Файл: Microcontroller based applied digital control (D. Ibrahim, 2006).pdf

and expensive plants, such as large chemical processing plants or oil refineries. Since the introduction of microprocessors in the early 1970s the cost and size of digital computers have been greatly reduced. Early microprocessors, such as the Intel 8085 or the Mostek Z80, were very limited and required several chips before they could be used as processing elements. The required chips were read-only memory (ROM) to store the user program, random-access memory (RAM) to store the user data, input–output (I/O) circuitry, A/D and D/A converters, interrupt logic, and timer circuits. By the time all these chips were put together the chip count, power consumption, and complexity of the basic hardware were considerable. These controllers were in the form of microcomputers which could be used in many medium and large digital control applications.

Interest in digital control has grown rapidly in the last several decades since the introduction of microcontrollers. A microcontroller is a single-chip computer, including most of a computer’s features, but in limited sizes. Today, there are hundreds of different types of microcontrollers, ranging from 8-pin devices to 40-pin, or even 64or higher pin devices. For example, the PIC16F877 is an 8-bit, 40-pin microcontroller with the following features:

operation up to 20 MHz;

8K flash program memory;

368 bytes RAM memory;

256 bytes electrically erasable programmable read-only memory (EEPROM) memory;

15 types of interrupts;

33 bits of parallel I/O capability;

2 timers;

universal synchronous–asynchronous receiver/transmitter (USART) serial communications;

10-bit, 8-channel A/D converter;

2 analog comparators;

33 instructions;

programming in assembly or high-level languages;

low cost (approximately $10 each).

Flash memory is nonvolatile and is used to store the user program. This memory can be erased and reprogrammed electrically. EEPROM memory is used to store nonvolatile user data and can be written to or read from under program control. The microcontroller has 8K program memory, which is quite large for control based applications. In addition, the RAM memory is 368 bytes, which again is quite large for control based applications.

1.3 CENTRALIZED AND DISTRIBUTED CONTROL SYSTEMS

Until the beginning of 1980s, computer control was strictly centralized. Usually a single large computer or minicomputer (e.g. the DEC PDP11 series) was used to control the plant. The computer, associated power supplies, input–output, keyboard and display unit were all situated in a central location. The advantages of centralized control are as follows:

It is easy to manage the computer.

Only one computer is used.

Less number of people are required.

program with all the necessary hardware components. It can be developed using the popular visual programming languages such as Visual C++ or Visual Basic. Good human–computer interface techniques should be employed in the design of the user interface. Alternatively, graphical programming languages such as Labview or VisiDaq can be used to create powerful, user-friendly SCADA systems.

In a SCADA system the user can have access to a graphical screen in order to monitor or change a setting in the plant. SCADA systems consist of both hardware and software and are usually implemented using personal computers (PCs). Typical hardware includes the computer, keyboard, touch screen, sensors and actuators. The software is in the form of a graphical user interface, where parts of the plant, sensor data and actuator data can all be displayed in various colours on a screen. The advantage of a SCADA system is that the user can easily monitor the status of the overall system. It is important that a SCADA system should be secure and password protected to avoid unauthorized access to the control screens.

1.5 HARDWARE REQUIREMENTS FOR COMPUTER CONTROL

1.5.1 General Purpose Computers

In general, although almost any digital computer can be used for digital control there are some requirements that should be satisfied before a computer is used for such an application. Today, the majority of small and medium scale DDC-type applications are based on microcontrollers which are used as embedded controllers. Applications where user interaction and supervisory control are required are commonly designed around the standard PC hardware.

As shown in Figure 1.8, a general purpose computer consists of the following basic building elements:

central processing unit (CPU);

program memory;

data memory;

input–output devices.

The CPU is the part which contains the arithmetic and logic unit (ALU), the control unit (CU) and the general purpose registers (GPR). The ALU consists of the logic circuitry necessary to carry out arithmetic and logic operations, for example to add or subtract numbers, to compare numbers and so on. Some ALU units are equipped to carry out multiplication and division and floating point mathematical operations. The CU supervises the operations within the CPU, fetches instructions from the program memory, decodes these instructions and controls the ALU and other parts of the computer so that the required operations can be implemented. The GPR are a set of fast registers which are generally used to carry out fast operations within the CPU.

