Condition Of Equipment Research Articles

Investigation of Abnormal Level Control Oscillations in a BWR Feedwater System

In the long-term operation of nuclear power plants, the aging of systems, structures, and components can lead to maintenance issues that must be dealt with to maintain cost-effective plant operations. One common issue affecting the currently operated boiling water reactors is the onset of unexpected level oscillations in feedwater heaters. This phenomenon can cause excessive cycling of drain valves and lead to premature failures. In this work, we develop a dynamic model of a set of feedwater heaters to determine the root cause of oscillations observed in an operating plant. Simulation results of various transient scenarios were used to investigate the effects of the controller parameters, boundary conditions, and possible valve and instrument issues. The analysis led to the conclusion that the most likely causes of the observed self-sustained oscillations in the system are the nonlinear behaviors of the drain valve and the level transmitter induced by degraded equipment condition. A partial plug of the pressure line used for level sensing in the system can account for a significant deadtime in the level transmitter, a nonlinear effect shown to induce self-sustained oscillatory behaviors.

Partial Discharge Data Augmentation and Pattern Recognition Method Based on DAE-GAN

Accurate identification of partial discharge (PD) and its types is essential for assessing the operating conditions of electrical equipment. To enhance PD pattern recognition under imbalanced and limited sample conditions, a method based on a Deep Autoencoder-embedded Generative Adversarial Network (DAE-GAN) is proposed. First, the Deep Autoencoder (DAE) is embedded within the Generative Adversarial Network (GAN) to improve the realism of generated samples. Then, complementary PD data samples are introduced during GAN training to address the issue of limited sample size. Lastly, the model’s discriminator is fine-tuned with augmented and balanced training data to enable PD pattern recognition. The DAE-GAN method is used to augment data and recognize patterns in experimental PD signals. The results demonstrate that, under imbalanced and small sample conditions, DAE-GAN generates more authentic PD samples with improved probability distribution fitting compared to other algorithms, leading to varying levels of enhancement in pattern recognition accuracy.

Исследование влияния отклонений параметров моделей синхронных генераторов на результаты расчета динамической устойчивости электроэнергетической системы

One of the important tasks solved in planning the electric power system (EPS) operation is the assessment of its transient stability. Use of incorrect parameters for EPS element models, in particular synchronous generators (SG), contributes to the occurrence of errors when determining the operating conditions of power systems, and, therefore, can lead to a decrease in the reliability of EPS operation. This article is devoted to the study of the impact of errors in SG model parameters on the transient stability assessment results. The analysis presented in this article allows us to substantiate the relevance of the SG parameter estimation task using a “passive experiment”, such as synchronized phasor measurements. This research is based on the theory of electrical circuits, mathematical description of electrical machines and transients in EPS, as well as statistical analysis. Computational experiments have been carried out in the software package MATLAB using the simulation modeling environment “Simulink”. During the research, two different multi-machine EPS models have been used, including the well-known “IEEE 9-bus” system. To quantitatively assess the influence of SG parameter uncertainties on transient stability analysis, the values of the maximum permissible fault clearance time are determined, as well as the values of the maximum power output of generating equipment in prefault conditions. The authors have carried out a comprehensive analysis of the influence of deviations in SG model parameters on the maximum fault clearance time, as well as on the maximum power output of generating equipment in the prefault conditions for a given fault duration. During the study, two approaches have been considered; in each case, the EPS transient stability assessment has been performed almost automatically, and quantitative indicators have been computed for the influence of SG model parameter deviations on the outcome of EPS transient stability analysis. Analysis of the computational experiment results has demonstrated a significant influence of the SG model errors on the errors in calculating certain transient stability-related parameters. For the considered scenarios, the spread of errors in calculating the maximum fault clearance time reached 100 % relative to a base-case (errorless) result, and the error in calculating the maximum power output of generating equipment in prefault conditions for a given fault duration may reach 20 % relative to the errorless result.

