Chapter 1 : Introduction to Process Control

1.0 Introduction
This subject is about elements and methods of control system operation used in industry to control industrial processes. The basic objective is to regulate the value of some quantity so that it stays at a particular set point value despite external influences. Variations in proportions, temperature, flow, turbulence, and many other factors must be carefully and consistently controlled to produce the desired end product with a minimum of raw materials and energy. Process control technology is the tool that enables manufacturers to keep their operations running within specified limits and to set more precise limits to maximize profitability, ensure quality and safety.

1.1 Process
Process as used in the terms process control and process industry, refers to the methods of changing or refining raw materials to create end products. The raw materials, which either pass through or remain in a liquid, gaseous, or slurry (a mix of solids and liquids) state during the process, are transferred, measured, mixed, heated or cooled, filtered, stored, or handled in some other way to produce the end product. Process industries include the chemical industry, the oil and gas industry, the food and beverage industry, the pharmaceutical industry, the water treatment industry, and the power industry.

1.2 Types of Process Control 1.2.1 Human-Aided Control

A human can regulate the level using a sight tube, S, to compare the level, h, to the objective, H, and adjust a valve to change the level. The height can be regulated apart from the input flow rate by visual .

Figure 1.1- Human Aided Control

1

1.2.2 Automatic Control

Figure 1.2- Automatic Control Machines, electronics or computers replace the operations of the human. A sensor is added to measure the value of the level and convert it to proportional signal, s. This signal is provided as input to a controller that perform the functions of the human and provide the signal, u to change the valve setting via an actuator connected to the valve by mechanical linkage.

1.2.3 Servo Control Its objective is slightly different from process control. Here the objective is to force the controlled variable to follow variation of the reference value. In an industrial robot arm, servomechanism force the arm to follow a path from point A to B.

Figure 1.3- Servo System 1.2.4 Discrete-State Control System This type of control system concerned with controlling a sequence of events rather than regulation or variation of individual variable. The starting and stopping of events is a discrete based system because the event is either true or false ( i.e. started or stopped, opened or closed, on or off). PLCs are used to implement this type of control system.
2

1.3 Process Control Block Diagram and Definitions

Figure 1.4 Typical Process control block diagram Process: The physical system we are attempting to control or measure. Examples: water filtration system, molten metal casting system, steam boiler, oil refinery unit, power generation unit. Process Variable, or PV: The specific quantity we are measuring in a process. Examples: pressure, level, temperature, flow, electrical conductivity, pH, position, speed, vibration. Setpoint, or SP: The value at which we desire the process variable to be maintained at. In other words, the target value of the process variable. Primary Sensing Element, or PSE: A device that directly senses the process variable and translates that sensed quantity into an analog representation (electrical voltage, current, resistance;mechanical force, motion, etc.). Examples: thermocouple, thermistor, bourdon tube, microphone, potentiometer, electrochemical cell, accelerometer. Transducer: A device that converts one standardized instrumentation signal into another standardized instrumentation signal, and/or performs some sort of processing on that signal. Often referred to as a converter and sometimes as a relay. Examples: I/P converter (converts 4-20 mA electric signal into 3-15 PSI pneumatic signal), P/I converter (converts 3-15 PSI pneumatic signal into 4-20 mA electric signal), square-root extractor (calculates the square root of the input signal).

