Algorithm Vector Research Articles

Research and Evaluation of a New Structural Damage Identification Method Based on a Refined Genetic Algorithm

In order to solve the problem of structural damage location and degree identification, the weighted mean of vectors algorithm (INFO), a high-performance optimization algorithm, was first introduced to identify structural damage. By comparison with the refined genetic algorithm (RGA), the accuracy and advantages of INFO are analyzed and evaluated. An objective function is constructed by combining the dynamic response transfer ratio without modal analysis. The INFO and RGA algorithms are used to optimize the objective function for damage identification. The simulation results verify the effectiveness of the three damage identification methods. The results show that the identification effect of the INFO algorithm can reach nearly 100% without noise influence, and the anti-noise ability is the strongest. Among the three algorithms, the damage identification accuracy of the INFO algorithm is the highest, followed by the RGA algorithm and the GA algorithm.

State-of-health estimation of lithium-ion batteries using multiple correlation analysis-based feature screening and optimizing echo state networks with the weighted mean of vectors

Lithium-ion batteries (LIBs) are widely employed in electric vehicles (EVs) and energy storage systems owing to their high energy density, low self-discharge rate, and superior longevity. As a vital evaluation index for the degradation state of LIBs, the state of health (SOH) must be accurately estimated. This study adopts an electrochemical impedance spectroscopy (EIS)-based equivalent circuit model (ECM) to estimate battery SOH. The internal dynamic electrical behaviors of LIBs are decoupled by the distribution of relaxation times (DRT) method, whereby the critical degradation features are extracted from the DRT curves and the fitted ECM parameters. Subsequently, referring to the ensemble learning approach (ELA), the multiple correlation analysis (MCA) is utilized to identify the most pertinent degradation features. The battery data are subjected to a comprehensive analysis with five types of correlations to ascertain the optimal number and combination of health indicators (HIs). Furthermore, the weighted mean of vectors (INFO) algorithm is employed to optimize the hyperparameters in the echo state network (ESN) model for SOH estimation. The echo state networks model optimized with the weighted mean of vectors algorithm (INFO-ESN) is verified with eight types of LIBs under different operating conditions. Experimental results of the proposed method manifest that the root mean square error (RMSE) and mean absolute error (MAE) of the battery SOH range from 0.32 % to 1.06 %, thereby verifying the accuracy and robustness of the proposed method.

Comparative Evaluation of Pre-Trained Language Models for Biomedical Information Retrieval.

Finding relevant information in the biomedical literature increasingly depends on efficient information retrieval (IR) algorithms. Cross-Encoders, SentenceBERT, and ColBERT are algorithms based on pre-trained language models that use nuanced but computable vector representations of search queries and documents for IR applications. Here we investigate how well these vectorization algorithms estimate relevance labels of biomedical documents for search queries using the OHSUMED dataset. For our evaluation, we compared computed scores to provided labels by using boxplots and Spearman's rank correlations. According to these metrics, we found that Sentence-BERT moderately outperformed the alternative vectorization algorithms and that additional fine-tuning based on a subset of OHSUMED labels yielded little additional benefit. Future research might aim to develop a larger dedicated dataset in order to optimize such methods more systematically, and to evaluate the corresponding functions in IR tools with end-users.

Topography-based and vectorized algorithm for extracting physical quantities in 3D-SPH form and its application in debris-flow entrainment modeling

Extraction of physical quantities such as flow depth and velocities, is one of the major purposes of geophysical flow numerical modeling and critical for estimating consequent impact forces and sediment entrainment. It is simple in nature for mesh-based models but presenting challenges in three-dimensional smoothed particle hydrodynamics (SPH) schemes. The difficulties lie in the substantial number of particles and their uneven spatial-temporal distribution, particularly over complex topography. Inspired by our previous surface cell (SC) -based approach, we propose a novel topography-based and vectorized algorithm that significantly enhances the ability to extract physical quantities over complex topography. In the proposed algorithm, geomorphologic characteristics are mathematically represented by topographical normal vectors. The correlations of physical quantities with distinct coordinate descriptions are established through the vectorization concept, ultimately leading to effective extraction of physical quantities in SPH form over complex topography. This algorithm provides an important tool to incorporate topography-linked physical models within discretized frameworks. To validate its effectiveness, we employed the algorithm to integrate the debris-flow entrainment law with our previous HBP-SPH model, utilizing the 2010 Yohutagawa debris-flow event in Japan as a case study. The results demonstrate a good agreement between the numerical simulation and on-site observation. Discussion regarding the applicability and limitation of the algorithm concludes the paper.

