Mohammad Ghani, Ph.D.

I’m a researcher at Universitas Airlangga
I’m interested in Data-Driven Dynamical Systems
Ph.D. in Applied Mathematics

Invited Reviewer of MethodsX
Issued by Elsevier · Apr 2025

Conferences

Book Front ICATAM: October 17-18 2024 · Proceedings

Publications

Diphtheria transmission prediction by Extended Kalman Filter
22 March 2025 · Mohammad Ghani

Ocean wave prediction using Long Short-Term Memory (LSTM) and Extreme Gradient Boosting (XGBoost) in Tuban Regency for fisherman safety
2 November 2024 · Riswanda Ayu Dhiya’ulhaq, Anisya Safira, Indah Fahmiyah, Mohammad Ghani

Forecasting The Number of Foreign Tourism Visits to Indonesia using Seasonal Autoregressive Integrated Moving Average (SARIMA) and Holt-Winters Approach
1 November 2024 · Indah Fahmiyah, Lidya Septi Andini, Mohammad Ghani

Spatial impact on inflation of Java Island prediction using Autoregressive Integrated Moving Average (ARIMA) and Generalized Space-Time ARIMA (GSTARIMA)
17 July 2024 · Anisya Safira, Riswanda Ayu Dhiya’ulhaq, Indah Fahmiyah, Mohammad Ghani

Kalman filter based on a fractional discrete-time stochastic augmented CoVid-19 model
24 April 2024 · Mohammad Ghani, Dwi Rantini, Maryamah

Discussions

Estimation of Angle Velocities and Positions for ARM model (Shoulder joint and Elbow joint) using Extended Kalman Filter

Screenshot 2024-12-04 172437 [Source of Fig. 1]

EKF_ARM_Model_2

ARM model is shown below:
$M(\theta)\ddot{\theta}+C(\theta,\dot{\theta})\dot{\theta}+B\dot{\theta}=\tau$
$\theta=[\theta_1;\theta_2],\;\dot{\theta}=[\dot{\theta}_1;\dot{\theta}_2]=[\omega_1;\omega_2],\;\ddot{\theta}=[\dot{\omega}_1;\dot{\omega}_2]$
$M=[\alpha+2\beta\cos(\theta_2)\;\;\;\delta+\beta\cos(\theta_2);\delta+\beta\cos(\theta_2)\;\;\;\delta]$
$C(\theta,\dot{\theta})=[-\beta\dot{\theta}_2\sin(\theta_2)\;\;\;-\beta(\dot{\theta}_1+\dot{\theta}_2)\sin(\theta_2);\beta\dot{\theta}_1\sin(\theta_2)\;\;\;0]$
$B=[b_{11}\;\;\;b_{12};b_{21}\;\;\;b_{22}]$
$\tau_{\theta_1}=\tau_{\theta_2}=0.5*\cos(t)$
$\alpha=l_1+l_2+m_1L_{g1}^2+m_2(L_{1}^2+L_{g2}^2)$
$\beta=m_2L_{1}L_{g2}$
$\delta=l_2+m_2L_{g2}^2$

Calculation using MAPLE
Detailed description of dynamic systems
Code Arm_Model_EKF.m
Code System.m

fminsearch VS lsqcurvefit to estimate the parameters of infection rate and recovery rate for the SIR model

$\left\{\begin{matrix}&\dot{S}=-\beta\cdot S\cdot I,\\&\dot{I}=\beta\cdot S\cdot I-\alpha\cdot I,\\&\dot{R}=\alpha\cdot I.\end{matrix}\right.$

Initial values of parameters (before being estimated by fminsearch and lsqcurvefit):
$\beta=1:infection\;rate$
$\alpha=0.2:recovery\;rate$

fminsearch lsqcurvefit

Estimated parameters by fminsearch and lsqcurvefit:
$\beta=1.6926:infection\;rate$
$\alpha=0.4485:recovery\;rate$

Main part of fminsearch: fminsearch(@error_sum_of_squares_fminsearch,param)
Main part of lsqcurvefit: lsqcurvefit(@sir_rhs_lsqcurvefit,param,tspan,data)

param is the parameters of infection rate and recovery rate

tspan is time (number of data)

The data is based on the article entitled “On parameter estimation approaches for predicting disease transmission through optimization, deep learning and statistical inference methods”:
data=[3;6;25;73;222;294;258;237;191;125;69;27;11;4] and [S0,I0,R0]=[760,3,0]

Estimation of parameters infection rate and recovery rate for the SIR model using Extended Kalman Filter

WORKFLOW_001

$\left\{\begin{matrix}&\dot{S}=-\beta\cdot S\cdot I,\\&\dot{I}=\beta\cdot S\cdot I-\alpha\cdot I,\\&\dot{R}=\alpha\cdot I.\end{matrix}\right.$
$\beta:infection\;rate$
$\alpha:recovery\;rate$

SIR_model

estimate_state_SIR

estimate_parameter_SIR

Code SIR_Model_EKF.m

Estimations of parameters for the Mass-Spring-Damper Dynamical Systems using Extended Kalman Filter (Case 1: for the estimation of mass (m) only, Case 2: for the estimations of mass (m), damping (b), and spring (k))

WORKFLOW2_001

$\left\{\begin{matrix}&\dot{x}_1=x_2,\\&\dot{x}_2=-(k/m)\cdot x_1-(b/m)\cdot x_2+(1/m)\cdot F(t).\end{matrix}\right.$
$k:spring$
$b:damper$
$m:mass$
$F:force\;and\;assumed\;to\;be\;constant$