High Computational Accuracy Research Articles

Grad-CAM-Based Explainable Artificial Intelligence Related to Medical Text Processing.

The opacity of deep learning makes its application challenging in the medical field. Therefore, there is a need to enable explainable artificial intelligence (XAI) in the medical field to ensure that models and their results can be explained in a manner that humans can understand. This study uses a high-accuracy computer vision algorithm model to transfer learning to medical text tasks and uses the explanatory visualization method known as gradient-weighted class activation mapping (Grad-CAM) to generate heat maps to ensure that the basis for decision-making can be provided intuitively or via the model. The system comprises four modules: pre-processing, word embedding, classifier, and visualization. We used Word2Vec and BERT to compare word embeddings and use ResNet and 1Dimension convolutional neural networks (CNN) to compare classifiers. Finally, the Bi-LSTM was used to perform text classification for direct comparison. With 25 epochs, the model that used pre-trained ResNet on the formalized text presented the best performance (recall of 90.9%, precision of 91.1%, and an F1 score of 90.2% weighted). This study uses ResNet to process medical texts through Grad-CAM-based explainable artificial intelligence and obtains a high-accuracy classification effect; at the same time, through Grad-CAM visualization, it intuitively shows the words to which the model pays attention when making predictions.

Theoretical study on hydrogen abstraction reactions from pentane isomers by NO2

The rate constants for the H-abstraction reactions of C5H12 + NO2 are calculated by high-accuracy electronic structure calculations and transition state theory (TST). The investigation includes three pentane isomers, neopentane, n-pentane, and iso-pentane, each of which leads to the formation of three different HONO isomers, namely cis-HONO, trans-HONO, and HNO2. The lowest energy conformers of all relevant species are determined through geometry optimization, frequency analysis, and one-dimensional hindered rotor scanning at the M06-2X/6-311++G(d,p) level of theory, and are further re-optimized using the B2PLYP-D3/cc-pVTZ method. The single point energies of all the reactants, products, and transition states are obtained at the DLPNO-CCSD(T)/CBS level of theory. Traditional transition state theory was employed to derive the rate constants, which were then fitted to obtain the kinetic parameters in the Arrhenius expression. Simulation results show improved performance in ignition delay times and oxidation species prediction, when the calculated reaction rate constants are implemented in the literature n-pentane kinetic models. The results offer enhanced comprehension of the chemistry of C5H12 + NO2, and will aid in the understanding of the mutual reactions of large hydrocarbon fuels and NOx.

A novel second‐order mixture importance sampling strategy for system reliability analysis

AbstractReliability analysis and evaluation of systems with multiple failure modes is extremely challenging. In this paper, a novel, efficient, and accurate system reliability method is proposed based on the second‐order approximation and the mixture important sampling (mixture‐IS) method. The proposed method does not require the construction of surrogate models and integrates the advantages of second‐order approximation and mixture‐IS strategies, which can greatly reduce the computational effort of estimating the system failure probability as well as ensuring the accuracy. The accuracy and efficiency of the proposed method are verified by several numerical and engineering problems, and comparisons are made against other existing reliability methods. The results indicate that the method can effectively reduce the number of limit state function calls and achieve high computational accuracy.

Accurate GW frontier orbital energies of 134 kilo molecules

HOMO and LUMO energies are critical molecular properties that typically require high accuracy computations for practical applicability. Until now, a comprehensive dataset containing sufficiently accurate HOMO and LUMO energies has been unavailable. In this study, we introduce a new dataset of HOMO/LUMO energies for QM9 compounds, calculated using the GW method. The GW method offers adequate HOMO/LUMO prediction accuracy for diverse applications, exhibiting mean unsigned errors of 100 meV in the GW100 benchmark dataset. This database may serve as a benchmark of HOMO/LUMO prediction, delta-learning, and transfer learning, particularly for larger molecules where GW is the most accurate but still numerically feasible method. We anticipate that this dataset will enable the development of more accurate machine learning models for predicting molecular properties.

