Multi-scale Characteristics Research Articles

Multi-scale texture filtering-based quantitative machining mechanism characterisation for ultra-precision machined surfaces

Ultra-precision machined surfaces with nanometric surface texture have realised widespread applications. However, it is still challenging to quantitatively characterise the complex generation mechanisms due to the coupling of the machining factors. To address this issue, this work developed a multi-scale texture filtering-based method for ultra-precision machined surfaces. First, experiments were conducted including diamond turning, diamond milling, and ultra-precision grinding. Next, power spectrum density was utilised to reveal the multi-scale characteristics of surface texture generation. Then, the discrete wavelet transform was proposed for multi-scale texture filtering. Finally, root mean square height Sq was evaluated on S-F surfaces according to ISO 25178-2 and the machining mechanisms were quantitatively characterised by the contribution percentages of each scale. The results demonstrated that surface texture generation was dominated by multi-scale machining mechanisms to manifest as cutting residual, tool mark, material effect, spindle vibration, and ductile-brittle transition. Surface texture characterisation quantitatively indicated that small-scale features (cutting residuals and material effects) are crucial factors at low material removal volume (MRV). Due to the increased MRV, there is a transition since the middle/large-scale features (tool marks, spindle vibration, and ductile-brittle transition) gradually play their important role. This work provides novel thoughts to quantify machining mechanisms by multi-scale surface texture metrology for ultra-precision machining.

Human-in-the-loop Multi-objective Bayesian Optimization for Directed Energy Deposition with in-situ monitoring

Directed Energy Deposition (DED) is a free-form metal additive manufacturing process characterized as toolless, flexible, and energy-efficient compared to traditional processes. However, it is a complex system with a highly dynamic nature that presents challenges for modeling and optimization due to its multiphysics and multiscale characteristics. Additionally, multiple factors such as different machine setups and materials require extensive testing through single-track depositions, which can be time and resource-intensive. Single-track experiments are the foundation for establishing optimal initial parameters and comprehensively characterizing bead geometry, ensuring the accuracy and efficiency of computer-aided design and process quality validation. We digitized a DED setup using the Robot Operating System (ROS 2) and employed a thermal camera for real-time monitoring and evaluation to streamline the experimentation process. With the laser power and velocity as inputs, we optimized the dimensions and stability of the melt pool and evaluated different objective functions and approaches using a Response Surface Model (RSM). The three-objective approach achieved better rewards in all iterations and, when implemented in a real setup, allowed to reduce the number of experiments and shorten setup time. Our approach can minimize waste, increase the quality and reliability of DED, and enhance and simplify human-process interaction by leveraging the collaboration between human knowledge and model predictions.

Multi-scale reservoir response characteristics of coalbed methane development through stress relief via horizontal well cavity completion in tectonically deformed coals

China’s coalbed methane (CBM) resources hold significant potential. However, the widespread presence of soft and low-permeability tectonically deformed coal (TDC) presents a challenge to the efficient development of CBM. The development of CBM through stress relief via horizontal well cavity completion (HWCC) offers a promising innovative solution to this problem. To elucidate the multi-scale reservoir response characteristics of CBM development through stress relief via HWCC in TDCs, this article comprehensively employs particle discrete element and finite element numerical simulation methods, and carries out the numerical simulation studies of cyclic seepage at the specimen scale, HWCC at the coal seam scale, and CBM development at the coal seam group scale. The main findings are summarized as follows. Unlike the stress-compression specimens, which exhibited overall failure, the differential distribution of pressure gradients within the specimens during the gas seepage experiments resulted in the development of microcracks concentrated at the bottom of the specimens. With the escalation of coal body structural damage, the permeability stress compression coefficient during the elastic deformation stage shows an increasing trend. Upon entering the plastic deformation stage, the specimen’s permeability shifts from a decreasing to an increasing trend. Increasing the lower limit pressure ratio, upper limit pressure ratio, and cycle rate of gas injection pressure could accelerate the failure efficiency of the specimens. Increasing the lower limit pressure ratio has a more pronounced effect on the TDC specimens, while increasing the upper limit pressure ratio has a more significant impact on the primary undeformed and cataclastic coal specimens. In TDCs, the phenomenon of wellbore collapse and cavity expansion becomes more pronounced. The cavity volume tends to increase with the increase in the initial well diameter, in-situ stress, and alternating bottom-hole pressure differences. The HWCC could enhances reservoir permeability and production in both the cavity completed and adjacent coal seams. For two horizontal wells with cavity completions in the same layer, excessively close well spacing is unfavorable for fostering inter-well interference and reservoir pressure drop. When the HWCC is applied to thin coal seams, the enhancement of permeability and production is limited. Reasonable optimization of drainage rates could effectively mitigate the reservoir permeability velocity-sensitive damage. The research findings extend related theories and provide a robust reference for subsequent practical engineering applications.

