Outputs measured values

In order to calibrate and adjust the scale standards are used, which associate a precisely defined output measurement value to a measured quantity. 

Compensation of the measured value for conditions within the instrument, such as compensating the output of a pressure transmitter for the temperature within the transmitter. Smart transmitters are much less affected by temperature and pressure variations than conventional transmitters. 

Knowledge of the output or actual value This must be measured by a feedbaek sensor, again in a form suitable for the eontroller to understand. In addition, the sensor must have the neeessary resolution and dynamie response so that the measured value has the aeeuraey required from the performanee speeifieation. 

A simple control system, or loop, is illustrated in Figure 6.1. The temperature T, of the water at Y is measured by means of a thermocouple, the output of which is fed to a controller mechanism. The latter can be divided into two sections (normally housed in the same unit). In the first (the comparator), the measured value (To) is compared with the desired value (Td) to produce an error (e). where ... 

The graphioal output from the computer shows the process flowsheet, with several separation units and projeoted equipment and operating costs. It also flags information that is uncertain because it had to use thermodynamio data extrapolated from measured values. At the engineer s request, the oomputer shows several alternative flowsheets it had considered, indicates their projected costs, and tells why it eliminated eaoh of them. Some of the flowsheets were eliminated because of high cost, others beoause they were oonsidered unsafe, others because the startup procedures would be difficult, and still others because they were based on uncertain extrapolation of experimental data. 

As mentioned above, the random character of the input and output variables are of importance with regard to the calibration model and its estimation by calculus of regression. Because of the different character of the analytical quantity x in the calibration step (no random variables but fixed variables which are selected deliberately) and in the evaluation step (random variables like the measured values), the closed loop of Fig. 6.1 does not correctly describe the situation. Instead of this, a linear progress as shown in Fig. 6.2 takes place. 

The output layer, producing the output quantities (the measured values y )... 

This curve passes through the origin because if there is no temperature difference between the junctions there is no potential generated. It rises as a near linear curve over the range of commonly measured values. The output voltage is small (0.04-0.06 mV. °C 1) and so signal amplification is often needed. 

In view of the conflict between the reliability and the cost of adding more hardware, it is sensible to attempt to use the dissimilar measured values together to cross check each other, rather than replicating each hardware individually. This is the concept of analytical i.e. functional) redundancy which uses redundant analytical (or functional) relationships between various measured variables of the monitored process e.g., inputs/outputs, out-puts/outputs and inputs/inputs). Figure 3 illustrates the hardware and analytical redundancy concepts. 

It is usually not possible or practical to control a given system input at an exact level in practice, most inputs are controlled around set levels within certain factor tolerances. Thus, controlled system inputs exhibit some variation. Variation is also observed in the levels of otherwise constant system outputs, either because of instabilities within the system itself (e.g., a system involving an inherently random process, such as nuclear decay) or because of the transformation of variations in the system inputs into variations of the system output (see Figures 2.17 and 2.18) or because of variations in an external measurement system that is used to measure the levels of the outputs. This latter source of apparent system variation also applies to measured values of system inputs. 

The sensor output can be used to test the validity of processing models such as the Loos-Springer model [30]. Sensor measured values of t] can be compared with the Loos-Springer model predictions. Figure 4.14 is a comparison of the model s predictions and the measured values at the sixty-fourth ply. Agreement in the viscosity s time dependence and magnitude with the predictions of models is essential if the model is to be verified and used with confidence. 

Measurements on the breadboard configuration were made at several output capacitor values. The results are shown in Fig. 4.74. The SPICE model was simulated at V5n = V DC (85 volts AC), with the... 

The circuit was constructed in the laboratory. The values of Rl, R2, Cl, and C2 are the actual measured values of the components used in the circuit. A Tektronix TDS 340A digital real-time oscilloscope was used to record the output data, as shown in Fig. 8.2. The duty cycle was calculated to be 66.7%, and the output voltages oscillated from 5 V to 10 V. The simulated data, from IsSpice, Micro-Cap V, and PSpice, are shown in Figs. 8.3, 8.4, and 8.5, respectively. Table 8.1 illustrates the variances in the simulated output data. 

The calibration of more sophisticated apparatus has also been fated with additional problems arising from the difficulty of directly reaching the actual measured values. The software which so efficiently transforms the data can give rise to concern as to what has happened between the transducer and the final output. As mentioned earlier, the software itself requires verification which is often not an easy task. 

A certain mass flowmeter (see Section 6.2.3) was tested (calibrated) by comparing the readings given by the instrument G with true (known) values GT of the flow of a gas as measured by the instrument in a 0.15 m ID pipeline. True and measured values are compared in Fig. 6.61 and Table 6.17. Estimate the errors in the flowmeter due to bias and imprecision. Assume that variations in the input and output of the instrument are normally distributed. 

Obtaining these desired output parameters involves two steps. In the first step, tests are conducted on the target system in order to obtain a set of measured values for macroscopic properties such as temperature, pH, ultrasonic attenuation, etc. In the second, the previously measured data are analysed in order to compute the desired microscopic properties. Such analysis requires three tools a model dispersion, a prediction theory and an analysis engine. 

The normal operation of the furnace at this heal output is specified by a primary air input between approximately 15 and 18 10 mVs. Caused by the high nitrogen content of the chipboards, the NO, emissions normally are in the range between 250 and 400 mg/Nm. The average value of all the measurements is nearly 300 mg/Nm. (All emission data of this paper are related to an oxygen content of the flue gas of 13 %). The results of the normal operation of the furnace can be seen as the big cluster of measurement values in the right upper comer of the diagram in Fig 3. 

---