Cgs Units Research Articles

Attention Induced Dual Convolutional-Capsule Network (AIDC-CN): A deep learning framework for motor imagery classification

In recent times, Electroencephalography (EEG)-based motor imagery (MI) decoding has garnered significant attention due to its extensive applicability in healthcare, including areas such as assistive robotics and rehabilitation engineering. Nevertheless, the decoding of EEG signals presents considerable challenges owing to their inherent complexity, non-stationary characteristics, and low signal-to-noise ratio. Notably, deep learning-based classifiers have emerged as a prominent focus for addressing the EEG signal decoding process. This study introduces a novel deep learning classifier named the Attention Induced Dual Convolutional-Capsule Network (AIDC-CN) with the specific aim of accurately categorizing various motor imagination class labels. To enhance the classifier’s performance, a dual feature extraction approach leveraging spectrogram and brain connectivity networks has been employed, diversifying the feature set in the classification task. The main highlights of the proposed AIDC-CN classifier includes the introduction of a dual convolution layer to handle the brain connectivity and spectrogram features, addition of a novel self-attention module (SAM) to accentuate the relevant parts of the convolved spectrogram features, introduction of a new cross-attention module (CAM) to refine the outputs obtained from the dual convolution layers and incorporation of a Gaussian Error Linear Unit (GELU) based dynamic routing algorithm to strengthen the coupling among the primary and secondary capsule layers. Performance analysis undertaken on four public data sets depict the superior performance of the proposed model with respect to the state-of-the-art techniques. The code for this model is available at https://github.com/RiteshSurChowdhury/AIDC-CN.

Self-powered human movement measurement with a unidirectional ratchet driven wearable energy harvester

A large amount of energy is generated during human movement which is mostly not effectively utilized. To collect human movement energy and utilize it rationally, a wearable piezoelectric-electromagnetic energy harvester (WPEEH) driven by a unidirectional ratchet is proposed, which adopts the principle of up-conversion to convert mechanical energy into electrical energy. The electromagnetic power generator unit is used to obtain energy to power the sensing circuit, while piezoelectric generator unit can be used for both energy generation and self-powered sensing. A spring push rod and a ratchet structure are used to convert the swinging of the human limb into the unidirectional rotation of the device to enhance its energy harvesting efficiency. The theoretical model of the WPEEH is established, and the experimental test platform is built. The results show that the WPEEH is able to generate a maximum power of 17.2 mW at an oscillation frequency of 1.5 Hz. In the actual wearable test, the proposed energy harvester is able to supply the harvested energy to the low-power electronic devices through capacitors, and it enables to light up 42 LEDs as well as to drive the digital thermo-hygrometer to work continuously and stably. Apart from having excellent properties of a high output power and low drive frequency, the proposed energy harvester also exhibits a high energy density (7.1 × 10–5 W/cm3). What’s more, the speed of human movement can be predicted and calculated for different individuals by testing the results of the swing frequency with a maximum error of 2.71 %. The investigation proves that the proposed WPEEH can measure human movement and has great potential in the field of power supply for low-power electronic devices.

AI-enhanced backpack with double frequency-up conversion vibration energy converter for motion recognition and extended battery life

With growing interest in physical activities, wearable device sales have been increasing each year due to their help in tracking movement. However, keeping these devices running for a long time is still a key challenge. This paper presents an AI-enhanced backpack equipped with a hybrid vibration energy converter that integrates piezoelectric (PE) and electromagnetic (EM) technologies, featuring a double frequency-up conversion (FUC) mechanism. The PE unit enables efficient motion recognition, while the EM unit extends battery life via kinetic energy harvesting. The double FUC mechanism boosts human motion signal strength and resolution by converting sub-Hertz frequencies to hundreds of Hertz. This makes the device particularly effective in environments with performance constraints and interference. By adopting deep learning techniques, the system maintains a high motion recognition accuracy of 97.33 % even under low data sampling rates and significant noise interference. The EM unit effectively harvests kinetic energy, generating an average output power of 4.5 mW during activities such as running at 4.5 Hz, despite of the device’s compact size. Additionally, a power management circuit with human motion-stimulated switch optimizes battery life by managing the system’s sleep and active states. This innovative energy management approach ensures that during human activity, the system conserves energy by remaining in sleep state for at least 30 % of the time, significantly extending battery life and enhancing the backpack’s suitability for outdoor applications.