Multi-scale structure and reinforcement mechanisms of multi-element composite ceramic coatings

Ceramic coating is of significant importance for improving metallic materials in terms of mechanical properties or durability, which is commonly encountered in the field of machinery, civil engineering, and aerospace, etc. However, ceramic materials optimization is a great challenge due to the complex multi-scale design from atomic to macro level. This paper investigates the multiscale characteristics of multi-element composite ceramic coating, including electronic properties, 3D pore structure, and engineering performances and gives their bottom-up connections, via first-principles calculations and multiscale experiments. The doping of Ti, Zr, and Ce in the alumina-based ceramic crystal improves the overlap of electronic clouds between oxygen and metal atoms, which modifies the atomic charges and enhances the ionic bonding. In terms of microstructure, it reveals the mechanisms of phase transformation toughening effect of ZrO2 and the grain refinement and grain boundary purification effects of CeO2, which facilitates the amelioration of pore structure and macro mechanical properties. It provides multiscale information on the phase stability of multi-element alumina-based ceramics, shedding light on the fundamental atomic level mechanisms that play a crucial role in customizable functional designs

Rolling bearing fault diagnosis based on acoustic-vibration data fusion and mode decomposition combined with the crested porcupine optimization algorithm

Variational Mode Decomposition (VMD) is extensively utilized in the domain of rotating machinery fault diagnosis. Nevertheless, the reliance on empirical parameter tuning and the limitations inherent in traditional single-signal (either vibration signals or acoustic signals) fault diagnosis methods present substantial challenges. These include incomplete information and high sensitivity to noise. To address these shortcomings, a new method is introduced that integrates multi-modal sensor data and implements an adaptive VMD approach, enhanced through the Crested Porcupine Optimization Algorithm (CPO) to automatically optimize the key parameter. Then, uses optimized VMD to decompose the time series into intrinsic mode functions of different frequencies to capture the multi-scale characteristics of data. Finally, the decomposed acoustic-vibro signal components are fed into the CNN-BiGRU-AT network, which learns the spatio-temporal dependencies to enhance the accuracy of fault classification. The experimental results conducted across diverse high-noise environments demonstrate that this method’s superior noise robustness compared to single-modality sensor approaches， highlighting the method's potential for machinery fault detection in noisy conditions.

RSPS-SAM: A Remote Sensing Image Panoptic Segmentation Method Based on SAM

Satellite remote sensing images contain complex and diverse ground object information and the images exhibit spatial multi-scale characteristics, making the panoptic segmentation of satellite remote sensing images a highly challenging task. Due to the lack of large-scale annotated datasets for panoramic segmentation, existing methods still suffer from weak model generalization capabilities. To mitigate this issue, this paper leverages the advantages of the Segment Anything Model (SAM), which can segment any object in remote sensing images without requiring any annotations and proposes a high-resolution remote sensing image panoptic segmentation method called Remote Sensing Panoptic Segmentation SAM (RSPS-SAM). Firstly, to address the problem of global information loss caused by cropping large remote sensing images for training, a Batch Attention Pyramid was designed to extract multi-scale features from remote sensing images and capture long-range contextual information between cropped patches, thereby enhancing the semantic understanding of remote sensing images. Secondly, we constructed a Mask Decoder to address the limitation of SAM requiring manual input prompts and its inability to output category information. This decoder utilized mask-based attention for mask segmentation, enabling automatic prompt generation and category prediction of segmented objects. Finally, the effectiveness of the proposed method was validated on the high-resolution remote sensing image airport scene dataset RSAPS-ASD. The results demonstrate that the proposed method achieves segmentation and recognition of foreground instances and background regions in high-resolution remote sensing images without the need for prompt input, while providing smooth segmentation boundaries with a panoptic segmentation quality (PQ) of 57.2, outperforming current mainstream methods.

