main.cpp

/*
 * main.cpp
 *
 * This file implements the main-function.
 * Here, the user has to choose what to simulate by commenting things in (by placing \\ at the start of the line) and out (by removing the \\)
 *
 * Copyright (c) 2019, The Chancellor, Masters and Scholars of the University
 * of Oxford, VITO nv, and the 'Slide' Developers.
 * See the licence file LICENCE.txt for more information.
 */

//!< Include header files
#include "slide.hpp"
#include "../benchmark/benchmarks.hpp"
#include "../examples/examples.hpp"

#include <ctime>
#include <thread>
#include <array>
#include <tuple>
#include <random>
#include <cmath>
#include <iomanip>

int main()
{
  /*
   * This function decides what will be simulated.
   * In the code below, you have to uncomment the line which you want to execute (to uncomment, remove the //!<in front of the line)
   * and comment all the other lines in the blocks with FUNCTION CALLS (to comment in, add //!<in front of the lines)
   */

  //!< print that you start simulations
  //!< slide::tests::unit::test_all();
  std::cout << "Start simulations" << std::endl;

  //!< Make a clock to measure how long the simulation takes
  slide::Clock clk;
  //!< Read the values for the matrices for the spatial discretisation of the solid diffusion PDE.
  //!< these values are calculated by the MATLAB-script 'modelSetup.m', which writes them to csv files.
  //!< Here, we only need to read these csv files

  //!< slide::Model_SPM M; //!<Initialised at constructor.
  //!<  Look for extra settings at Constants.hpp (such as verbosity)

  //!< Choose the prefix, a string with which the names of all subfolders with the outputfiles will start (the folders will be called prefix_xxxx).
  //!< This is typically a number, and the reason why we use it is that results of previous simulations are not overwritten.
  //!< i.e. for one simulation, use pref "0", then the next simulation could have pref "1".
  //!< this ensures the results of the second simulation are written in different files and folders, rather than overwriting the files from the first simulation
  const std::string pref = "0"; //!< todo
                                //!<  will be used to name a folder with the results, so must comply with the naming convention on your operating system
                                //!<  letters, numbers, underscores and dots are fine
                                //!<  don't use spaces
                                //!<  don't use forbidden characters, e.g. for Windows the following are not allowed < > : " \ | / ? *
                                //!<  https://docs.microsoft.com/en-us/windows/desktop/fileio/naming-a-file

  //!< Choose which cell should be used for the simulations
  //!< const int cellType = slide::cellType::KokamNMC;
  //!< which cell to use for the simulation.
  //!< 0 	high power Kokam NMC cell (18650)
  //!< 1 	high energy LG Chem NMC cell (18650)
  //!< 2 	user cell (template class provided for where the user can define his own parameters)