Microstructure evolution and magnetic characteristics of a novel high entropy alloy produced by mechanical alloying and spark plasma sintering

In this research, CoMoMnNiV high entropy alloys were successfully prepared by mechanical alloying (MA) and spark plasma sintering (SPS). The microstructure, phases and magnetic properties of the as-milled powders and bulk sample were examined, employing X-ray diffraction, differential scanning calorimeter (DSC), scanning electron microscopy, and Vibrating Sample Magnetometer (VSM) techniques. The resultant phase following MA exhibits a dual-phase microstructure comprising BCC and FCC solid solutions, with individual crystal dimensions below 18 nm. Thermal stability assessment via DSC reveals the alloy robustness up to 1216 °C. The SPS was performed on the MA samples at 980 °C and 50 MPa pressure, and their density was determined to be 89.64 %. The sample subjected to a milling duration of 40 h demonstrated a saturation magnetization of 25.977 electromagnetic units per gram (emu/g) and a coercivity of 371.5 Oersteds (Oe).

Multiple-ResNet GAN: An enhanced high-resolution image generation method for translation from fundus structure image to fluorescein angiography.

Fluorescein angiography (FA) is a diagnostic method for observing the vascular circulation in the eye. However, it poses a risk to patients. Therefore, generative adversarial networks have been used to convert retinal fundus structure images into FA images. Existing high-resolution image generation methods employ complex deep network models that are challenging to optimize, which leads to issues such as blurred lesion boundaries and poor capture of microleakage and microvessels. In this study, we propose a multiple-ResNet generative adversarial network (GAN) to improve model training, thereby enhancing the ability to generate high-resolution FA images. First, the structure of the multiple-ResNet generator is designed to enhance detail generation in high-resolution images. Second, the Gaussian error linear unit (GELU) activation function is used to help the model converge rapidly. The effectiveness of the multiple-ResNet is verified using the publicly available Isfahan MISP dataset. Experimental results show that our method outperforms other methods, achieving better quantitative results with a mean structural similarity of 0.641, peak signal-to-noise ratio of 18.25, and learned perceptual image patch similarity of 0.272. Compared with state-of-the-art methods, the results showed that using the multiple-ResNet framework and GELU activation function can improve the generation of detailed regions in high-resolution FA images.

Effect of stability state transition of variable potential well in tri-hybridized energy harvesters

Energy harvesting is a crucial technology for self-powering wireless sensor nodes and mobile terminals. In addressing this, we utilized a variable potential well to create a tri-hybrid energy harvester that integrates piezoelectric effect, electromagnetic induction, and triboelectric effect. The compression and stretching of a compressible magnet-spring oscillator alter the system’s potential well, thereby promoting the operation of the piezoelectric unit (PU), electromagnetic unit (EU), and triboelectric unit (TU). A coupled electromechanical model using Hamilton’s principle was established. By calculating the static potential energy and restoring force, we discussed the stability state transition, revealing that when the magnet distance is less than 38 mm, the system transitions from a mono-stable to a bi-stable state. The experimental study investigated the impact of excitation and structural parameters on voltage and power. A high excitation level enhances all three units. The TU demonstrates excellent electrical performance under high potential energy in the bi-stable state, while the PU performs best in mono- and bi-stable switching structure. The EU shows relatively balanced electrical performance across various conditions. These findings offer new insights into enhancing energy harvesting capabilities for external circuits.

Joint coordinate attention mechanism and instance normalization for COVID online comments text classification.

