Convergence Tolerance Research Articles

An improved meta-heuristic method for optimal optimization of electric parking lots in distribution network

In this study, a stochastic multi-objective structure for optimization of the intelligent electric parking lots (EPLs) is implemented in the distribution network for minimizing the power losses annual costs, power purchased from the main grid, unsupplied energy of subscribers, cost of vehicles to the grid as well as minimizing the network voltage deviations considering battery degradation cost (BDC) and network load uncertainty (NLUn). In this research, the unscented transformation method (UTM) is used for NLUn modeling and this method is easily applicable and has a low computational cost. An improved meta-heuristic algorithm named improved fire hawks optimization (IFHO) is utilized for decision variables finding defined as the site and size of the EPLs in the distribution network. The conventional fire hawks optimization (FHO) algorithm is inspired by the fire hawks foraging behavior and in this research, the Taylor-based neighborhood technique (TBNT) is used to reduce the dependency and the possibility of becoming trapped in local optimal. To evaluate the proposed methodology, the simulations are implemented in three scenarios (1) EPLs optimization without BDC and NLUn based-UTM, (2) EPLs optimization with BDC and without NLUn, and (3) EPLs optimization with BDC and NLUn. The results of the third scenario considering BDC and NLUn showed that the EPLs optimization integrated with a multi-objective framework by finding the EPL's optimal size and capacity in the network via the IFHO has reduced the annual losses, voltage deviations, ENS cost, and substation cost by 21.06%, 12.15%, 70.82%, and 39.10%, respectively compared to the base distribution network. Additionally, the results demonstrated that incorporating the BDC and NLUn, the annual losses, voltage oscillations, ENS cost, grid cost, and EPLs have increased in comparison with the EPLs optimization without BDC and NLUn based-UTM. In addition, the TBNT based-IFHO superiority has been confirmed in different scenarios by achieving better values of the objectives and also obtaining the convergence process with lower convergence tolerance and higher convergence accuracy.

First-principle calculations of the electronic, vibrational, and thermodynamic properties of nitrogen-rich salt of 3,6-dinitramino-1,2,4,5-tetrazine [(NH4)2(DNAT)

Energy-containing materials such as explosives have attracted considerable interest recently. In the field of high-energy materials, tetrazine and its derivatives can largely meet the requirements of high nitrogen content and oxygen balance. Nitrogen-rich energetic salts are important research subjects. Nitrogen-rich salt of 3,6-dinitramino-1,2,4,5-tetrazine is a high-energy nitrogen-rich material, but there are few related studies. This paper systematically studies the crystal structure and electronic, vibrational, and thermodynamic properties of (NH4)2(DNAT). The lattice parameters of (NH4)2(DNAT) are observed to align well with the experimental values. The properties of electrons are analyzed by band structure and density of states (DOS). The phonon dispersion curves indicate that the compound is dynamically stable. The vibrational modes of bonds and chemical groups are described in detail, and the peaks in the Raman and infrared spectra are assigned to different vibration modes. Based on the vibration characteristics, thermodynamic properties such as enthalpy (H), Helmholtz free energy (F), entropy (S), Gibbs free energy (G), constant volume heat capacity (CV), and Debye temperature (Θ) are analyzed. This article can pave the way for subsequent work or provide data support to other researchers, promoting further research. In this study, we utilized the density functional theory (DFT) for our calculations. The exchange-correlation potential and van der Waals interactions were characterized based on the GGA-PBE + G function calculation. We obtained Brillouin zone integrals using Monkhorst-Pack k-point grids, with the k-point of the Brillouin zone set to a 2 × 2 × 2 grid. During the self-consistent field operation, we set the total energy convergence tolerance to 5 × 10-6eV per atom. The cut-off energy for the calculation was established at 830eV. Additionally, the states of H (1s1), C (2s2 2p2), N (2s2 2p3), and O (2s2 2p4) were treated as valence electrons in our study.

