Auswahl der wissenschaftlichen Literatur zum Thema „DMRG algorithme“

We propose a new method for the calculation of thermodynamic properties of one-dimensional quantum systems by combining the TMRG approach with the corner transfer-matrix method. The corner transfer-matrix DMRG method brings reasonable advantage over TMRG for classical systems. We have modified the concept for the calculation of thermal properties of one-dimensional quantum systems. The novel QCTMRG algorithm is implemented and used to study two simple test cases, the classical Ising chain and the isotropic Heisenberg model. In a discussion, the advantages and challenges are illuminated.

Since its inception, the DMRG method has been a very powerful tool for the calculation of physical properties of low-dimensional strongly correlated systems. It has been adapted to obtain dynamical properties and to consider finite temperature, time-dependent problems, bosonic degrees of freedom, the treatment of classical problems and non-equilibrium systems, among others. We will briefly review the method and then concentrate on its latest developments, describing some recent successful applications. In particular we will show how the dynamical DMRG can be used together with the Dynamical Mean Field Theory (DMFT) to solve the associated impurity problem in the infinite-dimensional Hubbard model. This method is used to obtain spectral properties of strongly correlated systems. With this algorithm, more complex problems having a larger number of degrees of freedom can be considered and finite-size effects can be minimized.

We benchmark a variant of the recently introduced density matrix renormalization group (DMRG)-X algorithm against exact results for the localized random field XX chain. We find that the eigenstates obtained via DMRG-X exhibit a highly accurate l-bit description for system sizes much bigger than the direct, many-body, exact diagonalization in the spin variables is able to access. We take advantage of the underlying free fermion description of the XX model to accurately test the strengths and limitations of this algorithm for large system sizes. We discuss the theoretical constraints on the performance of the algorithm from the entanglement properties of the eigenstates, and its actual performance at different values of disorder. A small but significant improvement to the algorithm is also presented, which helps significantly with convergence. We find that, at high entanglement, DMRG-X shows a bias towards eigenstates with low entanglement, but can be improved with increased bond dimension. This result suggests that one must be careful when applying the algorithm for interacting many-body localized spin models near a transition. This article is part of the themed issue ‘Breakdown of ergodicity in quantum systems: from solids to synthetic matter’.

The density-matrix renormalization group (DMRG) method has established itself over the last decade as the leading method for the simulation of the statics and dynamics of one-dimensional strongly correlated quantum lattice systems. The DMRG is a method that shares features of a renormalization group procedure (which here generates a flow in the space of reduced density operators) and of a variational method that operates on a highly interesting class of quantum states, so-called matrix product states (MPSs). The DMRG method is presented here entirely in the MPS language. While the DMRG generally fails in larger two-dimensional systems, the MPS picture suggests a straightforward generalization to higher dimensions in the framework of tensor network states. The resulting algorithms, however, suffer from difficulties absent in one dimension, apart from a much more unfavourable efficiency, such that their ultimate success remains far from clear at the moment.

AbstractIn this paper, the concept of the DMRG minimization scheme is extended to several important operations in the TT-format, like the matrix-by-vector product and the conversion from the canonical format to the TT-format. Fast algorithms are implemented and a stabilization scheme based on randomization is proposed. The comparison with the direct method is performed on a sequence of matrices and vectors coming as approximate solutions of linear systems in the TT-format. A generated example is provided to show that randomization is really needed in some cases. The matrices and vectors used are available from the author or at http://spring.inm.ras.ru/osel

Device-to-device (D2D) technology is a promising technique in terms of being capable of providing efficiency, decreased latency, improved data rate, and increased capacity to cellular networks. Allocating power to users in order to reduce energy consumption and maintain quality of service (QoS) remains a major challenge in D2D communications. In this paper, we aim to maximize the throughput of D2D users and cellular users subject to QoS requirements and signal-to-interference-plus-noise ratio (SINR). To this end, we propose a resource and power allocation approach called optimal power allocation and delay minimization based on the conflict graph (OP-DMCG) algorithm that considers optimal power allocation for D2D multi-users in the cellular uplink channels and minimization of the total network delay using conflict graphs. Based on the simulations presented in this paper, we show that the proposed OP-DMCG algorithm outperforms the greedy throughput maximization plus (GTM+), delay minimization conflict graph (DMCG), and power and delay optimization based uplink resource allocation (PDO-URA) algorithms in terms of both total network throughput and total D2D throughput.