Random walk algorithms for solving nonlinear chemotaxis problems

Abstract Random walk based stochastic simulation methods for solving a nonlinear system of coupled transient diffusion and drift-diffusion equations governing a two-component chemotaxis process are developed. The nonlinear system is solved by linearization, the system is evolved in time, by small time steps, where on each step a linear system of equations is solved by using the solution from the previous time step. Three different stochastic algorithms are suggested, (1) the global random walk on grid (GRWG), (2) a randomized vector algorithm (RVA) based on a special transformation of the original matrix to a stochastic matrix, and (3) a stochastic projection algorithm (SPA). To get high precision results, these methods are combined with an iterative refinement method.

Vectfem: a generalized MATLAB-based vectorized algorithm for the computation of global matrix/force for finite elements of any type and approximation order in linear elasticity

In this paper, we introduce a new vectorized MATLAB-based algorithm for efficient serial computation of global matrix/force arising from finite element method (FEM) for meshes of any type and approximation order in linear elasticity. Because for-loops in MATLAB are very slow, we propose a modified process that takes advantage of vectorization and sparse assembly to achieve good performance while using the same memory as the standard algorithm. For this purpose, by using good programming practices, the implementation of this scheme is succinctly described and can be integrated into any MATLAB package dealing with FEM. Specifically, attention is paid to the calculation of the triplet (row index, column index, matrix components) as well as the assembly of the global stiffness matrix, mass matrix and force vector. Additionally, an extension of the proposed approach for Mindlin plate theory and functionally graded materials is outlined. Finally, the accuracy of this strategy is verified on selected numerical tests after comparing the obtained results with those of ABAQUS. In terms of performance, the study conducted on a set of meshes considering the standard algorithm and two other well-known MATLAB vectorized algorithms revealed that: (i) for a 2D beam problem meshed with P1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$P_{1}$$\\end{document}-triangle elements, a speedup of about 8 and 15 is achieved with sparse and fsparse, respectively. (ii) for a 3D plate problem meshed with P1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$P_{1}$$\\end{document}-tetrahedral elements, a speedup of about 4 and 8 is achieved with sparse and fsparse, respectively. When compared to ABAQUS performance, the proposed scheme results in a computational time that is about five times smaller.

Deep Sketch Vectorization via Implicit Surface Extraction

We introduce an algorithm for sketch vectorization with state-of-the-art accuracy and capable of handling complex sketches. We approach sketch vectorization as a surface extraction task from an unsigned distance field, which is implemented using a two-stage neural network and a dual contouring domain post processing algorithm. The first stage consists of extracting unsigned distance fields from an input raster image. The second stage consists of an improved neural dual contouring network more robust to noisy input and more sensitive to line geometry. To address the issue of under-sampling inherent in grid-based surface extraction approaches, we explicitly predict undersampling and keypoint maps. These are used in our post-processing algorithm to resolve sharp features and multi-way junctions. The keypoint and undersampling maps are naturally controllable, which we demonstrate in an interactive topology refinement interface. Our proposed approach produces far more accurate vectorizations on complex input than previous approaches with efficient running time.

Blood DNA methylation in post-acute sequelae of COVID-19 (PASC): a prospective cohort study

DNA methylation integrates environmental signals with transcriptional programs. COVID-19 infection induces changes in the host methylome. While post-acute sequelae of COVID-19 (PASC) is a long-term complication of acute illness, its association with DNA methylation is unknown. No universal blood marker of PASC, superseding single organ dysfunctions, has yet been identified. In this single centre prospective cohort study, PASC, post-COVID without PASC, and healthy participants were enrolled to investigate their symptoms association with peripheral blood DNA methylation data generated with state-of-the-art whole genome sequencing. PASC-induced quality-of-life deterioration was scored with a validated instrument, SF-36. Analyses were conducted to identify potential functional roles of differentially methylated loci, and machine learning algorithms were used to resolve PASC severity. 103 patients with PASC (22.3% male, 77.7% female), 15 patients with previous COVID-19 infection but no PASC (40.0% male, 60.0% female), and 27 healthy volunteers (48.1% male, 51.9% female) were enrolled. Whole genome methylation sequencing revealed 39 differentially methylated regions (DMRs) specific to PASC, each harbouring an average of 15 consecutive positions, that differentiate patients with PASC from the two control groups. Motif analyses of PASC-regulated DMRs identify binding domains for transcription factors regulating circadian rhythm and others. Some DMRs annotated to protein coding genes were associated with changes of RNA expression. Machine learning support vector algorithm and random forest hierarchical clustering reveal 28unique differentially methylated positions (DMPs) in the genome discriminating patients with better and worse quality of life. Blood DNA methylation levels identify PASC, stratify PASC severity, and suggest that DNA motifs are targeted by circadian rhythm-regulating pathways in PASC. This project has been funded by the following agencies: NIH-AI173035 (A. Jaitovich and R. Alisch); and NIH-AG066179 (R. Alisch).

