Journal of Artificial Intelligence in Electrical Engineering
Volume:5 Issue: 20, Winter 2017

In this paper, recurrent neural network is used for path planning in the joint space of the robot with obstacle in the workspace of the robot. To design the neural network, first a performance index has been defined as sum of square of error tracking of final executor. Then, obstacle avoidance scheme is presented based on its space coordinate and its minimum distance between the obstacle and each of robot links and proper inequality equations have been derived which describe the qualification of obstacle avoidance. Moreover, nonlinear optimization problem with nonlinear constraint functions has been derived by considering the joint physical limits. In order to design the neural network, equivalent problem of projection theorem has been converted to quadratic programming by using Kuhn-Tucker 1optimality conditions. Based on the projection theorem, model of recurrent neural network has been determined which is as a first order differential equation. Simulation results show the good performance of the proposed method by applying the proposed algorithm on the seven-degree freedom PA-10 robot.

Considering the high consumption of the air compressors, a control system of screw compressor is designed and implemented to deal with energy saving and localization of mentioned compressor. In this paper the variable speed drive (VSD) control algorithm based on proportional-integral-derivative (PID) controller is optimized to decrease power consumption and more stable motor speed and outlet pressure. Automatic PI factors adjustment according to system behavior is goal of this control system. To show the validity of the proposed algorithm, we simulated P and I changes. The results of simulation and practical test on a GA111 Atlas Copco compressor were established which demonstrate that the proposed algorithm provides system stability improvement and as a consequence depreciation reduction and energy saving were achieved.

One of the important tasks of robots is object recognition without human intervention. The main aim of this study is developing an ultrasonic sensor to recognize the materials of thin wall objects. This study is based on the physics rules in ultrasonic that ultrasonic velocities, longitudinal and transverse velocities, are altered in different materials because of differences in their elastic modulus. Obviously applying forces by robot to an object walls should be in harmony with their elastic modules and kind of materials, especially thin wall objects. The novelty of this research is related to using an especial probe for omitting the effect of near field by using a delay line made of a column of water in plastic membrane. The ultrasonic sensors with delay lines make it possible to detect the material of thin wall objects. The measured ultrasonic velocities are compared with standard ones to detect objects materials. The results of experiments are very promising and confirm that it is possible to detect plates made of St 11, copper, with 1 m thickness

In this article a boost dc-dc converter with the ability to remove the input current ripple is given for specific work cycle.Ability to remove the input current ripple in this structure is obtained without increasing the number of elements compared to other structures.In addition, the converter provided without the need for a large cycle of work or boost transformers produced high voltage gain without utilizing high values of duty cycle or boosting transformers. These features make the converter provided for the transfer of low voltage power sources, such as renewable resources. The converter operation has been investigated in the working modes and the relations related to this converter have been calculated. Ultimately the structure is provided in the PSCAD software and simulation results are presented to validate the performance and relationships obtained for this structure.

Many real systems are among underactuated stimulation systems. The number of actuators in such systems is less than their freedom degree. Translational oscillator with rotational actuator (TORA) is an underactuated system in cascaded form. In the present paper, this system has been investigated as a case study and then a sliding mode controller has been designed for it. This is due to the fact that among non-linear controlling methods, sliding mode control is appropriate to achieve global stabilization. Also it is robust against external disturbances and parametric uncertainties in the system under control. In this paper the efficiency of simulation has been investigated.

In island mode, when the grid is separated from the rest of the system, the dispersed generation system acts to provide a constant voltage to the local load in the voltage control mode. In this paper .Sliding mode controls method based on the second-order sliding surface and to improve sliding mode control and increase robustness .The system is controlled by an adaptive control against disturbances. In this regard, to overcome chattering and uncertainty from the derivative of the control signal as a controller, which ultimately results in a control signal with integration, up to Adaptive sliding mode controller in the presence of unwanted and uncertain interference and interconnection of grid stability and improvement. Guaranteed output and tracking of the system as well as reducing unnecessary high frequency oscillations in the entire system slowly. According to the simulation results, it is seen that the inverter output waveform has a good ability to follow the waveforms of the voltage and refered three-phase current. This paper is simulated by MATLAB software.