The results of improving the algorithms for diagnosing hail on the Earth surface and its size based on the DMRL-C network and numerical prediction data are presented. The algorithms are implemented as part of an automated technology that operates in real time, with the presentation of results in a database and in the form of maps. A two-hour (relative to the observation period) animation of the diagnosed hail zones is provided. An algorithm for identifying the phase state of precipitation in a cloud is implemented, which made it possible to refine the diagnosis of hail in late spring and early autumn, rejecting cases with snow and ice pellets. The results of verifying the probability of detection of hail are given. It is concluded that the results of improving hail diagnosis using radar data will make it possible to refine data on cases of hail reaching the Earth surface in the European part of Russia, supplementing the already existing ones, according to the information of the network of meteorological and remote observations, which is also of practical importance for producing more accurate storm warnings about the occurrence of the phenomenon. Keywords: hail, diagnosis, phase state of precipitation in a cloud, radar data, DMRL-C network, automated technology

Over the last decade, the density-matrix renormalization group (DMRG) has emerged as the most powerful method for the simulation of strongly correlated one-dimensional (1D) quantum systems. Input from quantum information has allowed to trace the method's performance to the entanglement properties of quantum states, revealing why it works so well in 1D and not so well in 2D; it has allowed to devise algorithms for time-dependent quantum systems and, by clarifying the link between DMRG and Wilson's numerical renormalization group (NRG), for quantum impurity systems.

The purposes are to manage human resource data better and explore the association between Human Resource Management (HRM), data mining, and economic management. An Ensemble Classifier-Decision Tree (EC-DT) algorithm is proposed based on the single decision tree algorithm to analyze HRM data. The involved single decision tree algorithms include C4.5, Random Tree, J48, and SimpleCart. Then, an HRM system is established based on the designed algorithm, and the evaluation management and talent recommendation modules are tested. Finally, the designed algorithm is compared and tested. Experimental results suggest that C4.5 provides the highest classification accuracy among the single decision tree algorithms, reaching 76.69%; in contrast, the designed EC-DT algorithm can provide a classification accuracy of 79.97%. The proposed EC-DT algorithm is compared with the Content-based Recommendation Method (CRM) and the Collaborative Filtering Recommendation Method (CFRM), revealing that its Data Mining Recommendation Method (DMRM) can provide the highest accuracy and recall, reaching 35.2% and 41.6%, respectively. Therefore, the data mining-based HRM system can promote and guide enterprises to develop according to quantitative evaluation results. The above results can provide a reference for studying HRM systems based on data mining technology.

La seguente tesi presenta l'implementazione in codice C++ di un algoritmo per l'evoluzione temporale del DMRG, basato sull'approssimazione di Trotter per primi vicini. Il corretto funzionamento del codice è stato controllato calcolando un ristema risolubile esattamente, la catena di fermioni liberi senza spin (con condizioni al contorno aperte); in seguito il nuovo algoritmo è stato comparato con l'evoluzione temporale del DMRG basata sul metodo di Runge-Kutta. L'analisi degli errori ha mostrato come, per brevi periodi di tempo, il metodo di Runge-Kutta sia il più adatto fra i due, mentre per periodi di media durata il metodo di Trotter offra prestazioni migliori. Le evoluzioni temporali per tempi elevati sono attualmente al di là della portata di entrambi gli algoritmi.