Optimizing feedforward neural networks using a modified weighted mean of vectors: Case study chemical datasets

This paper proposes a modified version of the weighted mean of vectors algorithm (mINFO), which combines the strengths of the INFO algorithm with the Enhanced Solution Quality Operator (ESQ). The ESQ boosts the quality of the solutions by avoiding optimal local values, verifying that each solution moves towards a better position, and increasing the convergence speed. Furthermore, we employ the mINFO algorithm to optimize the connection weights and biases of feedforward neural networks (FNNs) to improve their accuracy. The efficacy of FNNs for classification tasks is mainly dependent on hyperparameter tuning, such as the number of layers and nodes. The mINFO was evaluated using the IEEE Congress on Evolutionary Computation held in 2020 (CEC’2020) for optimization tests, and ten chemical data sets were applied to validate the performance of the FNNs classifier. The proposed algorithm’s results have been evaluated with those of other well-known optimization methods, including Runge Kutta optimizer’s (RUN), particle swarm optimization (PSO), grey wolf optimization (GWO), Harris hawks optimization (HHO), whale optimization algorithm (WOA), slime mould algorithm (SMA) and the standard weighted mean of vectors (INFO). In addition, some improved metaheuristic algorithms. The experimental results indicate that the proposed mINFO algorithm can improve the convergence speed and generate effective search results without increasing computational costs. In addition, it has improved the FNN’s classification efficiency.

Unveiling suspicious phishing attacks: enhancing detection with an optimal feature vectorization algorithm and supervised machine learning

IntroductionThe dynamic and sophisticated nature of phishing attacks, coupled with the relatively weak anti-phishing tools, has made phishing detection a pressing challenge. In light of this, new gaps have emerged in phishing detection, including the challenges and pitfalls of existing phishing detection techniques. To bridge these gaps, this study aims to develop a more robust, effective, sophisticated, and reliable solution for phishing detection through the optimal feature vectorization algorithm (OFVA) and supervised machine learning (SML) classifiers.MethodsInitially, the OFVA was utilized to extract the 41 optimal intra-URL features from a novel large dataset comprising 2,74,446 raw URLs (134,500 phishing and 139,946 legitimate URLs). Subsequently, data cleansing, curation, and dimensionality reduction were performed to remove outliers, handle missing values, and exclude less predictive features. To identify the optimal model, the study evaluated and compared 15 SML algorithms arising from different machine learning (ML) families, including Bayesian, nearest-neighbors, decision trees, neural networks, quadratic discriminant analysis, logistic regression, bagging, boosting, random forests, and ensembles. The evaluation was performed based on various metrics such as confusion matrix, accuracy, precision, recall, F-1 score, ROC curve, and precision-recall curve analysis. Furthermore, hyperparameter tuning (using Grid-search) and k-fold cross-validation were performed to optimize the detection accuracy.Results and discussionThe findings indicate that random forests (RF) outperformed the other classifiers, achieving a greater accuracy rate of 97.52%, followed by 97.50% precision, and an AUC value of 97%. Finally, a more robust and lightweight anti-phishing model was introduced, which can serve as an effective tool for security experts, practitioners, and policymakers to combat phishing attacks.