The majority of extant methodologies for text classification prioritize the extraction of feature representations from texts with high degrees of distinction, a process that may result in computational inefficiencies. To address this limitation, the current study proposes a novel approach by directly leveraging label information to construct text representations. This integration aims to optimize the use of label data alongside textual content. The methodology initiated with separate pre-processing of texts and labels, followed by encoding through a projection layer. This research then utilized a conventional self-attention model enhanced by instance normalization (IN) and Gaussian Error Linear Unit (GELU) functions to assess emotional valences in review texts. An advanced self-attention mechanism was further developed to enable the efficient integration of text and label information. In the final stage, an adaptive label encoder was employed to extract relevant label information from the combined text-label data efficiently. Empirical evaluations demonstrate that the proposed model achieves a significant improvement in classification performance, outperforming existing methodologies. This enhancement is quantitatively evidenced by its superior micro-F1 score, indicating the efficacy of integrating label information into text classification processes. This suggests that the model not only addresses computational inefficiencies but also enhances the accuracy of text classification.

MCGnet with multi-level gated unit transformations: A time series prediction-based model for unknown pattern abnormality detection

The task of anomaly detection in data patterns remains challenging due to the diverse of fault patterns, which often render existing models ineffective when encountered with unknown or unseen data patterns in the training data. Furthermore, insufficient samples, the presence of chaotic behavior, unbalanced anomaly patterns are potential limiting factors in the practical application. While prediction error-based approaches attempt to address these challenges, the discriminability of abnormal patterns in existing scenarios is not yet sufficient, thus limiting the ability to identify abnormal phenomena. To overcome these limitations, we proposed a novel model called MCGnet. The backbone architecture of MCGnet employed a novel design of continuous multi-layer circular convolution transformations. This design enhances the expressive capability of the model by expanding the differences among data patterns. Additionally, we introduced in parallel the newly proposed temporal-frequency gated Gaussian unit, which generates richer feature representations. Subsequently, the obtained feature representations were fed into an explicit causal network to learn complex patterns of data along temporal dependencies. The MCGnet not only solves high-precision time series prediction problems but also significantly improves the discriminability between normal and abnormal patterns. Since the proposed model is accurately modeled using only a small amount of normal data, it bypasses the limiting factors such as insufficient fault samples and sample imbalance, and is able to effectively deal with various unknown data patterns. The accuracy of the proposed method was verified through simulation data (Duffing system) and its potential for engineering applications was demonstrated through real-world data (rotor system).

An integrated micromachined flexible ultrasonic-inductive sensor for pipe contaminant multiparameter detection

Pipe contaminant detection holds considerable importance within various industries, such as the aviation, maritime, medicine, and other pertinent fields. This capability is beneficial for forecasting equipment potential failures, ascertaining operational situations, timely maintenance, and lifespan prediction. However, the majority of existing methods operate offline, and the detectable parameters online are relatively singular. This constraint hampers real-time on-site detection and comprehensive assessments of equipment status. To address these challenges, this paper proposes a sensing method that integrates an ultrasonic unit and an electromagnetic inductive unit for the real-time detection of diverse contaminants and flow rates within a pipeline. The ultrasonic unit comprises a flexible transducer patch fabricated through micromachining technology, which can not only make installation easier but also focus the sound field. Moreover, the sensing unit incorporates three symmetrical solenoid coils. Through a comprehensive analysis of ultrasonic and induction signals, the proposed method can be used to effectively discriminate magnetic metal particles (e.g., iron), nonmagnetic metal particles (e.g., copper), nonmetallic particles (e.g., ceramics), and bubbles. This inclusive categorization encompasses nearly all types of contaminants that may be present in a pipeline. Furthermore, the fluid velocity can be determined through the ultrasonic Doppler frequency shift. The efficacy of the proposed detection principle has been validated by mathematical models and finite element simulations. Various contaminants with diverse velocities were systematically tested within a 14 mm diameter pipe. The experimental results demonstrate that the proposed sensor can effectively detect contaminants within the 0.5−3 mm range, accurately distinguish contaminant types, and measure flow velocity.

