On-line analyzers

On-line analysis is often more expensive and difficult to set up initially but can be more accurate and rehable if performed properly. On-line analyzers can also be used to provide real-time control of a process through a secondary variable such as severity or conversion, as opposed to controlling a primary variable, such as temperature (36,52). 

On-line analyzers Add appropriate on-line analyzers to (1) monitor the status of the process, (2) detect problems at the incipient stage, and (3) take appropriate action to minimize effects of problems while still in initial phase of development. 

Process engineering Process engineer In-line/On-line analyzer integration Process understanding (IPOs, P and IDs, etc.) Facilitate risk assessments Process understanding Process Instrument integration... 

Most on-line analyzers are installed as permanent fixtures. Bearing this in mind, environmental issues associated with temperature and vibration become more critical, and the reqnirement for conformance to... 

Depending on the PAT application, there could be additional, or different, objectives to the experiment. For example, one might not want to build a quantitative regression model for an on-line analyzer, but rather perform an exploratory analysis to identify which of many design variables has the greatest affect on the analyzer response, hi this case, a set of tools called screening designs [22] can be quite useful. 

Complaints regarding the method s performance These could reflect a wide range of method attributes, including accuracy, stability, repeatability and precision. The most useful observations tend to come from the true customers of the method for on-line analyzers these are usually the process operators and engineers that rely on the method for control purposes. 

Improved x-data pretreatment Newly discovered effects in the on-line analyzer data could be more effectively filtered out using a different pretreatment method. More effective pretreatment would then reduce the burden on the modeling method to generate an effective model. 

Figure 12.33 Time series plot of reduced V and Q statistics associated with the application of an analyzer-specific PCA model to on-line analyzer data - covering a period of approximately 4 months.

Step 3 Feasibility assessment Detemiine (via literature search, consultation, etc.) suitable candidate methods and assess the feasibility of these methods. The purpose of this step is to collect enough information to accurately define (or scope ) and prioritize the project. The amount of information required may vary considerably, depending on the magnitude of the proposed project. Significantly less information is required to develop a new method for an existing off-line analyzer than is required to justify spending several million dollars on a major installation of on-line analyzers. (The capabilities and limitations of NIR process analyzers are discussed in Section 15.2.4.)... 

Step 5 Off-line method or analyzer development and validation This step is simply standard analytical chemistry method development. For an analyzer that is to be used off-line, the method development work is generally done in an R D or analytical lab and then the analyzer is moved to where it will be used (QA/ QC lab, at-line manufacturing lab, etc.). For an analyzer that is to be used on-line, it may be possible to calibrate the analyzer off-line in a lab, or in situ in a lab reactor or a semiworks unit, and then move the analyzer to its on-line process location. Often, however, the on-line analyzer will need to be calibrated (or recalibrated) once it is in place (see Step 7). Off-line method development and validation generally includes method development and optimization, identification of appropriate check samples, method validation, and written documentation. Again, the form of the documentation (often called the method or the procedure ) is company-specific, but it typically includes principles behind the method, equipment needed, safety precautions, procedure steps, and validation results (method accuracy, precision, etc.). It is also useful to document here which approaches did not work, for the benefit of future workers. 

Step 7 On-line analyzer calibration Calibrating an analyzer entirely on-line is a last resort, for reasons discussed in Section 15.2.6. It is preferable to do at least the initial work to calibrate the analyzer off-line, or to transfer to the on-line analyzer a method developed on an off-line analyzer or on another on-line analyzer. However, sometimes this is not possible. On-line analyzer calibration is similar to standard analytical chemistry method development, except that getting sufficient variation in sample composition to build a robust calibration model may be difficult or may take a long time. (Ways to address the challenges involved in on-line calibration are discussed in Section 15.2.6 and in Chapter 14.)... 

Occasionally attempts to justify a process analyzer are made on the basis that it will reduce the number of grab samples that need to be taken from the process to be analyzed, and therefore reduce the sample load on the QC lab. This is not usually a good justification for an on-line analyzer for two reasons (1) The reduction in lab analyses rarely offsets the cost of analyzer installation and operation (2) An on-line analyzer relies on high quality lab data for calibration, validation, and long-term model maintenance, so the reference method can never be eliminated entirely. 

Will the measurement be made on-line or off-line For on-line analyzers, is there a grab sample port nearby Is the sample stable once it is withdrawn from the process (See Chapter 3 for more information on sampling issues.) Only an on-line analyzer can enable real-time process control. However, an on-line analyzer implementation will generally be more expensive, more complex, and involve more people in more functions than an off-line one. 