An inversion model for burnup history of spent fuels in pebble bed HTGR based on optimization algorithm

The measurement of transuranic isotopes in spent nuclear fuel is a crucial technique for nuclear material accounting and nuclear non-proliferation. In a pebble bed high-temperature gas-cooled reactor, numerous fuel spheres circulate through the reactor core multiple times, making it unfeasible to document the power histories of individual fuel spheres. This complexity renders it impossible to employ traditional methods, such as solving the forward burnup equations used in conventional reactors, to calculate the transuranic isotopes.In this approach, optimization algorithms were coupled with the data obtained from the BUrnup Measurement System to establish a power history inversion model, which was used to solve for the power history of fuel spheres and subsequently calculate the transuranic isotopes.The paper also compared the pros and cons of coupling the inversion model with different optimization algorithms and evaluated models using various schemes of activity vector. This process helped determine the most suitable optimization algorithm and scheme of activity vector. Sensitivity analysis revealed that the established inversion model exhibited a degree of tolerance to noise in input data and the underlying physical model.

The New Method for Increasing the Efficiency of Vectorization of BLAS Operations

The issue of increasing the efficiency of software for computing architectures that support vector extensions of the command system is considered. Modern compilers can perform automatic vectorization of calculations, convert programs from a scalar representation to a vector implementation. The paper analyzes the effectiveness of automatic vectorization performed by modern compilers, discusses the problems inherent in automatic vectorization. A new algorithm for vectorization of calculations is proposed, which allows to significantly increasing the efficiency of the resulting software.

Sarcasm Text Detection on News Headlines Using Novel Hybrid Machine Learning Techniques

One of the biggest problems with sentiment analysis systems is sarcasm. The use of implicit, indirect language to express opinions is what gives it its complexity. Sarcasm can be represented in a number of ways, such as in headings, conversations, or book titles. Even for a human, recognizing sarcasm can be difficult because it conveys feelings that are diametrically contrary to the literal meaning expressed in the text. There are several different models for sarcasm detection. To identify humorous news headlines, this article assessed vectorization algorithms and several machine learning models. The recommended hybrid technique using the bag-of-words and TF-IDF feature vectorization models is compared experimentally to other machine learning approaches. In comparison to existing strategies, experiments demonstrate that the proposed hybrid technique with the bag-of-word vectorization model offers greater accuracy and F1-score results.

Enhancing Tumor Localization in Brain MRI Images Using a Scalable Systematic Vector Algorithm for Result-Oriented Accuracy

This paper introduces a novel foundational architecture incorporating two primary algorithms. The Partially Enhanced Linear Mean Analysis Algorithm for significant vectorization and the Vertically Enhanced Tensor-Load Interface Algorithm for structural vectorization. These algorithms are meticulously designed to optimize the extraction of valuable information from complex data sets, enabling efficient and insightful historical data analysis. The performance of these algorithms has been rigorously compared with existing significant vectorization and structural methods, demonstrating superior accuracy in mean value estimation. Simulated results provide robust evidence of the high-dimensional capability and throughput of the proposed architecture. Incorporating tools such as Microsoft Excel for data organization, MATLAB for advanced computational analysis, and Python for algorithmic implementation and automation further enhances the framework’s efficiency and precision. Imminent work on this architecture could expand its applicability into multi-functional systems, potentially integrating a multi-role tape framework. Such an expansion would be particularly beneficial for identifying brain tumors with greater accuracy and precision, contributing significantly to medical diagnostics. The proposed system represents a promising advancement in applying computational tools for complex data analysis and structural vectorization.

Robust kernel extreme learning machines with weighted mean of vectors and variational mode decomposition for forecasting total dissolved solids

A stable and accurate forecast of water quality parameters is crucial for planning and managing future investment programs. A well-known water quality indicator is the total dissolved solids (TDS), which measures the number of metals, minerals, and salts dissolved in a particular volume of water. This paper introduces an innovative approach, which combines the kernel extreme learning machine (KELM) with the robust weight mean of vectors (INFO) algorithm. The proposed technique stands out for its strategic integration of Boruta-XGBoost (B-XGB) as an exceptional feature selection method, along with variational mode decomposition (VMD) to decompose the input variables and enhance the accuracy of forecasting. The integrated model is termed V-KELM-INFO. INFO is an essential component that incorporates a complex exploration process. It utilizes a weighted mean of vectors, a vector combining operator to enhance population variety, and a localized search operator to expedite convergence. INFO, a potent tool, is included in the V-KELM training stage. It efficiently extracts optimum parameters and dramatically improves the accuracy of monthly TDS predictions at the Idenak station in southwest Iran. Also, the multi-criteria decision-making (MCDM) approach of weighted aggregated sum product assessment (WASPAS) is used to rank models. The V-KELM-INFO model was chosen based on the WASPAS results and statistical metrics (R = 0.962, RMSE = 57.84, WHD = 7.01, and U95 = 160.74) as the best model to forecast the TDS. In addition, the superiority of the proposed model was assessed against the metaheuristic-based V-KELM models, comprising the V-KELM grey wolf optimizer (V-KELM-GWO), V-KELM slime mold algorithm (V-KELM-SMA), and V-KELM equilibrium optimizer (V-KELM-EO). This work provided insight into improving the accuracy of modeling-based methodologies and spurred water quality modeling technology to develop sustainable and clean practices.

