Strong Self-adaptability Research Articles

Rolling Element Bearing Fault Diagnosis Based on Multi-objective Optimized Deep Auto-encoder

Abstract Bearing fault diagnosis holds significant importance, with widespread attention focused on enhancing its accuracy and efficiency. Existing diagnostic methods based on deep learning and transfer learning typically tackle this issue by introducing new function modules and diagnostic strategies, such as attention mechanism, adversarial domain adaptation, etc. However, most studies do not consider the structure and hyperparameters optimization of the network to improve the diagnostic performance of the network itself. To address this limitation, a novel multi-objective optimized deep auto-encoder (MODAE) is proposed in this paper. The optimal network structure and hyperparameters is determined by a multi-objective particle swarm optimization algorithm. Crucially, the method is based on a data-driven approaches to automatically search for network structures with stronger generalization and feature extraction capabilities to address engineering problems in different scenarios. Finally, this method is examined in both multi-fault classification diagnosis and transfer diagnosis scenarios, demonstrating strong self-adaptability through experimental results. In comparison with typical deep learning fault diagnosis methods, the proposed method demonstrates higher diagnostic accuracy and superior generalization ability.

Self-sustaining personal all-day thermoregulatory clothing using only sunlight

The human body must stay within a certain temperature range for comfort and safety. However, challenges for thermoregulatory clothing exist for harsh application scenarios, such as full day/night cycles, frigid polar regions, and space travel. We developed a flexible and sustainable personal thermoregulatory clothing system by integrating a flexible organic photovoltaic (OPV) module to directly acquire energy from sunlight and bidirectional electrocaloric (EC) devices. The flexible OPV-EC thermoregulatory clothing (OETC) can extend the human thermal comfort zone from 22°–28°C to 12.5°–37.6°C with a fast thermoregulation rate. The low energy consumption and high efficiency of the EC device allows for 24 hours of controllable and dual-mode thermoregulation with 12 hours of sunlight energy input. This self-powered wearable thermoregulatory platform has a simple structure, compact design, high efficiency, and strong self-adaptability with sunlight as the sole energy source.

Spatial structure and evolutionary logic of urban agglomerations based on remote sensing data

City clusters are important spatial carriers for countries to participate in global competition. The continuous monitoring of a city's status can be performed through sensors and processors applied within the real-world infrastructure. Based on the DMSP/OLS and NPP-VIIRS night-time light remote sensing data of the Beijing–Tianjin–Hebei (BTH) and Boswash (BW) urban agglomerations from 2000 to 2019, this study used the methods of dynamic threshold and spatial statistical SDE, with the support of GIS technology, to portray the spatial morphology and evolutionary logic of BTH and BW. The analysis used urban life entity theory. The results showed that BTH is in the middle stage of growth and development, with large differences in internal development, centered on the two major growth poles of Beijing and Tianjin. It is expanding in a circular wave-like pattern in all directions, characterized by fast metabolism and strong self-adaptation. BW is in the mature stage, with its overall development level well ahead of BTH's. The cities within are developed in a balanced manner, with a ribbon-like corrugated pattern in all directions, and its development is characterized by slow metabolism and strong self-adaptability.

MDFI regulates fast-to-slow muscle fiber type transformation via the calcium signaling pathway

Muscle fiber is the basic unit of skeletal muscle with strong self-adaptability, and its type is closely related to meat quality. Myod family inhibitor (Mdfi) has the function of regulating myogenic regulatory factors during cell differentiation, but how Mdfi regulates muscle fiber type transformation in myoblasts is still unclear. In the present study, we constructed overexpressing and interfering with Mdfi C2C12 cell models by lipofection. The immunofluorescence, quantitative real-time PCR (qPCR), and western blot results show that the elevated MDFI promoted mitochondrial biogenesis, aerobic metabolism and the calcium level by activating CaMKK2 and AMPK phosphorylation and then stimulated the conversion of C2C12 cells from fast glycolytic to slow oxidative type. In addition, after inhibiting IP3R and RYR channels, the higher MDFI reversed the blockage of calcium release from the endoplasmic reticulum by calcium channel receptor inhibitors and increased intracellular calcium levels. Therefore, we propose that the higher MDFI promotes muscle fiber types conversion through the calcium signaling pathway. These findings further broaden our understanding of the regulatory mechanism of MDFI in muscle fiber type transformation. Furthermore, our results suggest potential therapeutic targets for skeletal muscle and metabolic-related diseases.