Incentive-based demand response under incomplete information based on the deep deterministic policy gradient

Incentive-based demand response (IBDR), as an important measure to encourage the users to participate in the demand-side management, is commonly modeled as the Stackelberg game with the complete information. However, it is difficult to acquire the users' complete information due to privacy protections. In this paper, a Markov decision process (MDP) game model is proposed to address IBDR under the incomplete information, which is based on a deep deterministic policy gradient algorithm (DDPG). Considering differences on the users' load, the K-means method is used to classify different users according to the daily load rate and the peak-to-valley difference, such that different types of user load will have different incentive prices to participate in demand responses. The proposed DDPG algorithm can improve the calculation efficiency of the Nash equilibrium solution of the IBDR under the incomplete information, as it can deal with the multi-dimensional continuous state and action spaces. Simulation results show that the proposed approach can achieve the Nash equilibrium under the incomplete information and has the higher calculation accuracy and the lower calculation time in comparison to the Q learning method.

Integrated approach to analyzing the stress-strain state of the tubbing erector manipulator mechanism based on the SolidWorks program

Purpose. The research purpose is todevelop and substantiate a method for computer analysis of the stress-strain state that occurs in the tubbing erector manipulator mechanism with several degrees of freedom, using modern computing complexes based on the finite-element method. Methods. The research uses software complexes, with the help of which a calculation-theoretical apparatus has been created, which is intended for calculating dynamic and static loads that occur in the elements of the tubbing erector manipulator mechanism. The results of modeling the stress-strain state in the parts and nodes of this mechanism, obtained using the SolidWorks Motion and SolidWorks Simulation programs, have been compared. The errors in these results have been estimated. Findings. The effectiveness of the proposed analysis method, based on the combination of SolidWorks applications, in particular, SolidWorks Motion and SolidWorks Simulation, has been proven on the example of modeling the most complex design of the UT62 tubbing erector mechanism.The essence of the method is that initially, using the SolidWorks Motion application, the laws of motion of the motors are set. The tubbing erecting process itself is divided into discrete moments of time with the selection of parts in which the stress fields are calculated and with the determination of the maximum equivalent stress, using the Simulation Setup tool. The change in these stresses during the erecting cycle is analyzed, and if it is necessary to refine the data, the discrete time step of the mechanism motion is reduced. Specific information about the stress-strain state of the part is obtained in the SolidWorks Simulation application, using data imported from the SolidWorks Motion program about gravitational and inertial forces that occur in the tubbing erector mechanism. Originality. The scientific novelty of the developed method of computer analysis for the mechanism of the manipulator of the UT62 type tubing stacker is manifested in a comprehensive approach to modeling the stress-strain state, which includes the integration of data from SolidWorks Motion and SolidWorks Simulation programs for accurate determination of stresses, taking into account both gravitational forces and dynamic loads, which ensures high accuracy and complexity of calculations in real operating conditions of the mechanism. Practical implications. The proposed method can be used for modeling the motion of mechanisms of different complexity, in particular, in the design of domestic tubbing erectors.

Feature selection using bio-inspired optimization for IoT intrusion detection and prevention system

Nowadays smart Internet of Things (IoT) devices are used briskly, and these devices communicate with each other via wireless medium. However, this increase in IoT devices has resulted in a rise of security issues associated with the IoT system. Therefore, an intrusion detection and prevention system (IDPS) is used to locate and report any malicious activity. The IDPS's feature selection (FS) task is necessary to improve the data quality and decrease the data used for classifying intrusive traffic. Therefore, this paper proposes a novel FS method that hybridizes improved salp swarm algorithm and harris hawk optimization algorithm. The XGBoost classifier is used for classifying reduced network traffic. Proposed system demonstrates high accuracy and low computation time, surpassing other related approaches used for the IDPS feature selection task.

Fast Magnetic Field Prediction Based on Hybrid Subdomain Method and Multiobjective Optimization Design for Interior Permanent Magnet Synchronous Machines

Benefiting from the prosperity of new energy industry, interior permanent magnet synchronous machines (IPMSM) have ushered in unprecedented development opportunities. To further suppress defects and enhance engineering practicality, fast magnetic field prediction based on hybrid subdomain method and multiobjective optimization design for IPMSM are studied. First, fast no-load magnetic field prediction model based on hybrid subdomain method for IPMSM is innovatively proposed. By combining master/slave subdomain division method and lumped parameter magnetic equivalent circuit, complex rotor structure and saturation effects of magnetic bridges and cores can be comprehensively considered. Prediction model has extremely high calculation accuracy and speed. After that, based on the proposed prediction model, a multiobjective optimization model combining Taguchi method and nondominated sorting genetic algorithm II (NSGA-II) is implemented. Model optimization efficiency has been greatly improved scientifically and effectively. Finally, the efficiency, advancement, innovation, and engineering practicability of studied prediction and optimization models are well verified through a series of finite element analysis (FEA) and prototype experiments. In addition, the generality and universality of this study are verified for other types of IPMSM.