Recent developments on multiscale simulations for rheology and complex flow of polymers

AbstractThis review summarized the multiscale simulation (MSS) methods for polymeric liquids. Since polymeric liquids have multiscale characteristics of monomeric, mesoscopic, and macroscopic flow scales, MSSs that relate different hierarchical levels are adequate to reproduce flow properties accurately. Our review includes pioneering studies to the most advanced MSS studies on rheology predictions and flow simulations of polymeric liquids. We discuss two major types of MSS methods: the bottom-up and model-embedded MSS methods. The former method mainly connects all-atom molecular dynamics models and mesoscopic models to predict rheological properties. In contrast, the latter method, where a microscopic or mesoscopic model is embedded in a macroscopic computational domain, is designed to predict macroscopic flow properties. Finally, we also discuss MSS methods using machine learning techniques. Graphical abstract

Prediction model of permeability index for blast furnace based on WD-NL-transformer

The permeability index is an important variable to reflect the operating state of a blast furnace quickly, intuitively and comprehensively. Given the multi-scale, nonlinear and massive characteristics of the blast furnace smelting process data, a prediction model of permeability index based on the WD-NL-transformer (wavelet de-noising-nonlinear-transformer) method was proposed. The key variables affecting permeability index were selected from multiple perspectives by Pearson correlation coefficient and copula entropy. The extreme outliers were optimised using box plots and Lagrange linear interpolation. A data set was constructed after wavelet de-noising to build the prediction model with nonlinear processing capability, time-series prediction capability, and early stopping method. The results show that the MSE of WD-NL-transformer is 0.0093, with the RMSE of 0.0086. Meanwhile, the R2 is 0.9859, and the running time is 19.59 s, which means the model has great prediction accuracy and inference speed. The proposed model could provide theoretical support for advanced control of permeability index, which is of practical significance to ensure the stability of blast furnace smelting and accelerate the metallurgical process of intelligent automation.

Characterizing the Multiscale Knock Energy of the in-Cylinder Pressure of Compound Combustion Engines Fueled with Dimethyl Ether.

A two-cylinder, direct injection, and four-stroke naturally aspirated diesel engine is reformed to the compound combustion engine fueled with dimethyl ether. The wavelet energy spectra of the in-cylinder pressure with normal combustion and knocking combustion in a compound combustion engine are experimentally studied. The effects of the port fuel quantity, engine load, and engine speed on them are analyzed. The multiscale knock energy characteristics of the in-cylinder pressure are investigated based on wavelet energy and the Shannon entropy-energy ratio. The result shows that the in-cylinder pressure oscillation tends to be violent with the increase of port fuel quantity, the peak in-cylinder pressure increases, and its crank angle phase advances. With the increase in port fuel quantity, the wavelet energy of high-frequency detail signals d1, d2, and d3 obtained from wavelet decomposition all increases and the Shannon entropy-energy ratio decreases. The high-frequency detail signal d1 is more sensitive than the other detail signals. The frequency band of 5-10 kHz is the knock characteristic frequency band. The energy of the detail signal d1 increases significantly during knocking combustion, and the oscillation range enlarges and moves forward. The wavelet energy of detail signal d1 is the largest, and the Shannon entropy-energy ratio is the smallest at different brake mean effective pressures and different engine speeds. The effect of brake mean effective pressure and engine speed on the values is not obvious, and the port fuel quantity is the main factor.

Advances in Multiscale Modeling and Mechanical Properties Characterization of 3D‐Braided Composites

The 3D‐braided composites have been widely used in aerospace and other fields due to the excellent mechanical properties, designability, and resistance to delamination. Accurately analyzing and evaluating mechanical properties of braided composite structures is the key, which successfully designs related structural components. However, 3D‐braided composites have complex internal meso–microscopic structures, and directly establishing a solid structure model to predict mechanical behavior poses certain challenges. The multiscale research methods are a type of approach that studies the cross‐scale structural characteristics at macro–meso–microscopic scales and couples relevant scales into a whole. Its emergence provides the possibility for evaluating the overall structural performance of 3D‐braided composites. In this article, first, the multiscale analysis models of 3D‐multidirectional‐braided composites are reviewed, and then the theoretical and numerical research progress on the static and dynamic multiscale mechanical performance of regular components of 3D‐braided composites is summarized. In addition, due to the distortion and variation of geometric shapes of 3D‐braided composites on the meso–microscopic structure; therefore, also in this article, the research progress on multiscale mechanical performance characteristics of special‐shaped components is summarized. Finally, the key problems in the multiscale research of 3D‐braided composites are presented.