A hybrid harvester using a magnetically coupled pendulum structure for limb movement

ABSTRACT Energy harvesting and conversion devices can provide power solutions for wearable devices. In this paper, a piezo-electromagnetic hybrid energy harvester (PEMHEH) using a magnetically-coupled pendulum structure for low-frequency limb movement is proposed. PEMHEH consists of a piezoelectric generator (PEG) unit including a main piezoelectric transducer, an auxiliary piezoelectric transducer (APT), and an electromagnetic generator (EMG) unit. It can reduce the starting frequency and increase the output performance by adopting a slidable mass block, the APT, and the rational magnetic pole arrangement. Its theoretical model is established, and magnetic simulation and experimental tests are carried out. The results show it can be started at 0.4 Hz excitation, and the output performance is maximized at 2 Hz. The maximum voltages of the piezoelectric generator and electromagnetic generator units are 5.29 V and 457 mV, and the maximum power is 286.16 μW (30 kΩ) and 11.25 μW (2 kΩ). The PEMHEH can light up LED boards at a 1 Hz excitation frequency. It has the potential to power wearable devices.

An LCC-CS bilateral matching network for high efficiency energy conversion in high-frequency piezoelectric rotary ultrasonic machining system

The processing efficiency and machining quality of piezoelectric rotary ultrasonic machining (PRUM) system are closely related to its electromagnetic energy conversion units and electroacoustic energy conversion units. Generally, the high efficiency and high-power capacity of PRUM systems are limited by several energy conversion devices including the capacitive piezoelectric transducer (PT) load, loosely coupled rotating transformer (RT), and non-soft switching inverter. This article improves the overall performance of high-frequency PRUM system by bilateral matching network. On the rotating side, the compared results of five different static matching schemes show that the CS matching scheme ensures stable and efficient vibration of capacitive PTs even under incomplete matching. On the stationary side, a dynamic LCC matching scheme is adopted to keep the output impedance angle of H-bridge inverter in the weak inductive state. Finally, the proposed high-order matching network in this article makes the efficiency of high-frequency PRUM system over 90 % steadily.

Improving Surgical Scene Semantic Segmentation through a Deep Learning Architecture with Attention to Class Imbalance.

This article addresses the semantic segmentation of laparoscopic surgery images, placing special emphasis on the segmentation of structures with a smaller number of observations. As a result of this study, adjustment parameters are proposed for deep neural network architectures, enabling a robust segmentation of all structures in the surgical scene. The U-Net architecture with five encoder-decoders (U-Net5ed), SegNet-VGG19, and DeepLabv3+ employing different backbones are implemented. Three main experiments are conducted, working with Rectified Linear Unit (ReLU), Gaussian Error Linear Unit (GELU), and Swish activation functions. The applied loss functions include Cross Entropy (CE), Focal Loss (FL), Tversky Loss (TL), Dice Loss (DiL), Cross Entropy Dice Loss (CEDL), and Cross Entropy Tversky Loss (CETL). The performance of Stochastic Gradient Descent with momentum (SGDM) and Adaptive Moment Estimation (Adam) optimizers is compared. It is qualitatively and quantitatively confirmed that DeepLabv3+ and U-Net5ed architectures yield the best results. The DeepLabv3+ architecture with the ResNet-50 backbone, Swish activation function, and CETL loss function reports a Mean Accuracy (MAcc) of 0.976 and Mean Intersection over Union (MIoU) of 0.977. The semantic segmentation of structures with a smaller number of observations, such as the hepatic vein, cystic duct, Liver Ligament, and blood, verifies that the obtained results are very competitive and promising compared to the consulted literature. The proposed selected parameters were validated in the YOLOv9 architecture, which showed an improvement in semantic segmentation compared to the results obtained with the original architecture.

