Plant data

Determination of separation efficiencies from pilot-plant data is also affected by axial dispersion. Neglecting it yields high or values. Literature data for this parameter have usually not been corrected for this effect. 

The Tokuyama Soda single-step catalyst consists of a zirconium phosphate catalyst loaded with 0.1—0.5 wt % paHadium (93—97). Pilot-plant data report (93) that at 140°C, 3 MPa, and a H2 acetone mole ratio of 0.2, the MIBK selectivity is 95% at an acetone conversion of 30%. The reactor product does not contain light methyl substituted methyl pentanes, and allows MIBK recovery in a three-column train with a phase separator between the first and second columns. 

The theoretical models caimot predict flux rates. Plant-design parameters must be obtained from laboratory testing, pilot-plant data, or in the case of estabhshed apphcations, performance of operating plants. 

During process development, a model can be developed as soon as a conceptual flow sheet has been formulated. This model can be updated as more information about the process is obtained. Even at an early stage in the project, the model can be used to assess the preliminary economics of the process and the effect of technological changes on these economics. The model can aid in interpreting pilot-plant data and allows the study of many process alternatives. 

At times, it is possible to build an empirical mathematical model of a process in the form of equations involving all the key variables that enter into the optimisation problem. Such an empirical model may be made from operating plant data or from the case study results of a simulator, in which case the resultant model would be a model of a model. Practically all of the optimisation techniques described can then be appHed to this empirical model. 

Numeric-to-numeric transformations are used as empirical mathematical models where the adaptive characteristics of neural networks learn to map between numeric sets of input-output data. In these modehng apphcations, neural networks are used as an alternative to traditional data regression schemes based on regression of plant data. Backpropagation networks have been widely used for this purpose. 

The development of a dynamic model from plant data is time consuming, typically requiring one to three weeks of around-the-clock plant tests. 

Implementation Issues A critical factor in the successful application of any model-based technique is the availability of a suitaole dynamic model. In typical MPC applications, an empirical model is identified from data acquired during extensive plant tests. The experiments generally consist of a series of bump tests in the manipulated variables. Typically, the manipulated variables are adjusted one at a time and the plant tests require a period of one to three weeks. The step or impulse response coefficients are then calculated using linear-regression techniques such as least-sqiiares methods. However, details concerning the procedures utihzed in the plant tests and subsequent model identification are considered to be proprietary information. The scaling and conditioning of plant data for use in model identification and control calculations can be key factors in the success of the apphcation. 

Results of rigorous calculations and comparison to plant data, when possible, are shown in Figs. 13-95, 13-96, and 13-97. Plant temperatures are in good... 

FIG. 13-95 Comparison of computed stage temperatures with plant data for the example of Fig. 13-94. 

Scaling Up from Laboratory or Pilot-Plant Data. 14-19... 

As stated above, the design of an RDC contactor usually involves the performance of pilot tests due to the large number of factors whicF can influence performance. These pilot plant data must then be scaled-up to Rill commercial size. The following procedure is recommended. 

SOURCE Plant data and calculated design values from Rase, Chemical Reac-... 

Verneuil, VS. Jr., P. Yang, and F. Madron, Banish Bad Plant Data, Chemical Engineeiing Piogiess, October 1992, 45-51. (Gross-error detection overview)... 

The three vertices are the operating plant, the plant data, and the plant model. The plant produces a product. The data and their uncertainties provide the histoiy of plant operation. The model along with values of the model parameters can be used for troubleshooting, fault detection, design, and/or plant control. 

The vertices are connected with hues indicating information flow. Measurements from the plant flow to plant data, where raw measurements are converted to typical engineering units. The plant data information flows via reconciliation, rec tification, and interpretation to the plant model. The results of the model (i.e., troubleshooting, model building, or parameter estimation) are then used to improve plant operation through remedial action, control, and design. 

History The histoiy of a plant forms the basis for fault detection. Fault detection is a monitoring activity to identify deteriorating operations, such as deteriorating instrument readings, catalyst usage, and energy performance. The plant data form a database of historical performance that can be used to identify problems as they form. Monitoring of the measurements and estimated model parameters are typic fault-detection activities. 