Orifice Versus Converging-Nozzle Grid Turbulence: A Wavelet Perspective

Grids such as perforated plates are of fundamental importance in flow turbulence study and are commonly utilised to promote mixing. An orificed perforated plate (OPP) and its reversed counterpart, the converging-nozzle perforated plate (CNPP), were applied to produce quasi-isotropic turbulence inside a wind tunnel. The three orthogonal velocity components were measured using a triple hotwire at 10D downstream of the perforated plate for Reynolds numbers, ReD, 18,700 and 28,400, where D is the diameter of the perforated holes. The statistics of the grid-generated turbulence was analysed using the time-averaged local velocity profile and turbulence intensity, which revealed a more homogeneous distribution of the flow field with a higher level of turbulence for the OPP. Fourier and wavelet analyses were employed to investigate the energy of the eddies as a function of frequency and multiscale characteristics of the fluctuating velocity, respectively. At ReD = 18,700, the turbulent energy remains prominently with large-scale vortical structures which are non-intermittently present in the flow for both perforated plates. The thickness of the converging channels of the CNPP appears to provide the venue for spawning intermittent fluctuations. At higher ReD 28,400, the effect of this intermittent behaviour becomes evident for the CNPP, leading to a multiscale distribution of turbulent energy.

Study of perforated well productivity by a multi-scale coupling flow model with Darcy-brinkman-Navier/Stokes

The study of oil and gas flow patterns after perforation completion is essential for well productivity prediction and solutions optimization. Due to the multi-scale flow characteristics of production fluids after perforation completion, we have developed a multi-scale coupled flow model incorporating Darcy, Brinkman, and Navier/Stokes equations based on mass conservation and pressure equilibrium principles. The oil and gas flow after perforation completion is simulated, and the multi-scale coupled model is solved by numerical methods, revealing the flow law of oil and gas in the multi-scale space formed by perforation. Additionally, sensitivity analysis on factors influencing perforation completion productivity was conducted. Results indicate that the most influential parameters on well productivity, in descending order, are: blocking perforation permeability > perforation diameter > perforation density > perforation depth > perforation phase > crushed zone thickness. The selection of appropriate internal flow models for perforations is critical for accurate well productivity estimation. Given the significant impact of perforation permeability on completion productivity, measures should be taken to prevent perforation blockage during production, and clear blockages promptly upon detection. This research provides a theoretical foundation for rational selection of perforation methods, gun bullets, perforation parameters, and completion techniques.

Structure and function of skin barrier lipids: Effects of hydration and natural moisturizers in vitro

Lipid membranes play a crucial role in regulating the body’s water balance by adjusting their properties in response to hydration. The intercellular lipid matrix of the stratum corneum (SC), the outermost skin layer, serves as the body’s primary defense against environmental factors. Osmolytes, including urocanic acid (UCA) and glycerol, are key components of the natural moisturizing factor that help the SC resist osmotic stress from dry environments. This study examines the effects of UCA and glycerol (each at 5 mol %) on isolated human SC lipids. For this, different techniques were employed, offering complementary information of the system’s multiscale characteristics, including humidity-scanning quartz crystal microbalance with dissipation monitoring, infrared spectroscopy, x-ray diffraction, electrical impedance spectroscopy, and studies of water loss and permeability. Our results show that UCA increases water sorption and makes lipid films more liquid-like at high relative humidity, without significantly altering the lipid lamellar structure, chain order, or orthorhombic chain packing. Lipid films containing UCA exhibited higher water loss and significantly higher model drug permeability compared to lipid films without UCA. Further, incorporation of UCA resulted in kinetically faster changes in electrical properties upon contact with aqueous solution compared with control lipids. These observations suggest that UCA reduces lipid cohesion in regions other than the acyl chain-rich leaflets, which may impact SC desquamation. In contrast, glycerol did not influence the hydration or permeability of the SC lipid matrix. However, it increased the proportion of orthorhombic domains at high humidities and slowed the kinetics of the hydration process, as evidenced by slower changes in the dielectric properties of the lipid film. These findings suggest that glycerol enhances lipid cohesion rather than increasing water uptake, which is typically the expected function of humectants. Consequently, UCA and glycerol appear to have distinct roles in maintaining epidermal homeostasis.