Research on Sensorless Control of Permanent Magnet Synchronous Motor Based on Novel Sliding Mode Observer

In this paper, an improved SMO system is proposed to solve the problems of jitter and poor interference immunity of the conventional SMO system, the system is improved by replacing and adding three new simulation modules. Starting from the ideas of the “quasi-sliding module” and “boundary layer” of sliding mode control theory, by replacing the symbolic function in the conventional observer with an exponential segmentation function, which can effectively suppress the jitter phenomenon. And for different speed motor control application scenarios, a variable sliding mode gain coefficient that can be dynamically changed with the motor speed is used. For the system’s anti-jitter performance to be improved, a variable sliding mode gain coefficient that can be dynamically changed with the motor speed is introduced, which has stronger self-adaptability in different speed scenarios. An improved SMO system is formed by adding three new and improved modules, and this basis is tested experimentally on the MATLAB/Simulink simulation platform. The SMO with the addition of three new modules can be tested by simulation experiments to know that it has stronger anti-disturbance and self-adaptive capabilities under change scenarios such as sudden changes in load torque and sudden changes in speed.

Value conflicts in grassroots environmental management from a network perspective: a case study of crop residue management in Harbin, China.

Open burning of crop residue is a hot issue in Asia and has attracted widespread attention. However, this attention rarely extends to the complex interactions between multi-stakeholders in the governance process, which is precisely the focus of today's environmental governance dilemma. Harbin is a major grain-producing area in China, the annual air pollution caused by the open burning of crop straw is more prominent than in other parts of China, and the conflicting relationships among multi-stakeholders are also typical. Taking Harbin as a case, this study quantifies the complex relationships among stakeholders through value demands conflicts and constructs a value conflict network in the context of straw governance. Through the analysis of the network nodes and relationships, we found that grassroots governments and farmers are the core of the conflict, while public and higher-level governments, as supervisory subjects, are marginalized. The multiple identities and value demands of the grassroots government, as well as cost and technology constraints, are the main reasons for the governance dilemma. In addition, the grassroots government in different scenario dimensions has different conflict resolution strategies, and it has a strong self-adaptation ability in the embedded value conflict network and can influence and reshape other stakeholders. These findings highlight the critical role of the grassroots government in crop residue governance, add to the research paradigm on grassroots environmental management from a multiple-stakeholder participation perspective, and provide a theoretical and methodological basis to formulate effective strategies.

Methods to Improve the Efficiency of Rural Physical Education Teaching Resources Allocation and Utilization in the Context of Artificial Intelligence.

School physical education is an important part of school education, and physical education resources are the overall measure of a school's physical education level. Rural schools are limited by environmental constraints, especially the lack of physical education resources verification. Therefore, the allocation and utilization efficiency improvement of rural physical education resources are an important element of rural physical education construction. With the mature development of artificial intelligence technology, it has great potential for application in the field of cultural education. Artificial intelligence technology has the advantages of ease of use, strong self-adaptation, and excellent generalization ability, and it is selected for data processing in this paper. Using artificial intelligence technology, the study of sports informatization to help rural sports development and rational utilization of resources can help rural sports education development, rural sports industry prosperity, rural sports talent revitalization, rural sports governance system improvement, and governance capacity enhancement. The artificial intelligence raindrop algorithm, which can automate the statistics of all relevant factors and derive results based on data fusion, was selected as the core algorithm of this paper. Therefore, this paper proposes an intelligent algorithm for allocating and improving the efficiency of rural sports teaching resources so as to improve the informatization level of the rural sports industry, promote industrial integration based on local needs, accelerate the “online urban-rural integration” in rural sports education, rural fitness guidance, and rural recreation and medical care, and rely on multiple entities. We will support the new model of sports informatization to empower rural sports teaching development by relying on the collaboration of multiple entities.

