br />CO<br />Maximum speed<br />in 4 deadtimes<br /> is critical speed<br />DCV<br />0.63*DCV<br />DCO<br />Time (seconds)<br />qo<br />tp2<br />to<br />or<br />observed <br />total loop<br />deadtime<br />Self-regulating process<br />open loop<br />negative feedback time constant<br />
  58. 58. Response to change in controller output with controller in manual<br />% Controlled Variable (CV) <br />or<br />% Controller Output (CO)<br />CV<br />Ki = { [ CV2/ Dt2 ] - [ CV1/ Dt1 ] } / DCO<br />Integrating process gain (%/sec/%)<br />CO<br />DCO<br /> ramp rate is<br />DCV2/ Dt2<br /> ramp rate is<br />DCV1 / Dt1<br />Time (seconds)<br />qo<br />observed <br />total loop<br />deadtime<br /> Integrating Process Open Loop Response<br />Maximum speed<br />in 4 deadtimes<br /> is critical speed<br />Wireless Trigger Level > noise<br />Wireless <br />Default<br />Update <br />Rate<br />
  59. 59. t’<br />t’<br />o<br />p2<br /> Runaway Process Open Loop Response<br />Response to change in controller output with controller in manual<br />% Controlled Variable (CV) <br />or<br />% Controller Output (CO)<br />Kp = DCV / DCO <br />Runaway process gain (%/%)<br />Acceleration<br />For safety reasons, tests are <br />terminated after 4 deadtimes<br />1.72*DCV<br />Maximum speed<br />in 4 deadtimes<br /> is critical speed<br />DCV<br />DCO<br />Noise Band<br />Time (seconds)<br />q<br />or<br />observed <br />total loop<br />deadtime<br />o<br />runaway process<br />open loop<br />positive feedback time constant<br />
  60. 60. Kc<br />Ti<br />Td<br /> Loop Block Diagram (First Order Approximation)<br />Delay<br />Gain<br />Lag<br />qL<br />Delay <=> Dead Time<br />Lag <=>Time Constant<br />KL<br />tL<br />DV<br />Load Upset<br />Delay<br />Delay<br />Lag<br />Lag<br />Gain<br />Lag<br />Delay<br />Gain<br />qv<br />tp1<br />qp2<br />tp2<br />tv<br />Kpv<br />qp1<br />Kmv<br />MV<br />Process<br />Valve<br />PV<br />Hopefullytp2is the largest lag in the loop<br />For integrating processes: Ki = Kmv * (Kpv / tp2 ) * Kcv<br />100% / span<br />CO<br />%<br />Local<br />Set Point<br />PID<br />½ of Wireless Default Update Rate<br />%<br />CV<br />%<br />Delay<br />Lag<br />Delay<br />Lag<br />Lag<br />Gain<br />tc1<br />tm2<br />qm2<br />tm1<br />qm1<br />Kcv<br />qc<br />tc2<br />Lag<br />Delay<br />Controller<br />Measurement<br />First Order Approximation: qo @ qv + qp1 + qp2 + qm1 + qm2 + qc + tv + tp1 + tm1 + tm2 + tc1 + tc2<br />(set by automation system design for flow, pressure, level, speed, surge, and static mixer pH control)<br />
  61. 61. Ultimate Limit to Loop Performance<br />Peak error is proportional to the ratio of loop deadtime to 63% response time<br />(Important to prevent SIS trips, relief device activation, surge prevention, and RCRA pH violations)<br />Total loop deadtime<br />that is often set by<br />automation design <br />½ of Wireless Default Update Rate<br />is additional deadtime<br />Largest lag in loop<br />that is ideally set by<br />large process volume<br />Integrated error is proportional to the ratio of loop deadtime squared to 63% response time<br />(Important to minimize quantity of product off-spec and total energy and raw material use)<br />
  62. 62. Practical Limit to Loop Performance<br />Peak error decreases as the controller gain increases but is essentially the <br />open loop error for systems when total deadtime >> process time constant<br />Open loop error for<br />fastest and largest<br />load disturbance<br />Integrated error decreases as the controller gain increases and reset time decreases <br />but is essentially the open loop error multiplied by the reset time plus signal <br />delays and lags for systems when total deadtime >> process time constant<br />Rise time (time to reach a new setpoint) is inversely proportional to controller gain<br />
  63. 63. Fastest Controller Tuning (reaction curve method*)<br />* - Ziegler Nichols method closed loop modified<br />to be more robust and less oscillatory<br />For self-regulating processes: <br />Near integrator (tp2 >> qo): <br />Deadtime dominant (tp2 << qo): <br />1.0 for Enhanced PID if Wireless Default <br />Update Rate > Process Response Time ! <br />For integrating processes: <br />For runaway processes:<br /> Near integrator (t’p2 >> qo): <br />These tuning equations provide maximum<br />disturbance rejection but will cause<br />some overshoot of setpoint response <br />
  64. 64. Nomenclature<br /><ul><li>DCV = change in controlled variable (%)
  65. 65. DCO = change in controller output (%)
  66. 66. Kc= controller gain (dimensionless)
  67. 67. Ki= integrating process gain (%/sec/% or 1/sec)
  68. 68. Kp= process gain (dimensionless) also known as open loop gain
  69. 69. DV = disturbance variable (engineering units)
  70. 70. MV = manipulated variable (engineering units)
  71. 71. PV = process variable (engineering units)
  72. 72. DSP = change in setpoint (engineering units)
  73. 73. SPff= setpoint feedforward (engineering units)
  74. 74. Dt = change in time (sec)
  75. 75. Dtx = execution or update time (sec)
  76. 76. qo = total loop dead time (sec)
  77. 77. tf = filter time constant or well mixed volume residence time (sec)
  78. 78. tm = measurement time constant (sec)
  79. 79. tp2 = primary (large) self-regulating process time constant (sec)
  80. 80. t’p2 = primary (large) runaway process time constant (sec)
  81. 81. tp1 = secondary (small) process time constant (sec)
  82. 82. Ti= integral (reset) time setting (sec/repeat)
  83. 83. Td= derivative (rate) time setting (sec)
  84. 84. Tr= rise time for setpoint change (sec)
  85. 85. to= oscillation period (sec)
  86. 86. l = Lambda (closed loop time constant or arrest time) (sec)
  87. 87. lf = Lambda factor (ratio of closed to open loop time constant or arrest time)</li>

×