Multi-scale flow atomization characteristics of Jatropha biodiesel swirl liquid film breakup

The characteristics of the spray and droplets produced by liquid film breakup are crucial for understanding the atomization process in industrial furnace atomizers. This study employed high-speed camera technology to capture images of Jatropha biodiesel (JB) spray at various pressures and Ohnesorge numbers (Oh). Image processing methods were employed to analyze the evolution morphology and proper orthogonal decomposition (POD) of fuel swirl spray, revealing the instability, breakup length, spray cone angle, fractal characteristics of the liquid film breakup zone, and droplet distribution characteristics. The results indicated that the JB jet evolution progressed through stages: cylindrical jet, torsion liquid film, open swirl, complete swirl, and fully developed mode. The evolution and disintegration of the liquid film exhibited a complex structure with ligaments, perforations, and droplet separation. Flow disturbances induced the Klystron effect, characterizing microscopic droplet motion. As spray pressure increased, the growth rate of the disturbance wave increased, the breakup length decreased, the atomization cone angle enlarged, and the fractal dimension (FD) of the liquid film breakup zone increased. The droplet size distribution followed a log-normal distribution, with a substantial increase in the number of small droplets as injection pressure increased. This study provides valuable insights into the multi-scale flow atomization of biodiesel in industrial furnace, and is of great significance for the design, operation, and optimization of spraying systems in various industrial applications.

Effects of multi-scale turbulent motions on aerodynamic performance of wind turbine under sand-laden conditions

Wind turbine installation in the desert and Gobi regions offers a promising approach for meeting long-term energy demands. However, the effect of multi-scale characteristics in sand-laden atmosphere flows on wind turbine aerodynamic performance has not been evaluated. In this study, wind velocity data collected from the Qingtu Lake Observation Array (QLOA) were employed to address this gap. Results show that up to 58% of the total turbulent kinetic energy (TKE) is accounted for by very large-scale motions (VLSMs), which make up a considerable portion of the TKE. The contributions of the large-scale motions (LSMs) and the small-scale motions (SSMs) to TKE are 36% and 6%, respectively. The contribution of multi-scale turbulent motions to the aerodynamic loads of wind turbine under sand-laden conditions has been quantified for the first time. The comparison demonstrates that while LSMs and SSMs exhibit a rapid drop in their contributions to wind turbine loads with height, VLSMs show a rapid increase. Wavelet analysis revealed a strong correlation between VLSMs and power, thrust, and blade root flapwise moment at periods ranging from 256 to 1024 s. This correlation weakens as the streamwise length scale of the turbulent coherent structure decreases. This study provides essential insights for optimizing wind turbine design and site selection in sand-laden environments.

Multi-scale neural network for accurate determination of the ash content of coal flotation concentrate using froth images

Flotation concentrate quality is strongly correlated with its froth. Therefore, froth images can be used to determine coal concentrate quality. Earlier studies on using coal flotation froth images to determine concentrate ash content ignore the multiscale characteristics of the images. This paper proposes a multiscale neural network (MSNet) to tackle this challenge. The MSNet model can process images with a wide range of resolutions and extract the image features from multiple scales for quality prediction. The model is validated using laboratorial and industrial datasets. Results show that the MSNet correlates with the laboratorial and industrial samples with R2 values of 0.9430 and 0.8288, respectively. The accuracy was further improved by transfer learning technique, resulting in R2 values of 0.9572 and 0.8647 for laboratorial and industrial samples, respectively. These results exhibit the good adaptability of the proposed MSNet. They also indicate that our MSNet is promising in real-world applications, promoting cleaner and more efficient coal production

Multi-scale modeling: Finite element analysis of the thermophysical properties of carbon/carbon composites considering manufacturing defects and porosity at high temperature

Carbon/carbon (C/C) composites have strict requirements for their thermal properties in extreme environments, especially at high temperature, where their properties are crucial for long-term life. The structure of C/C composites is exceptionally complex and exhibits multi-scale characteristics, and their thermophysical properties are closely related to temperature. Therefore, exploring their thermal response and heat transfer mechanisms through experimental method is relatively costly. This paper constructs a multi-scale finite element model to investigate the influence of structure and manufacturing defects on performance, and analyzes the thermophysical properties of the composites at High-temperature. By constructing a micro-representative volume element (Micro-RVE) that includes fibers, matrix, and pores, and using a steady-state heat transfer analysis method, the influence of material phase distribution on performance is studied. At the same time, a Meso-RVE reflecting the layering form, needle punching effect and manufacturing defects is established, and the transient heat transfer analysis method is used to investigate the influence of the structural form on performance. Finite element analysis shows that, the simulation values of the three C/C composites of unidirectional fiber bundles, short-chopped fiber felts, and needle punched C/C composites show good consistency with the experimental results in the published literature. This paper uses numerical methods to calculate the thermophysical properties of C/C composites from room temperature to 900°C and explores their variation with temperature. The constructed multi-scale model provides an accurate and effective method for predicting the thermophysical properties of C/C composites and also provides new perspectives and insights for thermal-mechanical coupling analysis and structural design.

