This paper investigates the identification of a permanent magnet synchronous motor (PMSM) velocity servo system based on deterministic learning theory. Unlike most of the existing studies, this study does not identify the system parameters, but rather the system dynamics. System dynamics is the fundamental knowledge of the PMSM system and contains all the information about the system parameters, various uncertainties, and the system structure. The accurate modeling of the various uncertainties is important to improve the control performance of the controller. In this study, the dynamics of the PMSM system containing various uncertainties are identified based on the system state. Firstly, the system state of the PMSM is measured, and then a suitable RBF neural network is designed based on it. The RBF neural network is used to construct a state estimator that takes the motor system as input. The weights of the RBF neural network are updated using the Lyapunov-based weights. As the weights converge, a constant RBF neural network can be obtained, which contains complete information about the system parameters and the various uncertainties of the motor system. We use the proposed method to identify the simulated and real-time PMSM velocity servo systems separately, and the identification results show the effectiveness and feasibility of the proposed method.

Although the active disturbance rejection controller can obtain good control performance without relying on specific model information, it targets integer-order systems. Fractional-order characteristics are commonly existed in practical systems. For fractional-order systems, it is more targeted to use the order information of the fractional-order model to design the active disturbance rejection controller, so as to obtain better control performance. A fractional active disturbance rejection controller composed of FOESO and FOPID (IDE-FOPID-FOESO) is proposed in this paper. The fractional-order extended state observer (FOESO) is designed based on the order information and the nonlinear state error feedback is replaced by the fractional-order PID controller (FOPID) whose parameters are obtained by the improved differential evolution algorithm (IDE). For IDE algorithm, the basis vector is randomly selected from the optimal individual population in the mutation strategy, and the scaling factor and cross-probability factor are adaptively adjusted according to the information of the successfully mutated individual in the search process to improve the exploration and mining capabilities of the algorithm. The simulation results show that the IDE algorithm can obtain the better parameters of FOPID faster compared with traditional DE algorithm and the IDE-FOPID-FOESO controller can be better applied to fractional-order systems with better control performance.

With the continuous increasing of system modeling accuracy and control performance requirements, the fractional‐order models are applied to describe the dynamic characteristics of the real systems more accurately over traditional integer‐order models. Traditional identification algorithms for fractional‐order system require stringent information on the derivatives of the objective function and initial values of the parameters, which is difficult in practical applications. The differential evolution (DE) algorithm has the advantage of simple implementation and superior performance for solving multi‐dimensional complex optimization problems which is suitable for the identification of fractional‐order systems. Aiming at improving the solution accuracy and convergence speed, an improved DE (IDE) algorithm for fractional‐order system identification is proposed. The movement strategy of the marine predator algorithm (MPA) and the mutation strategy of randomly selecting one from the optimal individual population as the basis vector are both adopted in this proposed IDE algorithm. The parameter information of the successfully mutated individual is archived during the iteration process. The scaling factor and crossover probability factor are adaptively adjusted according to the archived information to enhance the best solution accuracy obtained in the search process. By testing 13 sets of unimodal and multimodal functions with high‐performance optimization algorithms, the accuracy performance of the proposed IDE solution is verified. The IDE algorithm is applied to identify the parameters of the fractional‐order system of permanent magnet synchronous motor with simulation and experiment. The identification results clearly show the effectiveness of the proposed methodology.

In order to solve the problems of grouting diffusion and invisible formation process in shield slurry film forming test, transparent soil and transparent slurry were introduced to carry out shield mud film test based on shield grouting model. Using laser imaging combined with CCD camera (high-speed industrial camera) rapid capture photography technology, through a series of visible physical model tests of slurry diffusion and mud film development under different working conditions, the temporal and spatial conditions of slurry diffusion were observed and recorded in real time to explore the slurry diffusion law and the formation mechanism of mud film in slurry shield excavation face.The phenomenon shows that the smaller the grouting pressure is, the more favorable the formation of mud film is.The mud penetration distance is generally proportional to the grouting pressure.The infiltration distance of the grout increases with time, and the infiltration distance of the grout is similar at each moment under the same grouting pressure. The leachate volume and time show a quadratic function roughly under the same density and mud pressure, and the increase rate of the leachate volume begins to slow down with time.