Research on Fault Diagnosis Method of Rolling Bearing Based on Feature Optimization and Self-Adaptive SVM

Aiming at the problems of the difficulty of extracting the fault sensitive features in the mixed domain of bearings under complex working conditions and the lack of self-adaptability of support vector machine (SVM) parameters, to solve these problems, we investigate a diagnosis method combining optimal feature selection and self-adaptive SVM. Firstly, the optimal feature space through multicluster feature selection (MCFS) is constructed. Then, the differential evolution (DE) strategy is used to improve the search performance of gray wolf optimizer (GWO) to improve the diagnosis accuracy of SVM and make it strong self-adaptability. The experimental results show that, with a small number of sensitive features containing significant category differences, the proposed method not only guarantees the construction of the optimal feature space under the minimum feature dimension, but also greatly improves the accuracy of fault recognition. Simultaneously, compared with the traditional feature reduction method (PCA), the proposed MCFS combined diagnosis model improves the accuracy from 98.5% to 100% with 1.5%. Concurrently, compared with MPA-SVM, GWO-SVM, and PSO-SVM, the convergence performance is improved by 61.54%, 78.26%, and 92.64%, respectively. It can seek the best classification performance in the shortest time. The effectiveness and superiority of the proposed method are fully verified.

Application of Multilayer Neural Network in Sports Psychology

The research on multilayer neural network theory has developed rapidly in recent years. It has parallel processing capabilities and fault tolerance and has aroused the interest of many researchers. The neural network has made great progress in the field of control, especially in model identification and control. It has been quickly applied in the fields of device design, optimized operation, and fault analysis and diagnosis. Neural network control, as an automated control technology in the 21st century, has been fully proved by theories and practices at home and abroad, and it is very useful in complex process control. Sports psychology is a discipline that studies the psychological characteristics and laws of people engaged in sports, and it is also a new development in sports science. The main task of sports psychology is to study people’s psychological processes when participating in sports, such as feeling, perception, appearance, thinking, memory, emotion, and characteristics of will and its role and significance in sports. An important feature of multilayer neural networks is to achieve results that match the expected output through network learning. It has strong self-learning, self-adaptability, and fault tolerance. The multilayer neural network system evaluation method is unique with its extraordinary ability to deal with complex nonlinear problems and does not involve human intervention. This article presents a multilayer neural network algorithm, which classifies the samples of athletes, and studies the physical education training process, the psychological characteristics of related personnel in sports competitions, such as the psychological characteristics of the formation of sports skills, and the psychological training of athletes before the game.

Research on Batch Evaluation of Real Estate Price Based on XGBoost

Nowadays, there is an upsurge of research in the field of artificial intelligence around the world. As artificial intelligence slowly penetrates and is applied to all walks of life, the application of artificial intelligence to the field of real estate price batch evaluation has gradually become a reality. As the core area of artificial intelligence-machine learning, it has strong self-organization, self-adaptation and learning capabilities, and is very good at dealing with nonlinear mapping relations. It is a good choice to apply the ideas and methods of artificial intelligence to the field of real estate price batch evaluation.This paper uses Python web crawler technology to obtain Nanjing's second-hand housing transaction information data set from Shell.com, and cleans the second-hand housing transaction information data set to obtain a standardized data set. Then data processing is performed on the standardized data set, which mainly includes the quantitative processing of sub-type variables to obtain a standardized sample data set. Next, we build a real estate price batch evaluation model based on XGBoost, tune the main parameters of the regression model, and perform 5-fold cross-validation. Finally, we compare and analyze the XGboost regression model with the random forest regression model and the multiple regression model, and find that the XGBoost regression model has considerable advantages compared with the traditional multiple regression model, and is more suitable for building real estate price batch evaluation models; moreover, The real estate price batch evaluation model based on XGBoost is slightly better than the real estate price batch evaluation model based on random forest, and the regression prediction results are more robust.