A fractional derivative aerodynamic model at high angles of attack

A fractional derivative aerodynamic model is put forward based on Grünwald-Letnikov fractional derivative in this paper. In order to express the hysteresis property of aerodynamic coefficients, a fractional derivative state-space equation is developed from Goman’s differential equation. A fast and high-accuracy calculation method of Grünwald-Letnikov fractional derivative is proposed. Then, the model is demonstrated by several experiments data from literatures. The simulation results show that the presented fractional derivative aerodynamic model can preferably express the nonlinear unsteady aerodynamic characteristics of airfoils and aircrafts. The proposed modelling framework is meaningful to the designation of flight and control system.

Efficient Protocol for Computing MCD Spectra in a Broad Frequency Range Combining Resonant and Damped CC2 Quadratic Response Theory.

Coupled cluster response theory offers a path to high-accuracy calculations of spectroscopic properties, such as magnetic circular dichroism (MCD). However, divergence or slow convergence issues are often encountered for electronic transitions in high-energy regions with a high density of states. This is here addressed for MCD by an implementation of damped quadratic response theory for resolution-of-identity coupled cluster singles-and-approximate-doubles (RI-CC2), along with an implementation of the MCD term from resonant response theory. Combined, damped and resonant response theory calculations provide an efficient strategy to obtain MCD spectra over a broad frequency range and for systems that include highly symmetric molecules with degenerate excited states. The protocol is illustrated by application to zinc tetrabenzoporphyrin in the energy region of 2-8 eV and comparison to experimental data. Timings are reported for the resonant and damped approaches, showing that a greater part of the calculation time is consumed by the construction of the building blocks for the final MCD ellipticity. A recommendation on how to use the procedure is outlined.

Modeling of gas emission in coal mine excavation workface: a new insight into the prediction model.

Coalbed methane (CBM) is a clean energy source, but its utilization is inefficient due to the complexity and low accuracy of its emission prediction model. In this research, we constructed a mathematical model of gas emission from the excavation workface, and combined the experimental results to propose a new model for accurate and concise prediction. The new model was validated in the field workface and compared with the traditional prediction model. Moreover, the sensitivity of gas emission parameters and the participation ratio of gas emission sources were analyzed. The study results show that the new model has higher calculation accuracy than the old model prediction, with an average error reduction of 4.693%. In the excavation workface, the coal fall gas emission conforms to the negative power function equation, and the coal wall gas emission conforms to the negative exponential function equation. In the early stage of excavation, the proportion of coal fall gas emission is higher than that of coal wall gas emission, and the peak proportion reaches 58.5%. In the later stage, the proportion of coal fall gas emission gradually decreases to below 30%. The order of the sensitivity of gas emission parameters is coal wall gas initial velocity > coal fall gas decay coefficient > coal fall gas initial velocity > coal wall gas decay coefficient. The new model is successfully applied in engineering, which helps to improve the efficiency of coal mine gas disaster control and utilization.

Method for controlling phytoplankton distribution in fresh open water

ABSTRACT This paper presents a processed phytoplankton dataset from Lake Zerendi and other Kokshetau Upland lakes, Kazakhstan. The study aimed to develop an automated method for controlling the distribution of phytoplankton in fresh open water using a gradient descent approach with the search for a starting point for research. The paper also presents the reconstructed biomass indicators for changes in water temperature, with model calibration based on results from 5 years of surface sampling. A special feature of the proposed computer data processing is the cubic spline interpolation, which allows a higher calculation accuracy. This method reduces the sample size required, providing a comprehensive picture of the condition of the infrastructure and allowing for a more efficient response to emerging ecosystem issues.

