My Project
Loading...
Searching...
No Matches
List of all members | Static Public Member Functions | Static Public Attributes
Opm::ReservoirCoupling::Seconds Struct Reference

Utility class for comparing double values representing epoch dates or elapsed time. More...

#include <ReservoirCoupling.hpp>

Static Public Member Functions

static bool compare_eq (double a, double b)
 Determines if two double values are equal within a specified tolerance.
 
static bool compare_gt (double a, double b)
 Determines if a is greater than b within the specified tolerance.
 
static bool compare_gt_or_eq (double a, double b)
 Determines if a is greater than b within the specified tolerance.
 
static bool compare_lt_or_eq (double a, double b)
 Determines if a is less than or equal to b within the specified tolerance.
 

Static Public Attributes

static constexpr double abstol = 1e-15
 Absolute tolerance used for comparisons.
 
static constexpr double reltol = 1e-15
 Relative tolerance used for comparisons.
 

Detailed Description

Utility class for comparing double values representing epoch dates or elapsed time.

This class is used to compare double values that represent:

  • Epoch dates (seconds since the Unix epoch)
  • Elapsed time (seconds since the start of the simulation)
Note
When comparing against the start of a report step or similar, it is important not to miss these points due to numerical issues. Communication between master and slave processes is based on these specific points in time.
Epoch values in this century (2000-2100) fall within the range [1e9, 4e9]. A double variable cannot represent such large values with high precision. For example, the date 01-01-2020 corresponds to 1.5778368e9 seconds, and adding 1e-7 seconds to this value does not change it. Microseconds (1e-6) are approximately the smallest time units that can be represented accurately for such numbers.
Report steps appear to have a maximum resolution of whole seconds. See the stepLength() function in Schedule.cpp in the opm-common module, which returns the step length in seconds.

Member Function Documentation

◆ compare_eq()

bool Opm::ReservoirCoupling::Seconds::compare_eq ( double  a,
double  b 
)
static

Determines if two double values are equal within a specified tolerance.

Two values a and b are considered equal if:

\[ |a - b| \leq \text{tol} = \text{abstol} + \text{reltol} \times \max(|a|, |b|) \]

For example, if abstol = reltol = 1e-15:

  • If |a| and |b| are below 1, the absolute tolerance applies.
  • If a and b are above 1, the relative tolerance applies.
Note
For epoch dates between 01-01-2000 and 01-01-2100, epoch values are in the range [1e9, 4e9]. Therefore, $ 10^{-15} \times 10^9 = 10^{-6} $, meaning differences smaller than one microsecond will be considered equal.
This approach is not accurate for numbers close to zero, but such comparisons are not expected.
Parameters
aFirst double value.
bSecond double value.
Returns
True if a and b are considered equal within the tolerance.

◆ compare_gt()

bool Opm::ReservoirCoupling::Seconds::compare_gt ( double  a,
double  b 
)
static

Determines if a is greater than b within the specified tolerance.

Parameters
aFirst double value.
bSecond double value.
Returns
True if a is greater than b.

◆ compare_gt_or_eq()

bool Opm::ReservoirCoupling::Seconds::compare_gt_or_eq ( double  a,
double  b 
)
static

Determines if a is greater than b within the specified tolerance.

Parameters
aFirst double value.
bSecond double value.
Returns
True if a is greater than b.

◆ compare_lt_or_eq()

bool Opm::ReservoirCoupling::Seconds::compare_lt_or_eq ( double  a,
double  b 
)
static

Determines if a is less than or equal to b within the specified tolerance.

Parameters
aFirst double value.
bSecond double value.
Returns
True if a is less than or equal to b.
The documentation for this struct was generated from the following files: