Extraction Unit Research Articles

Electric Trucks Pantograph-Catenary Interaction Condition Monitoring Method Based on Semantic Segmentation Network and Linear Fitting

Abstract The pantograph–catenary system (PCS) of dual source electric trucks is an important way for the vehicle to obtain electricity from the grid. Power quality depends on the reliability of the PCS contact. It is therefore necessary to monitor the PCS by detecting the contact point (CPT) between the pantograph and the catenary. However, due to structural variations between train PCS and electric truck PCS, existing CPT detection methods designed for trains are not suitable for electric trucks. To solve this problem, this paper proposes a CPT detection method that combines semantic segmentation and linear fitting. First, a lightweight pantograph and contact line segmentation network (PCSN) is proposed to extract the pantograph and contact line regions. The PCSN comprise a lightweight encoder, Cross-Attention Fusion decoder and Hybrid Atrous Spatial Pyramid Pooling (HASPP). The lightweight encoder comprise mainly of focus and lightweight feature extraction unit. The cross-attention fusion decoder is used to fuse low-level and high-level features, and the HASPP is used to obtain multi-scale context
information. Second, we use position correction and least squares linear fitting algorithm to detect the CPT of electric trucks. Experimental results show that the proposed method's detection error is concentrated around ±5 pixels. In terms of processing speed, the proposed method can reach a high speed of 76.9 FPS (Frames Per Second) on RTX 3080 desktop computer, 47.13 FPS on CPU I9-12900, and 11.63 FPS on embedded device Jeston TX2.

Efficient recovery of organic phase from crud in solvent extraction unit of Sarcheshmeh copper complex in Iran by powerful water-absorbent reagents

The formation of stable multi-phase emulsions (crud) is a serious problem in almost all commercial solvent extraction (SX) plants. Crud is a mixture of organic and aqueous phases, air, and organic or inorganic solid particles, often found to distribute itself randomly throughout the separating phases of SX plants. Crud formation is encountered in almost all SX units in metal extraction plants. It causes the loss of solvent and thus increases the process costs. Therefore, the removal of crud and recovery of the expensive organic phase including diluent and extracting reagents from it is of great importance. In this research, the applicability of a range of powerful water absorbent reagents was studied for the recovery of organic phase from the crud formed in copper extraction/electrowining unit of Sarchashmeh copper complex (IRAN). The reagents included gypsum, carbomer, sodium polyacrylate, sodium sulfate, magnesium sulfate and calcium chloride. The results showed that gypsum, carbomer and sodium polyacrylate are effective additives which let the organic phase to be recovered with up to 95 % yield. To ensure that gypsum does not cause a negative effect on the copper loading and stripping of the recovered solvent, the extraction/stripping properties of the organic phase separated with gypsum was compared with the organic phase separated by centrifugation without additives. Surface tension analysis of the gypsum-recovered organic phase also confirmed its suitability for recovery of organic phase from crud. Finally, the calcium content of gypsum-recovered organic phase was measured and compared with that of the organic phase separated by centrifugation without any additive.

Bipolar Membrane Capacitive Deionization for the Selective Capture of Lithium Ions from Brines and Conversion to Lithium Hydroxide

Meeting the increasing demand for lithium in vehicle electrification and renewable energy storage requires innovations in lithium-ion (Li+) separations. Traditional solar evaporation methods for lithium recovery are slow and consume tremendous volumes of water and secondary chemicals (acids and bases). This study introduces a bipolar membrane capacitive deionization (BPM-CDI) unit for direct lithium extraction and LiOH production without the external addition of acids and bases. Utilizing de-lithiated lithium-iron-phosphate (LFP) coated carbon cloth electrodes, the BPM-CDI unit demonstrates selective Li+ capture over competing ions. Molecular dynamics simulations and H-cell experiments elucidate pH inversion mechanisms during Li+ release, yielding LiOH. The BPM-CDI platform efficiently removes Li+ from synthetic brines featuring 8x higher Mg2+ concentrations (200 ppm Mg2+) and 26x higher Na+ concentrations (682 ppm Na+), achieving a LiOH concentration of 124 ppm (36 ppm Li+) after 8 cycles of recirculation. Post-mortem analysis confirms electrode integrity and stability. BPM-CDI integrated with selective electrodes is a promising electrochemical separation-reactor platform for lithium recovery while producing LiOH.

Sub-diffraction-limited single-photon 3D imaging based on domain features extraction network at kilometer-scale distance

Single-photon imaging holds extensive application value, yet its imaging quality is often constrained by low signal-to-background ratios (SBRs). In this paper, we designed a network based on the statistical priors of photons, corresponding to the Gaussian statistical prior along the photon time axis, the Poisson statistical prior of photon counts, the varying receptive field requirements for different SBRs, and the deconvolution for solving the photon point spread function (PSF). We designed multiple modules to handle different photon statistical priors and introduced Selective Kernel Network (SKNet) as the feature extraction unit. This enables our method to accurately restore the true spatial distribution of photons. Compared to other methods, we place greater emphasis on the impact of neighborhood characteristics on photons and feature extraction under low SBR. In addition, we designed FoV Module to do deconvolution for PSF at sub-pixel scanning scenarios for sub-diffraction imaging. Experimental results demonstrate our methods, in particularly, achieved the best implementation results to date at low SBR data, which can achieve denoising while preserving more detail features and our methods also achieved twofold sub-diffraction imaging at a distance of 1.5 km using collected real-world data, thereby showing its immense potential to improve the performance of single-photon imaging systems for kilometer-scale imaging.

A clustered federated learning framework for collaborative fault diagnosis of wind turbines

Data-driven approaches demonstrate significant potential in accurately diagnosing faults in wind turbines. To enhance diagnostic performance and reduce communication costs in federated learning with data heterogeneity among different clients, we introduce a clustered federated learning framework to wind turbine fault diagnosis. Initially, a lightweight multiscale separable residual network (LMSRN) model is proposed for each local client. The LMSRN model integrates a multiscale spatial feature derivation unit and a depthwise separable feature extraction unit. Subsequently, to tackle data heterogeneity among clients, canonical correlation coefficients of representations are extracted from the intermediate layers of local LMSRN models, and a representational canonical correlation clustering (RCCC) method is proposed to assess the similarity of local LMSRN models and group them into clusters. Finally, a global model is trained for each cluster. Real-world wind turbine data experiments showcase the superior performance of the proposed clustered federated learning framework over traditional methods in terms of diagnostic accuracy and computational speed. Additionally, the optimal choice of the number of clusters is also discussed.

Comparison of dynamic tests of the modified purex process flowsheet using 30 and 45% TBF in isoparafin

Results of centrifugal contactor rig trials of the flowsheet of the first solvent extraction cycle in protective glove boxes using simulated solutions of spent nuclear fuel and 30 and 45% TBP in isoparaffin as a solvent are compared. It was shown that an increase in the TBP concentration to 45% resulted in deterioration of the quality of the products obtained by the same flowsheets with 30% TBP in Isopar L as a solvent. Comparison of experimental and computer simulated profiles of component distribution across the stages of extraction units demonstrated satisfactory agreement in nitric acid and uranium concentrations. At the same time, it was found that simulation model of zirconium, technetium, and neptunium extraction requires improvement considering interaction of these elements with complexing agents.

Simulation and Economic Analysis of Helium Extraction Process from Natural Gas

The investment estimation of the helium extraction project from natural gas is a crucial step in economically obtaining helium from both domestic and international projects. This article employs Aspen HYSYS to simulate the process and estimate the investment levels of Linde and Exxon Mobil integrated helium extraction processes. We investigate the influence of feed composition and processing capacity on investment costs and product returns. The results indicate that higher helium content of feed correlates with increased equipment investment costs and total capital cost (CAPEX), and that the Linde integrated process is significantly more sensitive to changes in helium content of feed than the Exxon Mobil integrated process. As the helium content of feed rises, the product returns of the two processes are evidently improved, leading to reduced investment payback periods. Both techniques exhibit favorable payback periods when the feed helium content exceeds 0.5 vol%. Nevertheless, elevated nitrogen content in the feed notably escalates the equipment investment costs and total capital costs. Furthermore, an increase in the processing capacity of feed gas leads to a nonlinear increase in total capital costs and annual operating costs. However, the cost per unit of helium extraction diminishes with increasing capacity. In general, the Linde integrated process requires higher separation energy consumption in comparison with the Exxon Mobil integrated process at similar processing capacities. Moreover, the sensitivity analysis shows that helium breakeven price is strongly affected by the price of both LNG and feed gas.