A piezoelectric and electromagnetic hybrid energy harvester based on two-stage magnetic coupling

Traditional piezoelectric energy harvesters cannot adapt to environmental energy with low frequency, wideband, and randomness because of the narrow working bandwidth. The introduction of magnetic coupling can improve the narrow-band characteristics of piezoelectric energy harvesting structures. To explore more efficient broadband methods, this paper designs a piezoelectric and electromagnetic hybrid energy harvester (PEEH) based on two-stage magnetic coupling to realize the expansion of piezoelectric energy harvesting bandwidth. In this paper, the magnetic pole alternating frequency is changed and the output of the structure is tested, and the operating frequency bands under different alternating frequencies are compared according to the test results. It can be seen from the experimental results that the output performance of the structure is obviously improved by increasing the AC frequency of the magnetic pole while reducing the optimal frequency point of the structure. When the number of magnets in the rotor magnet array is 6, increasing the pole alternating frequency can effectively improve the output performance of the electromagnetic unit when the rotor speed is low, and can also increase the maximum output voltage and energy harvest bandwidth of the piezoelectric unit by 189% and 150%, respectively. Therefore, by increasing the alternating frequency of magnetic poles, the output voltage of the structure can be improved, and the bandwidth of the structure can be broadened.

Scientific and Technical Basis for the Development of an Induction Heating Unit for Milk Pasteurization

Introduction. Induction heating is a preferred heating technique for industrial, medical and consumer systems, because it has a number of advantages over traditional heat transfer methods. The advantages include energy efficiency, heating rate, safety of operation, cleanliness of the process, low metal consumption, simple design, and precise control of the temperature of the heated raw materials. An induction heating unit is especially important for farms involved in processing of milk and producing milk-based products. Aim of the Study. The study is aimed at developing a prototype unit for long-term pasteurization of milk using a container heated by induction currents and at selecting optimal operating conditions for the developed prototype unit. Materials and Methods.There was used 3D modeling in the KOMPAS-3D computeraided design system to develop the main components of a milk pasteurization prototype unit with induction heating. The container for raw materials, stirrer and lid are made of stainless steel AISI 304 and AISI 430. The inductor is a frame made of polymer material with a litz wire arranged in a spiral manner. The body of the prototype unit is made of aluminum composite material. The developing and debugging of the electronic circuit of the prototype unit power part was carried out with the use of the design program Proteus 7.10. The microcontroller Mega 2560 was used to make the power part of the electromagnetic induction generation unit. The controlled temperature was monitored by using the waterproof temperature sensor DS18B20. A thermal imager was used to visualize the propagation of the thermal field over the surface of the container walls. Results. The structure diagram of the developed prototype unit with induction heating for long-term pasteurization of milk is presented. The article gives grounds to the use of the necessary elements and actuators in the unit for pasteurization of milk in a container heated by induction currents. There are presented a diagram of the developed power part for the prototype unit and the results of testing it when heating containers made of various materials. An algorithm has been developed to control the operation and PID regulation of the milk pasteurization in an experimental unit with the use the Raspberry Pi microcomputer. The graphs of transient processes when changing the coefficients of PID temperature control are presented. Discussion and Conclusion. When testing the induction heating principle on stainless steels of different compositions, it has been concluded that for the efficiency of heating the container, there is required a ferromagnetic steel pad welded on top of the main container made of food-grade stainless steel. The developed system of inductors made it possible to create a prototype unit with two heating zones depending on the volume of processed raw materials that is important for small farms engaged in processing milk and producing milk-based products.

