Crusher load control system and self-organizing fuzzy controller (2)

(B) modify the control table through the self-organizing fuzzy controller The system uses the following steps:

(1) The performance measurement takes “deviation” (e(nT)=Y _n -S), and “deviation variation” (e(nT)=e _n -e _n-1 ) is two parameters to measure the output characteristics and Deviation of the desired characteristics is shown in Figure 3. Calculate the correction amount P(nT) required for the output characteristics by the same method as the fuzzy control table. The "deviation" e(nT), "deviation change" e(nT), correction amount P (nT) is divided into the following levels:
e(nT): -6,-5,-4,-3,-2,-1,-0, +0,1,2,3,4,5,6;
e(nT): -6,-5,-4,-3,-2,-1,0,1,2,3,4,5,6;
P(nT): -6,-5,-4,-3,-2,-1,0,1,2,3,4,5,6;
The magnitude of the correction amount P(nT) is determined according to whether e(nT) is at or above a given value, and e(nT) is a trend toward a given value or away from a given value.
(2) The control quantity correction system is a system with single input-single-output pure hysteresis, and the control quantity correction should be (the problem of lag quantity is not considered for the time being)
r(nT)=KP(nT)
Here, since the control amount and the output are both normalized, the coefficient K=1, then r(nT)=KP(nT)
Since the system has a large amount of hysteresis, as mentioned above, the control at time (nT-9T) has an effect on the system performance at nT, so the correction amount of the control amount here should be Ï…(nT-9T)=u(nT-9T) +r(nT) (1)
(3) Modify the control table by self-organizing fuzzy control

[next]

2) When changing the state: The control table ΠnT is always corrected according to the above method.
After the above method is continuously modified, a self-organizing fuzzy control table with better performance can be obtained. The control table can be modified and used each time it is sampled, so that the capacity and speed requirements of the computer can be reduced, and can be realized by a microcomputer and a single board machine.
Different scale factors GF, GE, EU and different sampling periods in the system have certain influence on the characteristics, as shown in Figure 4. When GE increases, the rate of increase increases and overshoot occurs, and GE overshoots. It can be seen from Fig. 4(b) that GE also affects the rising speed and overshoot of the characteristic curve. Therefore, in actual work, it is necessary to properly determine GE, GE and EU.

Figure 5 shows the effect of different sampling periods on the characteristics. It can be seen that the sampling period is small, the system characteristics rise rapidly, but the overshoot amount increases; when the sampling period is large, the system characteristics rise slowly, but it is beneficial to reduce the steady state error.
The system adopts the above self-organizing fuzzy control technology and obtains better control effects. The maximum allowable power of the crusher drive motor is 200 kW. When using PI or PID regulator, it can only be given above 150 kW. It is still impossible to avoid the occurrence of production stop accidents. With the self-organizing fuzzy controller, it is allowed to be given at 195 kW, which increases the productivity by about 13%, and avoids the accident that the motor is overloaded and the production is stopped.

Cast aluminum alloys have a lower density than cast iron and cast steel, while the specific strength is higher. Therefore, the use of aluminum alloy castings under the same load conditions can reduce the weight of the structure, so in the aviation industry and power machinery and transport machinery manufacturing, aluminum alloy castings have been widely used.