Finite element modeling of electric vehicle power battery pack and its influence analysis and test research at body-in-white stage

Compared with the design of traditional fuel vehicles, the design of electric vehicles has its uniqueness, consisting mainly in that the body design must be able to adapt to the new power system and its layout. Power battery pack is an important factor affecting the body design of electric vehicles. In order to study the modeling of power battery packs and its impact on body performance, it was proposed to use the finite element method for modeling the power battery pack and analyzing its influence on the performance of the body-in-white. Based on a certain electric vehicle, the basic body-in-white model and the body-in-white model with the mass point addition and the refined model of the power battery pack were established, the static stiffness of the body-in-white simulation results of different models were compared and analyzed, and finally the effectiveness of the simulation model was verified through bench tests. Based on this effective model, the influence of the power battery pack on the modal and strength of the body-in-white was analyzed. The results show that the established refined model of the power battery pack has higher computational accuracy. Besides, the power battery pack can significantly increase the static stiffness of the electric vehicle body-in-white, effectively optimize the first-order torsion frequency and the first-order front cabin yaw frequency of the body-in-white, reduce the first-order bending frequency of the body-in-white, and greatly increase the risk of static strength failure of the body-in-white. In the setting of body performance goals and structural development, the influence of the power battery pack cannot be ignored.

Dynamic mechanical properties and failure mechanism of magnesium aluminum spinel based on experiments and peridynamic simulations

Experiments and peridynamic simulations of the split-Hopkinson pressure bar (SHPB) have been conducted to investigate the dynamic mechanical behaviors of magnesium aluminate (Mg-Al) spinel. The high-speed camera and scanning electron microscopy were used to observe the failure process and the fracture surface. In our modified model, the crystal orientation and grain boundary strength were explicitly quantified, accompanied by the dual-horizon peridynamic technique, resulting in high calculation accuracy and efficiency. Then effects of the stress state and strain rate on the failure mechanism of spinel have been investigated with the damage evolution process. Micro-pores tend to be generated at the intersection of grain boundaries, while cracks will propagate along these micro-pores, leading to trans-granular and inter-granular failures. Finally, the microstructure effects on the mechanical properties of spinel have been systemically investigated. The strength and distribution of grain boundaries were found to significantly affect the strength and crack propagation of Mg-Al spinel.

Analysis and Prediction of Image Quality Degradation Caused by Diffraction of Infrared Optical System Turning Marks

This paper addresses the issue of reduced image quality due to annular turning marks formed by single-point diamond turning (SPDT) during the processing of metal-based mirrors and infrared lenses. An ideal single-point diamond turning marks diffraction action model to quantitatively analyze the impact of turning marks diffraction on imaging quality degradation is proposed. Based on this model, a fast estimation algorithm for the optical modulation transfer function of the system under turning marks diffraction (TMTF) is proposed. The results show that the TMTF algorithm achieves high computational accuracy, with a relative error of only 3% in diffraction efficiency, while being hundreds of times faster than rigorous coupled wave analysis (RCWA). This method is significant for reducing manufacturing costs and improving production efficiency, as it avoids the problem of being unable to compute large-size optical systems due to computational resource and time constraints.

Research and Hardware Implementation of a Reduced-Latency Quadruple-Precision Floating-Point Arctangent Algorithm

In the field of digital signal processing, such as in navigation and radar, a significant number of high-precision arctangent function calculations are required. Lookup tables, polynomial approximation, and single/double-precision floating-point Coordinate Rotation Digital Computer (CORDIC) algorithms are insufficient to meet the demands of practical applications, where both high precision and low latency are essential. In this paper, based on the concept of trading area for speed, a four-step parallel branch iteration CORDIC algorithm is proposed. Using this improved algorithm, a 128-bit quad-precision floating-point arctangent function is designed, and the hardware circuit implementation of the arctangent algorithm is realized. The results demonstrate that the improved algorithm can achieve 128-bit floating-point arctangent calculations in just 32 cycles, with a maximum error not exceeding 2×10−34 rad. It possesses exceptionally high computational accuracy and efficiency. Furthermore, the hardware area of the arithmetic unit is approximately 0.6317 mm2, and the power consumption is about 40.6483 mW under the TSMC 65 nm process at a working frequency of 500 MHz. This design can be well suited for dedicated CORDIC processor chip applications. The research presented in this paper holds significant value for high-precision and rapid arctangent function calculations in radar, navigation, meteorology, and other fields.

Fast and accurate computation of polar harmonic Fourier moments for image description.