Study of the Accuracy of the Ronen Method at Material Interfaces

The Ronen Method (RM) is a nonlinear iterative scheme that demands successive resolutions of the diffusion equation, where local diffusion constants are modified to reproduce more accurate estimates of the neutron currents by a transport operator. The methodology is currently formulated using the formalism of the collision probability method for evaluation of the current. The RM was recently tested on a complete suite of one-dimensional (1-D) multigroup benchmark problems. Small differences in the flux (less than 2%) were reported at material interfaces and close to the vacuum boundary with respect to the reference solution from transport. This work investigates first a possible numerical equivalence between transport and diffusion in some representative 1-D problems from the same benchmark test suite. The equivalence is sought with optimal diffusion coefficients computed using reference transport solutions that allow for adjusting the diffusion model. The RM, which attempts to obtain such equivalent diffusion coefficients without knowing the reference solution, is then compared to the optimal coefficients. The accuracy of the flux distribution at material interfaces is investigated for different approximations of the vacuum boundary and by decreasing progressively the RM convergence tolerance set in the iterative scheme. Using tighter convergence criteria, the RM calculates more accurate flux distributions at all material interfaces, regardless of the value of the diffusion coefficient and the extrapolated distance set at the beginning of the iterative scheme. Maximum flux deviations are remarkably reduced when the RM convergence tolerance is set to eight or more significant digits, leading to improvements in the flux deviation of two orders in magnitude and providing numerical proof for equivalence with transport in the tested configurations.

A stabilised Benders decomposition with adaptive oracles for large-scale stochastic programming with short-term and long-term uncertainty

Benders decomposition with adaptive oracles was proposed to solve large-scale optimisation problems with a column-bounded block-diagonal structure, where subproblems differ only in the right-hand side and cost coefficients. Adaptive Benders reduces computational effort significantly by iteratively building inexact cutting planes and valid upper and lower bounds. However, Adaptive Benders and standard Benders may suffer severe oscillation when solving degenerate models. Therefore, we propose stabilising Adaptive Benders with the level method and adaptively selecting which subproblems to solve each iteration for more accurate information. In addition, we propose a dynamic level method to improve the robustness of stabilised Adaptive Benders by adjusting the level set each iteration. We compare stabilised Adaptive Benders with the unstabilised versions of Adaptive Benders with one subproblem solved per iteration and standard Benders on a multi-region long-term power system investment planning problem with short-term and long-term uncertainty. The problem is formulated as multi-horizon stochastic programming. Four algorithms were implemented to solve linear programming with up to 1 billion variables and 4.5 billion constraints. The computational results show that: (a) for a 1.00% convergence tolerance, the proposed stabilised method is up to 113.7 times faster than standard Benders and 2.1 times faster than unstabilised Adaptive Benders; (b) for a 0.10% convergence tolerance, the proposed stabilised method is up to 45.5 times faster than standard Benders and unstabilised Adaptive Benders cannot solve the largest instance to convergence tolerance due to severe oscillation and (c) dynamic level method makes stabilisation more robust.

On the number of subproblem iterations per coupling step in partitioned fluid‐structure interaction simulations

AbstractIn literature, the cost of a partitioned fluid‐structure interaction scheme is typically assessed by the number of coupling iterations required per time step, while ignoring the internal iterations within the nonlinear subproblems. In this work, we demonstrate that these internal iterations have a significant influence on the computational cost of the coupled simulation. Particular attention is paid to how limiting the number of iterations within each solver call can shorten the overall run time, as it avoids polishing the subproblem solution using unconverged coupling data. Based on systematic parameter studies, we investigate the optimal number of subproblem iterations per coupling step. In addition, this work proposes a new convergence criterion for partitioned algorithms that is based solely on the number of subproblem iterations required to reach the subproblem residual tolerances and therefore does not require any additional convergence tolerance for the coupling loop.

Comparative Study of the Gompertz and Logistic Growth Models on the Prevalence and Fatality of Covid-19 Pandemic in Nigeria

This study models the prevalence and fatality of the Covid-19 pandemic in Nigeria from February 2020 to July 2022. It is a comparative study of two prominent models: The Gompertz and Logistic population growth models. The data for this study was obtained from the website of Our World in Data, OWID (https//www.ourworldindata.org). The Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) were employed to compare the performance of the models, and the number of iterations before convergence and convergence tolerance for each model was also put into consideration. The study revealed that the Gompertz population growth model provides a better fit compared to the logistic growth in modelling the cumulative covid-19 cases and cumulative covid-19-related deaths in Nigeria. From the models, we obtained important features of the pandemic, such as the growth rate and asymptotes.

