Journal of Electronic Research and Application

An Intelligent Recognition Method for Radar Comb Spectrum Jamming Based on Dual-Channel Deep Convolutional Network

Kuo Wang — 2026-03-26

This paper presents a deep learning method to recognize comb spectrum jamming in radar systems. Unlike traditional methods requiring manual feature extraction, our approach learns features directly from signal data. We built a dataset of radar echoes with four comb jamming types and five non-comb interference types. A dual-channel method creates 2D images preserving both magnitude and phase information from the signal spectrum. A CNN classifier with convolutional blocks, batch normalization, and dropout achieves 99.75% accuracy with 1.5% false alarm rate after only 7 training epochs.

Real-Time Electricity Price Prediction and Trading Signal Generation Using Ensemble Tree-Based Machine Learning Models: A Comparative Study on the Spanish Electricity Market

Yirui Liu — 2026-03-26

Accurate real-time electricity price forecasting is of critical importance for market participants seeking to optimize energy procurement, dispatch scheduling, and arbitrage strategies in liberalized electricity markets. However, existing forecasting approaches suffer from several key limitations: (1) conventional statistical models fail to capture the complex nonlinear interactions among generation mix, load demand, and temporal variables that collectively drive price dynamics; (2) single-model approaches lack robustness and are sensitive to overfitting, limiting their generalizability across diverse market conditions; (3) the interpretability of black-box prediction models remains insufficient, hindering the practical deployment of data-driven forecasting systems in operational decision-making. To address these challenges, this study proposes a comprehensive machine learning framework based on six tree-based ensemble models for hourly electricity price prediction in the Spanish electricity market. The proposed framework introduces three key contributions: (1) a systematic feature engineering pipeline incorporating lagged price variables, rolling statistics, and calendar-based temporal encodings; (2) a rigorous comparative evaluation of Decision Tree, Random Forest, Extra Trees, Gradient Boosting, XGBoost, and LightGBM under identical experimental conditions; (3) a SHAP-based interpretability analysis that quantifies feature contributions and interaction effects at both global and local levels. Experimental results on the ENTSO-E Spanish market dataset demonstrate that XGBoost achieves the best overall predictive performance, with an R² of 0.9660 and MAE of 1.5631 €/MWh.

Enhancing Tea Leaf Disease Classification with Cross- Attention Fusion and Magnitude-Aware Linear Attention

Jiaxin Zhu — 2026-04-10

Accurate tea leaf disease classification in real-world scenarios is hindered by complex backgrounds and the loss of fine-grained lesion details during CNN down sampling. To address this, we propose ResNet50-Dual-Fusion. It integrates a Cross-Attention Feature Fusion module (CAmodule) to adaptively reconstruct tiny lesion edges via cross-spatial interaction between shallow and deep features. Furthermore, a Magnitude-Aware Linear Attention (MALA) module with 2D Rotary Position Embedding (RoPE) is introduced to rectify magnitude neglect, effectively suppressing background noise. Evaluated on a 5,276-image dataset, our model achieves 85.96% accuracy (+3.00% over the baseline), outperforming architectures like ViT and Swin-Tiny. Grad-CAM visualizations confirm its superior lesion localization, providing a robust paradigm for automated crop disease diagnosis.

A Tibetan Speaker Verification Method Based on the Improved MFA-NConformer Model

Yitong Gong — 2026-04-24

MFA-conformer methods are widely used in English and Chinese speaker recognition. Theoretically language-independent but practically language-related, Tibetan speaker recognition currently relies on traditional models with poor performance. To address this, we adopt MFA-conformer as the basic framework and propose improvements: integrating 1D depth-wise separable convolution and channel attention into the conformer feed-forward network, fusing multi-block features, and adding an intra-class correlation regularizer to GE2E loss. Experiments show the improved model reduces the equal error rate (EER) compared with the conformer baseline.