Optimizing planting density enhances the multi-scale structural characteristics and in vitro digestibility of maize starch via modulating the size distribution of granules

The modification of starch through agricultural practices is becoming increasingly significant for producing healthy foodstuffs and raw materials for industrial applications, consequently gaining momentum in academic research. This study examined how three different planting densities influenced the distribution of granule sizes, multi-scale structural characteristics, and in vitro digestibility of maize starch. The results showed that planting density significantly enhanced grain yield and relative crystallinity, and significant increases were also observed in the contents of both rapidly and slowly digestible starch. The surface- and volume-weighted mean diameter of granules significantly increased under the medium level (6.75 × 104 plants ha−1), and then decreased under high planting density level. As planting density level increased, the amylose content, peak viscosity, and hardness varied from 23.3 to 26.4 %, from 1962 to 2659 mPa·s, and from 129.3 to 307.6 g, respectively. However, no change was found in crystalline structure of maize starch. These results indicated that optimizing planting density could effectively improve grain yield and starch characteristics of maize, with the best effect under the level of 6.75 × 104 plants ha−1.

Improved Population Mapping for China Using the 3D Building, Nighttime Light, Points-of-Interest, and Land Use/Cover Data within a Multiscale Geographically Weighted Regression Model

Large-scale gridded population product datasets have become crucial sources of information for sustainable development initiatives. However, mainstream modeling approaches (e.g., dasymetric mapping based on Multiple Linear Regression or Random Forest Regression) do not consider the heterogeneity and multiscale characteristics of the spatial relationships between influencing factors and populations, which may seriously degrade the accuracy of the prediction results in some areas. This issue may be even more severe in large-scale gridded population products. Furthermore, the lack of detailed 3D human settlement data likewise poses a significant challenge to the accuracy of these data products. The emergence of the unprecedented Global Human Settlement Layer (GHSL) data package offers a possible solution to this long-standing challenge. Therefore, this study proposes a new Gridded Population Mapping (GPM) method that utilizes the Multiscale Geographically Weighted Regression (MGWR) model in conjunction with GHSL-3D Building, POI, nighttime light, and land use/cover datasets to disaggregate population data for third-level administrative units (districts and counties) in mainland China into 100 m grid cells. Compared to the WorldPop product, the new population map reduces the mean absolute error at the fourth-level administrative units (townships and streets) by 35%, 51%, and 13% in three test regions. The proposed mapping approach is poised to become a crucial reference for generating next-generation global demographic maps.

VMD-ATT-LSTM electricity price prediction based on grey wolf optimization algorithm in electricity markets considering renewable energy

Electricity price prediction is essential for the optimal dispatch in power markets, with accurate prediction models being critical for efficient power system operations and market trading decisions. Deep learning networks, with their powerful nonlinear modeling capabilities, have shown promising results in electricity price forecasting. However, their design techniques, especially the selection of network parameters, remain challenging. This indicates that the optimization and exploration of deep learning networks in electricity price forecasting models require further investigation. This paper innovatively proposes a forecasting model that uniquely integrates Variational Mode Decomposition (VMD), Grey Wolf Optimization (GWO), Attention Mechanism (ATT), and Long Short-Term Memory Network (LSTM), optimizing the model from three different perspectives. First, during the data preprocessing phase, the training set is subjected to VMD to reduce noise, thereby enhancing the capture of multi-scale characteristics inherent in electricity price time series. The ATT layer is integrated to adaptively allocate weights, enhancing the model's focus on significant features. The GWO is applied to optimize hyperparameters of the LSTM, accelerating convergence and improving iteration accuracy, thereby reducing model error. A series of experiments were conducted using multiple regional electricity price datasets, evaluated with several metrics including RMSE. The results validated the effectiveness of the proposed three modules in improving the performance of the time series network, with VMD making the most significant contribution. Among all models, VMD-GWO-ATT-LSTM consistently outperformed others, demonstrating its effectiveness and robustness in electricity price forecasting.