Negative impact of vibration on process pipelines of a compressor station with electrically driven gas pumping units

Relevance. In gas industry, the amount of equipment that has outlived its intended service life is increasing every year. In accordance with the law “On Industrial Safety of Hazardous Production Facilities,” compressor stations are dangerous objects, the reliable operation of which largely determines the condition of main gas pipelines. Technological pipelines of a compressor station are important objects of this system. They are subject to strict requirements due to the action of static and dynamic loads on them. Reliability of compressor stations is based on the results of a study of the condition of the object and the main factors affecting increased wear of equipment. The solution to wear problem is timely monitoring of equipment, in particular level measurement and determining the causes of pipeline vibrations using vibration diagnostics methods, and frequency and modal analysis methods. Aim. To study the causes of increased dynamic loads on the technological piping that arise during the operation of electric and gas pumping units at a compressor station using vibration monitoring methods. Methods. Vibration monitoring methods to analyze the causes of increased vibration values in the process piping of a compressor station with an electric gas pumping unit. Results and conclusions. The authors have measured vibration in a wide frequency band and assessed the technological condition of the main equipment of the compressor station with EGPA-6.3/8200-56/1.44-R based on the regulatory documentation STO Gazprom 2-2.3-324-2009. The main frequencies of the root-mean-square value of the vibration velocity, at which the highest vibration values were observed, were identified, and a modal analysis was performed using ANSYS software. Based on the results of natural studies and modal analysis, it was concluded about the occurrence of resonant high-frequency oscillations, multiple of the rotor rotation frequency.

An Extended Method Based on the Geometric Position of Salient Image Features: Solving the Dataset Imbalance Problem in Greenhouse Tomato Growing Scenarios

Machine vision has significant advantages in a wide range of agricultural applications; however, acquiring a large number of high-quality image resources is often challenging in actual agricultural production due to environmental and equipment conditions. Therefore, crop image augmentation techniques are particularly important in crop growth analysis. In this paper, greenhouse tomato plants were used as research subjects to collect images of their different fertility stages with flowers and fruits. Due to the different durations of each fertility period, there is a significant difference in the number of images collected. For this reason, this paper proposes a method for balanced amplification of significant feature information in images based on geometric position. Through the geometric position information of the target in the image, different segmentation strategies are used to process the image and supervised and unsupervised methods are applied to perform balanced augmentation of the image, which is combined with the YOLOv7 algorithm to verify the augmentation effect. In terms of the image dataset, the mixed image dataset (Mix) is supplemented with mobile phone images on top of in situ monitoring images, with precision increased from 70.33% to 82.81% and recall increased from 69.15% to 81.25%. In terms of image augmentation, after supervised balanced amplification, the detection accuracy is improved from 70.33% to 77.29%, which is suitable for supervised balanced amplification. For the mobile phone dataset (MP), after amplification, it was found that better results could be achieved without any amplification method. The detection accuracy of the mixed dataset with different data sources matching the appropriate amplification method increased slightly from 82.81% to 83.59%, and accurate detection could be achieved when the target was shaded by the plant, and in different environments and light conditions.

Calibration of a hybrid model for HVAC systems for fault data generation

The application of automated fault detection and diagnosis algorithms has proven to be an effective way to increase the energy efficiency of buildings. While data-driven strategies have demonstrated their potential in developing such algorithms, they require a set of historical data that represents both healthy and faulty equipment operation, which is not a common scenario in real facilities. This article presents a methodology for the generation of synthetic data on both healthy and faulty operation for building heating and air conditioning equipment, which is especially useful for supervision and failure detection in the case of replication and expansion of buildings in tertiary or residential building planning programmes. This work investigates the automatic calibration of a typical air-handling unit hybrid model, composed of the detailed parametric representation of the system components and its control sequence coupled with a simplified thermal load. The model is then extended with fault models that resemble commonly encountered component faults to generate operation data in faulty scenarios. The proposed framework is validated in a real use case with the calibration of an air-handling unit system and its control sequence using 15 days of data gathered from the building management system. The results obtained from the injection of seven typical faults are then compared with the fault-free simulations to highlight and validate their impact on key operating variables. The results prove the capabilities of this methodology to support database generation in the fields of fault detection and advanced maintenance of buildings’ air conditioning systems for building replicas.

Mass exodus? Police officer turnover intention and organizational commitment in New York City