Short-term Photovoltaic Power Prediction based on Sparse Representation Method

In the research of solar power prediction, providing accurate prediction data in real time is one of the most effective means to enhance the capacity of wind power acceptance and improve the power reliability and economy. The existing prediction models based on statistical methods are often unavoidable in data preprocessing and model training stage, and their adaptive ability needs to be improved. Considering that the sparse coding method does not require model training, and has the characteristics of high solving efficiency and strong self-adaptability, an online solar energy prediction model using sparse coding is proposed.Firstly, the historical time series data is composed of input-output pairs with delay, and the dictionary is respectively constructed in atomic form. Then, the sparse weight is calculated for the delay input data vector to be predicted, and the corresponding predicted output is obtained by borrowing the dictionary. Taking the actual solar power data of Alberta, Canada as sample, the simulation was carried out in MATLAB. The simulation results show that the model can accurately predict the solar power and improve the effectiveness and practicability of the prediction

Modeling of the HL-2A plasma vertical displacement control system based on deep learning and its controller design

The modeling and control of the plasma equilibrium response is still one of the more important research areas in tokamak discharge experiments. Although theoretically, first principles can predict the plasma instability, how to build a physical model for accurate prediction is still a challenging problem. Therefore, a deep learning method is proposed to model the plasma vertical displacement system in the HL-2A tokamak experiment, whose method expands the modeling strategy for tokamak plasma control systems. Through the training of a large number of high-dimensional experimental data, the obtained deep neural network model in this paper has a higher precision prediction ability. Additionally, to illustrate the significance of the predictive model in controller design, a data-driven adaptive control algorithm is proposed to replace the traditional proportional-integral-derivative control algorithm for controlling the vertical displacement of plasma. The simulation results showed that the proposed algorithm had less adjustable parameters, strong self-adaptability, and effective control for the HL-2A plasma vertical displacement.

Design method for a bionic wrist based on tensegrity structures

The traditional bionic upper limb structure design is limited by the motion pair and cannot guarantee the flexibility of the mechanical structure. The tensegrity structure has the characteristics of high deformability, strong self-adaptability, and resistance to multi-directional impact. According to the biological characteristics of the upper limbs of the human body, an anatomical study is performed on the upper limb wrist joints that achieve adduction/abduction, flexion/extension, to obtain the relationship between the movements of the related bones and muscles, and to simplify the shape and structure of the wrist. Equivalent mapping of a mechanical model based on two-bar tensile properties. Through the contraction and stretching of the spring, the movement characteristics of the human muscles are realised, and the optimised bionic upper limb wrist tensioning robot without motion pair is further obtained. Adams simulation is used to verify that the bionic tensile wrist can simulate the change movement of the human wrist. The experimental platform was built and a physical prototype was made and the prototype was tested. The results show that the bionic tensile wrist can realise the adaptive motion characteristics of the human wrist well and stably, which proves the validity and feasibility of this design method.

Fault Detection Based on the Generalized S-Transform With a Variable Factor for Resonant Grounding Distribution Networks

Fault detection in resonant grounding (RG) distribution networks remains a challenge due to weak fault signals, extremely complex fault conditions, and unstable intermittent arc grounding faults. This paper addresses this issue by applying generalized S-transform (GST) with a variable factor to conduct denoising of transient zero-sequence currents based on threshold filtering followed by time-frequency distribution filtering in sequence. Meanwhile, this paper proposes a comprehensive multi-criteria faulty feeder detection method based on the transient zero-sequence current polarity (criterion 1), the energy relative entropy (criterion 2), and the total transient current energy (criterion 3). Here, criteria 2 and 3 are based on the time-frequency representation of the GST. The performances of the proposed denoising and faulty feeder detection methods are evaluated under single line to ground faults based on simulations conducted using a modeled 10 kV RG networks with overhead and cable mixed lines in addition to reasonably sophisticated permanent and intermittent arc discharge models to ensure that the simulations faithfully represent actual complex working conditions. Compared with existing method, simulation experiments and field test show that the method proposed in this paper provide a better denoising effect with stronger self-adaptability, higher detection accuracy, and a faster calculation speed.

A Bregman adaptive sparse-spike deconvolution method in the frequency domain

To improve the anti-noise performance of the time-domain Bregman iterative algorithm, an adaptive frequency-domain Bregman sparse-spike deconvolution algorithm is proposed. By solving the Bregman algorithm in the frequency domain, the influence of Gaussian as well as outlier noise on the convergence of the algorithm is effectively avoided. In other words, the proposed algorithm avoids data noise effects by implementing the calculations in the frequency domain. Moreover, the computational efficiency is greatly improved compared with the conventional method. Generalized cross validation is introduced in the solving process to optimize the regularization parameter and thus the algorithm is equipped with strong self-adaptation. Different theoretical models are built and solved using the algorithms in both time and frequency domains. Finally, the proposed and the conventional methods are both used to process actual seismic data. The comparison of the results confirms the superiority of the proposed algorithm due to its noise resistance and self-adaptation capability.

