QAL 3
QAL 3 functionality is available EmiDAS Continuous Emissions Monitoring Software
- Overview
- EmiDAS Control Charts
- EmiDAS QAL3 Table
- CUSUM Chart
- CUSUM Chart Example
- Shewhart Chart
- Shewhart Chart Example
- EWMA Chart
- EWMA Chart Example
Overview
The use of QAL 3 (Quality Assurance Level 3) Control Charts to ensure the correct functioning of remote analysers is a very useful tool used by operators that come under EN14181 / IED.
The charts provide an ongoing analysis of instrument zero and span calibration checks to ensure correct instrument operation and to ensure that any drift of measurement is within the specified drift allowable for the instrument, this drift allowance is derived from the QAL 1 data for the instrument being monitored.
Periodically, instruments will go through a zero and span calibration check where zero and span calibration gases are briefly applied to the analyser. No actual calibration is carried out, it is the response and accuracy of the instrument that is being monitored.
Typically an automatic calibration check may be carried out weekly, EmiDAS then records the resultant readings for zero and span for each measurement, the data can then be displayed in tabulated form and QAL3 Control Charts generated to give the user a quick and easy overview of system performance.
The baseline value is set by the user and corresponds to the concentration of the calibration gases being applied to the instrument during the QAL3 check. Errors in the calibration check or instrument drift which can be caused by a number of factors are easily identified by the user and the cause can then be investigated.
Control Charts
This table summarises the types of QAL3 Control Charts:-
EmiDAS QAL3 Table
Tabulated QAL3 Data for 9 Weekly Calibration Checks for Nitric Oxide (NO) Measurement
Some parts of the EmiDAS tabulated data table can be edited by the user and comments added for specific calibration checks or actual calibrations of the instrument.
This is an effective and easy way to track previous calibration information and results, and to identify past service or maintenance activities on the system.
The AMS reset column identifies when the instrument calibration has been reset (i.e. the instrument has been re-calibrated to certified zero and span calibration gases).
Note that the instrument should only be re-calibrated if the control charts have alerted the user to a problem and the cause of instrument drift or failure has been identified and rectified as per the manufacturer’s recommendations by a qualified instrumentation engineer.
CUSUM Chart
A CUSUM chart is a time-weighted control chart that displays the cumulative sums (CUSUMs) of the deviations of each measurement value from the baseline value.
The baseline value is set to the concentration of the audit gas being applied to the measurement.
CUSUM control charts contain control limits, these limits define the boundaries of expected common cause (random) variation around the baseline and are identified by the red lines on the x-axis in the next example.
Because it is cumulative, even minor drifting of the measurement will lead to steadily increasing or decreasing cumulative deviation values. Therefore, this chart is especially useful in detecting slow shifts away from the baseline value due to factors such as:-
- electrochemical cell degradation
- gas cell leakage in analysers using Gas Cell correlation
- calibration problems
- degradation of span calibration gases and so on.
If a trend develops upward or downward, it indicates that the measurement mean has shifted, and you should look for special causes. Sites operating under IED (EN14181) have strict limits imposed under which only a certain amount of consecutive measurement points are allowed to fall on one side of the baseline. If this occurs it indicates a bias in the measurement and appropriate action should be taken by the operator. The next CUSUM chart identifies a slight negative bias over the period when the 9 measurements were taken.
CUSUM Chart Example
QAL3 CUSUM Chart for above SPAN check data for Nitric Oxide (NO) measurements
Shewhart Chart
Shewhart charts are quite good at detecting large changes in the measurement mean or variance.
The control charts use the audit gas concentration as the baseline, this value is entered during set up of the system and when new audit gases are used to replace suspect or expired ones.
Because the relative distances from the baseline are taken into consideration, Shewhart charts are a more sensitive way of detecting whether the observed variation is due to common or special causes. Shewhart control charts also contain control limits, these limits define the boundaries of expected common cause (random) variation around the baseline.
If a process is stable (i.e. data points are randomly arranged within the control limits) then this indicates that the instrument is in control and operating within its specifications and no action is necessary.
Readings outside the control limits or consistent runs of measurement points on one side of the baseline should be interpreted as a signal of some change in the measurement. When such signals exist, action should be taken to identify and eliminate the cause. When no such signals are present, no changes to the process control variables (i.e. "tampering") are necessary or desirable.
Sites operating under IED (EN14181) have strict limits imposed under which only a certain amount of consecutive measurement points are allowed to fall on one side of the baseline. If this occurs it indicates a bias in the measurement and appropriate action should be taken by the operator.
Shewhart Chart Example
Shewhart Chart for above SPAN check data for Nitric Oxide (NO) measurements
EWMA Chart
What is an EWMA chart?
An EWMA control chart is a time-weighted control chart that plots the exponentially weighted moving averages.
EWMA charts are especially suited to monitor processes that exhibit a drifting mean over time, or for detecting small shifts in instrument readings.
What are plotted points based on?
The plot points are based on individual readings. When you plot individual readings, exponentially weighted moving averages are calculated from those individual observations.
By default, the moving range is of length 1, since consecutive points have the highest chance of being alike. You can also change the length of the moving range.
Guidelines for selecting the weight for a EWMA chart
The calculations for each point on a EWMA chart include information from the previous points. The points are weighted based on a user-specified weighting factor. An advantage of EWMA charts is that they are not greatly affected when a small or large value enters the calculation.
By changing the weight (also called lambda or λ) and the width of the control limits, you can detect a shift of almost any size. Because of this, EWMA charts are often used to detect small shifts away from the target. Usually, you use smaller weights to detect smaller shifts. Monitoring data can be selected and exported in real-time for use in 3rd party systems at time periods of the user’s choice.
EWMA Chart Example
EWMA Chart for above SPAN check data for Nitric Oxide (NO) measurements.