PurposeThis study assesses the relationship between job satisfaction, perceived organizational support and workplace factors on officer turnover intention within an urban, municipal police organization.Design/methodology/approachUsing data from an online survey of New York City Police Officers (n = 1,823), both bivariate analysis and logistic regression models were utilized to assess the salience of police officer job satisfaction, perceived organizational support and perception of six workplace domains, including financial compensation, environmental factors, professional fulfillment, work/quality of life balance, treatment from management and occupational risk, on predicting turnover intention.FindingsThe cross-sectional study finds that job satisfaction, financial factors (salary, benefits and retirement benefits) and fulfillment predict lower levels of turnover intention (i.e. higher levels of organizational commitment). Work–life balance and environmental factors (cleanliness of work environment and condition of equipment) predict higher levels of turnover intention. Both perceptions of organizational support and occupational risk, while significant in the bivariate models, were not significantly associated after accounting for other factors. There is no evidence that officer perception of public support or the risk of being injured/killed at work were related to officer turnover intention.Research limitations/implicationsThe current study is limited by its focus on only one police department and its use of cross-sectional data, which may limit the generalizability of the results to agencies that differ in size and type and do not allow for assessment of causality.Practical implicationsOfficer turnover intention may be reduced by increasing financial compensation, improving the work environment and promoting a healthy work–life balance.Originality/valueThe study contributes to a growing body of research on police officer voluntary turnover by evaluating established predictors along with workplace factors in an urban police department: the setting where officer turnover intention is hypothesized to be the greatest.

Unsupervised anomaly detection in shearers via autoencoder networks and multi-scale correlation matrix reconstruction

As the main equipment of coal mining production, the anomaly detection of shearer is important to ensure production efficiency and coal mine safety. One key challenge lies in the limited or even absence of labeled monitoring data for the equipment, coupled with the high costs associated with manual annotation. Another challenge stems from the complex structure of the mining machines, making it difficult to reflect the overall operational state through local anomaly detection. Consequently, the application of decoupled local anomaly detection for mining machines in practical production remains challenging. This paper presents an unsupervised learning-based method for detecting anomalies in shearer. The method includes a module for constructing a Multi-scale Correlation Matrix (MSCM) of mining machine operating conditions, as well as the CNN-ConvLSTM Autoencoder (C-CLA) network. The module for constructing an MSCM enhances the representation of interrelationships between various features of the equipment from different perspectives using multiple correlation analysis methods. The C-CLA network integrates convolutional and convolutional recurrent neural networks, with the convolutional structure extracting local spatial features and the ConvLSTM structure further capturing information from different time scales and feature scales, thereby enhancing the model’s perceptual capabilities towards changes in equipment status. Finally, shearer anomaly detection is achieved through the analysis of reconstructed residual matrices. The rationality and practicality of the proposed method have been validated on our dataset, and the model’s generalization capability has been verified through repeated experiments in similar scenarios. However, due to variations in the working environment of different mining faces and differences in equipment models, implementing detection on other mining faces often requires retraining the model with new data. Furthermore, we compared our method with other anomaly detection techniques, and our detection efficiency was superior by approximately 3%. This method effectively detects anomalies in the shearer.

Research on Through-Flame Imaging Using Mid-Wave Infrared Camera Based on Flame Filter.

High-temperature furnaces and coal-fired boilers are widely employed in the petrochemical and metal-smelting sectors. Over time, the deterioration, corrosion, and wear of pipelines can lead to equipment malfunctions and safety incidents. Nevertheless, effective real-time monitoring of equipment conditions remains insufficient, primarily due to the interference caused by flames generated from fuel combustion. To address this issue, in this study, a through-flame infrared imager is developed based on the mid-wave infrared (MWIR) radiation characteristics of the flame. The imager incorporates a narrowband filter that operates within the wavelength range of 3.80 μm to 4.05 μm, which is integrated into conventional thermal imagers to perform flame filtering. This configuration enables the radiation from the background to pass through the flame and reach the detector, thereby allowing the infrared imager to visualize objects obscured by the flame and measure their temperatures directly. Our experimental findings indicate that the imager is capable of through-flame imaging; specifically, when the temperature of the target exceeds 50 °C, the imager can effectively penetrate the outer flame of an alcohol lamp and distinctly capture the target's outline. Importantly, as the temperature of the target increases, the clarity of the target's contour in the images improves. The MWIR through-flame imager presents considerable potential for the real-time monitoring and preventive maintenance of high-temperature furnaces and similar equipment, such as detecting the degradation of refractory materials and damage to pipelines.

Device-free crowd counting using multi-link WiFi CSI for occupant-driven energy management of HVAC systems

ABSTRACT Accurate estimation of the number of occupants in a building’s thermal zone is a crucial input for the dynamic estimation of the cooling load of a heating, ventilation, and air conditioning equipment. This data, in conjunction with an intelligent control algorithm, can be judiciously employed for enhancing occupant comfort and achieving substantial energy savings through demand-driven operation. In this work, we propose a WiFi channel state information (CSI) based passive people counting model that utilizes a wavelet-based signal denoising scheme and an artificial neural network (ANN) model for automatic feature extraction. The application of the proposed model on a public dataset shows more than 99% accuracy in detecting up to eight people in three rooms of varying sizes. The proposed model was also validated on CSI samples collected from a laboratory environment with five occupants, using ESP32, a low cost and low power IoT device. The results show that the proposed model can perform people counting with an accuracy of up to 98%, using single- and multi-link measurements.

