Performance estimate a case history

A temporary extra heat load was placed on an olefin plant cooling tower one winter. The operations people asked the question, Will the cooling tower make it next summer with the extra load Of course the tower would deliver the heat removal, so the real question was, What will the cold water temperature be next summer  

The author rated the tower and then used the GPSA nomograph in the previous section titled Performance to determine the next summer s maximum cold water temperature. When the author proudly presented the results the operation people, of course, then asked, Will the plant run at necessary production at your calculated cold water temperature  

To simulate the next summer s condition the plant was run at the desired production rate and two cooling tower fans were turned off. It turned out that the cold water temperature rose to slightly above that predicted for the next summer. A thorough inspection of critical temperatures and the plant s operation indicated that the plant would barely make it the next summer. Process side temperatures were at about the maximum desired, with an occasional high oil temperature alarm on the large machines. 

The decision was made to take a shutdown that spring to clean the water side of critical coolers. The following summer we were glad that we had done so. 

---

They performed an extensive case study to demonstrate the use of automatic history matching to reservoir characterization. For example, if the estimated permeability of a particular zone is unrealistically small compared to geological information, there is a good chance that an impermeable barrier is present. Similarly if the estimated porosity of a zone approaches unrealistically high values, chances are the zone of the reservoir should be expanded beyond its current boundary. 

The course has a total amount of 3 ECTS. The instructor s presentations related to the case histories discussed herein take a total of approximately 1,5 contact hours. Students are further requested to work independently on the Yacambii-Quibor case history (see below), to review the paper and to write a short essay with a summary and with their personal opinions and thoughts about it. This is estimated to take, approximately, an additional 4 hours of the student s time. Furthermore, in this course, the case histories discussed herein also serve as an example from which students can build to broaden the scope of discussion. In particular, students are asked to work independently to prepare and deliver a short presentation (of approximately 10-15 minutes) in relation to other geotechnical failure case histories, where the term failure is employed in a broad sense, to indicate cases in which performance was not as expected during design . 

Seed (1987) first proposed to back-analyze liquefaction failures to estimate liquefied shear strengths. However, as discussed by Olson and Stark (2(X)2), there were a few limitations and inconsistencies in the back-analyses performed by Seed (1987). Seed and Harder (1990) expanded the work by Seed (1987), and their resulting correlation is stiU used in practice. Olson (2001) nearly doubled the number of back-analyzed case histories (from 17 to 33) used by Seed and Harder (1990) and performed kinetics analysis for 10 of the 33 case histories with sufficient documentation. This database has been widely used by numerous researchers (e.g., Idriss and Boulanger 2(X)7 Robertson 2010) for evaluating liquefied shear strengths and strength ratios. 

Since the likelihood integral is intractable for the nonlinear case, most software packages approximate it with the assumption being that the approximation is adequate. History has shown that these approximations perform well under most circumstances in pharmacokinetics and any bias introduced by the approximation is of acceptable magnitude. One test is to fit the final model with a more accurate estimation routine, i.e., fit the final model using Laplacian estimation if the model was developed with FOCE, and compare the two model fits. If the approximation is adequate, the parameters obtained from the two models will not be substantially different. Note that it is inappropriate to compare objective function values under the two estimation models. 

For TRXRD measurements to be meaningful, special attention must be devoted to the questions posed, to the manner in which the experiments designed to answer these questions are performed, and to the proper choice of materials and complete documentation of experimental protocols and results. As an example, in the case of a T-jump experiment, documentation should include direction, rate and magnitude of the jump, initial and final temperatures, and the thermal history of the sample. Ideally, in-sample temperature should be recorded at or close to the interrogating X-ray beam throughout the course of the transition. The manner in which the characteristic transition time is determined must be adequately described and, where possible, an estimate made of the number of molecules transforming per unit of time and volume. A measure of the sensitivity of the detection system to the presence of intermediates or minor and disordered phases would enormously benefit data evaluation. 

Histories of probable poisoning with aflatoxins are described above in relation to the toxic episodes reported in Northwest India and Kenya. The case described below is of interest in that a full medical investigation and follow-up was performed which indicated that the patient remained well up to 14 years post-exposure. In addition, this case is extremely unusual in that a reasonably accurate estimate of exposure to aflatoxin can be made. 

---