Implementation of the hottest general fuzzy contro

2022-08-09
  • Detail

The realization of general fuzzy controller on PLC

1 introduction

fuzzy control in modern control system can easily solve complex problems such as nonlinearity, time-varying, large lag, strong coupling, variable structure, harsh end conditions and so on, which are common in industrial fields. With its high reliability, convenient programming, anti harsh environment, powerful function and other characteristics, programmable controller solves the problems of reliability, safety, flexibility, convenience, economy and so on, which are widely concerned in the field of industrial control. The combination of the two can be widely used in practical engineering. This paper studies several algorithms of general fuzzy controller implemented on PLC, and realizes fuzzy control by off-line calculation, table lookup and interpolation

in order to meet the requirements of different actuators for the form of control quantity, the algorithm of incremental/position fuzzy control output is adopted. When the incremental fuzzy control output is used, the undisturbed switching between manual and automatic can be realized. In order to eliminate the oscillation caused by frequent actions, a fuzzy control algorithm with dead zone is adopted. In addition, there is a fuzzy quantization rounding link in the general look-up table fuzzy controller, that is, when the error E and the error change rate EC are not equal to the fuzzy language value (such as Nb, nm, NS, Zo, PS, PM or Pb), e and EC are rounded. At this time, the control quantity u found in the look-up table can only approximately reflect the fuzzy control rules, resulting in errors. Due to the existence of quantization error, not only the output U of the fuzzy controller can not accurately reflect its control rules, but also the regulation dead zone will be caused. In the steady-state stage, the system will produce steady-state error, and even flutter. The bivariate three-point interpolation method proposed in this paper can fundamentally eliminate the quantization error and adjustment dead zone, and overcome the steady-state error and steady-state flutter caused by quantization error. Figure 1-1 shows the basic structure of the general fuzzy controller

2 design and implementation of general fuzzy controller on PLC

Figure 2-1 block diagram of incremental output fuzzy control system

; Determine the membership function type of each fuzzy variable; Fuzzification of precise input and output variables; Formulate fuzzy control rules; Determine the fuzzy reasoning algorithm; Defuzzification of fuzzy output variables; Save the calculation results in the required format to generate a query table

the two-dimensional fuzzy controller, which is widely used in practical applications and has the largest selection of models of 5 tons, mostly selects the controlled variables and the input given deviation E and deviation change rate EC as the input variables, because it has been able to strictly reflect the dynamic characteristics of the input variables in the controlled process, and can meet most engineering needs. At the same time, it is simpler than the three-dimensional fuzzy controller in calculation, and the fuzzy control rules are easy to understand. For the multivariable fuzzy controller, the decoupling characteristics of the fuzzy controller itself can be used to realize decoupling in the controller structure through the decomposition of fuzzy relational equations, that is, a multi input multi output (MI-MO) fuzzy controller can be decomposed into several multi input single output (mi-so) fuzzy controllers, so the design method of the single variable fuzzy controller can be adopted. This paper studies the design of two-dimensional general fuzzy controller. In order to make it easier for the user to decide whether to output incrementally or positionally, the output variable takes the change U of the adjustment amount, which is also conducive to reducing the system deviation by adjusting the change U of the adjustment amount

because the control quality of the fuzzy controller is affected by the output mode of the controller, the positional output is provided for different controlled objects. The following will focus on PPS, peek, PI and high temperature resistant PA output and incremental output. The disadvantage of position plastic output algorithm is that the output U (k) corresponds to the actual maintenance position of the actuator. If the computer fails, it will cause a large change in the position of the actuator due to the large change of U (k). If the incremental algorithm is adopted, the computer outputs the control increment Δ Figure 2-1 shows the block diagram of the incremental output fuzzy control system. The control quantity of the actual position of the valve, that is, the accumulation of the control quantity increment, is calculated by the formula U (k) = u (k-1) + Δ U (k) is completed by executing software

the realization of fuzzy control algorithm is obtained through fuzzy reasoning, but the result is a fuzzy vector, which cannot be directly used to control the controlled object, and must be converted into an accurate quantity acceptable to the actuator. The non fuzzy outputs of all possible input states are calculated in the same way to form a query table as shown in table 2-1. This table is stored in the computer program in the form of data modules. When a group of inputs is given, the corresponding output values can be found out by this table. This method integrates complex fuzzy calculation into the query table, saves calculation time in actual use, and makes the control simple and clear. 2.2 part of the design

the total control table of the fuzzy controller obtained by the computer off-line operation is debugged and modified repeatedly by the system, and finally stored in the PLC system memory in the form of data module, which is managed by a subroutine that queries the table. The flow of the query subroutine is shown in Figure 2-2. In the figure, fielde, fieldec and fieldu represent the universe of error E, error change rate EC and control quantity u respectively. According to the flow chart, the controller can be adjusted manually and automatically, and the output mode can be incremental and position output. If the output mode is selected as incremental output, the impact free switching from manual adjustment mode to automatic adjustment mode can be realized

Copyright © 2011 JIN SHI