Study on the Continuous Whole Process Preparation of Peptizable Pseudoboehmite by High Gravity Strengthening Reaction of CO2 and Heat Transfer.

The preparation of highly peptizable pseudoboehmite faces challenges such as slow reaction rate, prolonged aging times leading to poor peptization performance, and difficulty in achieving continuous production. In this study, a novel approach was proposed involving a high gravity reactor to enhance carbonation reaction control for pseudoboehmite nucleation, a high gravity steam heating process, and a continuous aging process in pipelines. By coupling these three processes, the continuous whole process production of highly peptizable pseudoboehmite under high gravity conditions was achieved. First, the effect of high gravity reactor enhancement on the reaction and the resulting solid phase was investigated. Second, the impact of aging temperature and time on the solid phase formed at different reaction end points was studied using high gravity and steam heating of the liquid phase. Furthermore, the influence of synthesis conditions on the peptization performance of pseudoboehmite was extensively examined. Finally, the nucleation and growth mechanisms of pseudoboehmite under high gravity conditions were analyzed. It was found that increasing high gravity factor, CO2 gas flow rate, and CO2 content accelerated carbonation reaction rate, promoting pseudoboehmite crystal nucleation. Increasing aging temperature and time facilitated the growth of pseudoboehmite nuclei and improved peptization performance. The high gravity device altered the gas-liquid phase contact state of traditional kettle-type equipment, reducing the reaction time and heating time from minutes to seconds, thus achieving the continuous whole process production of pseudoboehmite with a peptization index of 100% from sodium aluminate solution by kiln flue gas.

Neural network and layer-wise relevance propagation reveal how ice hockey protective equipment restricts players' motion.

Understanding the athlete's movements and the restrictions incurred by protective equipment is crucial for improving the equipment and subsequently, the athlete's performance. The task of equipment improvement is especially challenging in sports including advanced manoeuvres such as ice hockey and requires a holistic approach guiding the researcher's attention toward the right variables. The purposes of this study were (a) to quantify the effects of protective equipment in ice hockey on player's performance and (b) to identify the restrictions incurred by it. Twenty male hockey players performed four different drills with and without protective equipment while their performance was quantified. A neural network accompanied by layer-wise relevance propagation was applied to the 3D kinematic data to identify variables and time points that were most relevant for the neural network to distinguish between the equipment and no equipment condition, and therefore presumable result from restrictions incurred by the protective equipment. The study indicated that wearing the protective equipment, significantly reduced performance. Further, using the 3D kinematics, an artificial neural network could accurately distinguish between the movements performed with and without the equipment. The variables contributing the most to distinguishing between the equipment conditions were related to the upper extremities and movements in the sagittal plane. The presented methodology consisting of artificial neural networks and layer-wise relevance propagation contributed to insights without prior knowledge of how and to which extent joint angles are affected in complex maneuvers in ice hockey in the presence of protective equipment. It was shown that changes to the equipment should support the flexion movements of the knee and hip and should allow players to keep their upper extremities closer to the torso.

Imitational Modeling of SPP Electrical Equipment in Various Climatic Conditions

Imitational Modeling of SPP Electrical Equipment in Various Climatic Conditions

Green Management for Disasters Using an Inflatable Tent Prototype with a Solar Power Plant

Given the ever-increasing number of natural disasters, disaster management must be understood and implemented by all parties in a manner that includes green management and technology. This research begins with the problem of disaster preparedness management which currently uses less environmentally friendly facilities and infrastructure, including using tents with conventional structures and technological systems that are less flexible and require electricity with a gasoline-fueled generator. The aim of this research is to create a green technology-based disaster response facility in the form of an inflatable tent prototype for disaster preparedness using solar electricity. The method used in this study is experimental, designing and manufacturing prototype inflatable tents that use solar power plants. Tests are carried out on several variables: strength of tarpaulins, speed in construction and disassembly, comfort of air temperature, and performance of solar power generation. The results are of the experimental method are: the design and application of two prototype inflatable tents dimensions 6x9 square meters with a capacity of 20 patients, a 3,600 WP power plant using solar power, and a VRLA battery with a voltage capacity of 48 volts and a current of 200 Ah that is capable storing 4,800 watts of electricity. Thus, these devices meet electricity needs for air conditioning, lighting, and medical equipment in inflatable tents without using electricity from the state electricity company or gasoline generators. Keywords: disaster, inflatable tent, solar power generation