Design and Analysis of a Moment of Inertia Adjustment Device

The vibration frequency characteristics of a rotor system are directly related to its moment of inertia. In this paper, a moment of inertia adjustment device is proposed to adjust the frequency characteristics of a rotor system and better reduce vibration by changing the moment of inertia. First, a mathematical model of the moment of inertia and the temperature field are established. A finite element simulation model of the electromagnetic field of the electromagnetic control unit in the device is established. The influence of current and air gap on the electromagnetic forces is discussed. Then, the validity of the finite element simulation for the electromagnetic control unit is verified using experimental results. In addition, the variations in the displacement and force of the moving mass and the moment of inertia of the device with speed are analyzed. The results show that the proposed moment of inertia adjustment device can be used to significantly adjust the moment of inertia, which provides a reference for better controlling vibrations in rotor systems. Finally, a finite element simulation model for an electromagnetic field analysis of the electromagnetic control unit in the device is established. The results show that the maximum temperature of the electromagnetic control device is 332 K in 60 min, which is in accordance with the requirements.

Electromagnetic shielding Janus membrane for direct current‐type pressure and sliding sensing

AbstractFacing increasing electromagnetic radiation pollution, soft sensors with electromagnetic interference (EMI) shielding is highly desirable. Here, flexible electromagnetic shielding [polyacrylonitrile/Fe3O4/polyaniline]//[multiwalled carbon nanotube/polyvinyl pyrrolidone] (denoted [PAN/Fe3O4/PANI]//[MWCNT/PVP]) Janus membrane is reported for pressure and sliding sensing. The [PAN/Fe3O4/PANI] layer of Janus membrane as the force and sliding sensing unit can generate direct current‐type (DC‐type) voltage signals by conducting polymer/metal Schottky junction, whereas [MWCNT/PVP] layer serves as main electromagnetic shielding unit. Under the vertical force and horizontal sliding of aluminum (Al) foil, it can output DC‐type voltage signals based on the Schottky contact. The performance of [PAN/Fe3O4/PANI]//[MWCNT/PVP] Janus membrane for pressure and sliding sensing is appropriately explored and assessed. Moreover, the Janus membrane exhibits excellent shielding effects with average EMI shielding effectiveness (SE) of 37.54 dB in 8.2–12.4 GHz (X‐Band) frequency ranges. Thus, based on the asymmetrical double‐layer structure, soft DC‐type sensor with EMI shielding is achieved, showing utilization potential in wearable electronics and intelligent robots.

A Lightweight convolutional medical segmentation algorithm based on ConvNeXt to improve UNet

In recent years, UNet and its derivative networks have gained widespread recognition as major methods of medical image segmentation. However, networks like UNet often struggle with Point-of-Care (POC) healthcare applications due to their high number of parameters and computational complexity. To tackle these challenges, this paper introduces an efficient network designed for medical image segmentation called MCU-Net, which leverages ConvNeXt to enhance UNet. 1) Based on ConvNeXt, MCU-Net proposes the MCU Block, which employs techniques such as large kernel convolution, depth-wise separable convolution, and an inverted bottleneck design. To ensure stable segmentation performance, it also integrates global response normalization (GRN) layers and Gaussian Error Linear Unit (GELU) activation functions. 2) Additionally, MCU-Net introduces an enhanced Multi-Scale Convolution Attention (MSCA) module after the original UNet’s skip connections, emphasizing medical image features and capturing semantic insights across multiple scales. 3)The downsampling process replaces pooling layers with convolutions, and both upsampling and downsampling stages incorporate batch normalization (BN) layers to enhance model stability during training. The experimental results demonstrate that MCU-Net, with a parameter count of 2.19 million and computational complexity of 19.73 FLOPs, outperforms other segmentation models. The overall performance of MCU-Net in medical image segmentation surpasses that of other models, achieving a Dice score of 91.8% and mIoU of 84.7% on the GlaS dataset. When compared to UNet on the BUSI dataset, MCU-Net shows an improvement of 2% in Dice and 2.9% in mIoU.