Cette thèse présente de nouveaux algorithmes numériques et effectue une étude approfondie de certaines méthodes numériques existantes pour relever les défis de haute dimension résultant de la résolution de l'équation de Schrödinger électronique en chimie quantique. En se concentrant sur deux problèmes spécifiques, notre approche implique l'identification et l'exploitation des symétries et des structures de rang faible au sein de matrices et de tenseurs. Le premier problème abordé dans cette thèse concerne l'évaluation numérique efficace de la composante à longue portée du potentiel de Coulomb à séparation de portée et des intégrales à deux électrons à longue portée, un tenseur du quatrième ordre qui intervient dans de nombreuses méthodes de chimie quantique. Nous présentons deux nouvelles méthodes d'approximation. Cela est réalisé en s'appuyant sur l'interpolation Chebyshev, des règles de quadrature Gaussienne combinées à des approximations de rang faible ainsi que des méthodes rapides multipolaires (FMM). Ce travail offre une explication détaillée de ces approches et algorithmes introduits, accompagnée d'une comparaison approfondie entre les méthodes nouvellement proposées. Le deuxième problème abordé concerne l'exploitation des symétries et des structures de rang faible pour dériver des représentations efficaces en train de tenseurs des opérateurs impliqués dans l'algorithme DMRG. Cet algorithme est une méthode d'optimisation itérative précise utilisée pour résoudre numériquement l'équation de Schrödinger indépendante du temps. Ce travail vise à comprendre et interpréter les résultats obtenus par les communautés de physique et de chimie, et cherche à offrir des perspectives théoriques nouvelles qui, selon nos connaissances, n'ont pas reçu une attention significative auparavant. Nous menons une étude approfondie et fournissons des démonstrations, si nécessaire, pour explorer l'existence d'une représentation particulière en train de tenseurs, creuse par blocs, de l'opérateur Hamiltonien et de sa fonction d'onde associée. Cela est réalisé tout en maintenant les lois de conservation physiques, manifestées sous forme de symétries de groupe dans les tenseurs, telles que la conservation du nombre de particules. La troisième partie de ce travail est dédiée à la réalisation d'une bibliothèque prototype en Julia, pour l'implémentation de DMRG qui est conçue pour le modèle d'opérateur Hamiltonien de la chimie quantique. Nous exploitons ici la représentation en train de tenseurs, creuse par blocs, de l'opérateur et de la fonction d'onde (fonction propre). Avec ces structures, notre objectif est d'accélérer les étapes les plus coûteuses de la DMRG, y compris les contractions de tenseurs, les opérations matrice-vecteur, et la compression de matrices par décomposition en valeurs singulières tronquée. De plus, nous fournissons des résultats issus de diverses simulations moléculaires, tout en comparant les performances de notre bibliothèque avec la bibliothèque ITensors de pointe, où nous démontrons avoir atteint une performance similaire
This thesis presents novel numerical algorithms and conducts a comprehensive study of some existing numerical methods to address high-dimensional challenges arising from the resolution of the electronic Schrödinger equation in quantum chemistry. Focusing on two specific problems, our approach involves the identification and exploitation of symmetries and low-rank structures within matrices and tensors, aiming to mitigate the curse of dimensionality. The first problem considered in this thesis is the efficient numerical evaluation of the long-range component of the range-separated Coulomb potential and the long-range two-electron integrals 4th-order tensor which occurs in many quantum chemistry methods. We present two novel approximation methods. This is achieved by relying on tensorized Chebyshev interpolation, Gaussian quadrature rules combined with low-rank approximations as well as Fast Multipole Methods (FMM). This work offers a detailed explanation of these introduced approaches and algorithms, accompanied by a thorough comparison between the newly proposed methods. The second problem of interest is the exploitation of symmetries and low-rank structures to derive efficient tensor train representations of operators involved in the Density Matrix Renormalization Group (DMRG) algorithm. This algorithm, referred to as the Quantum Chemical DMRG (QC-DMRG) when applied in the field of quantum chemistry, is an accurate iterative optimization method employed to numerically solve the time-independent Schrödinger equation. This work aims to understand and interpret the results obtained from the physics and chemistry communities and seeks to offer novel theoretical insights that, to the best of our knowledge, have not received significant attention before. We conduct a comprehensive study and provide demonstrations, when necessary, to explore the existence of a particular block-sparse tensor train representation of the Hamiltonian operator and its associated eigenfunction. This is achieved while maintaining physical conservation laws, manifested as group symmetries in tensors, such as the conservation of the particle number. The third part of this work is dedicated to the realization of a proof-of-concept Quantum Chemical DMRG (QC-DMRG) Julia library, designed for the quantum chemical Hamiltonian operator model. We exploit here the block-sparse tensor train representation of both the operator and the eigenfunction. With these structures, our goal is to speed up the most time-consuming steps in QC-DMRG, including tensor contractions, matrix-vector operations, and matrix compression through truncated Singular Value Decompositions (SVD). Furthermore, we provide empirical results from various molecular simulations, while comparing the performance of our library with the state-of-the-art ITensors library where we show that we attain a similar performance