Transmitter: A device that translates the signal produced by a primary sensing element (PSE) into a standardized instrumentation signal such as 3-15 PSI air pressure, 4-20 mA DC electric current, Fieldbus digital signal packet, etc., which may then be conveyed to an indicating device, a controlling device, or both. Lower- and Upper-range values, abbreviated LRV and URV, respectively: the values of process measurement deemed to be 0% and 100% of a transmitters calibrated range. For example, if a temperature transmitter is calibrated to measured a range of temperature starting at 300 degrees Celsius and ending at 500 degrees Celsius, 300 degrees would be the LRV and 500 degrees the URV. Controller: A device that receives a process variable (PV) signal from a primary sensing element (PSE) or transmitter, compares that signal to the desired value for that process variable (called the setpoint), and calculates an appropriate output signal value to be sent to a final control element (FCE) such as an electric motor or control valve. Final Control Element, or FCE: A device that receives the signal from a controller to directly influence the process. Examples: variable-speed electric motor, control valve, electric heater. It provide the required changes in the controlled variable to bring it to the setpoint. Automatic mode: When the controller generates an output signal based on the relationship of process variable (PV) to the setpoint (SP). Manual mode: When the controllers decision-making ability is bypassed to let a human operator directly determine the output signal sent to the final control element. Measurement: Information about the variable itself. It involves conversion of the variable into corresponding analog of the variable such as pneumatic pressure, voltage or current or digitally encoded signal. A sensor makes the initial measurement and energy conversion. Further transformation is or signal conditioning maybe needed to complete the measurement. Error Detector: The difference between actual controlled variable value and a setpoint. It has both magnitude and polarity. Loop : The signal flow forms a complete circuit from process through measurement, error detection, controller and final control element. In most cases it is called a feedback loop because we determine an error and feed back a correction to the process.

1.3.1 Physical Diagram of a process control loop

Figure 1.5 Physical Implementation

1.4 Control System Evaluation

A control system need to be evaluated to know how well it is working. The variable used to measure the performance is the error signal e(t). Since the value of the controlled variable vary with time, the error also changes with time. The objective of a control system is make the error exactly zero but this cannot be perfectly achieved and there will always be error. The evaluation will be on how large the error is and how it varies with time. Control system objective System should be stable System should provide best steady state regulation System should provide best transient regulation.

1.4.1 Stability
Figure shows that before the control system is on, the controlled variable drifts in a random fashion and not regulated. After control system is on, the variable is forced to adopt the setpoint value. Some time later the variable begins to show signs of instability and starts to oscillate.

Figure 1.6 Controlling Process variable

1.4.2 Steady State Regulation

Best steady state regulation simply means that the steady state error should be a minimum. Generally a setpoint has a allowable deviation, c. This means that variation of the variable within this band are expected and acceptable. External influences that tend to cause drift of the value beyond the allowable deviation are corrected by the control system.

1.4.3 Transient Regulation

It specifies how a control system react to sudden change in setpoint and bring it to the new setpoint value. The sudden change can also occur in other processes that affects the controlled variable .

Figure 1.7- Transient Effect to PV

1.5 Data Representation

Specification indicates the range of the variables involved, such as temperature between 20 to 120, or controller signal for valve from fully closed to fully open. Two analog standard are in common use: a. Current between 4 to 20 mA for electrical system b. Gas pressure between 3-15 psi for pneumatic system These signals are used to transmit variables information over some distance.

Figure 1.8 Standard data transmission

Example Temperature range between 20 to 120 C is linearly converted to the standard current range of 4 to 20 mA. What current will result from 66 C? What temperature does 6.5 mA represent?

1.6 Practical Example of Measurement and Control System

Boiler Water Level Control System Steam boilers are very common in industry, principally because steam power is so useful. Common uses for steam in industry include doing mechanical work (e.g. a steam engine moving some sort of machine), heating, producing vacuums (through the use of steam eductors), and augmenting chemical processes (e.g. reforming of natural gas into hydrogen and carbon dioxide). Making steam continuously, however, is a little more complicated. An important variable to measure and control in a continuous boiler is the level of water in the steam drum (the upper vessel in a water-tube boiler). In order to safely and efficiently produce a continuous flow of steam, we must ensure the steam drum never runs too low on water, or too high. If there is not enough water in the drum, the water tubes may run dry and burn through from the heat of the fire. If there is too much water in the drum, liquid water may be carried along with the flow of steam, causing problems downstream.

