scholarly journals Design of Estimator for Computing Yaw and Pitch for a Twin Rotor MIMO system

2020 ◽  
Vol 14 ◽  
Keyword(s):  
Neural Network ◽  
Mimo System ◽  
Backpropagation Algorithm ◽  
Luenberger Observer ◽  
The Neural Network ◽  
Output System ◽  
System Information ◽  
Artificial Neural Network Ann ◽  
Twin Rotor Mimo System ◽  
Twin Rotor

Performance of any system is identified through the observation of significant system parameters. Required parameters have to be measured using suitable sensors. But in some scenarios, it is difficult to measure some of the parameters due to issues in the placement of sensors. In such cases, estimators are developed to measure the parameters indirectly. In this paper, an attempt is made to develop an estimator to monitor the value of pitch and yaw of a twin-rotor multi input multi output system. The observer is developed using two methods one using Luenberger’s equations and the other using an Artificial Neural Network (ANN). For training the neural network model, the backpropagation algorithm is used. Tests have been conducted to analyze and compare the behavior of both observers. From the results, it is evident that a Luenberger observer performs better when sufficient system information is available and ANN observer performs better when inadequate system information is available

scholarly journals Design of a soft sensing technique for measuring pitch and yaw angular positions for a Twin Rotor MIMO System

F1000Research ◽  
2021 ◽  
Vol 10 ◽  
pp. 342
Author(s):  
Sneha Nayak ◽  
Sravani Vemulapalli ◽  
Santhosh Krishnan Venkata ◽  
Meghana Shankar
Keyword(s):  
Neural Network ◽  
Kalman Filter ◽  
Optical Sensors ◽  
Mimo System ◽  
Real Life ◽  
Absolute Error ◽  
Luenberger Observer ◽  
Fan Blade ◽  
Twin Rotor Mimo System ◽  
Twin Rotor

Background: This paper presents a soft sensor design technique for the estimation of pitch and yaw angular positions of a Twin Rotor MIMO System (TRMS). The objective of the proposed work was to calculate the value of pitch and yaw angular positions using a stochastic estimation technique.  Methods: Measurements from optical sensors were used to measure fan blade rotations per minute (RPM).  The Kalman filter, which is a stochastic estimator, was used in the proposed system and its results were compared with those of the Luenberger observer and neural network. The Twin Rotor MIMO System is a nonlinear system with significant cross-coupling between its rotors.  Results: The estimators were designed for the decoupled system and were applied in real life to the coupled TRMS. The convergence of estimation to the actual values was checked on a practical setup. The Kalman filter estimators were evaluated for various inputs and disturbances, and the results were corroborated in real-time.  Conclusion:  From the proposed work it was seen that the Kalman filter had at least Integral Absolute Error (IAE), Integral Square Error (ISE), Integral Time Absolute Error (ITAE) as compared to the neural network and the Luenberger based observer.

Neural network NARMA-L2 control of a Twin Rotor MIMO System

2019 ◽  
Author(s):  
Senoussaoui Abderrahmene ◽  
Chenafa Mohammed ◽  
Kacimi Abderrahmane ◽  
Hocine Rachida
Keyword(s):  
Neural Network ◽  
Mimo System ◽  
Twin Rotor Mimo System ◽  
Twin Rotor

scholarly journals Data Mining using Artificial Neural Network POS_NEG Rules

2013 ◽  
Vol 11 (6) ◽  
pp. 2709-2714
Author(s):  
Pushkar Shinde ◽  
Dr. Varsha Patil
Keyword(s):  
Neural Network ◽  
Data Mining ◽  
Sensitivity Analysis ◽  
Artificial Neural Network ◽  
Backpropagation Algorithm ◽  
The Neural Network ◽  
Artificial Neural ◽  
Using Data ◽  
Artificial Neural Network Ann ◽  
Sensitivity Method

Diabetes patients are increasing in number so it is necessary to predict , treat and diagnose the disease. Data Mining can help to provide knowledge about this disease. The knowledge extracted using Data Mining can help in treating and preventing the disease. Artificial Neural Network (ANN) can be used to create an classifier from the data. The neural network is trained using backpropagation algorithm The knowledge stored in the neural network is used to predict the disease. The knowledge stored in neural network is extracted using Pos-Neg sensitivity method. The knowledge extracted is in form of sensitivity analysis to analyze the disease and in turn help in treating the disease.

Adaptive Active Force Control Application to Twin Rotor Mimo System

2013 ◽  
Vol 393 ◽  
pp. 688-693 ◽  
Author(s):  
Hanif Ramli ◽  
Wahyu Kuntjoro ◽  
M.S. Meon ◽  
K.M Asraf K. Ishak
Keyword(s):  
Force Control ◽  
Control Algorithm ◽  
Mimo System ◽  
Active Force ◽  
Optimum Control ◽  
Active Force Control ◽  
Non Linear ◽  
Artificial Neural Network Ann ◽  
Twin Rotor Mimo System ◽  
Twin Rotor

This paper reports a current study on modeling and simulation of adaptive Active Force Control (AFC) based scheme embedded with an artificial neural network (ANN) and/or fuzzy logic (FL) in response manipulations of the twin rotor multi-input multi-output (MIMO) system (TRMS). TRMS is well known for its non-linear behaviour and common classical control scheme such as Proportional-Integral-Derivative (PID) would not be adequate to compensate disturbances. The disturbances in this case were as the results of non-linear external and internal parametric changes, namely angular momentum and couple reactions between the two axes of TRMS. The adaptive control algorithm was proposed in both pitch and yaw to generate an optimum control gain for both responses, simulated viz. MATLAB/SIMULINK software Package. The ANN and FL were integrated into the scheme and act as optimum control algorithm in catalyzing the performance of the TRMS. The results from hybrid conditions of PID-AFC, PID-AFC-ANN and PID-AFC-FL respectively were observed and analyzed. From performance evaluation, PID-AFC-FL scheme has demonstrated a potentially robust and effective manipulating capability in trajectory tracking.

Sign in / Sign up

Export Citation Format

Share Document