The present thesis deals with a theoretical study of electronic structures in -conjugated molecular materials with focus on their application in organic elec-tronics. We also discuss a modified and efficient symmetrized DMRG algorithm for studying excited states in these systems. In recent times, organic conjugated systems have emerged as potential candidates in a wide range of fascinating fields by virtue of their tunable electronic properties, easy processability and low cost. Tunability in the electronic and optical properties primarily are centered on the or-dering and nature of the low-lying excited states. Probing these important excited states also demands development of efficient and adaptable techniques. Chapter 1 provides a basic overview of conjugated organic polymers which have been utilized over decades in diverse fields as in organic light emitting diodes (OLED), organic solar cells (OSC) and non-linear optical (NLO) devices. These systems also contribute significantly to theoretical understanding as they pro vide important insights of one and quasi-one dimensional systems. In this chapter, we have given basic description of the electronic processes in OLED and OSC along with a brief theoretical description of -conjugated organic systems. Chapter 2 gives an account of the numerical techniques which are necessary for the study of low-dimensional strongly correlated systems like -conjugated sys-tems. For this purpose, effective low-energy model Hamiltonians viz. Huckel,¨ Hubbard and Pariser-Parr-Pople Hamiltonians are discussed. Exact diagonalization technique within the diagrammatic valence bond (DVB) basis and density matrix renormalization group (DMRG) technique are discussed in details. We have also given brief accounts of the methods employed to study real-time dynamics. A short description of different computational techniques for the study of NLO properties in -conjugated systems is also provided. Engineering the position of the lowest triplet state (T1) relative to the first excited singlet state (S1) is of great importance in improving the efficiencies of organic light emitting diodes and organic photovoltaic cells. In chapter 3, we have carried out model exact calculations of substituted polyene chains to understand the fac-tors that affect the energy gap between S1 and T1. The factors studied are backbone dimerization, different donor-acceptor substitutions and twisted backbone geome-try. The largest system studied is an eighteen carbon polyene which spans a Hilbert space of about 991 million in the triplet subspace. We show that for reverse inter-system crossing (RISC) process, the best choice involves substituting all carbon sites on one half of the polyene with donors and the other half with acceptors. Singlet fission (SF) is a potential pathway for significant enhancement of efficiency in OSC. In chapter 4, we study singlet fission in a pair of polyene molecules in two different stacking arrangements employing exact many-body wave packet dy-namics. In the non-interacting model, SF is absent. The individual molecules are treated within Hubbard and Pariser-Parr-Pople (PPP) models and the