The program memory of a general purpose computer is usually an external unit and attached to the computer via the data bus and the address bus. A bus is a collection of conductors which carry electrical signals. The data bus is a bidirectional bus which carries the data to be sent or received between the CPU and the other parts of the computer. The size of this bus is 8 bits in most microprocessors and microcontrollers. Some microcontrollers have data buses that are 16 or even 32 bits wide. Minicomputers and mainframe computers usually have 64 or even higher data widths. The address bus is a unidirectional bus which is used to address the peripheral

in digital control applications. Most microcontrollers have the built-in circuits necessary for computer control applications. For example, a microcontroller may have A/D converters so that the external signals can be sampled. They also have parallel input–output ports so that digital data can be read or output from the microcontroller. Some devices have built-in D/A converters and the output of the converter can be used to drive the plant through an actuator (e.g. an amplifier). Microcontrollers may also have built-in timer and interrupt logic. Using the timer or the interrupt facilities, we can program the microcontroller to implement the control algorithm accurately.

Microcontrollers have traditionally been programmed using the assembly language of the target device. As a result, the assembly languages of the microcontrollers manufactured by different firms are totally different and the user has to learn a new language before being able program a new type of device. Nowadays microcontrollers can be programmed using highlevel languages such as BASIC, PASCAL or C. High-level languages offer several advantages compared to the assembly language:

It is easier to develop programs using a high-level language.

Program maintenance is much easier if the program is developed using a high-level language.

Testing a program developed in a high-level language is much easier.

High-level languages are more user-friendly and less prone to making errors.

It is easier to document a program developed using a high-level language.

In addition to the above advantages, high-level languages have some disadvantages. For example, the length of the code in memory is usually larger when a high-level language is used, and the programs developed using the assembly language usually run faster than those developed using a high-level language.

In this book, PIC microcontrollers are used as digital controllers. The microcontrollers are programmed using the high-level C language.

1.6 SOFTWARE REQUIREMENTS FOR COMPUTER CONTROL

Computer hardware is nowadays very fast, and control computers are generally programmed using a high-level language. The use of the assembly language is reserved for very special and time-critical applications, such as fast, real-time device drivers. C is a popular language used in most computer control applications. It is a powerful language that enables the programmer to perform low-level operations, without the need to use the assembly language.

The software requirements in a control computer can be summarized as follows:

the ability to read data from input ports;

the ability to send data to output ports;

internal data transfer and mathematical operations;

timer interrupt facilities for timing the controller algorithm.

All of these requirements can be met by most digital computers, and, as a result, most computers can be used as controllers in digital control systems. The important point is that it is not justified and not cost-effective to use a minicomputer to control the speed of a motor, for example. A microcontroller is much more suitable for this kind of control application. On

10 INTRODUCTION

the other hand, if there are many inputs and many outputs, and if it is required to provide supervisory tasks as well then the use of a minicomputer can easily be justified.

The controller algorithm in a computer is implemented as a program which runs continuously in a loop which is executed at the start of every sampling time. Inside the loop, the desired reference value is read, the actual plant output is also read, and the difference between the desired value and the actual value is calculated. This forms the error signal. The control algorithm is then implemented and the controller output for this sampling instant is calculated. This output is sent to a D/A converter which generates an analog equivalent of the desired control action. This signal is then fed to an actuator which in turn drives the plant to the desired point.

The operation of the controller algorithm, assuming that the reference input and the plant output are digital signals, is summarized below as a sequence of simple steps:

Repeat Forever

When it is time for next sampling instant

Read the desired value, R

Read the actual plant output, Y

Calculate the error signal, E = R − Y

Calculate the controller output, U

Send the controller output to D/A converter

Wait for the next sampling instant

End

Similarly, if the reference input and the plant output are analog signals, the operation of the controller algorithm can be summarized as:

Repeat Forever

When it is time for next sampling instant

Read the desired value, R, from A/D converter

Read the actual plant output, Y , from the A/D converter

Calculate the error signal, E = R − Y

Calculate the controller output, U

Send the controller output to D/A converter

Wait for the next sampling instant

End

One of the important features of the above algorithms is that once they have been started they run continuously until some event occurs to stop them or until they are stopped manually by an operator. It is important to make sure that the loop is run continuously and exactly at the same times, i.e. exactly at the sampling instants. This is called synchronization and there are several ways in which synchronization can be achieved in practice, such as:

using polling in the control algorithm;

using external interrupts for timing;

using timer interrupts;