In recent years, experiments using a computer to control the temperature of the resistance furnace have solved the problems that could not be realized in the analog controller. Explore a series of improved algorithms and apply them to practice; also tried multi-layer mapping network, network neuron control algorithm, using neural network to solve practical problems, applied to control electric furnace temperature system examples; successful improvement The remote control method is also designed based on the control. The hardware is the material basis of the computer control system. It realizes the monitoring of the parameters of the production process or the controlled object under the coordination of the software, and completes the control of the production process or the controlled object.
The hardware of the process computer control system is mainly composed of a host, an external device, a process input and output device, a human-machine contact device, and a communication device: a central processing unit and an internal memory (the host is the core of the control system. The host inputs according to the process) All kinds of information sent by the equipment reflecting the working status of the production process in real time, as well as the predetermined control algorithm, automatically perform information processing and calculation, select the corresponding control strategy in time, and immediately send control commands to the production process through the process output device. .
The nickel-chromium-nickel-silicon thermocouple used in this experiment is used as a temperature measurement component, which converts the temperature into a millivolt-level voltage signal. The input and output interface circuit used in the furnace temperature control system is a multi-function interface card, which is completed and functioned, that is, the temperature signal (analog quantity) obtained by the thermocouple is converted into a digital quantity, sent to the computer, and the calculated amount of the computer is calculated. Converted to analog quantities and sent to the solid state relay for temperature control. In this experiment, the thermocouple conditioning board is used to amplify the millivolt-level signal sent by the thermocouple into a voltage signal to meet the requirements of the input and output signals of the board. The system uses a zero-crossing AC solid relay as a power regulator to connect with the furnace. By controlling the on/off of the solid relay to control whether the furnace of the resistance furnace is energized or not, the furnace temperature control is performed. Adopt adaptive Ethernet, server, client.
Software is a general term for various programs. The advantages and disadvantages of software are not only related to the function of hardware, but also related to the control quality and management level of the computer to the production process. The software is divided into system software and application software. System software generally includes assembly language, advanced algorithm language, operating system, database system, communication network software and diagnostic programs. Process computer control designers need to master the use of system software to better compile application software. The control interface and control program are used in this system. The language is used for computational simulation of neural networks. The application software is the control and management program that the system designer prepares for a certain production process. Its advantages and disadvantages directly affect the control quality and management level. The application software in this system mainly includes: process input and output program, process control program, human-machine interface program, print display program and the like. The process control program is the core of the application software, and is a concrete implementation of the control algorithm based on classical or modern control theory.
Characteristics of Computer Control System In computer control systems, the amount of control is usually analog, and the input and output of the computer itself are digital. Therefore, most computer control systems have a digital analog hybrid structure. The basic block diagram of the computer control system includes two parts: the computer in the working and discrete state and the production process in the continuous working state.
The hardware component block diagram of the computer control system can clearly see the process of the computer obtaining the signal, which is done by the converter. The process from analog signal to discrete analog signal is sampling, which is the sampling period. Obviously, it is necessary to select the sampling period reasonably, and the signal transformation and transmission in the computer control system of the damage graph is too large.
The process of changing from a discrete analog signal to a digital signal is called quantization, which uses a set of binary digits to approximate the sampled analog signal value. Obviously, the process of conversion is a quantitative process.
After the computer introduces the control system, due to its fast operation speed, high precision, large storage capacity, and its powerful computing function and programmability, it can not only complete conventional algorithms, but also complete adaptive control, optimal control, Self-learning control, intelligent control, etc. It can be said that without a computer, the application of modern control theory is hard to imagine.
In the computer process control system, the control law is implemented by using software. Modifying a control law is complicated or simple. It only needs to modify the software. Generally, it does not need to change the hardware, so it is easy to implement complex control rules and control schemes. Online modifications make the system very flexible and adaptable.
Control algorithm In the computer control system, the control law is realized by computer program. Some problems that could not be realized in the analog controller are solved after the introduction of the computer. A series of improved algorithms are developed: the integral separation control algorithm, Limit weakening control algorithm, incomplete differential control algorithm, differential advance control algorithm and dead zone control algorithm, fuzzy control algorithm, neural network control algorithm, etc. Explain, for example, the following: the purpose of the integral separation control algorithm to introduce the integral link in the ordinary digital controller is mainly to eliminate the static difference and improve the accuracy. However, when the process starts, ends or increases or decreases the set value, it will cause the integral of the operation to accumulate, eventually causing a large overshoot of the system, and even causing the system to oscillate, which is absolutely not allowed in some production processes. .
The language neural network toolbox is used to analyze the structure and algorithm of the neural network and the related simulation experiments. The process of each control is formed into a sample set, and then the neural network is used to simulate the control. Finally, the training has been used. Neural networks to predict the effects of control; and to achieve human-computer interaction. Digital control is one of the most commonly used control methods in the production process, and has been widely used in metallurgy, machinery, and chemical industries. The algorithm selection program is convenient to provide the user with a variety of control algorithms for selection. The algorithm selection function is implemented by the parameter setting form, and the main control is used. (Finish)
FUWA HEAVY INDUSTRY CO.,LTD , https://www.fuwacraneparts.com