interac-tion between them involves transfer terms, intersite electron repulsions and site-charge—bond-charge repulsion terms. Initial wave packet is construc ted from ex-cited singlet state of one molecule and ground state of the other. Time develop-ment of this wave packet under the influence of intermolecular interactions is fol-lowed within the Schrodinger¨ picture by an efficient predictor-corrector scheme. In unsubstituted Hubbard and PPP chains, 21A state leads to significant SF yield while the 11B state gives negligible fission yield. On substitution by donor-acceptor groups of moderate strength, the lowest excited state will have sufficient 2 1A char-acter and hence gives significant SF yield. Because of rapid internal c onversion, the nature of the lowest excited singlet will determine the SF contribution to OSC effi - ciency. Furthermore, we find the fission yield depends considerably on th e stacking arrangement of the polyene molecules. In chapter 5, we have given an account of a new modified algorithm for symmetry adaptation within symmetrized density matrix renormalization group (SDMRG) technique. SDMRG technique has been an efficient method for studying low-lying eigenstates in one and quasi-one dimensional electronic systems. However, SDMRG method until now, had bottlenecks involving construction of linearly in-dependent symmetry adapted basis states as the symmetry matrices in the DMRG basis were not sparse. Our modified algorithm overcomes this bottleneck. T he new method incorporates end-to-end interchange symmetry (C2), electron-hole symmetry (J) and parity or spin-flip symmetry (P) in these calculations. The one-to-one correspondence between direct-product basis states in the DMRG Hilbert space for these symmetry operations renders the symmetry matrices in the new ba-sis with maximum sparseness, just one non-zero matrix element per row. Using methods similar to those employed in exact diagonalization technique for Pariser-Parr-Pople (PPP) models, developed in the eighties, it is possible to construct or-thogonal SDMRG basis states while bypassing the slow step of Gram-Schmidt orthonormalization procedure. The method together with the PPP model which incorporates long-range electronic correlations is employed to study the correlated excited states of 1,12-benzoperylene. In chapter 6, we have studied the correlated excited states of coronene and ova-lene within Pariser-Parr-Pople (PPP) model employing symmetry adapted density matrix renormalization group technique. These polynuclear aromatic hydrocar-bons can be considered as graphene nanoflakes and study of their ele ctronic struc-tures will shed light on the electron correlation effects in these finite-size gr aphene analogues. The electron correlation effect usually diminishes on going from one-dimensional to higher-dimensional systems, yet, it is significant within these fin ite-size graphene derivatives where it depends on the molecular topology. We have characterized these low-lying energy states by calculating bond orders, spin den-sities in the lowest triplet state and two-photon absorption cross-sections for low-lying two-photon states. vi