Structural, elastic, electronic, and optical properties of NaAlSi3O8 and Al4[Si4O10](OH)8 from first-principles calculations

Based on the first-principles calculations, this paper investigates the structural, elastic, electronic, and optical properties of albite and kaolinite, respectively. It is determined that both of them show structural stability, mechanical stability, and brittleness by calculating formation enthalpy, phonon dispersion, elastic, and mechanically relevant properties. Both materials are insulators with an indirect bandgap. By calculating the TDOS and PDOS, the main characteristics of the electronic structure of NaAlSi3O8 come from O-2p and Si-3p states, O-2p, and Al-3p states hybridization, similar to Al4[Si4O10](OH)8. The covalence of Si-O bonds in NaAlSi3O8 is greater than Al-O bonds, and the covalent property sequence of Si-O bands in NaAlSi3O8 is Si2-O3 > Si1-O4 > Si2-O2 > Si1-O8 > Si1-O6 > Si3-O2 > Si3-O4. The optical anisotropy of NaAlSi3O8 and Al4[Si4O10](OH)8 is analyzed. First-principles density functional theory (DFT) calculation was carried out by the CASTEP computer program. The GGA-PW91 exchange-correlation was used. The energy convergence tolerance, the maximum force, and the maximum displacement were set in the calculation.

Improved Golden Jackal Optimization for Optimal Allocation and Scheduling of Wind Turbine and Electric Vehicles Parking Lots in Electrical Distribution Network Using Rosenbrock’s Direct Rotation Strategy

In this paper, a multi-objective allocation and scheduling of wind turbines and electric vehicle parking lots are performed in an IEEE 33-bus radial distribution network to reach the minimum annual costs of power loss, purchased grid energy, wind energy, PHEV energy, battery degradation cost, and network voltage deviations. Decision variables, such as the site and size of wind turbines and electric parking lots in the distribution system, are found using an improved golden jackal optimization (IGJO) algorithm based on Rosenbrock’s direct rotational (RDR) strategy. The results showed that the IGJO finds the optimal solution with a lower convergence tolerance and a better (lower) objective function value compared to conventional GJO, the artificial electric field algorithm (AEFA), particle swarm optimization (PSO), and manta ray foraging optimization (MRFO) methods. The results showed that using the proposed method based on the IGJO, the energy loss cost, grid energy cost, and network voltage deviations were reduced by 29.76%, 65.86%, and 18.63%, respectively, compared to the base network. Moreover, the statistical analysis results proved their superiority compared to the conventional GJO, AEFA, PSO, and MRFO algorithms. Moreover, considering vehicles battery degradation costs, the losses cost, grid energy cost, and network voltage deviations have been reduced by 3.28%, 1.07%, and 4.32%, respectively, compared to the case without battery degradation costs. In addition, the results showed that the decrease in electric vehicle availability causes increasing losses for grid energy costs and weakens the network voltage profile, and vice versa.

Re-understanding and improvement of the pseudo-spectral method for rapid footprint generation

A PS (pseudo-spectral) method with improved parameters is proposed for the rapid footprint generation of reusable vehicles. Three aspects of the method were explored, including the consistency of the discrete NLP (Nonlinear Programming) problem with the original problem, the impact of the NLP problems' convergence accuracy on the results, and the final convergence accuracy factors. The reasonable ranges of the IPOPT solver’s convergence tolerance are redefined through these explorations. Furthermore, the solution process of the footprint is partially optimized. Finally, the results show that the method can fully improve the computational efficiency of solving the footprint using the PS method. The average time of each trajectory is minor than 0.9s, showing the parameter-improved PS method’s superiority.

Three-dimensional X-FEM modeling of crack coalescence phenomena in the Smart CutTM technology

A new 3D X-FEM approach is proposed to model crack coalescence phenomena in the Smart Cut TM technology. An adaptive mesh is developed based on a 3D fractal mesh allowing a significant gain in computing time compared to regular meshes. In order to calculate stress intensity factors (SIFs), the term corresponding to internal pressure, that is normal loading on crack faces has been determined, added to the interaction integral and the results of SIFs in mode I have been compared to their analytical values with mean error less than 1%. An implicit algorithm is implemented to predict the evolution of the internal pressure for growing cracks of any shape; a constitutive law in pressure correlated with the equation of state of ideal gases has therefore been used. Compared to the analytical solution, numerically evaluated pressure is accurate with maximum errors less than 1% for relevant convergence tolerance. Finally, the coalescence of two cracks is taken into account in the model. Qualitative local stress analysis makes it possible to explain some phenomena experimentally observed in the Smart Cut TM such as neighboring effects of growing cracks, the outgrowth of nearby cracks or influence of the presence of concave areas at a front on growth. • A 3D X-FEM model has been proposed to model crack coalescence phenomena under variable pressure (Smart Cut TM process). • The interaction integral has been adapted to take into account the internal pressure when computing 3D stress intensity factors. • Prediction of the internal pressure of propagating cracks whatever are their shapes. • Analysis of neighboring effects by stress analysis on a two penny-shape crack coalescence model.