Continuous orthogonal moments are widely used in various image techniques due to their simplicity and good rotational invariance and stability. In recent years, numerous excellent continuous orthogonal moments have been developed, among which polar harmonic Fourier moments (PHFMs) exhibit strong image description capabilities. However, the numerical integration error is large in the calculation, which seriously affects the calculation accuracy, especially in higher-order calculation. In this paper, a continuous orthogonal moments-fast and accurate PHFM (FAPHFM) is proposed. It utilizes the polar pixel tiling technique to reduce numerical errors in the computation; this method particularly improves the accuracy of higher-order moments of traditional PHFMs. However, as accuracy increases, calculation complexity also increases. To address this issue, an eight-way symmetric/anti-symmetric calculation of the angular and radial functions was performed using the symmetry and anti-symmetry of traditional PHFMs, and clustering of pixels was performed as a way to improve the computational speed. The experimental results show that FAPHFMs perform better in image reconstruction (including noise), with higher computational accuracy, lower time complexity, and better image description ability.

Prediction of extreme response of an innovative HSR integral bridge subjected to crosswind and high-speed train

The randomness of responses in a train-bridge interaction (TBI) system induced by stochastic track irregularity and crosswind has been addressed. However, the extreme value distribution (EVD) and reliability prediction of TBI system are little known. Due to the complexity of TBI system, the prediction approach is critical to the issues of extreme value and reliability evaluation, involving the computational accuracy and efficiency. Thus, the Monto Carlo-based global maximum method, probability density evolution-based virtual time process method (VPM), and average conditional exceeding rate method (ACERM) are considered in this study. The optimal approach for the EVD prediction of bridge buffeting responses and the TBI system with or without the consideration of crosswinds is identified. An innovative 3 × 70 m integral bridge is chosen as the study case, which is an emerging type of bridge structures used in the Chinese high-speed railway lines. Its extreme responses under the single and dual stochastic excitations of track irregularity and wind turbulence are fully investigated in this study. The study results show that the VPM and ACERM are efficient and accurate to evaluate the extreme responses of a TBI system. The ACERM is recommended in assessing the train running safety due to its high computational efficiency and reasonable accuracy. When the crosswind is not a consideration, the high-speed train running safety can be ensured with high probability if the train travels on the integral bridge below the design speed of 350 km/h. However, the failure probability of wheel-rail lateral contact forces reaches 51% if predicted under the mean wind speed of 30 m/s and train speed of 120 km/h. Finally, the advantages of the integral bridge over conventional bridges (e.g., continuous girder bridge) are discussed.

Calculation of envelope area between grinding tool and curved surface

Typical aerospace curved surface parts have high contour accuracy, which directly affects the working performance and service life of the equipment. To improve the contour accuracy of these parts, the freestyle abrasive belt grinding technology can be employed, where the size of the envelope area determines the material removal volume. However, due to the complex contour of the curved surface and large contact deformation of the belt, the envelope area between the flexible belt and curved surface parts is complex and time-varying. The accurate calculation of the envelope area is still an open problem. To this end, a novel locally approximate distance fields-mass spring systems (LADF-MSS) method is proposed. Firstly, the improved distance field is established accurately: the octree is applied to search for the reference point where the curved surface is a local quadric approximation, and then the projection point is calculated based on the reference point. Secondly, the locally approximate distance field is combined with efficient mass-spring systems to extract the envelope area. The experimental results of grinding the curved surface show that the average area error and average centroid error of the envelope area are 0.7 mm2 and 1.08 mm respectively, and the average computing time is 7.5 min. Compared with the finite element method (FEM), the average calculation errors through utilizing the LADF-MSS method are reduced by 40.3% and 76.4% respectively, and calculation efficiency is increased by 166%. These results demonstrate that the proposed method offers both higher calculation accuracy and efficiency.

Influence of Parasitic Elements and Operating Conditions of Semiconductor Switches on Power Losses and the Junction Temperature of These Switches

This article presents the results of computer analysis of selected switching networks. In these analyses, the influence of selected parasitic components of electronic switches on the total and active power losses in these switches is considered. Analyses are performed using the SPICE software for two models of semiconductor switches: an ideal switch with RC parasitic components and the SPICE model of an IGBT. The influence of parasitic capacitances and resistances of these devices operating with the control signal of different parameters values on the total and active power dissipated in these switches is analyzed. On the basis of the obtained computations the average and peak-to-peak values of the junction temperature of electronic switches at the steady state are calculated using a compact thermal model. It is shown that parasitic elements visibly influence waveforms of the active and total power. It is proved that the simplified model using the total power in computations of the junction temperature makes it possible to obtain a high accuracy of computations only in a situation when the transistor operates with a resistive load. For an inductive load, such simplification can cause an unacceptably high computation error exceeding even 30%. Such an error is a result of big differences between the active and total powers during switching-on and switching-off processes.