Diffusion akantu (features/50-using-gitpod-in-gitlab) Browse

50-using…

heat_transfer_model_inline_impl.cc
View Options

			/**
			* @file heat_transfer_model_inline_impl.cc
			*
			* @author Guillaume Anciaux <guillaume.anciaux@epfl.ch>
			* @author Srinivasa Babu Ramisetti <srinivasa.ramisetti@epfl.ch>
			*
			* @date creation: Sun May 01 2011
			* @date last modification: Thu Jun 05 2014
			*
			* @brief Implementation of the inline functions of the HeatTransferModel class
			*
			* @section LICENSE
			*
			* Copyright (©) 2014 EPFL (Ecole Polytechnique Fédérale de Lausanne)
			* Laboratory (LSMS - Laboratoire de Simulation en Mécanique des Solides)
			*
			* Akantu is free software: you can redistribute it and/or modify it under the
			* terms of the GNU Lesser General Public License as published by the Free
			* Software Foundation, either version 3 of the License, or (at your option) any
			* later version.
			*
			* Akantu is distributed in the hope that it will be useful, but WITHOUT ANY
			* WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
			* A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
			* details.
			*
			* You should have received a copy of the GNU Lesser General Public License
			* along with Akantu. If not, see <http://www.gnu.org/licenses/>.
			*
			*/

			/* -------------------------------------------------------------------------- */
			inline FEEngine & HeatTransferModel::getFEEngineBoundary(std::string name) {
			return dynamic_cast<FEEngine &>(getFEEngineClassBoundary<MyFEEngineType>(name));
			}

			/* -------------------------------------------------------------------------- */
			inline UInt HeatTransferModel::getNbDataToPack(SynchronizationTag tag) const{
			AKANTU_DEBUG_IN();

			UInt size = 0;
			UInt nb_nodes = getFEEngine().getMesh().getNbNodes();

			switch(tag) {
			case _gst_htm_temperature:
			case _gst_htm_capacity: {
			size += nb_nodes * sizeof(Real);
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}

			AKANTU_DEBUG_OUT();
			return size;
			}

			/* -------------------------------------------------------------------------- */
			inline UInt HeatTransferModel::getNbDataToUnpack(SynchronizationTag tag) const{
			AKANTU_DEBUG_IN();

			UInt size = 0;
			UInt nb_nodes = getFEEngine().getMesh().getNbNodes();

			switch(tag) {
			case _gst_htm_capacity:
			case _gst_htm_temperature: {
			size += nb_nodes * sizeof(Real);
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}

			AKANTU_DEBUG_OUT();
			return size;
			}

			/* -------------------------------------------------------------------------- */
			inline void HeatTransferModel::packData(CommunicationBuffer & buffer,
			const UInt index,
			SynchronizationTag tag) const{
			AKANTU_DEBUG_IN();

			switch(tag) {
			case _gst_htm_capacity:
			buffer << (*capacity_lumped)(index);
			break;
			case _gst_htm_temperature: {
			buffer << (*temperature)(index);
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}

			AKANTU_DEBUG_OUT();
			}

			/* -------------------------------------------------------------------------- */
			inline void HeatTransferModel::unpackData(CommunicationBuffer & buffer,
			const UInt index,
			SynchronizationTag tag) {
			AKANTU_DEBUG_IN();

			switch(tag) {
			case _gst_htm_capacity: {
			buffer >> (*capacity_lumped)(index);
			break;
			}
			case _gst_htm_temperature: {
			buffer >> (*temperature)(index);
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}

			AKANTU_DEBUG_OUT();
			}

			/* -------------------------------------------------------------------------- */
			inline UInt HeatTransferModel::getNbDataForElements(const Array<Element> & elements,
			SynchronizationTag tag) const {
			AKANTU_DEBUG_IN();

			UInt size = 0;
			UInt nb_nodes_per_element = 0;
			Array<Element>::const_iterator<Element> it = elements.begin();
			Array<Element>::const_iterator<Element> end = elements.end();
			for (; it != end; ++it) {
			const Element & el = *it;
			nb_nodes_per_element += Mesh::getNbNodesPerElement(el.type);
			}