WP Nanocrystals on N,P Dual-Doped Carbon Nanosheets with Deeply Analyzed Catalytic Mechanisms for Lithium–Sulfur Batteries

WP Nanocrystals on N,P Dual-Doped Carbon Nanosheets with Deeply Analyzed Catalytic Mechanisms for Lithium–Sulfur Batteries

Application of Two Fuzzy Logic Systems to Complex-Type ZNN Models for the Drazin Inverse of Time-Dependent Complex-Value Matrix

In accordance with the advantages of zeroing neural network (ZNN) with the parallel processing character and fuzzy logic systems for calculating the uncertainties, two complex-type fuzzy ZNN (CtFZNN) models, which are mainly derived from two different limit forms of the Drazin inverse, are developed for solving the time-dependent complex-value Drazin inversion (TDCVDI) problem in this article. The most significant feature of the CtFZNN models is to use the improved fuzzy evolutionary formula, where the traditional constant or time-dependent factors are replaced by the fuzzy factors. For the non-noise or the noise disturbed CtFZNN models, the applied fuzzy factors are, respectively, generated from the single-input and single-output fuzzy logic system or the double-input and single-output fuzzy logic system. From the analytical discussions, it can conclude that the proposed CtFZNN models not only have finite-time convergence and inherent noise tolerance simultaneously, but also possess faster adaptive convergence rate even in a noisy environment. The presented theorems and the provided numerical simulations demonstrate the effectiveness of the proposed methods for addressing the TDCVDI problem, especially compared to the general ZNN model.

Heat and Mass Transport Analysis of MHD Rotating Hybrid Nanofluids Conveying Silver and Molybdenum Di-Sulfide Nano-Particles under Effect of Linear and Non-Linear Radiation

This article is an attempt to explore the heat transfer features of the steady three-dimensional rotating flow of magneto-hydrodynamic hybrid nanofluids under the effect of nonlinear radiation over the bi-directional stretching surface. For this purpose, two different nano-particles, namely silver (Ag) and molybdenum di-sulfide (MoS2), were selected. Three different conventional base fluids were utilized to form desired hybrid nanofluids such as water (H2O), engine oil (EO), and ethylene glycol (EG). We obtained steady three-dimensional highly nonlinear partial differential equations. These highly nonlinear partial differential equations cannot be solved analytically, so these equations were handled in MATLAB with the BVP-4C technique with convergence tolerance at 10−6. The graph depicts the effect of the magnetization effect, thermal radiation, and stretching ratio on rotating hybrid nanofluids. Additionally, the impact of thermal radiation on the heat coefficient of three different hybrid nanofluids is being investigated. The augmentation in magnetization decreases the primary velocity, whereas the increment in radiation enhances the primary velocity. The stretching ratio and the presence of higher magnetic forces increase the temperature profile. The concentration profile was enhanced with an increment in the magnetic field, stretching, and rotation ratio. The maximum Nusselt number was achieved for the Ag-MoS2/EO hybrid nanofluid. It was concluded that augmentation in nonlinear radiation enhances the heat transfer coefficient for the examined cases (I) and (II) of the hybrid nanofluids. The Nusselt number doubled for both the examined cases under nonlinear radiation. Moreover, it was discovered that Ag-MoS2/water produced the best heat transfer results under nonlinear radiation. Therefore, the study recommends more frequent exploration of hybrid nanofluids (Ag-MoS2/water) when employing nonlinear radiation to analyze the heat transfer coefficient.