An Improved Detection Method for Crop & Fruit Leaf Disease under Real-Field Conditions

Using deep learning-based tools in the field of agriculture for the automatic detection of plant leaf diseases has been in place for many years. However, optimizing their use in the specific background of the agriculture field, in the presence of other leaves and the soil, is still an open challenge. This work presents a deep learning model based on YOLOv6s that incorporates (1) Gaussian error linear unit in the backbone, (2) efficient channel attention in the basic RepBlock, and (3) SCYLLA-Intersection Over Union (SIOU) loss function to improve the detection accuracy of the base model in real-field background conditions. Experiments were carried out on a self-collected dataset containing 3305 real-field images of cotton, wheat, and mango (healthy and diseased) leaves. The results show that the proposed model outperformed many state-of-the-art and recent models, including the base YOLOv6s, in terms of detection accuracy. It was also found that this improvement was achieved without any significant increase in the computational cost. Hence, the proposed model stood out as an effective technique to detect plant leaf diseases in real-field conditions without any increased computational burden.

YOLO-ECG: multi-stage ECG arrhythmia classification using deep learning based YOLO network for portable monitoring

Cardiovascular disease is the leading cause of death and more than half million people were died around the world. However, cardiovascular health monitoring is crucial for effective heart disease diagnosis and management. In this paper, a novel deep learning-based YOLO-ECG model is proposed to ECG arrhythmia classification method for portable monitoring. Initially, the ECG signals are gathered using 12-lead electrodes in the real time and these signals are denoised using two-dimensional stationary wavelet transform (2D-SWT). In SWT, zeros are inserted between filter taps rather than decimal points to eliminate repetitions and increase robustness. The denoised ECG signals are fed into the deep learning-based YOLO network with Gaussian error linear unit (GELU) activation function for detecting the ECG abnormalities of arrythmia. ECG waveforms are analyzed for the local fractal dimension at each sample point before heartbeat waveforms are extracted within a set length window. A squeeze and excitation attention (SEAN) module is introduced in the YOLO network for selecting size of 1D convolution kernel, and the dimension is preserved during local cross-channel interactions, decrease network complexity and enhance model efficiency. The classification findings demonstrate that the proposed YOLO-ECG model performs better by ECG recordings from the MIT-BIH arrhythmia dataset. From the experimental analysis, the proposed YOLO-ECG model yields the overall accuracy of 99.16% for efficient classification of arrythmia ECG signals.

Low-temperature sintering of bauxite raw material with alkali as an alternative to the parallel Bayer sintering process

The aim is to develop an alternative technology of bauxite raw material processing based on low-temperature sintering of bauxite with caustic alkali, as well as to solve the issue of carbon footprint control at alumina refineries in the Urals. Laboratory tests were carried out by sintering artificial bemite and hematite with chemically pure caustic alkali at temperatures of 300, 500 and 700°C and their further leaching in weakly alkaline solutions. To study the phase, chemical, and particle size distribution of red muds after leaching, various physical and chemical methods of analysis were used, such as X-ray fluorescence, titration method, X-ray phase analysis, scanning electron microscopy, magnetometry with a vibrating sample. The Brunauer – Emmett – Teller method was used to determine the specific surface area and porosity. The study of the kinetics of the solid-phase reaction of the bemite interaction with caustic alkali has shown the kinetic interaction in the temperature range under study. Moreover, sintering of hematite at temperatures of 300 and 500°C and further leaching of the sinter with water resulted in mineralogical changes in the sludge with the production of a new mineral, maghemite, which possesses magnetic properties. When studying the magnetic properties of red mud of lowtemperature sintering of bauxite, we determined that the magnetization was as high as 19–20 electromagnetic units per g (at a sample density of 2.38 g/cm3) at a magnetic field of 10 kE. The specific surface area of these samples was 54.97 and 51.77 m2/g. The performed studies confirm that the proposed technology can be adapted for bauxite to produce highiron red slimes, thus contributing to the integrated processing of bauxite raw materials. In addition, ways to reduce carbon emissions at alumina refineries by eliminating the sintering operation with soda and limestone, which is accompanied by CO2 emission during decomposition of these compounds, can be studied.