The present thesis deals with a theoretical study of electronic structures in -conjugated molecular materials with focus on their application in organic elec-tronics. We also discuss a modified and efficient symmetrized DMRG algorithm for studying excited states in these systems. In recent times, organic conjugated systems have emerged as potential candidates in a wide range of fascinating fields by virtue of their tunable electronic properties, easy processability and low cost. Tunability in the electronic and optical properties primarily are centered on the or-dering and nature of the low-lying excited states. Probing these important excited states also demands development of efficient and adaptable techniques. Chapter 1 provides a basic overview of conjugated organic polymers which have been utilized over decades in diverse fields as in organic light emitting diodes (OLED), organic solar cells (OSC) and non-linear optical (NLO) devices. These systems also contribute significantly to theoretical understanding as they pro vide important insights of one and quasi-one dimensional systems. In this chapter, we have given basic description of the electronic processes in OLED and OSC along with a brief theoretical description of -conjugated organic systems. Chapter 2 gives an account of the numerical techniques which are necessary for the study of low-dimensional strongly correlated systems like -conjugated sys-tems. For this purpose, effective low-energy model Hamiltonians viz. Huckel,¨ Hubbard and Pariser-Parr-Pople Hamiltonians are discussed. Exact diagonalization technique within the diagrammatic valence bond (DVB) basis and density matrix renormalization group (DMRG) technique are discussed in details. We have also given brief accounts of the methods employed to study real-time dynamics. A short description of different computational techniques for the study of NLO properties in -conjugated systems is also provided. Engineering the position of the lowest triplet state (T1) relative to the first excited singlet state (S1) is of great importance in improving the efficiencies of organic light emitting diodes and organic photovoltaic cells. In chapter 3, we have carried out model exact calculations of substituted polyene chains to understand the fac-tors that affect the energy gap between S1 and T1. The factors studied are backbone dimerization, different donor-acceptor substitutions and twisted backbone geome-try. The largest system studied is an eighteen carbon polyene which spans a Hilbert space of about 991 million in the triplet subspace. We show that for reverse inter-system crossing (RISC) process, the best choice involves substituting all carbon sites on one half of the polyene with donors and the other half with acceptors. Singlet fission (SF) is a potential pathway for significant enhancement of efficiency in OSC. In chapter 4, we study singlet fission in a pair of polyene molecules in two different stacking arrangements employing exact many-body wave packet dy-namics. In the non-interacting model, SF is absent. The individual molecules are treated within Hubbard and Pariser-Parr-Pople (PPP) models and the interac-tion between them involves transfer terms, intersite electron repulsions and site-charge—bond-charge repulsion terms. Initial wave packet is construc ted from ex-cited singlet state of one molecule and ground state of the other. Time develop-ment of this wave packet under the influence of intermolecular interactions is fol-lowed within the Schrodinger¨ picture by an efficient predictor-corrector scheme. In unsubstituted Hubbard and PPP chains, 21A state leads to significant SF yield while the 11B state gives negligible fission yield. On substitution by donor-acceptor groups of moderate strength, the lowest excited state will have sufficient 2 1A char-acter and hence gives significant SF yield. Because of rapid internal c onversion, the nature of the lowest excited singlet will determine the SF contribution to OSC effi - ciency. Furthermore, we find the fission yield depends considerably on th e stacking arrangement of the polyene molecules. In chapter 5, we have given an account of a new modified algorithm for symmetry adaptation within symmetrized density matrix renormalization group (SDMRG) technique. SDMRG technique has been an efficient method for studying low-lying eigenstates in one and quasi-one dimensional electronic systems. However, SDMRG method until now, had bottlenecks involving construction of linearly in-dependent symmetry adapted basis states as the symmetry matrices in the DMRG basis were not sparse. Our modified algorithm overcomes this bottleneck. T he new method incorporates end-to-end interchange symmetry (C2), electron-hole symmetry (J) and parity or spin-flip symmetry (P) in these calculations. The one-to-one correspondence between direct-product basis states in the DMRG Hilbert space for these symmetry operations renders the symmetry matrices in the new ba-sis with maximum sparseness, just one non-zero matrix element per row. Using methods similar to those employed in exact diagonalization technique for Pariser-Parr-Pople (PPP) models, developed in the eighties, it is possible to construct or-thogonal SDMRG basis states while bypassing the slow step of Gram-Schmidt orthonormalization procedure. The method together with the PPP model which incorporates long-range electronic correlations is employed to study the correlated excited states of 1,12-benzoperylene. In chapter 6, we have studied the correlated excited states of coronene and ova-lene within Pariser-Parr-Pople (PPP) model employing symmetry adapted density matrix renormalization group technique. These polynuclear aromatic hydrocar-bons can be considered as graphene nanoflakes and study of their ele ctronic struc-tures will shed light on the electron correlation effects in these finite-size gr aphene analogues. The electron correlation effect usually diminishes on going from one-dimensional to higher-dimensional systems, yet, it is significant within these fin ite-size graphene derivatives where it depends on the molecular topology. We have characterized these low-lying energy states by calculating bond orders, spin den-sities in the lowest triplet state and two-photon absorption cross-sections for low-lying two-photon states. vi

