utils.hpp

#ifndef __BIO__UTILS_HPP__
#define __BIO__UTILS_HPP__
 
#include "Kokkos_Macros.hpp"
#include <Kokkos_Core.hpp>
#include <Kokkos_MathematicalConstants.hpp>
#include <Kokkos_Random.hpp>
#include <common/traits.hpp>
#include <mc/alias.hpp>
#include <mc/prng/prng.hpp>
namespace Models
{
 
  template <FloatingPointType F>
  static F consteval _get_phi_s_max(F density,
                                    F dl,
                                    F glucose_to_biomass_yield = 0.5)
  {
    // dl and density must be same unit, dl*density -> mass and y is mass yield
    return (dl * density) / glucose_to_biomass_yield;
  }
 
  template <FloatingPointType T>
  KOKKOS_INLINE_FUNCTION MC::Status
  check_div(const T l, const T lc)
  {
    return (l >= lc) ? MC::Status::Division : MC::Status::Idle;
  }
 
  namespace MolarMass
  {
    namespace GramPerMole
    {
 
      template <FloatingPointType T> constexpr T glucose = static_cast<T>(180);
      template <FloatingPointType T> constexpr T dioxygen = static_cast<T>(32);
      template <FloatingPointType T> constexpr T acetate = static_cast<T>(59);
 
      constexpr double co2 = 44;
      ;
    }; // namespace GramPerMole
 
    constexpr double glucose = GramPerMole::glucose<double> * 1e-3;
    constexpr double dioxygen = GramPerMole::dioxygen<double> * 1e-3;
    ;
    constexpr double acetate = GramPerMole::acetate<double> * 1e-3;
    ;
    constexpr double co2 = GramPerMole::co2 * 1e-3;
    ;
    [[deprecated]] constexpr double X = 113.1e-3;
  }; // namespace MolarMass
 
  template <FloatingPointType F>
  KOKKOS_INLINE_FUNCTION consteval F
  c_linear_density(F rho, F d)
  {
    return rho * F(Kokkos::numbers::pi) * d * d / F(4.);
  }
 
  static constexpr double tau_division_proba = 1e-7;
 
  KOKKOS_INLINE_FUNCTION bool
  check_probability_division(double d_t,
                             double gamma,
                             MC::pool_type random_pool)
  {
    (void)d_t;
    // const double proba_div =
    //     (1 - Kokkos::exp(-d_t / tau_division_proba)) * gamma;
 
    auto generator = random_pool.get_state();
    const double x = generator.drand(0., 1.);
    random_pool.free_state(generator);
 
    return x < gamma;
  }
 
  namespace GammaDivision
  {
    KOKKOS_INLINE_FUNCTION double
    threshold_linear(double lenght, double threshold, double upper_bound)
    {
      return (lenght <= threshold)
                 ? 0
                 : (lenght - threshold) / (upper_bound - threshold);
    }
 
    KOKKOS_INLINE_FUNCTION constexpr float
    logistic(float x, float xmax, float alpha)
    {
      // auto blna = alpha * std::log(xmax);
      // return 1 / (1 + std::exp(-alpha * std::log(x) + blna));
      // return x<xmax?1. / (1 + std::exp(alpha * std::log((x-xmax)/xmax ))):1.;
 
      const auto z = x < xmax ? std::pow(x / (xmax - x), alpha) : 1.;
      return z / (1. + z);
    }
 
  } // namespace GammaDivision
 
  KOKKOS_INLINE_FUNCTION bool
  almost_equal(double val, double val2, double tolerance = 1e-8)
  {
    return Kokkos::abs(val - val2) < tolerance;
  }
 
  KOKKOS_INLINE_FUNCTION constexpr bool
  almost_zero(double x, double tol = 1e-8)
  {
    return (-tol < x) && (x < tol);
  }
 
  template <typename... Args>
  KOKKOS_INLINE_FUNCTION double
  min_var(Args... args)
  {
    return (Kokkos::min)({ args... });
  }
  template <>
  KOKKOS_INLINE_FUNCTION double
  min_var()
  {
    return 0.;
  }
 
} // namespace Models
 
#endif