Figure 1.9 - Boiler

Table 1.1- Transmitted pneumatic data

8

Table 1.2- Controller signal

1. 7 Other Type of instruments 1.7.1 Indicator

An indicator gives a human operator a convenient way of seeing what the output of the transmitter is without having to connect test equipment (pressure gauge for 3-15 PSI, ammeter for 4-20 mA) and perform conversion calculations. Moreover, indicators may be located far from their respective transmitters, providing readouts in locations more convenient than the location of the transmitter itself. An example where remote indication would be practical is shown here, in a nuclear reactor temperature measurement system:

Figure 1.10- Indicator at a remote place

1.7.2 Recorders
Another common auxiliary instrument is the recorder (sometimes specifically referred to as a chart recorder or a trend recorder), the purpose of which is to draw a graph of process variable(s) over time. Recorders usually have indications built into them for showing the instantaneous value of the instrument signal(s) simultaneously with the historical values.

Figure 1.10 Recording Recorders become powerful diagnostic tools when coupled with the controllers manual control mode. By placing a controller in manual mode and allowing direct human control over the final control element (valve, motor, heater), we can tell a lot about a process.

Figure 1.11- Recording indicating delay

10

1.8 Additional Definitions

Some common terms and expression used to describe process control elements Transfer function Describes the relationship between the input and output for the block.

Accuracy Used to specify maximum overall error to be expected from a device. It can appear in several forms. a. Measured variable ; the accuracy is 2 C b. Percentage of instrument full scale(FS) reading. E.g. 5% FS in a 5V full scale range meter. c. Percentage of instruments span. Thus for a device measuring 3% of span for a 20 to 50 psi range of pressure the accuracy will be ( 0.03)(5020) = 0.9 psi d. Percentage of the actual reading. Thus for a 2% of reading voltmeter, we would have an inaccuracy of 0.04 V for a reading of 2V.

Example 1. A temperature sensor has a span of 20 250 C. A measurement results in a value of 55 C. Specify the error if the accuracy is a) 0.5% FS b) 0.75% of span c) 0.8% of reading. What is the possible temperature in each case? 2. A temperature sensor has a transfer function of 5mV/ C with an accuracy of 1%. Find the possible range of the transfer function. Suppose a reading of 27.5 mV results from this sensor, find the temperature that could provide this reading. Sensitivity It is a measure of the change in output of an instrument for a change of input. High sensitivity is desirable in instruments because measurements can be taken easily. The value of sensitivity is indicated by the transfer function e.g. 5mV/ C

11

Hysteresis Different reading results for a specific input depending on whether the input value is approached from higher or lower values.

Figure 1.11- Hysterisis

Resolution Minimum measurable value of the input variable Example A force sensor measures a range of 0 to 150N with a resolution of 0.1% FS. Find the smallest change in force that can be measured. Example. A sensor has a transfer function of 5mV/C. Find the required voltage resolution of the signal conditioning if a temperature resolution of 0.2 C is required.

Linearity Output is normally represented in some functional relationship to the input. A linear relationship is highly desirable. A straight line equation can be used y = mx + c. Example A sensor resistance changes linearly from 100 ohm to 180 ohm as temperature changes from 20 to 120 C. Find a linear equation relating resistance and temperature.

One of the specification of sensor output is the degree to which it is linear with the measured variable. A measure of sensor linearity is to determine the deviation of sensor output from a best fit straight line over a particular range. A common specification of linearity is the maximum deviation from a straight line expressed as percent of FS.
12

Consider a sensor that outputs a voltage as a function of pressure from 0 to 100psi with a linearity of 5% FS. It means that at some point on the curve, the deviation between actual pressure and linearly indicated pressure deviates by 5% of 100 psi, or 5 psi.

Figure 1.13 - Nonlinearity

1.9 Simple Statistics Techniques

Confidence of the value of a variable can be improved by statistical analysis of measurement. This is true where random errors in measurement cause a distribution of readings of the value of some variable. Two important values are the arithmetic mean and standard deviation Example A control system was installed to regulate the weight of potato chips dumped into bags. Given a sample of 15 bags drawn from the operation before and after the control system was installed, evaluate the success of the system. Do this by comparing the mean and standard deviation before and after. The bags should be 200g. Sample before: 201, 205, 197, 185, 202, 207, 215, 220, 179, 201, 197, 221, 202, 200, 195 Sample after : 197, 202, 193, 210, 207, 195, 199, 202, 193, 195, 201, 201, 200, 189, 197

13

14