Evaluation of Counter Current Horizontal Screw Extractor’s Performance in Determining Leaching Equilibrium of Clitoria ternatea Anthocyanins

Anthocyanin compounds as natural coloring substances contained in Clitoria ternatea flowers have higher antioxidant activity than anthocyanins from other flower extracts. Obtaining anthocyanin extract from Clitoria ternatea flower can be done using the solid-liquid extraction method. The objectives of this study include making a leaching equilibrium diagram of Clitoria ternatea flowers and determining the optimum L/S ratio and the number of equilibrium stages in the counter current horizontal screw extractor. In the maceration process, a leaching equilibrium diagram is produced, which is then used to design and determine the number of equilibrium stages. The equilibrium diagram of anthocyanins from Clitoria ternatea flowers with a gradient slope to the right is in accordance with the literature so that the equilibrium diagram can be used in the design of the extraction unit and the calculation of the theoretical stage of the multi-stage extraction unit. Based on the equilibrium diagram, the Number of Transfer Units (NTU) of 4 stages and the Height of the Transfer Unit value of this process is 0.105 m.

Advanced feature engineering in microgrid PV forecasting: A fast computing and data-driven hybrid modeling framework

This study introduces an innovative framework designed to forecast the fluctuating short-term generation of photovoltaic (PV) energy in isolated microgrids. The framework relies entirely on solar irradiance data obtained from weather stations located far from the target microgrid. It implements a sophisticated decomposition approach to effectively extract an enriched collection of features from the dataset. Subsequently, a machine learning-based clustering method further enhances the forecasting process by accurately separating the relevant data points. The approach utilizes an advanced two-stage Hybrid Data Linked Model (HDLM) architecture, integrating a Layered Recurrent Neural Framework (LRNF) for prediction and a pattern identification network unit for pattern extraction. This paper demonstrates significant improvements in both the accuracy and effectiveness of estimating PV generation, achieving a mean absolute error of 1.02, a root mean square error of 2.176, and an R-squared score of 0.991. Additionally, the method reduces computing time by 15% after finalizing the input features. A comparative analysis evaluates the superior forecasting capabilities of the HDLM in remote microgrids by benchmarking it against other advanced hybrid deep learning models. The findings highlight the HDLM’s potential to greatly enhance and revolutionize the management and operation of renewable energy systems in remote microgrids.

Online chatter detection via lightweight deep learning framework with efficient signal pre-processing

Deep learning methods have attracted considerable attention in the field of chatter detection due to their exceptional feature extraction and identification capabilities. However, it is challenging to adapt these methods to diverse machining conditions with limited sensor resources. In addition, the computational efficiency of these methods is compromised by their complexity, resulting in a reduced real-time performance for online chatter detection. To address them, this paper proposes a novel deep learning-based framework that combines an efficient signal pre-processing method with a novel lightweight parallel convolutional neural network (LPNN). A fast continuous wavelet transform algorithm is applied to the raw signal, effectively generating time–frequency maps that represent chatter characteristics. The LPNN incorporates the condensed block as a feature extraction unit, which is constructed using depthwise separable convolution. The introduction of the condensed block and the parallel structure increases the depth and width of the designed model while maintaining fewer parameters and higher computational efficiency. A transfer learning strategy is adopted to train the model, reducing the training time and data requirements for online chatter detection. Experimental results using large open-source turning datasets demonstrate the effectiveness of the proposed method in identifying chatter under various machining conditions. The proposed method outperforms other methods with a classification accuracy of 96.5% on the test set. Furthermore, the performance of the proposed method for online chatter detection is validated in our milling experiments, exhibiting enhanced real-time performance compared to other methods.

A Scoping Review on Complications in Modern Hair Transplantation: More than Just Splitting Hairs.

Hair restoration surgery (HRS) is a commonly performed elective procedure but to date lacks a review of the full scope of complications. To provide a comprehensive overview of observed complications associated with follicular unit extraction (FUE) and follicular unit transplantation (FUT). Randomized control trials, cohort studies, case series, and case reports published in 1985 or later on adults (age>18). Nontraumatic or autoimmune etiologies of alopecia were excluded as procedure indications. PubMed, EMBASE, Cochrane databases (last search December 31, 2022). Data-charting and extraction were independently performed with two reviewers using Covidence. Forty-three publications were included. Two large series reported the overall complication rate to be 1.2 and 4.7%. Common complications included bleeding requiring intervention (up to 8%), persistent numbness (up to 11%), infection (up to 11% with two reports of Kaposi varicelliform eruptions and one of mucormycosis), effluvium at donor and recipient sites (up to 4.1% and 6.5%, respectively). The most common donor-site complication was hypertrophic scarring/keloid formation after FUT (up to 15.1%). Complications at the recipient site, including crusting (up to 54.8%), frontal edema (up to 50%), and sterile folliculitis (up to 53.3%), tended to be poorly defined with a broad range of incidences. Serious complications associated with HRS are rare in the hands of experienced providers. However, comprehensive discussions of risk must be had with prospective patients as any complication in the context of an elective procedure may be significant and psychologically devastating for the individual patient. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Main features of a novel extraction-hydrogenation unit