Ideally, an on-line analyzer will be calibrated before it is installed in the process. It may be possible to accomplish this by calibrating it off-line with process grab samples and/or synthetic samples. It may be possible to install the analyzer in a lab-scale reactor, or in a semi-works or pilot plant. It may be possible to transfer to the on-line analyzer a method developed on an off-line analyzer or on another on-line analyzer (e.g. at a different plant site). However, sometimes none of these are possible and the analyzer will need to be calibrated on-line. The challenges of on-line model development (calibration) and validation, as well as approaches to dealing with them, are discussed below. For information related to calibration transfer issues, please see Chapter 12 of this book. 

Note that there are more sources of variability for accuracy than for precision. This is the basis of the common rule of thumb which states that the precision of an on-line analyzer will typically be about ten times better than its accuracy. 

The next generation system used an on-line analyzer to monitor the inhibitor concentration in the rundown continuously. This system would virtually have a vanishingly small probability of out of spec inhibitor. 

Raw Measurement Plot If an on-line analyzer were soning the containers, unknown 3 would have been rejected and the spectrum stored for further evaluation. The spectrum of this unknown is plotted in Figure 4.56 along with the average of the training set spectra for PVC and PET. Unknown 3 has features from both of these classes and, therefore, the classification results are not surprising. 

Process Operations. The operation of a modem chemical plant is typically computer controlled and does not involve any routine operator contact with the feedstock, intermediates, or product (see Process control). There are, however, a few actions the operators may need to take which can involve contact with process materials. Sampling of process streams is one such task. Whereas use of on-line analyzers has substantially reduced the need for operator-collected samples, the latter are necessary to check the on-line analyzer or wherever on-line analyzers are not used. Exposure during sampling can be very high if the sampling line is flushed by running a quantity of a volatile liquid out on the pad. On the other hand, exposure can be very low where the sample is collected in a bomb from a closed loop. Worker care in following prescribed practices is important. 

The differences between laboratory and on-line instruments are huge to the point where often one does not see that they are related. What are some of the commonalities and differences between an on-line analyzer and a laboratory piece of equipment Usually the name is the same. The laboratory is a temperature-controlled, reasonably safe environment with instrumentation operated by trained chemists. In contrast, the process is frequently outdoors so temperature control is what Mother Nature gives you. Many processes also... 

Chemical or physical analysis of materials in the process stream through the use of an in-line or on-line analyzer. 

Most on-line analyzers are installed as permanent fixtures. Bearing this in mind, environmental issues associated with temperature and vibration become more critical, and the requirement for conformance to safety standards may be enforced by the need for system certification. The latter requires an assessment of the working environment, and the potential for fire and/or explosion hazard, relative to the anticipated presence of flammable vapors or gases. Safety hazard and local electrical design code compliance, which includes CE Mark for Europe, Canadian Standards Association (CSA) for Canada,... 

Small details can impact how successfully an installation is viewed. The unforeseen window coatings and associated maintenance were not sufficient for the plant to stop using the analyzer, but they probably reduced the personnel s satisfaction with the technology. Since it is simply not possible to predict and investigate every possible issue during feasibility studies, owners of any kind of on-line analyzer must be prepared to consider the need to make changes to the system over time. 

Another assumption, which becomes apparent when one carefully examines the model (Equation 8.7), is that all of the model error (f) is in the dependent variable (y). There is no provision in the model for errors in the independent variable (x). In PAC, this is equivalent to saying that there is error only in the reference method, and no error in the on-line analyzer responses. Although this is obviously not true, practical experience over the years has shown that linear regression can be very effective in analytical chemistry applications. 

Once a chemometric model is built, and it is used to produce concentration or property values in real time from on-line analyzer profiles, the detection of outliers is a particularly critical task. This is the case for two reasons ... 

Case 3 indicates a means by which a qualitative model can be used to support a quantitative model, and was discussed in Section 8.4.3. Case 4 is a special application for PAC, in which the on-line analyzer can be set up to serve as health monitor for the process. 

Figure 8.39 shows a three-dimensional scatter plot of the first three PC scores obtained from a PCA analysis of 987 calibration spectra that were collected for a specific on-line analyzer calibration project. In this case, cluster analysis was done using the first six PCs (all of which cannot be displayed in the plot ) in order to select a subset of 100 of these samples for calibration. The three-dimensional score plot shows that the selected samples are well distributed among the calibration samples, at least when the first three PCs are considered. 

Step 6 Installation This is where the actual analyzer hardware is physically installed in its final location and started up. For both off-line and on-line analyzers, this may involve... 

---