Values target

As we have mentioned, the particular characterization task considered in this work is to determine attenuation in composite materials. At our hand we have a data acquisition system that can provide us with data from both PE and TT testing. The approach is to treat the attenuation problem as a multivariable regression problem where our target values, y , are the measured attenuation values (at different locations n) and where our input data are the (preprocessed) PE data vectors, u . The problem is to find a function iy = /(ii ), such that i), za jy, based on measured data, the so called training data. 

Figure 9-17. Outline of the procedure for supervised learning The output of the netw/ork is compared with the target value or vector, which yields the error The weights of the network are then adapted to reduce this error.

A key feature of MFC is that future process behavior is predicted using a dynamic model and available measurements. The controller outputs are calculated so as to minimize the difference between the predicted process response and the desired response. At each sampling instant, the control calculations are repeated and the predictions updated based on current measurements. In typical industrial applications, the set point and target values for the MFC calculations are updated using on-hne optimization based on a steady-state model of the process. Constraints on the controlled and manipulated variables can be routinely included in both the MFC and optimization calculations. The extensive MFC literature includes survey articles (Garcia, Frett, and Morari, Automatica, 25, 335, 1989 Richalet, Automatica, 29, 1251, 1993) and books (Frett and Garcia, Fundamental Process Control, Butterworths, Stoneham, Massachusetts, 1988 Soeterboek, Predictive Control—A Unified Approach, Frentice Hall, Englewood Cliffs, New Jersey, 1991). 

Each product is derived from individual pieces of material, individual components and individual assembly processes. The properties of these individual elements have a probability of deviating from the ideal or target value. In turn, the designer defines allowable tolerances on component characteristics in anticipation of the manufacturing variations, but more often than not, with limited knowledge of the cost... 

Modern equipment is frequently eomposed of thousands of eomponents, all of whieh interaet within various toleranees. Failures often arise from a eombination of drift eonditions rather than the failure of a speeifie eomponent (Smith, 1993). For example, typieally an assembly toleranee exists only to limit the degradation of the assembly performanee. Being off target may involve later warranty eosts beeause the produet is more likely to break down than one whieh has a performanee eloser to the target value (Vasseur et al., 1992). This again is related to manufaeturing variation problems, and is more diffieult to prediet, and therefore less likely to be foreseen by the designer (Smith, 1993). 

Further variations may arise from the working of the material during the manufaeturing proeess or from deliberate or unavoidable heat treatment (Bolz, 1981). In general, the undesirable and sometimes uneontrollable faetors that eause a funetional eharaeteristie to deviate from its target value are often ealled noise faetors and are defined below (Kapur, 1993) ... 

We now have a means of predieting the standard deviation multiplier z whieh ean be used in equation 3.2. However, z, the assembly toleranee standard deviation multiplier, must be estimated before this equation ean be satisfied. This is aehieved by setting a eapability requirement. The level of eapability required typieally by industry is Cp = 2 (Harry and Stewart, 1988 O Connor, 1991) whieh equates to 0.002 parts-per-million (ppm) (see Appendix II for a relationship between Cp, Cp and ppm). Note, this value is with no failure severity taken into aeeount, but is a blanket target value diffieult to realize in praetiee. It follows then for the overall assembly proeess eapability for a bilateral assembly toleranee ean be given from equation 3.7 as ... 

The variability or spread of the data does not always take the form of the true Normal distribution of course. There can be skewness in the shape of the distribution curve, this means the distribution is not symmetrical, leading to the distribution appearing lopsided . However, the approach is adequate for distributions which are fairly symmetrical about the tolerance limits. But what about when the distribution mean is not symmetrical about the tolerance limits A second index, Cp, is used to accommodate this shift or drift in the process. It has been estimated that over a very large number of lots produced, the mean could expect to drift about 1.5cr (standard deviations) from the target value or the centre of the tolerance limits and is caused by some problem in the process, for example tooling settings have been altered or a new supplier for the material being processed. 

By calculating where the process is centred (the mean value) and taking this, rather than the target value, it is possible to account for the shift of a distribution which would render Cp inaccurate (see Figure 3). Cp is calculated using the following equation ... 

By using the nearest tolerance limit, which is the tolerance limit physically closest to the distribution mean, the worst case scenario is being used ensuring that overopti-mistic values of process capability are not employed. In Figure 3, a — 1.5cr shift is shown from the target value for a Cp =1.5. Cp is a much more valuable tool than Cp because it can be applied accurately to shifted distributions. As a large percentage of distributions are shifted, Cp is limited in its usefulness. If is applied to a non-shifted Normal distribution, by the nature of its formula it reverts to Cp. 

Since the p is within 5% of the target value of. 29, then use the mean blade velocity, u , = 720 fps, as a final value and proceed with the sizing. 

In Example 10.9(b), if the target values for the outputs are d 2 = 0 and 1 22 = 1, ealeulate new values for the weights and biases using the baek-propagation algorithm. Assume a learning rate of 0.5 with no momentum term. 

When the heat load even locally exceeds 230 kW m - the target values for drum pressure, 160 bar (except for Si02), should be used for all boiler pressures. For feedwater, the recommended values for >67 bar should be... 

TARGET VALUES FOR THERMAL FACTORS AN OVERVIEW OF INTERNATIONAL STANDARDS 373... 

TABLE 6.1 Example of Target Values for a Paper Machine Hall Ventilation... 

Target Values for Acceptable Thermal Environments for Comfort... 

---