			#ifndef AKANTU_NDEBUG
			size += elements.getSize() * spatial_dimension * sizeof(Real); /// position of the barycenter of the element (only for check)
			// size += spatial_dimension * nb_nodes_per_element * sizeof(Real); /// position of the nodes of the element
			#endif

			switch(tag) {
			case _gst_htm_capacity: {
			size += nb_nodes_per_element * sizeof(Real); // capacity vector
			break;
			}
			case _gst_htm_temperature: {
			size += nb_nodes_per_element * sizeof(Real); // temperature
			break;
			}
			case _gst_htm_gradient_temperature: {
			size += getNbQuadraturePoints(elements) * spatial_dimension * sizeof(Real); // temperature gradient
			size += nb_nodes_per_element * sizeof(Real); // nodal temperatures
			// size += spatial_dimension * nb_nodes_per_element * sizeof(Real); // shape derivatives
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}

			AKANTU_DEBUG_OUT();
			return size;
			}

			/* -------------------------------------------------------------------------- */
			inline void HeatTransferModel::packElementData(CommunicationBuffer & buffer,
			const Array<Element> & elements,
			SynchronizationTag tag) const {
			#ifndef AKANTU_NDEBUG
			Array<Element>::const_iterator<Element> bit = elements.begin();
			Array<Element>::const_iterator<Element> bend = elements.end();
			for (; bit != bend; ++bit) {
			const Element & element = *bit;
			Vector<Real> barycenter(spatial_dimension);
			mesh.getBarycenter(element.element, element.type, barycenter.storage(), element.ghost_type);
			buffer << barycenter;
			}
			// packNodalDataHelper(mesh.getNodes(), buffer, elements);
			#endif

			switch (tag){
			case _gst_htm_capacity: {
			packNodalDataHelper(*capacity_lumped, buffer, elements, mesh);
			break;
			}
			case _gst_htm_temperature: {
			packNodalDataHelper(*temperature, buffer, elements, mesh);
			break;
			}
			case _gst_htm_gradient_temperature: {
			packElementalDataHelper(temperature_gradient, buffer, elements, true, getFEEngine());
			packNodalDataHelper(*temperature, buffer, elements, mesh);
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}
			}

			/* -------------------------------------------------------------------------- */
			inline void HeatTransferModel::unpackElementData(CommunicationBuffer & buffer,
			const Array<Element> & elements,
			SynchronizationTag tag) {
			#ifndef AKANTU_NDEBUG
			Array<Element>::const_iterator<Element> bit = elements.begin();
			Array<Element>::const_iterator<Element> bend = elements.end();
			for (; bit != bend; ++bit) {
			const Element & element = *bit;

			Vector<Real> barycenter_loc(spatial_dimension);
			mesh.getBarycenter(element.element, element.type, barycenter_loc.storage(), element.ghost_type);

			Vector<Real> barycenter(spatial_dimension);
			buffer >> barycenter;
			Real tolerance = 1e-15;
			for (UInt i = 0; i < spatial_dimension; ++i) {
			if(!(std::abs(barycenter(i) - barycenter_loc(i)) <= tolerance))
			AKANTU_DEBUG_ERROR("Unpacking an unknown value for the element: "
			<< element
			<< "(barycenter[" << i << "] = " << barycenter_loc(i)
			<< " and buffer[" << i << "] = " << barycenter(i) << ") - tag: " << tag);
			}
			}

			// Vector<Real> coords(spatial_dimension);
			// Real * nodes = getFEEngine().getMesh().getNodes().storage();
			// for (UInt n = 0; n < nb_nodes_per_element; ++n) {
			// buffer >> coords;
			// UInt offset_conn = conn[el_offset + n];
			// Real * coords_local = nodes+spatial_dimension*offset_conn;
			// for (UInt i = 0; i < spatial_dimension; ++i) {
			// if(!(std::abs(coords(i) - coords_local[i]) <= tolerance))
			// AKANTU_EXCEPTION("Unpacking to wrong node for the element : "
			// << element
			// << "(coords[" << i << "] = " << coords_local[i]
			// << " and buffer[" << i << "] = " << coords(i) << ")");
			// }
			// }
			#endif

			switch (tag){
			case _gst_htm_capacity: {
			unpackNodalDataHelper(*capacity_lumped, buffer, elements, mesh);
			break;
			}
			case _gst_htm_temperature: {
			unpackNodalDataHelper(*temperature, buffer, elements, mesh);
			break;
			}
			case _gst_htm_gradient_temperature: {
			unpackElementalDataHelper(temperature_gradient, buffer, elements, true, getFEEngine());
			unpackNodalDataHelper(*temperature, buffer, elements, mesh);