High performance multi-platform computing for large-scale image-based finite element modeling of bone

BackgroundImage-based finite element (FE) modeling of bone is a non-invasive method to estimate bone stiffness and strength. High-resolution imaging data as input allows for inclusion of bone microarchitecture but results in large amounts of data unsuitable for traditional FE solvers. Bone-specific mesh-free solvers have been developed over the past 20 years to improve on memory efficiency in simulated bone loading applications. The objective of this study was to provide linear performance benchmarking for a bone-specific, mesh-free solver (FAIM) using µCT and HR-pQCT image data on Mac, Linux, and Windows operating systems using both single- and multi-thread CPU and GPU processing. MethodsThe focus is on the linear gradient-descent solver using standardized uniaxial loading of bone models from µCT, and first- and second-generation HR-pQCT scans of the radius and tibia. Convergence, speedup, memory, and batch performance tests were completed using CPUs and GPUs on three laboratory-based systems with Windows, Linux, and Mac operating systems. ResultsAlthough varying by system and model size, time-per-iteration was as low as 0.03 s when an HR-pQCT-based radius model (6.45 million DOF) was solved with 3 GPUs. Strong scaling was achieved with GPU and CPU parallel processing, with strong parallel efficiencies when models were solved using 3 GPUs or ≤ 10 CPU threads. Errors in force, strain energy density, and Von Mises stress were as low as 0.1% when a convergence tolerance of 10−3 or smaller was used. ConclusionThe results of this study indicate that to maximize computational efficiency and minimize model solution times using FAIM software under the standardized tested conditions using µCT, XCT1 and XCT2 HR-pQCT image data, convergence tolerance set to 10−4, and 10 threads or 2 GPUs are sufficient for efficient solution times. Less strict convergence tolerances will improve solution times but will introduce more error in the outcome measures.

A Benchmark Study on Steepest Descent and Conjugate Gradient Methods-Line Search Conditions Combinations in Unconstrained Optimization

In this paper, it is aimed to computationally conduct a performance benchmarking for the steepest descent and the three well-known conjugate gradient methods (i.e., Fletcher-Reeves, Polak-Ribiere and Hestenes-Stiefel) along with six different step length calculation techniques/conditions, namely Backtracking, Armijo-Backtracking, Goldstein, weakWolfe, strongWolfe, Exact local minimizer in the unconstrained optimization. To this end, a series of computational experiments on a test function set is completed using the combinations of those optimization methods and line search conditions. During these experiments, the number of function evaluations for every iteration are monitored and recorded for all the optimization method-line search condition combinations. The total number of function evaluations are then set a performance measure when the combination in question converges to the functions minimums within the given convergence tolerance. Through those data, the performance and data profiles are created for all the optimization method-line search condition combinations with the purpose of a reliable and an efficient benchmarking. It has been determined that, for this test function set, the steepest descent-Goldstein combination is the fastest one whereas the steepest descent-exact local minimizer is the most robust one with a high convergence accuracy. By making a trade-off between convergence speed and robustness, it has been identified that the steepest descent-weak Wolfe combination is the optimal choice for this test function set.

Hybrid geometry trimming algorithm based on Integer Linear Programming for fluid flow topology optimization

In this work, a hybrid geometry trimming algorithm based on Topology Optimization of Binary Structures (TOBS) is proposed to design fluid flow devices with reduced seepage. The problem of solid reintroduction during optimization is solved by combining the geometry trimmed functional with the extended porous domain functional in a multi-objective function. The method is evaluated in the topology optimization of fixed-geometry fluidic diodes, which is a class of problems known to present difficulties related to high fluid penetration in the porous regions. Also, a new convergence criterion for TOBS is proposed to eliminate the necessity of selecting a convergence tolerance and to avoid running the optimization for cycles of the same designs. Fluidic diodes are designed for several Reynolds numbers (Re) to prove the effectiveness of the proposed formulation. The forward problem is solved by the Finite Element Method (FEM), and the sensitivities are calculated via automatic differentiation. The optimization problem is solved by Integer Linear Programming.

Design of an Artificial Intelligence Algorithm Teaching System for Universities Based on Probabilistic Neuronal Network Model

Intelligence is gradually becoming an important tool for solving difficult problems with the development of computers. This article takes the design of university teaching systems as the research context to establish an artificial intelligence network research and learning platform. A probabilistic process neuron network model is proposed, which combines the Bayesian probabilistic classification mechanism with the dynamic signal processing method of process neuron networks, and achieves dynamic classification based on Bayesian rules by adding a pattern unit layer to the feed-forward process neuron network as well as adopting a normalised exponential excitation function. Artificial intelligence prediction based on probabilistic neural networks is verified by MATLAB as having good convergence and fault tolerance as well as data processing capability. The article also analyses the functions of the university intelligent teaching system and realises the optimal design of the university intelligent teaching system.

