Abstract: In this paper, the Bee Algorithm is used to train a Neural Network. This is done by altering the connections and biases of the Neural Network (NN) so that the desired output from the input is obtained. The merging between the two concepts is tested to control an inverted pendulum which is a benchmark for testing control theories. The trained NN is used to stabilize the pendulum in its upright position. The NN is trained by comparing its response to that of a state feedback controller. The Bee Algorithm succeeded in training the NN for it to have the desired output. Moreover, the effect of changing the parameters of both the neural network and the bee algorithm is also studied.
Abstract: To cope with the problem of the degradation of actuation effectiveness caused by actuator deflection or fault in the attitude and orbit control system (AOCS) of spacecraft on-orbit, an attitude fault-tolerant and anti-disturbance control scheme is proposed based on a sliding mode iterative learning law, in which the pseudo control input is applied to design the sliding mode controller to ensure the AOCS tracks a reference trajectory precisely after some fault occurred; By analyzing the Lyapunov stability, a novel adaptive iterative learning law is developed, which in term of the tracking error, determine some parameters in controller on-line to address the actuator failure and external disturbance. Numerical simulation experiments show that the fault-tolerant and anti-disturbance controller can ameliorate actuation malfunction and compensate the influence of external disturbance effectively.
Abstract: In this paper, the design and construction of two-wheeled balancing scooter based on principle of the inverted pendulum that it must keep an angle of zero degree due to vertical line in time. The mechanical structure, electronic circuit and algorithm are developed to control motors and gyroscope sensor is used as detector of angle error due to vertical line. The mathematical model to describe the dynamic behavior of balancing system is analyzed, which stabilizes the handle angle of scooter in stable position. A basic implementation of the PID controller is conducted to compensate nonlinear behavior.
Abstract: As an important part of modern measurement science and technology, length measurement has significant value in application and research in industrial production and teaching practice. In the paper, the length measurement integrated system integrating voice broadcast function, keyboard operation functions and real-time display is designed; composition hardware is stated and flexible software programming is used to meet requirements of system design; finally, the system is tested.
Abstract: This paper devotes to address the problem of the singularity of the Stewart platform, which is different from the serial mechanism. An algorithm has been constructed; also, by solving a minimum problem, it finds a alternative path to instead the preliminary path, in which there are several singular configurations. At the end of this paper, a numerical case has been developed and gives two points on the preliminary path. By applying this algorithm, a singularity-free path can be found to avoid the singularities in the preliminary path and ill-condition. So, this algorithm can be proven practical.
Abstract: In this paper, stability analysis of an optical fiber amplifier gain controller is presented. The mathematical model of an EDFA (Eribium-Doped Fiber Amplifier) control system is described by two-time scaled system and thus singular perturbation approach can be applied to the design and analysis of the control system. In order to handle the channel variations such as add/drop, we adopt an internal model. The performance of the control system is theoretically analyzed based on a novel singular-perturbation-like approach.
Abstract: A road grade recognition method based on longitudinal acceleration was proposed after longitudinal dynamics analysis. The method based on longitudinal dynamics utilized a real-time engine output torque signal and the real-time vehicle speed signal to calculate road grade. The result of simulation and the vehicle field test showed that the method based on existing vehicle sensors was low cost, simple and feasible, it could identify road grade.
Abstract: PID control systems are the most commonly used control technology in industries. However, there are issues on control performances for the unstable process with time delays. In order to improve the control performances of PID control systems, a new two degree of freedom model driven PID control system is introduced in this paper and it is used to the unstable processes with time delay. The model driven PID control is capable of stabilizing with unstable processes by using PD feedback, regulating quickly for disturbance and tracking quickly to the change of set point. With case studies comparing with conventional PID control systems was done.
Abstract: Different from changeable chromosome length genetic algorithm, a new method based on firefly optimization algorithm is put forward to search a trains energy saving operation strategy under running disturbance condition in railway network. The algorithm is designed and validated using a referenced simulation case. Compared with other methods, it demonstrates this new FA-based algorithm has a better result in computation efficiency and may be considered to use in other energy efficient train operation models.