Current signal analysis using SW-GAT networks for fault diagnosis of electromechanical drive systems under extreme data imbalance

Abstract In complex industrial environments, ensuring the safe operation and effective maintenance of electromechanical equipment is of paramount importance. Intelligent fault diagnosis based on deep learning is currently the most popular data-driven method. However, conventional intelligent fault diagnosis techniques face several challenges: (1) Most diagnostic models rely heavily on analyzing vibration signals. However, vibration sensors are difficult to deploy in space-constrained environments, and vibration signals are frequently contaminated by strong noise. (2) The prevalence of class imbalance between normal and fault data in equipment condition monitoring can lead to model over-reliance on information from a few classes. (3) Traditional diagnostic models presuppose data independence, neglecting the coupling relationships between data. To address the aforementioned issue, this paper proposes a Self-Weighted Graph Attention Networks (SW-GAT) based on motor stator current signal analysis, aimed at solving the fault diagnosis problem of critical transmission components in electromechanical systems under severely imbalanced data scenarios. Firstly, the raw current data is preprocessed using Stacked Auto-Encoders (SAE), and then the decoded current frequency-domain data is utilized to construct graphical data, thereby enhancing the non-common features and weak fault information in the current signals. Secondly, by introducing the graph pooling attention mechanism into GAT, the model can more effectively focus on useful fault feature information within the graph data. Finally, a novel interclass adjustment loss function is designed to adaptively adjust and balance class weights, enabling the model to pay greater attention to minority class samples and thereby improving the recognition accuracy for minority class faults. Validating the proposed method on two cases and comparing it with other advanced approaches, our method achieved the highest accuracy among the compared methods.

Analysis of signals from air conditioner compressors with ordinal patterns and machine learning

Most machines are equipped with devices that monitor their operation. Air conditioners, in particular, are routinely monitored through various measurements. A desirable outcome of this monitoring is identifying when the device will likely require maintenance. In this study, we present the use of Ordinal Patterns, a symbolic transformation of time series, which allows for the visual assessment of the type of operation. Ordinal Patterns are chosen because they can transform intricate time series into simple and intuitive symbolic representations. The technique is visually appealing, generating points on a plane whose positions reveal hidden dynamics. This approach makes it easier to identify recurring or abnormal patterns in machine operations that may indicate wear or impending failure. Additionally, Ordinal Patterns allow for precise and understandable visualization of operational data, making interpreting results more accessible for professionals who may not be experts in data analysis. We compare two machines under different operational conditions with six measured variables. We analyze the expressiveness of the Ordinal Patterns and identify those variables that best differentiate the two machines. Furthermore, we incorporate machine learning algorithms, such as Artificial Neural Networks, Support Vector Machines, and Decision Trees, to evaluate and validate the effectiveness of Ordinal Patterns as discriminative features. Integrating machine learning methods with a symbolic transformation offers a robust approach for the early and accurate diagnosis of potential failures, enhancing the predictive maintenance of air conditioning equipment.

Planetary health in action: developing a heatwave vulnerability tool for primary care.

Heatwaves are becoming longer and more frequent. Despite the availability of open environmental data, little is operable and formatted for primary care use. Create a user-friendly online mapping tool to assess the vulnerability of communities to heatwaves for use by primary care practitioners. This study questioned what knowledge needed to be deployed, who needed to participate and how the knowledge should be shared. A participatory action-research project based on knowledge mobilization in France as part of the Green Data for Health Challenge. Knowledge was summarized on the factors most affecting heatwave vulnerability in a collaborative process, enabling a consensus on data variables and mobilised content for the online tool. Purposive sampling included primary care stakeholders with Regional Health Agencies (ARS), Public Health France, and data scientists. Nineteen participants participated in ten co-construction workshops, a brainstorming carousel strategy and five weekly co-design meetings between December 2022 and June 2023. The heatwave vulnerability variable was constructed using surface temperature, social deprivation, vegetation coverage, and presence of air conditioning equipment. Identified experts mobilized data on the national composite indicator at the communal level for heatwave morbidity. There is no standard platform for sharing environmental data in France. This co-creation study offers a new approach to incorporating environmental data on heatwaves into primary care consultations. We demonstrate the importance of knowledge mobilisation in primary care to bridge the research-practice gap. Integrating primary care records with environmental data may promote broader applications for planetary health research.