IoT Security Situation Prediction Based on AGWOOptimized BiGRU-ATTN

Menghao Niu — 2026-04-22

To address the complexity and variability of Internet of Things (IoT) security situation prediction, this paper proposes an IoT security situation prediction model based on an improved Grey Wolf Optimizer (AGWO) optimized Bidirectional Gated Recurrent Unit with an attention mechanism (BiGRU-ATTN). Aiming at the shortcomings of the standard Grey Wolf Optimizer, such as slow convergence and susceptibility to local optima, the algorithm is enhanced through chaotic mapping-based population initialization, a nonlinear adaptive convergence factor, and a fitness-weighted position updating strategy, thereby improving the global search capability and convergence speed. Moreover, a BiGRU network is employed to capture complex temporal correlations in security situation sequences, while an attention mechanism dynamically assigns different weights to key features. Finally, the improved grey wolf optimizer is used to optimize the hyperparameters of the BiGRU-ATTN network. Experimental results demonstrate that, compared with traditional methods, the proposed model achieves superior fitting performance and faster convergence.

A Zero-Dynamics Attack Detection Method for Offshore Wind Power Systems

Kaige Chen — 2026-04-22

With the increase in the scale and complexity of offshore wind power systems, zero-dynamics attacks pose a severe threat to the cyber security of such systems. Their concealment makes them difficult to detect using traditional output observation-based methods. To address this problem, this paper proposes a zero-dynamics attack detection framework integrating adaptive watermarking and Kalman filtering, which achieves effective attack identification by embedding an adaptive watermark into the system input and conducting residual analysis. Simulation results show that the proposed method can quickly detect zero-dynamics attacks without affecting the normal operation of the system.

Physics Informed Hybrid Quantum-Classical Dispatching for LargeScale Renewable Power Systems: A Noise-Resilient Framework

Fu Zhang — 2026-04-22

Rising renewable penetration introduces severe non-convexity in power dispatching, straining classical optimization. While variational quantum algorithms (VQAs) on NISQ devices offer combinatorial potential, “black-box” approaches struggle with scalability and grid constraints. We propose the physics-informed hybrid quantum-classical dispatching (PI-HQCD) framework to address these limitations. PI-HQCD maps power flow and storage constraints directly into a topology-aware Hamiltonian, shrinking the search space. A noise-adaptive regularization technique bounds the objective’s Lipschitz constant, ensuring convergence under measurement noise. Experiments on IEEE 39-bus and 118-bus systems show PI-HQCD outperforms stochastic dual dynamic programming (SDDP) in cost and renewable utilization. Theoretical analysis confirms our topology-aligned ansatz achieves gradient variance scaling, mitigating barren plateaus. This work bridges physical laws and quantum algorithms for next-generation grid operations.

Research on Flow Field Calibration Method and Accuracy Improvement for the Aerodynamic Performance Test Rig of Aircraft Engine Compressors

Cheng Lu — 2026-04-22

The aerodynamic performance test rig for aircraft engine compressors serves as a core ground test facility for conducting aerodynamic design verification, performance evaluation, stability analysis, and flow mechanism research on compressors. The accuracy of its flow field measurement results directly determines the reliability of key conclusions, such as compressor characteristic curves, adiabatic efficiency, stability boundaries, and interstage matching relationships. In the context of developing a new generation of compressors with high load, high efficiency, and wide stability margins, traditional methods relying on empirical debugging and local calibration struggle to meet the requirements for high-precision aerodynamic testing. This paper takes an axial-flow compressor aerodynamic performance test rig as the research object, systematically elaborates on the typical process and mainstream methods of flow field calibration, analyzes the primary sources of error affecting measurement accuracy, and proposes an integrated strategy for improving accuracy from aspects such as probe calibration, flow rate calibration, flow field uniformity correction, installation and environmental compensation, and traceability of measurements. Furthermore, it provides a comparative analysis of calibration effects based on engineering test data. The research results indicate that through systematic flow field calibration and multi-dimensional error correction, the uncertainty in flow rate measurement can be reduced to better than ± 0.3%, with significant improvements in the measurement accuracy of total pressure and flow angle. The flow field non-uniformity is controlled within 3%, providing reliable data support for ground testing of high-performance compressors.

