This paper organically combines fuzzy control algorithm and PID control algorithm to form a fuzzy PID control algorithm. This algorithm has strong adaptability and can automatically correct the PID control parameters according to changes in external conditions. 1. Design Idea Where: u(k) is the controller output at the k-th control; ec(k) is the deviation change at the k-th control; Kp(k) is the proportional coefficient of the controller at the k-th control; Ki(k) is the integral coefficient of the controller at the k-th control; d(k) is the derivative coefficient of the controller at the k-th control; u(k-1) is the integral sum of the controller at the (k-1)-th control, i.e., 2. Fuzzy PID Controller As shown in Figure 2-1, fuzzy PID control includes several important components such as parameter fuzzification, fuzzy rule inference, parameter defuzzification, and the PID controller. The computer calculates the deviation e(k) between the actual position and the theoretical position, as well as the current deviation change ec(k), based on the set input sp and feedback signal. Fuzzy inference is performed according to fuzzy rules, and then the fuzzy parameters are defuzzified to output the proportional, integral, and derivative coefficients of the PID controller. In addition, in order to compensate for the step changes caused by the gradation of general fuzzy control, the defuzzification output of the system is not the actual parameters of the controller, but the correction amount of the controller parameters. The actual parameters of the controller are where Cp, Ci, and Cd are the proportional correction coefficient, integral correction coefficient, and derivative correction coefficient, respectively. Kp0, Ki0, and Kd0 are called the initial values ​​of the control parameters. They are set by the user. Therefore, the user can make macroscopic adjustments to the control parameters, which can compensate for the error caused by ignoring the coupling relationship between parameters when simplifying fuzzy inference to a certain extent. This enhances the robustness of the system. According to the basic characteristics of PID control, the requirements for Kp, Ki, and Kd are different for different e(k) and ec(k): (1) When |e(k)| is large, Kp should be larger to eliminate the deviation as soon as possible and improve the response speed. To avoid overshoot, Ki and Kd should preferably be zero. (2) When the deviation is small, Kp should be reduced to continue to eliminate the deviation and prevent excessive overshoot and oscillation. Ki can be smaller. The value of Kd depends on |ec(k)|. (3) When e(k) and ec(k) have the same sign, the controlled variable changes in the direction of deviating from the given value; while when e(k) and ec(k) have opposite signs, the controlled variable changes in the direction of approaching the given value. Therefore, when the controlled variable approaches the given value, the proportional action with opposite signs hinders the integral action, thus avoiding integral overshoot and the resulting oscillation. However, when the controlled variable is far from approaching the given value and changes towards the given value, the control process will be slowed down due to the opposite signs of these two terms. When e(k) is large and ec(k) is negative, Kp takes a negative value, which can speed up the dynamic process of control. (4) When ec(k) is large, Kp should take a small value and Ki should take a larger value, and vice versa. (5) The derivative element is mainly used to control the deviation change ec(k), reduce overshoot, and overcome oscillation. In a multi-motor synchronous control system, it is not desirable for the speed to change rapidly, and the system overshoot is generally not too large. Therefore, in the specific design, the derivative coefficients are not fuzzy controlled, i.e., Kd = Kd0 and Cd = 1. Based on the above analysis, the universes of discourse for e(k), ec(k), and Cp are divided into 15 levels, denoted as -7, -6, -5, -4, -3, ..., +6, +7. The values ​​of the linguistic variables e(k), ec(k), and Cp are categorized into seven linguistic values: "negative large (NL)," "negative medium (NM)," "negative small (NS)," "positive large (PL)," "positive medium (PM)," "positive small (PS)," and "zero (Z)." The membership function is subjectively determined based on the above rules and experience. The inference rule adopts the form "IFAANDBTHENC," and the fuzzy relation is represented as follows: pk is the fuzzy relation matrix corresponding to the k-th rule; Mek is the fuzzy vector of the deviation value in the k-th rule; Meck is the fuzzy vector of the deviation change value in the k-th rule. The calculation process for the integral correction coefficient is similar and will not be elaborated further. Finally, the actual PID control parameters are calculated based on the correction coefficients and applied to the control system to ensure stable and reliable operation of the entire system. 3. Basic Scheme of Multi-Motor Synchronous Control: g represents the master motor speed set by the system. n1, n2, and n3 are the output speeds of the three motors. Using the principle of fuzzy PID control, the system is adjusted to ensure that the running speed of each motor quickly and stably approaches the master speed value. The synchronization control device uses an 89C51 microcontroller as the host and a reversible counter 193 as the phase-frequency discriminator. The reversible counter increments/decrements the pulses of the reference signal. Since the pulse count is the integral of the pulse signal frequency over time, when the reversible counter output is constant, the frequencies of the two signals are equal. The macro-control of the system is set by the host computer through the communication controller, while coordination is achieved by the counter frequency discrimination circuit using the fuzzy PID control algorithm. The specific hardware circuit mainly includes: sensor counting and direction pulse generation and width-fixing circuits, pulse leading edge staggering circuits, reversible counter frequency discrimination circuits, digital PWM generation circuits, and power amplifier drive circuits. Among them: the pulse generation fixed-width circuit can keep the width of the counter counting pulses the same; the pulse leading edge staggering circuit is to prevent the simultaneous arrival of the increment/decrement pulses of the reversible counter, which would affect the normal operation of the counter. 4 Application of Fuzzy PID in Multi-Motor Synchronous Control A series of experiments were conducted on the above system. First, the system's response capability to different master speeds was tested under constant external load. The experimental curves are shown in Figure 4-1. The thick solid line is the set master speed, the thin solid line is the speed response curve of a certain motor under the control of a general PID controller, and the thicker solid line is the speed response curve of the same motor under the control of a fuzzy PID controller. Second, Figure 4-2 shows the different control effects of the two controllers on the motor when the system master speed is constant, under the action of a step response and with different loads on the motor. Figure 4-2a shows the control effect of the fuzzy PID controller; Figure 4-2b shows the control effect of the general PID controller. (The thin solid line represents the curve when the motor is fully loaded, and the thick solid line represents the curve when the motor is half-loaded.) 5. Conclusion The application of the fuzzy PID controller in the multi-motor synchronous control device greatly enhances the robustness of the system and improves its dynamic response capability. This device has been applied in the Shandong Luneng pressure component inspection robot, and its operation has been stable and the control effect has been good over the past few months.