I don't understand your question. central difference method is the foundation of explicit analysis, it's nothing to do with abaqus. i suggest you to read more about explicit analysis, you should know what is explicit well first.
d^2(f)/dt = [f(t_n+1) - 2f(t_n) + f(t_n-1)]/(h^2) with h being the step size
to approximate the derivatives of a function with its differences and thus
replace a given differential equation with its difference equation. In this
manner you can turn a system of differential equations into a system of
algebraic equations in the value of function at various time steps and
therefore obtain an approximate solution. This method or variations of it are
used in commercial softwares to solve the set of governing differential
equations.
A stable time increment is the one that leads to the convergence of algorithm. So if the algorithm does not converge you should consider a smaller time step. If the algorithm does not converge by consecutively cutting time steps then you should expect some sort of instability in your model.
regarding the stability of explicit algorithms, topic is well discussed in:
Analysis of transient algorithms with particular reference to stability behavior
HUGHES, T J R
Computational methods for transient analysis (A84-29160 12-64). Amsterdam, North-Holland, 1983, p. 67-155
about your question
I don't understand your question. central difference method is the foundation of explicit analysis, it's nothing to do with abaqus. i suggest you to read more about explicit analysis, you should know what is explicit well first.
In reply to about your question by Hengxu Song
My question is what's the
My question is what's the role played by central differnce operator in dynamic explicit analysis?
Regards,
Shiva
In reply to My question is what's the by cmksiva
Central Difference
Dear Shiva,
The role is to use the expressions
df/dt = [f(t_n+1) - f(t_n)]/h or df/dt = [f(t_n) - f(t_n-1)]/h
d^2(f)/dt = [f(t_n+1) - 2f(t_n) + f(t_n-1)]/(h^2) with h being the step size
to approximate the derivatives of a function with its differences and thus
replace a given differential equation with its difference equation. In this
manner you can turn a system of differential equations into a system of
algebraic equations in the value of function at various time steps and
therefore obtain an approximate solution. This method or variations of it are
used in commercial softwares to solve the set of governing differential
equations.
Mohsen
stable time increment
Dear All,
Could anyone explain what is stable time increment in explicit, why it is used?
Regards,
Shiva
stable time increment
Dear All,
Could anyone explain what is stable time increment in explicit, why it is used?
Regards,
Shiva
In reply to stable time increment by cmksiva
Stable Time Increment
A stable time increment is the one that leads to the convergence of algorithm. So if the algorithm does not converge you should consider a smaller time step. If the algorithm does not converge by consecutively cutting time steps then you should expect some sort of instability in your model.
Mohsen
Stability of explicit algorithm
regarding the stability of explicit algorithms, topic is well discussed in:
Analysis of transient algorithms with particular reference to stability behavior
HUGHES, T J R
Computational methods for transient analysis (A84-29160 12-64). Amsterdam, North-Holland, 1983, p. 67-155