Sub-Pixel-Level Visual Inspection System for Dimensional Measurement of Ceramic Insulators Based on Halcon: Design and Implementation

Yuehua Cao — 2026-04-22

Aiming at the problems of low efficiency, poor accuracy consistency, and reliance on empirical judgment in the manual dimension inspection of ceramic insulators during the production process, a sub-pixel-level visual inspection system based on the Halcon platform was designed. Taking the 95-porcelain insulators with a 60 × 60 specification as the research object, a three-layer inspection architecture of “hardware acquisition–software processing–data output” was constructed. Through key technologies such as camera calibration, distortion correction, sub-pixel contour extraction, and template matching, the automatic measurement of three core dimensions of the insulator, namely height, width, and shed distance, was achieved. The experimental results show that the detection error of this system is controlled within the range of 0.5–1.2mm, the detection success rate reaches 99.2%, the detection time per sample is 2s, and the efficiency is 40% higher than that of traditional manual inspection. It can accurately meet the dimension inspection requirements of “GB/T 772-2005 Technical Conditions for Porcelain Insulators for High-voltage Overhead Lines”. This system requires no human intervention, and the detection results are stable and reliable. It provides an efficient solution for the on-line quality control in the production process of ceramic insulators and has important engineering application value.

Discussion on Data Privacy Protection Technologies in Cloud Computing Environment

Yixuan Dou — 2026-04-24

Cloud computing offers numerous benefits, including scalability, cost-effectiveness, and accessibility, making it an attractive solution for various organizations. However, the migration of sensitive data to cloud environments raises significant concerns regarding data privacy protection. This review paper provides a comprehensive overview of data privacy protection technologies in cloud computing. It begins by outlining the historical evolution of cloud computing and associated privacy challenges. The paper then delves into two core themes: access control mechanisms and data encryption techniques. Access control is explored in terms of attribute-based access control (ABAC), role-based access control (RBAC), and break-the-glass mechanisms. Encryption techniques are analyzed by covering homomorphic encryption, differential privacy and federated learning. The paper then offers a comparative analysis of these technologies, highlighting their strengths, weaknesses, and trade-offs in the cloud environment. Finally, the paper addresses the existing challenges and discusses future research directions, including the integration of artificial intelligence for enhanced privacy protection and the development of more robust and efficient encryption methods. This review aims to provide researchers and practitioners with a clear understanding of the current state-of-the-art in data privacy protection technologies for cloud computing and to identify potential avenues for future innovation.

Temporal-Spatial Evolution of Proton Beam Peak Energy and Its Correlation with Plasma Density

Lu Yang — 2026-04-22

In the fields of high-energy physics and particle acceleration, the peak energy of a proton beam is a core parameter for characterizing its energy properties. This paper presents a detailed discussion on the evolution of proton peak energy and its dependence on plasma density, combining theoretical research and simulations. The study integrates theoretical and simulation analyses to reveal that the peak energy of protons undergoes three distinct evolutionary stages: First, within a characteristic critical length, the variation in peak energy is independent of the channel density. Second, beyond this threshold length, the proton peak energy exhibits a rising trend over time, demonstrating a nearly linear increase with channel densities. Third, the proton peak energy does not increase indefinitely; it saturates before the protons reach the laser pulse front. Moreover, higher densities lead to earlier saturation of the peak energy. These findings provide an important foundation for future theoretical research on proton acceleration and the design of related experiments.

Research on Fault Location and Isolation Method of Power Distribution System Based on Intelligent Sensing

Yichi Zhang — 2026-04-22

Fault location and isolation in the power distribution system are the core links to ensure the reliability of power supply, and the traditional methods have problems such as insufficient positioning accuracy and slow isolation response in complex power grid structures. The introduction of intelligent sensing technology provides a new path for distribution network fault handling, and with the help of multi-source sensor data collection and deep integration of machine learning algorithms, the goal of accurate capture and rapid research and judgment of fault signals can be achieved. At the fault location level, a technical system including signal feature extraction, type recognition, multi-terminal fusion and single-phase grounding high-sensitivity positioning is constructed, and at the isolation level, adaptive criterion and distributed collaborative isolation scheme are proposed, which combines network reconstruction and multi-level protection coordination to improve power supply reliability. The simulation results show that the proposed method has better positioning accuracy and isolation speed, and has strong practical value in engineering applications.

