Smartlogic

Menu

תגית: Control and Automation

אוטומציה ובקרה – PID Control

Posted on by אביב צברי

אוטומציה ובקרה – PID Control

Between the measuring device and the final control element comes the controller. Its function is to receive themeasured output signal Ym (t) and after comparing it with the set point YSP to produce the actuating signal c (t) in such a way as to return the out put to the desired value YSP. Therefore the input to the controller is the error ε(t) = YSP –Ym (t), while its out put is c (t). The various types of continous feedback controllers differ in the way they relate ε (t) and c (t).

 The best feed back controller is the proportional – integral- derivative controller.

In the industrial practice it is commonly known as the proportional-plus-reset-plus-rate controller.

The actuating signal of this controller is given by the following mathematical equation.

c (t) = Kc ε (t) + Kc/Ʈ1 ∫0 t ε (t) dt + Kc ƮD dε/dt + Cs

Kc = proportional gain of the controller

Cs = controllers bias signal (i.e. its actuating signal when ε =0)

Ʈ1 = integral time constant OR reset time in minutes

TD = derivative time constant in minutes

Proportional = Kc ε (t)

Here the actuating out put c (t) is proportional to the error ε (t) = YSP –Ym (t),

It is clear that larger the gain Kc, the higher the sensitivity of controllers actuating signal to deviations (ε).

ie: Y SP =100, Ym(t) =96, hence error ε = 4.

If Kc=1, then controllers actuating signal, c = 4% to close/open, for TCV.

If Kc=3, then controllers actuating signal, c= 12% to close/open, for TCV

Integral = Kc/Ʈ1 ∫0 t ε (t) dt

The reset time, Ʈ1 is the time needed by the controller to repeat the initial proportional action change in its out put. Reset time, Ʈ1 is an adjustable parameter and some manufacturers do not calibrate their controllers in terms of Ʈ1, but in terms of its reciprocals, 1/ Ʈ1 (repeats per minute), which is known as the reset rate.

ie:

Reset rate=0.1, it means the reset time is = 1/0.1= 10 scans, Hence every 10 scans the controller will add the proportional action change (Kc ε (t)).

In most of the PID's time scans is configurable and called "update loop time" if it equals to 1 then every 10 seconds the controller will add the proportional action

Remark:

The integral term of a controller causes its output to continue changing as long as there is a non-zero error. Often the errors cannot be eliminated quickly, and given enough time they produce larger and larger values for the integral term, which in turn keeps increasing the control action until it is saturated ( ie: the valve is completely open or closed) &  called as integral wind up.

Derivative = Kc ƮD dε/dt

ƮD is the derivative time constant in minutes, with the presence of the derivative term, (dε/dt), the PID controller anticipates what error will be in the immediate future and applies a control action which is proportional to the current rate of change in the error. Due to this property, the derivative control action is referred to as anticipatory control.

Because of the forecasting action of the derivative parameter, the control valve will never "rest" it will always open and close to maintain the Setpoint therefore – this parameter must be used only for fast PID's such as compressed air control, Steam control, ETC

אילן שעיה Ilan Shaya 

,This guide was written by llan shaya, control and automation specialist

סמארט לוג’יק  מתכננת ומבצעת ביצוע פרויקטים הנדסיים בתחום הבקרה האוטומציה, בקרה מפעלית ובקרות תהליך, בקרות מבנה

ולידציה – GAMP-Test Execution

Posted on by אביב צברי

GAMP- Good Automated Manufacturing Practice -Test Execution

After a test is written, reviewed, perhaps rewritten, and finally approved, it is then ready for execution.

Before commencing GAMP  tests using individual Test Cases or Test Scripts, pre-requisites for the test phase should first be verified and recorded. For example:

  • Test environment hardware (HW) (e.g., serial numbers and calibration certificates if required)
  • Test software (SW) (e.g., SW baselines)
  • Data sets
  • User accounts
  • Personnel involved (e.g., documentation of names, positions and sample signatures and initials)
  • Availability of baselined documentation (including, most critically, Test Documentation and Procedures)
  • Where applicable, calibration of critical instrument inputs

Manual Test Execution

Tests should be carried out as follows:

  • Any pre-requisites for the test should first be checked, as indicated above.
  • The test is then executed following the test instructions given within the Test Script.
  • Each test should be run and the test data collected as test results.
  • The tester decides whether the acceptance criteria have been met and records whether the test has passed or failed, and then signs and dates the test results. Sometimes, a third category 'refer for review', or 'conditional pass', or 'pass with observation' is used for cases where the tester feels that an independent opinion is required.
  • Supporting documentary evidence required by the Test Case or Test Script should be collated.
  • If an incident occurs, it should be recorded on a test incident sheet (or within the test incident system) and retained as part of the test record. The key to dealing with incidents during Test Script execution is to accurately record the incident and retain sufficient information to help with future problem solution.
  • It is helpful to maintain a test progress summary for recording overall test results and number of test runs. Depending on the company test policy, this summary may be regarded purely as a status and scheduling tool, or may form part of the post-execution review, and may be included in the Validation Report as a GxP document.
  • After completion of all tests or a group of tests (e.g., at the end of the day), there should be a review of the progress. A review group should assess all tests and incidents.

Possible actions for failed tests and incidents are:

  • Repeat the test.
  • Apply a change via change control and, if necessary, repeat the test.
  • Abandon one, several, or all tests.
  • Review the result, and upgrade it to a 'pass' status (with a record of the rationale for the change in status).

The review group should decide which course of action  to take and what retesting is required, and document the justification for the action(s). In my next post I will elaborate more on: Automated Test Execution (and Computerized Test Management Tools).