A prediction model for rumor user propagation behavior based on sparse representation and transfer learning

This paper introduces a prediction model rooted in sparse representation and transfer learning, with the primary objective of predicting user behavior during rumor propagation. Users' behavior is dynamic, and rumor, rumor-refuting, and rumor-promoting messages interact dynamically, according to the model. Firstly, this paper proposes to compensate for the low performance of propagation prediction models due to data sparsity at the beginning of rumor topics' lifes. Transfer learning is used to compensate for the data sparsity problem at the beginning of the topic. To map the rumor topic space effectively, a low-rank dense vectorization algorithm based on sparse representation is proposed. Finally, to mine the potential impact of multiple types of information on users, a model based on three-party game theory is constructed. It considers the complex interaction between rumor, rumor-refuting, and rumor-promoting information in the propagation of rumor. Additionally, this paper develops a model of dynamic user behavior prediction using Rumor-Attention-Mechanism-Graph-Attention (RAM-GAT) to predict rumor propagation. Experiments demonstrate that our model can effectively mine the interaction influence between multiple types of information. It can accurately predict user behavior when initial data is insufficient. In addition, rumor propagation patterns and trends are revealed.

Enhanced Detection of Chronic Heart Disease: A Novel Approach With Extreme Gradient Boosting-Logistic Regression Algorithm

In recent years, biomedical healthcare applications for managing and treating chronic heart diseases have increased. Chronic heart disease is a serious and sometimes deadly ailment that affects millions of individuals globally. Therefore, timely and effective detection of heart disease is crucial in medicine, particularly in cardiology. These complications can lead to reduced quality of life, increased medical costs, and increased risk of death. Obesity, smoking, and high blood pressure are some of the risk factors that worsen chronic heart disease. However, challenges occur in accurately and sensitively classifying heart disease in performance evaluation. To resolve this problem, we propose an Extreme Gradient Boosting-Logistic Regression (EGBLR) algorithm to accurately and quickly predict chronic heart disease. Furthermore, pre-processing can be performed using the Normalized Minimum-Maximum Scaling Feature (NMMSF) technique to estimate missing or duplicate values. Next, the Inter-Quartile Range Optimization (IQRO) technique can analyze heart disease data to identify and remove outliers. After that, the Relief Feature Vector (RFV) algorithm can be used to find each feature weight. Finally, the proposed EGBLR method can be employed to detect normal or abnormal chronic heart diseases and increase accuracy in diagnosis. The evaluation of the proposed techniques can be predicted using performance matrices such as recall, sensitivity score, precision, false rate, and precision. The accuracy of the proposed EGBLR detection approach exceeds that of previous techniques. Moreover, the proposed method for diagnosing chronic heart disease can be easily adapted to biomedical healthcare applications.

Adaptive locally weighted support vector algorithm with asymmetrically parametric insensitive/margin model

In many fields such as finance and electrical load, locally weighted support vector regression (LWSVR) method can achieve better prediction results by constructing different sub-models for each prediction target. However, it selected a fixed and single nearest neighbor parameter of kglobal for each query, and adopted the traditional k-nearest neighbor (KNN) algorithm, which is suboptimal. The weights in LWSVR assigned to the neighboring sample points also affected the accuracy of the model predictions. In this paper, we propose an adaptive locally weighted support vector regression with asymmetrically parametric insensitive/margin model, called Asy-par-v-ALWSVR. Since the selection of similar sample for query points is significant. We propose a novel locally q-neighbour selection algorithm that adaptively selects the most suitable number of neighbours for each query point based on Mahalanobis distance, rather than using a fixed kglobal. This significantly improves prediction accuracy. The weights of neighbours for each query point in Asy-par-v-ALWSVR is based on the distance within the kernel-induced feature space. This effectively discerns the importance of samples during the training process. In particular, Asy-par-v-ALWSVR can handle skewed noise with heteroscedastic structure in regression problems. Finally, we evaluate the proposed model on a synthetic sinc dataset and 14 real datasets. The results demonstrate that Asy-par-v-ALWSVR outperforms six state-of-the-art SVR-based models.