Brain Tumor Segmentation from Multi-modality MRI Data Based on Tamura Texture

A segmentation algorithm of brain tumor MR image based on Tamura texture feature and BP Neural Network is proposed in this paper. Firstly, the local grayscale features of four modal MR images are combined with the Tamura texture metrics in the algorithm. The information in the image is extracted as much as possible. Then, the known samples are input into the BP Neural Network and classifier training is performed. Finally, other brain tumor images are processed with the trained BP Neural Network. The experiment was performed on the images of 30 patients. From the obtained data, the method proposed in this paper can segment the brain tumor region accurately and effectively and show strong self-adaptability to the difference of the brain tumor images.

Detecting Technology of Current Generator Blade Shape Based on Improved Canny Algorithm

Aiming at the shortcomings of the image excessively smoothing and the threshold value need to be determined in advance when the Canny algorithm is used for edge detection, an improved Canny algorithm is proposed to detect the blade chord length of the turbine generator. Selecting a proper Gaussian filter coefficient to smooth the original image and remove white noise effects while maintaining the edge information. The image gradient histogram after non-maximum suppression is processed with the Otsu method (the maximum between-class variance method) to adaptively set the high and low threshold. Using the improved canny edge detection method to detect the blade chord length and verify the modified blade. The results show that the hydrodynamic performance of the modified blade has been significantly improved, indicating that the improved canny edge detection algorithm can accurately detect the real edge of the blade, and has a strong self-adaptability. Its detection results are better than the traditional Canny algorithm.

An improved reinforcement-immune algorithm for agricultural resource allocation optimization

Due to the decreasing and limited agricultural resources, developing techniques for agricultural resource allocation optimization has received increasing attention in recent years. In this paper, an improved reinforcement-immune algorithm (RIA) is proposed to realize an efficient and intelligent optimization in agricultural resource allocation and routing problem of the delivery vehicles. By combining the strong self-adaptability and goal-driven performance of reinforcement learning with the antibody diversity and strong global search capability of immune optimization, RIA has better performance not only on the convergence property but also on the effectiveness of finding the optimal solution. First, we introduce Table Q as the action policy. Each element in Table Q means the action score that we choose delivery vehicle v to allocate resource t of supply center s to field i. Table Q is initialized according to the distance between the field and the supply center. Second, we update Table Q to learn good genetic information according to the optimal antibody after each iteration. Each antibody represents an allocation scheme. Third, at the stage of antibody mutation, we use Table Q to guide antibodies, which makes the algorithm adaptive and goal-driven. Finally, simulation results illustrate that the proposed algorithm is effective in improving the climbing performance, searching ability and finding the optimal solution.

Neuron adaptive control of induction motor drive system

Performance optimisation of AC motor variable frequency speed regulating system is studied in this paper. Neuron adaptive controller is presented to further improve the performance of asynchronous motor variable frequency speed regulating system when it works in V/F mode. Based on the analysis of structure and principle of neuron PID (Proportion-Integration-Differentiation) controller, the supervisory Hebb study rule is combined with adaptive PSD (Proportion, Sum and Differentiation) algorithm, which enhances the self-study ability of single neuron PID controller and makes the neuron PID controller gain-adaptive, realising parameter optimisation and self-tuning online. The variable frequency speed regulation control performances were tested on the experimental platform. The results show that, compared with other adaptive controllers, induction motor speed regulating system, which is controlled by neuron adaptive PSD, improves the dynamic performance and steady-state accuracy and has stronger self-adaptability and robustness. Therefore, the performance of speed regulating has been optimised.

The Fault Diagnosis of Electric Power Metering System Based on Momentum BP Neural Network

Momentum BP neural network is used to diagnose the fault of high voltage electric power metering system according to the advantage of strong self-adaptation. First of all, the information of 7 operation parameters of metering system is extracted as input values of the neural network, secondly, we collect samples to train and test BP network and use 4-bit binary number to represent the normal circumstance and 7 main fault types, The results show that the momentum BP network model is effective in the fault diagnosis of high voltage electric power metering system.