Research on Multi-Sensor Fusion Fault Diagnosis Method Based on Spatiotemporal Attention Mechanism

As industrial systems become increasingly complex, real-time monitoring and intelligent management of industrial equipment have become imperative. However, the limitations in coverage and accuracy of single sensors make it challenging to comprehensively characterize the operational state of equipment, leading to reduced system reliability and increased pressures on data transmission and storage. To address these challenges, this study presents a novel fault diagnosis method based on multi-sensor fusion using a spatio-temporal attention mechanism. Initially, one-dimensional convolutional neural networks (1D-CNN) are employed to extract features from raw signals, effectively capturing local characteristics and ensuring the integrity and validity of fault signals. Subsequently, the spatiotemporal attention mechanism adjusts the feature weights based on the temporal and spatial correlations of different sensors, as well as their respective importance, thereby capturing the spatio-temporal dependencies across multiple sensors and enhancing the efficacy of information fusion. Finally, the proposed method is validated through experiments on a nickel flash smelting furnace system. The results demonstrate that the method achieves a fault diagnosis accuracy exceeding 97.78%, significantly enhancing fault detection and decision-making performance.

ANALYSIS OF THE REGULATORY AND METHODOLOGICAL APPARATUS FOR CARRYING OUT TESTS OF ARMORED TANK AND VEHICLE EQUIPMENT TO THE INFLUENCE OF EXTERNAL MECHANICAL AND CLIMATE FACTORS

In the work is carried out an analytical study of normative and methodical support of the processes of testing armored and automobile vehicles to the influence of climatic and mechanical factors. Unsolved aspects of the testing are highlighted. The advantages and disadvantages of conducting tests of military equipment and its components on the influence of external mechanical and climatic factors have been identified for national standards of Ukraine and national standards harmonized with international standards in order to form the direction of further improvement of the regulatory and methodical apparatus to increase the accuracy and reliability of equipment evaluation in operating conditions. The need to improve test methods based on increasing the accuracy and reliability of equipment evaluation during laboratory tests is substantiated. The provisions of foreign military standards, which are used in Ukraine and the world as basic – STANAG 4370, MIL-STD-810 were analyzed. Based on the analysis of these standards, which include requirements for conducting tests, methods, procedures and acceptance criteria for equipment in operational conditions, the direction and method of improving test methods is determined. Prospective directions for ensuring standardization in the process of testing are outlined, in which there is a high need to maintain a high level of reliability and efficiency of armored and automobile vehicles in operational conditions. The need to adapt standards to modern challenges and threats has been established by optimizing test methods in conditions that are as close as possible to real ones, as well as using the latest technologies to improve the processes of monitoring the state of equipment.

Fault diagnosis of planetary gearboxes under small and imbalanced samples based on enhanced Siamese network with improved downsampling module

Due to the limitations of the operating conditions and equipment status, in practice, the normal data for planetary gearboxes are often sufficient, but the fault data are scarce and difficult to obtain, causing the fault diagnosis to face not only an imbalance problem but also a small sample problem. It leads to serious misdiagnosis and poor generalizability. To overcome above challenges, an enhanced Siamese network with improved downsampling module is proposed to reduce the negative impact of small and imbalanced datasets on fault diagnosis. Firstly, the Siamese network is combined with data enhancement to adaptively increase the fault training data with different small sample degrees. It can make the number of training data of each class get rid of small sample, while avoiding the excessive use of scarce fault data causing additional overfitting, so as to improve the generalization of the model. Among them, the data enhancement adopts the extreme noise expansion method proposed in this work, which can quickly generate samples that meet the requirements and help to improve the performance of the model in noisy environments. Secondly, adding the improved downsampling module to the Siamese network. The improved downsampling module introduces density-based spatial clustering of applications with noise (DBSCAN) and distribution ratio calculation, which can consider the distribution range and density of data when selecting. It can not only balance the training data, but also avoid serious information loss, and then reduce the probability of misdiagnosis. Finally, using six imbalanced planetary gearbox datasets with different small sample degrees for verification. The experimental results show that the proposed method can efficiently increase and balance the training data, which greatly improves the accuracy of fault diagnosis under the condition of small and imbalanced samples, especially when the small sample degree is very serious, the accuracy is improved by 57.76% on average.