Limited Data First, plant data are limited. Unfortunately, those easiest to obtain are not necessarily the most useful. In many cases, the measurements that are absolutely required for accurate model development are unavailable. For those that are available, the sensitivity of the parameter estimate, model evaluation, and/or subsequent conclusion to a particiilar measurement may be very low. Design or control engineers seldom look at model development as the primaiy reason for placing sensors. Further, because equipment is frequently not operated in the intended region, the sensitive locations in space and time have shifted. Finally, because the cost-effectiveness of measurements can be difficult to justify, many plants are underinstru-mented. 

The first two examples show that the interaction of the model parameters and database parameters can lead to inaccurate estimates of the model parameters. Any use of the model outside the operating conditions (temperature, pressures, compositions, etc.) upon which the estimates are based will lead to errors in the extrapolation. These model parameters are effec tively no more than adjustable parameters such as those obtained in linear regression analysis. More comphcated models mav have more subtle interactions. Despite the parameter ties to theoiy, tliey embody not only the uncertainties in the plant data but also the uncertainties in the database. 

Analysts must recognize the above sensitivity when identifying which measurements are required. For example, atypical use of plant data is to estimate the tray efficiency or HTU of a distillation tower. Certain tray compositions are more important than others in providing an estimate of the efficiency. Unfortunately, sensor placement or sample port location are usually not optimal and, consequently, available measurements are, all too often, of less than optimal use. Uncertainty in the resultant model is not minimized. 

The problem with plant data becomes more significant when sampling, instrument, and cahbration errors are accounted for. These errors result in a systematic deviation in the measurements from the actual values. Descriptively, the total error (mean square error) in the measurements is... 

An example adapted from Verneuil, et al. (Verneuil, V.S., P. Yan, and F. Madron, Banish Bad Plant Data, Chemical Engineeiing Progress, October 1992, 45-51) shows the impact of flow measurement error on misinterpretation of the unit operation. The success in interpreting and ultimately improving unit performance depends upon the uncertainty in the measurements. In Fig. 30-14, the materi balance constraint would indicate that S3 = —7, which is unrealistic. However, accounting for the uncertainties in both Si and S9 shows that the value for S3 is —7 28. Without considering uncertainties in the measurements, analysts might conclude that the flows or model contain bias (systematic) error. 

Recommendations Plant measurements should be adjusted to close the constraints of the process. This adjustment shoiild be done on a component or subcomponent (e.g., atomic) basis. The adjustments should be done recognizing (at a minimum) the uncertainty in the measurements. While sophisticated routines have been developed for reconciliation, the vagaries of plant measurements may make them unsuitable in most applications. The routines are no substitute for accurate, precise measurements. They cannot compensate for the uncertainties and hmited information typically found in plant data. 

Perry, S. G., Paumier, J. O., and Burns, D. J., Evaluation of the EPA Complex Terrain Dispersion Model (CTDMPLUS) with the Lovett Power Plant Data Base, pp 189-192 in "Preprints of Seventh Joint Conference on Application of Air Pollution Meteorology with AWMA," Jan. 14-18,1991, New Orleans, American Meteorological Society, Boston, 1991. Bums, D. ]., Perry, S. G., and Cimorelli, A. ]., An advanced screening model for complex terrain applications, pp. 97-100 in "Preprints of Seventh Joint Conference on Application of Air Pollution Meteorology with AWMA," Jan. 14-18, 1991, New Orleans, American Meteorological Society, Boston, 1991. 

For gas/liquid mass transfer, a data point is taken at a particular power level and gas rate. This point is obtained from a similar application or from laboratory or pilot plant data. 

Rigorous calculation results combined with plant data can be used to back calculate column tray efficiencies for... 

Supports the organization, entry, and analysis of plant data and field measurements of fugitive emissions. A menu-driven system. 

One can to outline a general approach for medium selection along with a test sequence applicable to a large group of filter media of the same type. There are three methods of filter media tests laboratory- or bench-scale pilot-unit, and plant tests. The laboratory-scale test is especially rapid and economical, but the results obtained are often not entirely reliable and should only be considered preliminary. Pilot-unit tests provide results that approach plant data. The most reliable results are often obtained from plant trials. 

---