			// // Real tolerance = 1e-15;
			// if (!Math::are_vector_equal(spatial_dimension,gtemp.storage(),it_gtemp[element.element].storage())){
			// Real dist = Math::distance_3d(gtemp.storage(), it_gtemp[element.element].storage());
			// debug::debugger.getOutputStream().precision(20);
			// std::stringstream temperatures_str;
			// temperatures_str.precision(20);
			// temperatures_str << std::scientific << "temperatures are ";
			// for (UInt n = 0; n < nb_nodes_per_element; ++n) {
			// UInt offset_conn = conn[el_offset + n];
			// temperatures_str << (*temperature)(offset_conn) << " ";
			// }
			// Array<Real>::matrix_iterator it_shaped =
			// const_cast<Array<Real> &>(getFEEngine().getShapesDerivatives(element.type, ghost_type))
			// .begin(nb_nodes_per_element,spatial_dimension);


			// AKANTU_EXCEPTION("packed gradient do not match for element " << element.element << std::endl
			// << "buffer is " << gtemp << " local is " << it_gtemp[element.element]
			// << " dist is " << dist << std::endl
			// << temperatures_str.str() << std::endl
			// << std::scientific << std::setprecision(20)
			// << " distant temperatures " << temp_nodes
			// << "temperature gradient size " << temperature_gradient(element.type, ghost_type).getSize()
			// << " number of ghost elements " << getFEEngine().getMesh().getNbElement(element.type,_ghost)
			// << std::scientific << std::setprecision(20)
			// << " shaped " << shaped
			// << std::scientific << std::setprecision(20)
			// << " local shaped " << it_shaped[element.element]);
			// }
			break;
			}
			default: {
			AKANTU_DEBUG_ERROR("Unknown ghost synchronization tag : " << tag);
			}
			}
			}

			/* -------------------------------------------------------------------------- */

			template<SolveConvergenceMethod cmethod, SolveConvergenceCriteria criteria>
			bool HeatTransferModel::solveStep(Real tolerance,
			UInt max_iteration) {
			Real error = 0.;
			return this->template solveStep<cmethod,criteria>(tolerance,
			error,
			max_iteration);
			}

			/* -------------------------------------------------------------------------- */
			template<SolveConvergenceMethod cmethod, SolveConvergenceCriteria criteria>
			bool HeatTransferModel::solveStep(Real tolerance, Real & error,
			UInt max_iteration,
			bool do_not_factorize) {
			//EventManager::sendEvent(HeatTransferModelEvent::BeforeSolveStepEvent(method));
			this->implicitPred();
			this->updateResidual();

			AKANTU_DEBUG_ASSERT(conductivity_matrix != NULL,
			"You should first initialize the implicit solver and assemble the conductivity matrix");

			bool need_factorize = !do_not_factorize;

			if (method == _implicit_dynamic) {
			AKANTU_DEBUG_ASSERT(capacity_matrix != NULL,
			"You should first initialize the implicit solver and assemble the mass matrix");
			}

			switch (cmethod) {
			case _scm_newton_raphson_tangent:
			case _scm_newton_raphson_tangent_not_computed:
			break;
			case _scm_newton_raphson_tangent_modified:
			this->assembleConductivityMatrix();
			break;
			default:
			AKANTU_DEBUG_ERROR("The resolution method " << cmethod << " has not been implemented!");
			}

			this->n_iter = 0;
			bool converged = false;
			error = 0.;
			if(criteria == _scc_residual) {
			converged = this->testConvergence<criteria> (tolerance, error);
			if(converged) return converged;
			}

			do {
			if (cmethod == _scm_newton_raphson_tangent)
			this->assembleConductivityMatrix();

			solve<GeneralizedTrapezoidal::_temperature_corrector>(*increment, 1., need_factorize);

			this->implicitCorr();

			if(criteria == _scc_residual) this->updateResidual();

