Problem Of Poor Adaptability Research Articles

An innovative approach to detecting the freshness of fruits and vegetables through the integration of convolutional neural networks and bidirectional long short-term memory network

Fruit and vegetable freshness testing can improve the efficiency of agricultural product management, reduce resource waste and economic losses, and plays a vital role in increasing the added value of fruit and vegetable agricultural products. At present, the detection of fruit and vegetable freshness mainly relies on manual feature extraction combined with machine learning. However, manual extraction of features has the problem of poor adaptability, resulting in low efficiency in fruit and vegetable freshness detection. Although exist some studies that have introduced deep learning methods to automatically learn deep features that characterize the freshness of fruits and vegetables to cope with the diversity and variability in complex scenes. However, the performance of these studies on fruit and vegetable freshness detection needs to be further improved. Based on this, this paper proposes a novel method that fusion of different deep learning models to extract the features of fruit and vegetable images and the correlation between various areas in the image, so as to detect the freshness of fruits and vegetables more objectively and accurately. First, the image size in the dataset is resized to meet the input requirements of the deep learning model. Then, deep features characterizing the freshness of fruits and vegetables are extracted by the fused deep learning model. Finally, the parameters of the fusion model were optimized based on the detection performance of the fused deep learning model, and the performance of fruit and vegetable freshness detection was evaluated. Experimental results show that the CNN_BiLSTM deep learning model, which fusion convolutional neural network (CNN) and bidirectional long-short term memory neural network (BiLSTM), is combined with parameter optimization processing to achieve an accuracy of 97.76% in detecting the freshness of fruits and vegetables. The research results show that this method is promising to improve the performance of fruit and vegetable freshness detection.

Observation and research of deep underground multi-physical fields—Huainan −848 m deep experiment

Compared with the surface, the deep environment has the advantages of allowing “super-quiet and ultra-clean”-geophysical field observation with low vibration noise and little electromagnetic interference, which are conducive to therealization of long-term and high-precision observation of multi-physical fields, thus enabling the solution of a series of geoscience problems. In the Panyidong Coal Mine, where there are extensive underground tunnels at the depth of 848 m belowsea level, we carried out the first deep-underground geophysical observations, including radioactivity, gravity, magnetic, magne-totelluric, background vibration and six-component seismic observations. We concluded from these measurements that (1) the background of deep subsurface gravity noise in the long-period frequency band less than 2 Hz is nearly two orders ofmagnitude weaker than that in the surface observation environment; (2) the underground electric field is obviously weaker thanthe surface electric field, and the relatively high frequency of the underground field, greater than 1 Hz, is more than two orders of magnitude weaker than that of the surface electric field; the east-west magnetic field underground is approximately the same asthat at the surface; the relatively high-frequency north-south magnetic field underground, below 10 Hz, is at least one order ofmagnitude lower than that at the surface, showing that the underground has a clean electromagnetic environment; (3) in additionto the high-frequency and single-frequency noises introduced by underground human activities, the deep underground spacehas a sig-nificantly lower background vibration noise than the surface, which is very beneficial to the detection of weakearthquake and gravity signals; and (4) the underground roadway support system built with ferromagnetic material interferesthe geomagnetic field. We also found that for deep observation in the “ultra-quiet and ultra-clean” environment, the existinggeophysical equipment and observation technology have problems of poor adaptability and insufficient precision as well asdata cleaning problems, such as the effective separation of the signal and noise of deep observation data. It is also urgent tointerpret and comprehensively utilize these high-precision multi-physics observation data.

A Method for Detecting Amplitude-Phase Joint Characteristic Parameters of Wireless Channel for Generating Key Parameters

Aiming at the problems of poor adaptability and low secret key generation rate of secret key generation scheme based on single characteristic parameter of wireless channel, a secret key generation method based on joint characteristic parameters of amplitude and phase of wireless channel is proposed. This method is based on the single-eavesdropping wireless fading channel model; the joint characteristic model of amplitude and phase of wireless fading channel is established; it detects and extracts the joint characteristic parameters of amplitude and phase of channel, and then the proposed characteristics are quantified by using the equal probability joint quantization strategy of amplitude and phase to generate the secret key random parameters. In this paper, the amplitude and phase joint feature parameter detection method of wireless channel can not only improve the generation rate of random secret key parameters but also make the eavesdropping party’s eavesdropping error rate closer to 0.5. The test results show that the proposed scheme can significantly improve the rate, reliability, and security of generating key random parameters.