  //!< Choose which degradation models to use for the simulations
  //!< The models are identified by a number, see DEG_ID in DEG_ID.hpp and below
  /* SEI_id 		identification for which model(s) to use for SEI growth
   * 					0	no SEI growth
   * 					1 	kinetic model only (Tafel kinetics)
   * 							ref: Ning & Popov, Journal of the Electrochemical Society 151 (10), 2004
   * 					2 	Pinson&Bazant model: linear diffusion + Tafel kinetics
   * 							ref: Pinson & Bazant, Journal of the Electrochemical society 160 (2), 2013
   * 					3	Christensen and Newman model: diffusion and kinetics model
   * 							ref: Christensen & Newmann, Journal of the Electrochemical Society 152 (4), 2005
   * SEI_porosity	integer deciding whether we reduce the volume fraction of active electrode material due to SEI growth
   *  				0	don't subtract anything
   * 					1	use correlation from Ashwin
   * 							ref: Ashwin, Chung, Wang, Journal of Power Sources 328, 2016
   * CS_id	identification for which model(s) to use for surface crack growth
   *  				0 	no surface cracking
   * 					1 	Laresgoiti's stress + crack growth model
   * 							ref: Laresgoiti, Kablitz, Ecker, Sauer, Journal of Power Sources 300, 2015
   * 					2 	Dai stress model + Laresgoiti crack growth
   * 							ref: Laresgoiti, Kablitz, Ecker, Sauer, Journal of Power Sources 300, 2015
   * 								 Dai, Cai, White, Journal of Power sources 247, 2014
   * 					3 	model based on Deshpande and Bernardi
   * 							ref: Deshpande & Bernardi,Journal of the Electrochemical Society 164 (2), 2017
   * 					4 	model from Barai
   * 							ref: Barai, Smith, Chen, Kim, Mukherjee, Journal of the Electrochemical Society 162 (9), 2015
   * 					5 	model from Ekstrom
   * 							ref: Ekstrom and Lindbergh, Journal of the Electrochemical Society 162 (6), 2015
   * CS_diffusion	integer deciding whether to reduce the negative diffusion constant due to surface cracks
   *  				0 	don't decrease diffusion
   * 					1	decrease according to Barai
   * 							ref: Barai, Smith, Chen, Kim, Mukherjee, Journal of the Electrochemical Society 162 (9), 2015
   * LAM_id  identification for which model(s) to use for loss of active material
   *  				0 	no LAM
   * 					1	Dai's stress model and Laresgoiti's correlation to relate the stress to LAM
   * 							ref: Dai, Cai, White, Journal of Power sources 247, 2014
   * 								 Laresgoiti, Kablitz, Ecker, Sauer, Journal of Power Sources 300, 2015
   * 					2	delacourt's correlation between abs(current density) and porosity
   * 							ref: Delacourt & Safari, Journal of the Electrochemical Society 159 (8), 2012
   * 					3 	Kindermann's model for cathode dissolution: tafel kinetics for increased porosity
   * 							ref: Kindermann, Keil, Frank, Jossen, Journal of the Electrochemical Society 164 (12), 2017
   * 					4 	Narayanrao's correlation between effective surface area and current density
   * 							ref: Narayanrao, Joglekar, Inguva, Journal of the Electrochemical Society 160 (1), 2012
   * pl_id 	identification for which model to use for li-plating
   *  				0 	no plating
   * 					1	Yang et al's model for thermodynamic plating (Tafel kinetics)
   * 							ref: Yang, Leng, Zhang, Ge, Wang, Journal of Power Sources 360, 2017
   */
  //!< For now, assume we want to include 'Pinson&Bazant'-SEI growth, 'Delacourt'-LAM, 'Kindermann'-LAM and 'Yang'-lithium plating
  // slide::DEG_ID deg;
  // deg.SEI_id.add_model(2); //!< Pinson & Bazant SEI growth
  // deg.SEI_porosity = 0;    //!< don't decrease the porosity (set to 1 if you do want to decrease the porosity)

  // deg.CS_id.add_model(0); //!< no surface cracks
  // deg.CS_diffusion = 0;   //!< don't decrease the diffusion coefficient (set to 1 if you do want to decrease the diffusion)

  // //!< there are 2 LAM models (Delacourt and Kindermann)
  // deg.LAM_id.add_model(2); //!< Delacourt LAM
  // deg.LAM_id.add_model(3); //!< Kindermann LAM

  // deg.pl_id = 1; //!< Yang lithium plating

  // std::cout << "Used ageing model: " << deg.print() << '\n';

  //!< Then the user has to choose what is simulated.
  //!< In the code below, uncomment the line which calls the function you want to execute (uncommenting means removing the //!<in front of the line)
  //!< and comment all the other lines (commenting means putting //!<in front of the line)

  auto numThreads = std::thread::hardware_concurrency(); //!<
  std::cout << "Available number of threads : " << numThreads << '\n';


  // Please see examples for using SLIDE. For previous version refer to SLIDE_v2 branch.
  using namespace slide;

  //!< Examples:
  // slide::examples::drive_cycle_artemis();
  // slide::examples::GITT_test();
  //!< Benchmarks:

  // slide::benchmarks::run_Cell_Bucket();
  // slide::benchmarks::run_Cell_ECM();
  // slide::benchmarks::run_Cell_SPM_1(1);
  // slide::benchmarks::run_Cell_SPM_2(1);
  // slide::benchmarks::run_LP_case_SmallPack();
  // slide::benchmarks::run_LP_case_MediumPack();
  // slide::benchmarks::run_LP_case_LargePack();

  // MATLAB ECM benchmarks:
  // slide::benchmarks::run_Cell_Bucket_single_default_pulse();
  // slide::benchmarks::run_Cell_Bucket_single_default_CCCV();

  // slide::benchmarks::run_Cell_ECM_single_default_pulse();
  // slide::benchmarks::run_Cell_ECM_single_default_CCCV();

  // slide::benchmarks::run_Cell_ECM_2_RC_single_default_pulse();
  // slide::benchmarks::run_Cell_ECM_2_RC_single_default_CCCV();

  // slide::benchmarks::run_Cell_ECM_parallel_3_default_pulse();

  // slide::benchmarks::run_Cell_ECM_parallel_3_default_CCCV();

  // slide::benchmarks::run_Cell_ECM_parallel_3_withRcontact_CCCV();

  // slide::benchmarks::run_Cell_ECM_series_3_withRcontact_CCCV();

  //!<*********************************************** END ********************************************************
  //!< Now all the simulations have finished. Print this message, as well as how long it took to do the simulations
  std::cout << "finished all simulations in " << clk << ".\n";
}