			converged = this->testConvergence<criteria> (tolerance, error);

			if(criteria == _scc_increment && !converged) this->updateResidual();
			// this->dump();

			this->n_iter++;
			AKANTU_DEBUG_INFO("[" << criteria << "] Convergence iteration "
			<< std::setw(std::log10(max_iteration)) << this->n_iter
			<< ": error " << error << (converged ? " < " : " > ") << tolerance);

			switch (cmethod) {
			case _scm_newton_raphson_tangent:
			need_factorize = true;
			break;
			case _scm_newton_raphson_tangent_not_computed:
			case _scm_newton_raphson_tangent_modified:
			need_factorize = false;
			break;
			default:
			AKANTU_DEBUG_ERROR("The resolution method " << cmethod << " has not been implemented!");
			}


			} while (!converged && this->n_iter < max_iteration);

			// this makes sure that you have correct strains and stresses after the solveStep function (e.g., for dumping)
			if(criteria == _scc_increment) this->updateResidual();

			if (converged) {
			//EventManager::sendEvent(HeatTransferModelEvent::AfterSolveStepEvent(method));
			} else if(this->n_iter == max_iteration) {
			AKANTU_DEBUG_WARNING("[" << criteria << "] Convergence not reached after "
			<< std::setw(std::log10(max_iteration)) << this->n_iter <<
			" iteration" << (this->n_iter == 1 ? "" : "s") << "!" << std::endl);
			}

			return converged;
			}

			/* -------------------------------------------------------------------------- */

			template <GeneralizedTrapezoidal::IntegrationSchemeCorrectorType type>
			void HeatTransferModel::solve(Array<Real> &increment, Real block_val,
			bool need_factorize, bool has_profile_changed){

			updateResidualInternal(); //doesn't do anything for static

			if(need_factorize) {
			Real c = 0.,e = 0.;

			if(method == _static) {
			AKANTU_DEBUG_INFO("Solving K inc = r");
			e = 1.;
			} else {
			AKANTU_DEBUG_INFO("Solving (c M + e K) inc = r");

			GeneralizedTrapezoidal * trap_int = dynamic_cast<GeneralizedTrapezoidal*>(integrator);
			c = trap_int->getTemperatureRateCoefficient<type>(time_step);
			e = trap_int->getTemperatureCoefficient<type>(time_step);

			// std::cout << "c " << c << " e " << e << std::endl;
			}


			jacobian_matrix->clear();
			// J = c M + e K
			if(conductivity_matrix)
			jacobian_matrix->add(*conductivity_matrix, e);

			if(capacity_matrix)
			jacobian_matrix->add(*capacity_matrix, c);

			#if !defined(AKANTU_NDEBUG)
			// if(capacity_matrix && AKANTU_DEBUG_TEST(dblDump))
			capacity_matrix->saveMatrix("M.mtx");
			#endif

			jacobian_matrix->applyBoundary(*blocked_dofs, block_val);

			#if !defined(AKANTU_NDEBUG)
			//if(AKANTU_DEBUG_TEST(dblDump))
			jacobian_matrix->saveMatrix("J.mtx");
			#endif
			solver->factorize();
			}

			// if (rhs.getSize() != 0)
			// solver->setRHS(rhs);
			// else

			solver->setOperators();

			solver->setRHS(*residual);

			// solve @f[ J \delta w = r @f]
			solver->solve(increment);

			UInt nb_nodes = temperature->getSize();
			UInt nb_degree_of_freedom = temperature->getNbComponent() * nb_nodes;

			bool * blocked_dofs_val = blocked_dofs->storage();
			Real * increment_val = increment.storage();

			for (UInt j = 0; j < nb_degree_of_freedom;
			++j,++increment_val, ++blocked_dofs_val) {
			if ((*blocked_dofs_val))
			*increment_val = 0.0;
			}

			}

			/* -------------------------------------------------------------------------- */

rAKA/src/model/heat_transfer/heat_transfer_model_inline_impl.cca65be22768cafeatures/50-using…

heat_transfer_model_inline_impl.ccView Options

rAKA/src/model/heat_transfer/heat_transfer_model_inline_impl.cc
a65be22768cafeatures/50-using…

heat_transfer_model_inline_impl.cc
View Options