The potential for coupling a continuous hydrogenation reactor with a continuous extraction unit was investigated through computer simulation of series of mixers/settlers coupled to a slurry reactor in a closed loop.The process has the potential to replace the regeneration of extraction solvents by distillation. Additionally, the system eliminates the need to treat the entire feedstock when refining a complex mixture. It is believed that the reaction of the extract retains valuable components in the feed that are not extracted and could undergo undesirable hydrogenation.The parameters for the simulation model were obtained from correlations and published data. The system was expected to exhibit a pseudo-first-order reaction rate in relation to the impurity concentration due to its low concentration in the extract. This is expected to result in high hydrogen-to-impurity ratios, which can be advantageous from a reaction driving force perspective.Solvent degradation by hydrogenation and solvent-impurity reactions are thought to be the main problems for the extraction-hydrogenation process.

A knowledge graph algorithm enabled deep recommendation system.

Personalized learning resource recommendations may help resolve the difficulties of online education that include learning mazes and information overload. However, existing personalized learning resource recommendation algorithms have shortcomings such as low accuracy and low efficiency. This study proposes a deep recommendation system algorithm based on a knowledge graph (D-KGR) that includes four data processing units. These units are the recommendation unit (RS unit), the knowledge graph feature representation unit (KGE unit), the cross compression unit (CC unit), and the feature extraction unit (FE unit). This model integrates technologies including the knowledge graph, deep learning, neural network, and data mining. It introduces cross compression in the feature learning process of the knowledge graph and predicts user attributes. Multimodal technology is used to optimize the process of project attribute processing; text type attributes, multivalued type attributes, and other type attributes are processed separately to reconstruct the knowledge graph. A convolutional neural network algorithm is introduced in the reconstruction process to optimize the data feature qualities. Experimental analysis was conducted from two aspects of algorithm efficiency and accuracy, and the particle swarm optimization, neural network, and knowledge graph algorithms were compared. Several tests showed that the deep recommendation system algorithm had obvious advantages when the number of learning resources and users exceeded 1,000. It has the ability to integrate systems such as the particle swarm optimization iterative classification, neural network intelligent simulation, and low resource consumption. It can quickly process massive amounts of information data, reduce algorithm complexity and requires less time and had lower costs. Our algorithm also has better efficiency and accuracy.

Nanoplastic Sample Cleanup by Micro-Electromembrane Extraction across Free Liquid Membranes.

Sample preparation techniques enabling the separation and cleanup of nanoplastics removing other components present in complex sample matrices are scarce. Herein, micro-electromembrane extraction (μ-EME) has been explored for this purpose based on the extraction of nanoplastic particles across a free liquid membrane (FLM). The extraction unit is based on a perfluoroalkoxy tube sequentially filled with the acceptor solution (20 μL 5 mM phosphate buffer, pH 10.7), FLM (10 μL 1-pentanol), and donor solution (20 μL sample/standard solution). Sulfonated polystyrene beads (200 nm particle size) were selected as a model mimicking negatively charged nanoplastics. At 700 V, nanoplastics transferred from the donor solution into the FLM before moving across the FLM into the acceptor solution. Quantitative nanoplastic measurements after μ-EME were performed by injecting the acceptor solution into a capillary electrophoresis system with diode array detection. μ-EME allowed the rapid nanoplastic sample cleanup, requiring an extraction time of just 90 s and obtaining a nanoplastic transfer yield through the FLM of 60% with RSD values below 9%. The μ-EME technique enabled the efficient sample matrix cleanup of nanoplastics spiked in different tea matrices. Nanoplastic transfer yield through the FLM for black tea and flavored tea matrices were 56% and 47%, respectively, with complete sample matrix removal of UV-absorbing compounds.