Research on User Behavior Analysis Based on Big Data Technology

Yunzhe Dai — 2026-04-22

Big data technology refers to the ability to efficiently extract high-value information from multiple sources and massive amounts of data. It is an important achievement in the development of information technology and has significant application value in the field of user behavior analysis. Against the backdrop of rapid development of the digital economy and industry transformation, the role of e-commerce in the market system is increasingly prominent, and the scale of platform users continues to expand. In order to promote high-quality and sustainable development of the e-commerce industry, e-commerce platforms urgently need to use precise marketing methods to provide personalized products and services according to user needs, thereby improving user conversion rates and platform operating efficiency. This article takes e-commerce users as the research object. Firstly, it elaborates on the data characteristics and types of e-commerce user behavior. Secondly, it summarizes the relationship between big data and user behavior analysis, as well as the application value of big data technology in e-commerce user behavior analysis. Finally, it proposes scientific and effective application strategies, aiming to provide reference for e-commerce platforms to achieve accurate recommendations, optimize service strategies, enhance user experience and market competitiveness by mining user consumption preferences, potential needs and behavioral characteristics.

Practical Application of Low-Cost Visual Inspection Systems in Industrial Robot Integration

Chang Qi — 2026-04-24

In the transformation of industrial automation to smart manufacturing, visual inspection systems as critical sensing technologies are hindered by their high costs and algorithmic complexity, impeding the intelligent upgrading of small and medium-sized enterprises. This study focuses on low-cost visual inspection systems, enhancing performance through the selection of domestic industrial cameras, optimization of OpenCV and lightweight deep learning model algorithms, and the use of a C++ parallel computing framework, thereby constructing a solution that balances accuracy and cost. Experiments demonstrate that the system achieves sub-millimeter-level positioning and highly reliable detection in scenarios such as assembly guidance and defect identification, significantly reducing hardware costs while maintaining millisecond-level response capabilities, providing a feasible path for the intelligent upgrading of small and medium-sized enterprises.

Research on RF OTA Test System Optimization During the New Product Introduction Phase of Consumer Electronics

Peng Zhao — 2026-04-24

Under the trend of high integration and multi-band compatibility in consumer electronics, RF over air testing during the introduction stage of new products faces problems such as large space occupation, poor consistency, and environmental interference. The traditional broadband antenna coupling scheme is difficult to meet the accuracy and efficiency requirements of the production line due to volume redundancy, manual alignment errors, and shielding box attenuation effects. This article proposes an automated OTA testing optimization scheme based on miniaturized dual-frequency monopole antennas, which systematically solves testing pain points through compact MDMA antenna design, six-axis precision control model, and statistical process control verification method. The experiment shows that this scheme reduces the standard deviation of path loss by 40%, reduces the false alarm rate of RF desensitization from 12% to 2.5%, shortens the testing time of a single device to 8 seconds in practical applications, and increases the production yield by 9%, providing an efficient solution for high integration RF testing.

Analysis of Fracture Failure of Connecting Bolts for Diaphragm Pump Valve Box Cover

Xuan Qi — 2026-04-22

Aiming to address the repeated fracture of connecting bolts in the feed valve box cover of a large Type A diaphragm pump, a three-dimensional model of the valve box assembly was first established. Subsequently, bolt strength was evaluated using three finite element analysis methods: axisymmetric analysis, cyclic symmetry analysis, and full 3D integral analysis. It was clarified that the first few threads bear the maximum stress, which is consistent with the actual fracture position. Combined with the analysis results, it is concluded that the main causes of bolt fracture are alternating load bearing, low safety margin of specifications, and the coarse thread design also affects its load-bearing capacity.

