org.apache.commons.math3.ode.events

## Interface EventHandler

• All Known Implementing Classes:
EventFilter

public interface EventHandler
This interface represents a handler for discrete events triggered during ODE integration.

Some events can be triggered at discrete times as an ODE problem is solved. This occurs for example when the integration process should be stopped as some state is reached (G-stop facility) when the precise date is unknown a priori, or when the derivatives have discontinuities, or simply when the user wants to monitor some states boundaries crossings.

These events are defined as occurring when a g switching function sign changes.

Since events are only problem-dependent and are triggered by the independent time variable and the state vector, they can occur at virtually any time, unknown in advance. The integrators will take care to avoid sign changes inside the steps, they will reduce the step size when such an event is detected in order to put this event exactly at the end of the current step. This guarantees that step interpolation (which always has a one step scope) is relevant even in presence of discontinuities. This is independent from the stepsize control provided by integrators that monitor the local error (this event handling feature is available for all integrators, including fixed step ones).

Since:
1.2
• ### Nested Class Summary

Nested Classes
Modifier and Type Interface and Description
static class  EventHandler.Action
Enumerate for actions to be performed when an event occurs.
• ### Method Summary

Methods
Modifier and Type Method and Description
EventHandler.Action eventOccurred(double t, double[] y, boolean increasing)
Handle an event and choose what to do next.
double g(double t, double[] y)
Compute the value of the switching function.
void init(double t0, double[] y0, double t)
Initialize event handler at the start of an ODE integration.
void resetState(double t, double[] y)
Reset the state prior to continue the integration.
• ### Method Detail

• #### init

void init(double t0,
double[] y0,
double t)
Initialize event handler at the start of an ODE integration.

This method is called once at the start of the integration. It may be used by the event handler to initialize some internal data if needed.

Parameters:
t0 - start value of the independent time variable
y0 - array containing the start value of the state vector
t - target time for the integration
• #### g

double g(double t,
double[] y)
Compute the value of the switching function.

The discrete events are generated when the sign of this switching function changes. The integrator will take care to change the stepsize in such a way these events occur exactly at step boundaries. The switching function must be continuous in its roots neighborhood (but not necessarily smooth), as the integrator will need to find its roots to locate precisely the events.

Also note that the integrator expect that once an event has occurred, the sign of the switching function at the start of the next step (i.e. just after the event) is the opposite of the sign just before the event. This consistency between the steps must be preserved, otherwise exceptions related to root not being bracketed will occur.

This need for consistency is sometimes tricky to achieve. A typical example is using an event to model a ball bouncing on the floor. The first idea to represent this would be to have g(t) = h(t) where h is the height above the floor at time t. When g(t) reaches 0, the ball is on the floor, so it should bounce and the typical way to do this is to reverse its vertical velocity. However, this would mean that before the event g(t) was decreasing from positive values to 0, and after the event g(t) would be increasing from 0 to positive values again. Consistency is broken here! The solution here is to have g(t) = sign * h(t), where sign is a variable with initial value set to +1. Each time eventOccurred is called, sign is reset to -sign. This allows the g(t) function to remain continuous (and even smooth) even across events, despite h(t) is not. Basically, the event is used to fold h(t) at bounce points, and sign is used to unfold it back, so the solvers sees a g(t) function which behaves smoothly even across events.

Parameters:
t - current value of the independent time variable
y - array containing the current value of the state vector
Returns:
value of the g switching function