Integrated RF-CNN-GRU ensemble for enhanced beef quality classification: A multi-modal approach

Ensuring food safety is crucial for health and economic sustainability. A key indicator of food quality is its smell, which can be assessed using an electronic nose (e-nose). Combining e-nose technology with machine learning methods has shown promise in enhancing classification accuracy. However, traditional methods such as Support Vector Machines (SVM), k-Nearest Neighbors (KNN), and Convolutional Neural Networks (CNN) often fall short in accurately distinguishing between fresh and spoiled beef.To address these limitations, this study proposes a novel hybrid model that integrates Random Forest, CNN for local feature extraction, and Gated Recurrent Unit (GRU) for global feature extraction. Utilizing data from 11 e-nose sensors, this model achieves an unprecedented accuracy of up to 0.9977 in differentiating fresh from spoiled beef across all 12 types of cuts. The hybrid model demonstrates superior performance metrics, including accuracy, precision, recall, F1 score, coefficient of determination (R2), root mean square error (RMSE), and residual predictive deviation (RPD), significantly outperforming traditional approaches in beef quality classification. By incorporating both local and global features, our approach offers innovative insights and marks a significant advancement in machine learning applications for food quality assessment.

Параметры шагающего устройства для добычи полезных ископаемых, рассредоточенных по морскому дну

The purpose of the work is to analyze the aggregates of complexes for the extraction of marine minerals. In particular, the variants of complexes, their advantages and disadvantages are considered. The complex is analyzed, which uses robotic pickers/reclaimers of the walking type to perform mining operations of ferromanganese nodules (FMN) or cobalt-bearing manganese crusts (CMC). Results. The variants of complexes for mining of bottom deposits are analyzed, advantages and disadvantages of different variants are indicated. The device for collecting minerals on the bottom, developed in SPMU, is described, laboratory studies of the model of such a device are considered, the main parameters are presented, the dependence of productivity on the size and time of operations of the cycle of work is indicated. It is determined that for minimal impact on the seabed biota it is necessary to use walking submersibles with an actuator for extraction in the form of rarefaction chambers or a grapple with an impactor (depending on the type of solid minerals). It is shown that the productivity of the walking underwater vehicle varies with the diameter of the rarefaction chamber plates: the optimum plate diameter is 1.5 m. Experimental study of hydrodynamic resistance of “foot” models with different midships cross-section has been carried out. According to the experimental results, the conclusion about the preferred design of the “foot” was made: the cross-section should be a streamlined shape with the smallest drag coefficient cx and have a continuous surface without holes. Experimental studies of tine models movement in water have shown that it is reasonable to use as tine parts the structures without holes, streamlined ellipsoidal shape. Calculations of productivity of mining units when using rope hoisting from the bottom to the ship have shown the necessity of using four or more hoisting devices for realization of the required annual productivity of more than 1 million tons per year. The stages of creating a workable extraction unit including a vessel, vertical lifting device and a system of deep-water collectors require very significant material and monetary costs, so they should be carried out in accordance with the R&D plan and tasks of the relevant ministry and agency and require coordination of efforts of investors, scientists and machine builders.

GLFER-Net: a polyphonic sound source localization and detection network based on global-local feature extraction and recalibration

Polyphonic sound source localization and detection (SSLD) task aims to recognize the categories of sound events, identify their onset and offset times, and detect their corresponding direction-of-arrival (DOA), where polyphonic refers to the occurrence of multiple overlapping sound sources in a segment. However, vanilla SSLD methods based on convolutional recurrent neural network (CRNN) suffer from insufficient feature extraction. The convolutions with kernel of single scale in CRNN fail to adequately extract multi-scale features of sound events, which have diverse time-frequency characteristics. It results in that the extracted features lack fine-grained information helpful for the localization of sound sources. In response to these challenges, we propose a polyphonic SSLD network based on global-local feature extraction and recalibration (GLFER-Net), where the global-local feature (GLF) extractor is designed to extract the multi-scale global features through an omni-directional dynamic convolution (ODConv) layer and multi-scale feature extraction (MSFE) module. The local feature extraction (LFE) unit is designed for capturing detailed information. Besides, we design a feature recalibration (FR) module to emphasize the crucial features along multiple dimensions. On the open datasets of Task3 in DCASE 2021 and 2022 Challenges, we compared our proposed GLFER-Net with six and four SSLD methods, respectively. The results show that the GLFER-Net achieves competitive performance. The modules we designed are verified to be effective through a series of ablation experiments and visualization analyses.