Experimental Investigation of the Velocity Distribution in the Tail Flame of an Inductively- Coupled-Plasma

Xiaobao Mao — 2026-04-22

To achieve an accurate description of an Inductively Coupled Plasma (ICP) source, numerous two-dimensional and three-dimensional numerical models have been developed. However, experimental validation of these models remains a major challenge. Compared with plasma temperature and electron number density, the measured gas flow velocity distribution is more direct and reliable, and can serve as an important criterion for assessing model validity. In this work, the experimental method was improved by optimizing the observation parameters of a high-speed camera, effectively suppressing the interference from the intense emission spectra in the normal analytical zone. For the first time, ion clouds formed by injected particles were directly observed. Five types of suspended particles were sequentially introduced, including Er₂O₃, Y₂O₃, and borosilicate glass particles with diameters of 10, 5, and 2 μm. The particle-flow following behavior was comparatively evaluated. Using a tracer method, the gas flow velocity distribution in the ICP tail plume was measured. Furthermore, with Y₂O₃as the tracer, the axial gas velocity distribution in the central channel was systematically measured under different radio-frequency (RF) powers and carrier gas flow rates. The results show that within the range of 2.5 mm < z < 12.5 mm, the axial gas velocity in the central channel of the tail plume exhibits a distinct plateau region. The axial gas velocity increases with increasing RF power, while showing weak sensitivity to variations in carrier gas flow rate. The present study provides experimental data on the axial gas velocity distribution, offering essential validation and correction benchmarks for numerical ICP models.

Dynamics of Electric Field Perturbation in Gold Nanobipyramids During Dual-Pulse Two-Photon Coherent Excitation

Qiong Li — 2026-04-22

Optical-based microwave electric field detection has emerged as a research hotspot due to its advantages of high spatial resolution and immunity to electromagnetic interference. However, existing techniques are often limited by their sensitivity or reliance on specialized fluorescent materials. Gold nanobipyramids (AuNBPs), serving as nanoprobes with tip-enhancement effects and a well-defined three-level system, exhibit high sensitivity in their two-photon photoluminescence (TPPL) process to phase perturbations and plasmon resonance changes induced by microwave fields. By establishing a quantitative mapping model between microwave intensity and TPPL signal strength, we achieved an absolute measurement of microwave field strength with a spatial resolution that breaks the 100-nanometer barrier. Through comparative analysis of microwave responses under different pulse delays, we reveal that the microwave field primarily modulates TPPL intensity by interfering with the coherent excitation pathway. The most significant response of TPPL intensity to microwave power was observed near the zero-delay point, where the quantum coherence is strongest.

Comparison and Application Analysis of Three Wireless Charging Methods

Shuqi Wang — 2026-04-22

With the rapid development of the electronics industry, increasingly stringent requirements have been placed on battery endurance and power transfer efficiency (PTE). To meet the demands of modern high-technology society, numerous research teams have invested substantial efforts in wireless charging technologies. At present, wireless charging mainly includes electromagnetic induction-based wireless charging, magnetic resonant coupling-based wireless charging, and microwave-based wireless charging. By comparatively analyzing the operating principles of these three approaches, this paper summarizes their respective advantages and disadvantages. Electromagnetic induction-based wireless charging is highly constrained by transmission distance and is therefore suitable only for short-range power transfer. Magnetic resonant coupling-based wireless charging enables relatively longer transmission distances; however, it poses potential safety risks, as resonance may occur between the charging equipment and conductive objects in the surrounding environment under certain conditions. Microwave-based wireless charging is well-suited for radio-frequency wireless power transfer (WPT) in the microwave band. Through frequency-band adjustments, it can be extended to long-distance wireless power transfer across multiple bands. In the future, improvements in coil stability, transmitter frequency tuning, and bandwidth expansion may further enhance the power transfer efficiency and application potential of wireless charging technologies.

Research on Optimization of FPGA Streaming Processing System for High-Bandwidth Radar Echo Data

Zhonghao Jiang — 2026-04-22

For the real-time processing scenario of high-bandwidth radar echo data, an optimization scheme for a streaming processing system based on FPGA is proposed. Focusing on the engineering implementation requirements under conditions of high throughput, low latency, and resource constraints, the overall system architecture is designed, and optimizations are carried out in three aspects: pipeline parallel computing, cache organization and memory access scheduling, and timing convergence under resource constraints. The system adopts a modular streaming data path, builds a collaborative mechanism between on-chip cache and computing units, reduces data transfer overhead, and improves the continuity and stability of the processing link. Experimental results show that the optimized system can operate stably under higher input bandwidth conditions, with improved throughput capacity and real-time processing performance, compressed critical path, and resource utilization remaining within a reasonable range. This research can provide a reference for the engineering implementation of high-bandwidth radar signal processing platforms.