Optimal design of hybrid renewable energy sources with battery storage using an efficient weighted mean of vectors algorithm

An improved algorithm called mINFO is developed, based on the modification of the original weIghted meaN oF vectOrs Optimizer (INFO). In this article, to enhance the original INFO method, an improved version called mINFO is presented by incorporating the sine-cosine function and balancing exploration/exploitation strategy through the control randomization parameter. This suggested mINFO algorithm is applied to obtain the optimal size design of two hybrid renewable power sources (HRPS), the first configuration consists of the photovoltaic (PV)/biomass system (BS)/fuel cells (FC) and the second configuration consists of the PV/BS/battery storage units. This HRPS are designed to generate the power needed to meet the loads supply on the campus of Aswan University in the Sahary region of Aswan city, Egypt. The proposed mINFO approach is developed to avoid the weaknesses of the original INFO method by enhancing the balance between the exploration and exploitation phases and preventing the conventional INFO method from getting stuck in local minima. In order to prove the efficiency and robust search capabilities of the developed mINFO algorithm, the mINFO algorithm is tested by using 10 IEEE CEC’20 test suites and its results are compared with the original INFO method and other well-known optimization algorithms. The proposed mINFO algorithm achieved the first rank in terms of the Friedman mean rank-sum test over F2, F3, F4, F6, F9, and F10, and the second rank after the original INFO algorithm over F1 and F7. Boxplot analysis is also used to show the features of data distribution. For the majority of functions, the boxplots generated by the mINFO algorithm have the narrowest ranges of values out of all comparison algorithms. As a result, for the majority of test functions, the mINFO algorithm shows more promise than its competitors. The main objectives of the suggested HRPS are to minimize energy cost (EC), excess energy (EXE), and maximize the power system's reliability (LPSP). By applying the mINFO algorithm to find the optimal sizing for the two configurations proposed in this study and by comparing the mINFO results with other algorithms such as the original INFO, Particle Swarm optimization (PSO), Runge Kutta optimization (RUN), Sooty Tern Optimization Algorithm (STOA), and Gray Wolf optimization (GWO). The proposed mINFO method shows the best results for the two suggested configurations compared to the other algorithms used by studying the systems for 100 iterations and 50 times run. For the PV/BS/battery units, the best optimal solution has achieved after 36 iterations with the lowest EC with 0.11676275 $/kWh, net present cost (NPC) with 3,420,036 $, EXE with 0.090282 and LPSP with 0.0419687649, while for the PV/BS/FC units, the optimal solution has achieved after 11 iterations with the minimal EC with 0.17996701 $/kWh, NPC with 5,271,318 $, EXE with 0.150022 and LPSP with 0.07598679.

IInception-CBAM-IBiGRU based fault diagnosis method for asynchronous motors

Aiming at the problems of insufficient extraction of asynchronous motor fault features by traditional deep learning algorithms and poor diagnosis of asynchronous motor faults in robust noise environments, this paper proposes an end-to-end fault diagnosis method for asynchronous motors based on IInception-CBAM-IBiGRU. The method first uses a signal-to-grayscale image conversion method to convert one-dimensional vibration signals into two-dimensional images and initially extracts shallow features through two-dimensional convolution; then the Improved Inception (IInception) module is used as a residual block to learning features at different scales with a residual structure, and extracts its important feature information through the Convolutional Block Attention Module (CBAM) to extract important feature information and adjust the weight parameters; then the feature information is input to the Improved Bi-directional Gate Recurrent Unit (IBiGRU) to extract its timing features further; finally, the fault identification is achieved by the SoftMax function. The primary hyperparameters in the model are optimized by the Weighted Mean Of Vectors Algorithm (INFO). The experimental results show that the method is effective in fault diagnosis of asynchronous motors, with an accuracy rate close to 100%, and can still maintain a high accuracy rate under the condition of low noise ratio, with good robustness and generalization ability.