Direct numerical simulations of tritium extraction in PbLi-based breeding blankets in the laminar–turbulent transition region

In a lead-lithium (PbLi) based breeding blanket, the bred tritium remains dissolved in the PbLi, which flows towards the Tritium Extraction Unit (TEU). In a TEU, tritium must be extracted at a fast enough rate that guarantees the plant’s self-sufficiency and safety. Different TEU concepts exist, one of the most promising being Permeation Against Vacuum (PAV), based on the extraction of tritium from the PbLi through a highly permeable membrane.Even in the so-called low velocity blanket concepts, PbLi is expected to flow at relatively high total mass flow rates. This means that the high tritium extraction efficiencies that safe operation requires must be obtained partitioning the flow into several extraction channels, but this solution increases both cost and complexity. However, less partition channels may be required should turbulence be induced in the flow. Since turbulence increases the flow mixing, it should favor tritium extraction. Hence, in the range of velocities where turbulent phenomena start —i.e., the transition region—, extraction efficiency is expected to grow rapidly with velocity due to turbulence acting as a new transport mechanism. Thus, a turbulent flow in the TEU may achieve the high extraction efficiencies required with a more moderate partitioning scheme.In this work, a PbLi flow in a prototypical PAV channel was modeled using Direct Numerical Simulations (DNS). To trigger turbulence, two different methods were implemented, namely an instability-inducing oscillating boundary condition, and a physical turbulator consisting of a geometrical obstacle. Both methods proved successful as they resulted in greater extraction efficiencies than those seen in the analogous laminar regimes. In fact, up to a 15% increase in extraction efficiency or up to a 5-fold increase in total extraction rate were obtained.

YOLOv8n-GAM: an improved surface defect detection network for hot-rolled strip steel

Production defects caused by irresistible factors such as process design problems or differences in steel properties in strip production affect the economic benefits of the enterprise and threaten production safety. Traditional defect detection methods are difficult to achieve real-time and high-precision detection, so developing surface defect detection methods based on deep learning is of great significance for strip production. In order to effectively improve the accuracy of the deep learning model in detecting surface defects on hot-rolled strip, in this work we propose a real-time detection model for surface defects on strip steel based on the YOLOv8n model. Firstly, the newly convolutional layer Con5v is designed to replace the original convolutional layer in the neck, and an attention mechanism is added in front of each Con5v to improve the algorithm’s ability to extract small target information. Secondly, an additional set of upsampled feature extraction units is added to the neck in order to enhance the spatial information of the feature map. Subsequently, a set of feature fusion units is incorporated and the convolutional layers in it are improved to provide better feature maps. Thirdly, the number of decoupling detection heads is increased to receive more high-quality features. The final experimental results show that YOLOv8n-GAM (YOLOv8 Nano Model with Global Attention Mechanism) achieves 81.4mAP and 82.0FPS on the NEU-DET dataset and 71.2mAP and 55.0FPS on the GC10-DET dataset, which are 5.7% and 6.9% higher than those of YOLOv8n, respectively. The model proposed in this paper achieves a comprehensive performance improvement in strip steel.

A Novel UNet 3+ Change Detection Method Considering Scale Uncertainty in High-Resolution Imagery

The challenge of detecting changes in high-resolution remote sensing imagery often stems from the difficulties in effectively extracting features and constructing appropriate change detection models considering the scale characteristics of ground objects. To solve these issues, we propose a novel UNet 3+ change detection method that considers the scale characteristics inherent in various land-cover change types. Our method includes three key steps: a multi-scale segmentation method, a class-specific UNet 3+ method, and an object-oriented change detection method based on UNet 3+. To verify the effectiveness of this method, we select two datasets for experiments and compare our proposed method with the UNet 3+ single-scale sampling method, the class-specific UNet 3+ single-scale sampling method, and the UNet 3+ multi-scale hierarchical sampling method. Our experimental results show that our proposed method has higher overall accuracy and F1, lower missed detection rate and false detection rate, and can detect more changes in ground features than other methods. To verify the scalability of this method, we compare this method with traditional change detection methods such as PCA-k-means, OCVA, a single-scale sampling method based on random forest, and a class-specific object-based method. Experimental results and accuracy indexes show that our proposed method better considers the scale characteristics of ground objects and achieves higher accuracy. Additionally, we compared our proposed method with other DLCD methods including LamboiseNet, BIT, CDNet, FCSiamConc, and FCSiamDiff. Our results show that our proposed method effectively considers edge information and has an acceptable time consumption. Our approach not only considers the full-scale characteristics of the feature extraction but also the scale characteristics of the change detection model. In addition, it considers a more practical feature extraction unit (object), making it more accurate.