Practices and Insights of Scientific Data Security Grading Management Based on the Entire Life Cycle

Yu Zhai — 2026-04-22

In the era of big data, scientific data has become a strategic resource for national scientific and technological innovation and economic and social development, and the importance of its security management has become increasingly prominent. Based on the theory of the entire life cycle management of scientific data, this paper deeply discusses the core connotation of data security grading management, and systematically analyzes the prominent problems existing in the current scientific data security management in terms of system connection, process coverage, technology adaptation, and rights protection. On this basis, the paper constructs a practical path of scientific data security grading management covering six stages: data planning, collection, storage, use, sharing, and destruction, and puts forward targeted implementation strategies. Research shows that scientific data security grading management based on the entire life cycle is not only a technical issue but also a systematic project involving system design, organizational collaboration, and cultural cultivation. It has important theoretical value and practical enlightenment for improving data governance capabilities and promoting the orderly opening and sharing of scientific data.

Study on Performance Optimization of Radiation Protection Materials Based on Nanotechnology

Zhengyang Yuxiong — 2026-04-22

With the continuous development of nuclear energy development, medical treatment and aerospace, radiation protection materials are developing towards light weight, high efficiency and high stability. In the past, traditional protective materials have been difficult to adapt to the needs of complex working conditions. Based on the application of nano-technology radiation protection materials, this paper discusses the preparation technology and engineering regulation of nano-radiation protection materials, and explores the optimization path of the performance of radiation protection materials based on nano-technology, which provides a reference for R&D and application of high-performance nano-radiation protection materials.

Integrated Volt-Energy Storage-Lighting System for Smart Road Lighting Based on Distributed MPPT: A Review of Hardware Design and Economic Analysis

Xiangran Chen — 2026-04-22

Smart street lights are the key carrier of smart cities and dual carbon goals. Aiming at the problems of high energy consumption of street lamps, extensive control, lack of dynamic adaptation capabilities of existing smart street lamps, and the vulnerability of centralized photovoltaics to shadows and module aging, this paper proposes a distributed MPPT photovoltaic-energy storage-lighting integrated system. The system hardware integrates high-precision sampling, STM32 main control, DC-DC conversion and lithium iron phosphate battery management; The software integrates traffic flow monitoring, people flow statistics and multi-factor linear regression dimming algorithm to realize perception and dynamic dimming. Through the three-layer architecture of perception-processing-cloud, the system not only improves the efficiency of photovoltaic power generation and adapts to complex lighting, but also has been preliminarily verified to have significant technical feasibility and economic value in terms of power generation efficiency, energy saving and cost reduction, and full life cycle costs.

Shadow Thermodynamics of an AdS Black Hole in Non-Commutative Geometry

Ying Zhu — 2026-04-24

In this paper, we innovatively adopt the shadow radius to investigate the thermodynamics of an AdS black hole with non-commutative geometry terms. First, via geodesic analysis, we establish a quantitative relationship between the shadow radius and the event horizon radius, and derive the shadow radius of the black hole as a function of the event horizon radius, which exhibits a positive correlation between the two quantities. Furthermore, within the shadow framework, we find that the stability and heat capacity of the black hole can be effectively represented through the shadow radius. Further analysis reveals that the results obtained using the shadow radius in revealing the black hole phase transition process are essentially consistent with those obtained using the event horizon. Based on this, we constructed the thermal profile for an AdS black hole incorporating non-commutative parameters. Within the framework of non-commutative geometry, for P < Pc , the temperature derived from the shadow radius exhibits a distinct N-shaped trend, which is in perfect agreement with that obtained from the event horizon radius. This result reveals that even in non-commutative spacetime, the phase transition process of AdS black holes can be effectively and intuitively characterized by the thermal profiles of their shadows.