Benchmarking of finite-difference time-domain method and fast multipole boundary element method for room acoustics

Compared to geometrical acoustics, wave-based methods which solve the wave equation either in the time domain or in the frequency domain are known for their high accuracy. However, their systematic use as professional room acoustic simulation tools is less popular due to the modelling effort and computational time requirements, especially in the case of complex scenarios. This paper aims at providing guidelines for the use of two wave-based methods in complex room acoustics simulations, namely the finite-difference time-domain (FDTD) method and the fast multipole boundary element method (FMBEM). Numerical experiments are conducted to address the convergence issues of the two solvers, more specifically, the selection of the convergence tolerance of the iterative solver in FMBEM and the temporal sampling frequency in FDTD. To evaluate the capability of the solvers in simulating complex scenarios, five cases with increasing complexity of material input data are presented. The results show that both solvers give close predictions for various room acoustics parameters. In addition, an uncertainty sensitivity study is performed in a case where experimental data is available. Large deviations between measured and simulated reverberation time reveal that typical material data-sets poorly represent the behaviour of real materials in a room acoustics context. Lastly, the efficiency of the two solvers is discussed. With parallelization implemented, both solvers can simulate sizeable room acoustic problems with good accuracy within a reasonable time.

An improved corona-virus herd immunity optimizer algorithm for network reconfiguration based on fuzzy multi-criteria approach

Reconfiguration of the distribution network to determine its optimal configuration is a technical and low-cost method that can improve different characteristics of the network based on multi-criteria optimization. In this paper reconfiguration of unbalanced distribution networks is presented with the objective of power loss minimization, voltage unbalance minimization, voltage sag improvement, and minimizing energy not supplied by the customers based on fuzzy multi-criteria approach (FMCA) using new improved corona-virus herd immunity optimizer algorithm (ICHIOA). The voltage unbalances and voltage sag is power quality criteria and also the ENS refers to the reliability index. Conventional CHIOA is inspired based on herd immunity against COVID-19 disease via social distancing and is improved using nonlinearly decreasing inertia weight strategy for global and local exploration improvement. The methodology is implemented as single and multi-objective optimization on 33 and 69 bus IEEE standard networks. Moreover, the performance of the ICHIOA in problem-solving is compared with some well-known algorithms such as particle swarm optimization (PSO), grey wolf optimizer (GWO), moth flame optimizer (MFO), ant lion optimizer (ALO), bat algorithm (BA) and also conventional CHIOA. The simulation results based on the FMCA showed that all criteria are improved with reconfiguration due to compromising between them while in single-objective optimization, some criteria may be weakened. Also, the obtained results confirmed the superiority of the ICHIOA in comparison with the other algorithms in achieving better criteria with lower convergence tolerance and more convergence accuracy. Moreover, the results cleared that the ICHIOA based on FMCA is capable to determine the best network configuration optimally to improve the power loss, voltage sag, voltage unbalance, and ENS in different loading conditions.

Efficient Alternating Least Squares Algorithms for Low Multilinear Rank Approximation of Tensors

The low multilinear rank approximation, also known as the truncated Tucker decomposition, has been extensively utilized in many applications that involve higher-order tensors. Popular methods for low multilinear rank approximation usually rely directly on matrix SVD, therefore often suffer from the notorious intermediate data explosion issue and are not easy to parallelize, especially when the input tensor is large. In this paper, we propose a new class of truncated HOSVD algorithms based on alternating least squares (ALS) for efficiently computing the low multilinear rank approximation of tensors. The proposed ALS-based approaches are able to eliminate the redundant computations of the singular vectors of intermediate matrices and are therefore free of data explosion. Also, the new methods are more flexible with adjustable convergence tolerance and are intrinsically parallelizable on high-performance computers. Theoretical analysis reveals that the ALS iteration in the proposed algorithms is q-linear convergent with a relatively wide convergence region. Numerical experiments with large-scale tensors from both synthetic and real-world applications demonstrate that ALS-based methods can substantially reduce the total cost of the original ones and are highly scalable for parallel computing.