Reinforcement learning (RL) has been developed for mean field games over graphs (G-MFG) in social media and network economics, in which the transition of agents between a node pair incurs an instantaneous reward. However, agents’ en-route choices on edges are largely neglected that incur an experienced reward depending on agents’ actions and population evolution along edges. Here we focus on a broader class of MFGs, named “dual MFG on graphs” (G-dMFG), which models two interacting MFGs, namely, one on edges and one at nodes over a graph. In this setting, agents select travel speed along edges and next-go-to edge at nodes for a minimum cumulative cost, which arises from the congestion effect when many agents compete for the same resource. This has various implications for autonomous driving navigation, spatial resource allocation, and internet packet routing. We establish formally that G-dMFG is a generic G-MFG, encompassing a more complex cost structure (that is nonseparable between states and actions) and with no need to pre-specify a termination time horizon. RL algorithms are designed to solve mean field equilibria (MFE) on large networks.

Based on an artificial compressibility method, the explicit Runge-Kutta time stepping finite difference algorithm was applied to steady, incompressible, Navier-Stokes equations. A two-dimensional analysis computer code in a generalized curvilinear coordinate system was developed and its accuracy has been compared to known numerical solutions. The algorithm has been accelerated using our new Distributed Minimal Residual (DMR) method, which allows each equation in the system to advance in time with its own optimal speed. The effectiveness of the DMR method was examined for a number of test cases. The accelerated algorithm offers substantial savings of the computing time.

Abstract Remarks in daily drilling reports (DDR) and daily mud reports (DMR) are an invaluable source of information for the analysis of ongoing operations and planning of future wells. These remarks are entered as free text and thus represent unstructured information, which requires expert knowledge for interpretation. Here, we aim to develop a machine learning algorithm to automatically detect events of interest and convert this information into a structured format. Several unscheduled events, such as losses, influx and stuck pipe, were selected to develop a prototype of our natural language processing (NLP) approach for daily DMR remarks. Data selection, data annotation and analysis workflow, and results are discussed in the manuscript.

Abstract To bridge the gap between computer-aided process planning and computer-aided fixture design, an automated setup planning and tolerance decomposition method is developed in this research. Directed graph is extended to represent feature/ dimension/tolerance relationships (FTG) and datum/machining feature relationships (DMG). According to different production schemes and manufacturing resource capabilities, setup planning principles and algorithms are explored to automatically extract DMG from FTG. Under the true positioning frame (ANSI Y14.5), tolerance decomposition models are concluded to partition a tolerance into interoperable machining errors, such as locating error, tool alignment error, random process error. The setup plan with allowable locating tolerance specifications is the basis for fixture design. The proposed method is verified in a genuine manufacturing enterprise.

Abstract Hybrid Manufacturing (HM) combining Additive Manufacturing (AM) and subtractive machining (SM) technologies have recently been introduced and have the potential to address the shortcomings of AM, such as the poor surface finish and requires post-processing of the support structures. One such example of an HM machine is the DMG Mori Lasertec 65. These 5-axis HM machines allow for rapid deposition of material during additive manufacturing and address the issues of feature resolution, surface finish, and tolerances by subtractive machining. Additionally, these processes allow for the creation of complex geometries not possible with standard 5-axis machining. Process planning for HM is a reasonably complex manual task and could benefit from automation. Critical steps in process planning are the decomposition of the part into additive and subtractive features, sequencing all features and assigning the tool-paths for these features. This paper presents algorithms for decomposing the part and sequencing the additive and subtractive features in an automated manner, paving the way for a fully automated system for HM. Examples of a wide range of parts demonstrating the capability of the algorithm are presented.

Breast cancer is one of the most common type of cancer that affects woman in the World. It affects millions of women every year killing hundreds of thousands yearly. Early detection of this disease is essential for improving chances of cure and recovery of the patients, thus, one of the most important factors in patients’ treatment is early detection. Mammography, the golden standard for detection this disease is not always 100% effective, which means that it is not always recommended.In this sense, infrared imaging is a promising technique that might be used as a complementary examination technique in a computer-aided diagnosis system. In this work we used genetic algorithms and convolutional neural networks to classify images from the DMR-IR public database. We analyzed both the entire image and only the breast region. Best results were F1-score of 0.92 for entire images and 0.90 for breast regions