Study on Clamping Mechanism of Internal and External Variable Diameter Lifting Tool for Offshore Foundation Pile

Large marine foundation piles are an important part of offshore structural pile foundations, and their lifting operations have always been a major problem in the construction and construction of marine structures. Based on IHC’s bilateral marine foundation pile spreader, this paper proposes a structural scheme of “internal and external clamping type variable diameter marine foundation pile spreader”. It solves the problem of poor adaptability of spreaders to foundation piles of the same specification and different pipe diameters. At the same time, this article has conducted in-depth research on the two clamping methods of friction clamping and wedge tooth embedded clamping. Through experiments, it is found that under the same lateral load, the load capacity of the wedge teeth tightening is three times that of the friction clamping. Aiming at the embedding and clamping method of the wedge teeth of the spreader, first of all, the influence of the tooth profile angle of the wedge teeth on their embedding performance was studied by the plastic mechanics slip line field theory and Abaqus simulation analysis. Subsequently, the elastic mechanics theory and Abaqus simulation analysis were used to study the stress characteristics of the wedge teeth during the lifting process, and the internal stress distribution was obtained. The article aims to provide a reference for the design of spreaders in actual projects.

Nanoparticle foaming agents for major gas fields in China

The conventional foaming agents have the problems of poor adaptability and high cost during the application in different types of gas fields, especially in high temperature, high salinity, high acidic gas and high condensate oil and gas fields. In this study, the Gemini foaming agent was used as the main agent to enhance foaming and foam stability of the foaming agent, the grafted nanoparticles were used as foam stabilizer to further improve the foam stability, and the characteristic auxiliaries were added to make the foaming agent suitable for different types of gas reservoirs. Two types and six subtypes of nanoparticle foaming agents have been prepared for the main gas fields of China. The experimental evaluation results show that the overall temperature resistance, salinity resistance, H2S resistance, CO2 resistance and condensate resistance of the nanoparticle foaming agents can reach 160°C, 250 000 mg/L, 100 mg/L, 100% and 40%, respectively. The new foaming agents have been used in 8685 wells in China. Compared with conventional foaming agent, the average gas flow rate per well increased by 62.48%, the pressure difference (casing-tubing) decreased by 18.9%, and the cost dropped by 45%. The effect of reducing cost and increasing efficiency is obvious.

A new data preprocessing method for 3D reconstruction of pavement

ABSTRACT Three-dimensional (3D) reconstruction of pavement surface based on the laser scanning can be used to extract the elevation information for evaluating the damage degree of the pavement. This technology is of great significance to road safety as well as the advances of smart transportation. The preprocessing procedures of the laser-scanned point cloud serve as the basis for accurate surface reconstruction, which includes outlier removal, smoothing and sampling. In view of the limitations of the existing preprocessing methods for pavement applications, this paper proposes a new preprocessing method. First, a local density-based outlier removal algorithm is used to solve the problem of poor adaptability or long computation time of the regular outlier removal algorithms. And then, a continuous algorithm based on the moving least squares method is used to smooth and sample the points, solving the problem of repeated degradation of the data. Finally, the field tests on three types of pavements are carried out to verify the feasibility of the proposed preprocessing method. The accuracy of the reconstructed surfaces of the pavements is improved by the new preprocessing method.

An adaptive edge-detection method based on histogram

In order to solve the problems of poor adaptability when setting threshold and the high probability of detecting pseudo-edges in the existing methods of edge detection, the paper proposes an adaptive edge-detection method based on histogram. Multi-scale wavelet transform is used to preprocess the image, the image details are highlighted obviously, and it also can avoid the effect of manual setting filter coefficients. Difference of gray values between the pixels of local area are used to calculate the gradients comprehensively, it extends the gradient direction to four directions. When calculating the gradient of edge pixel, the four directions make the expression of the gradients of edge points more perfect and avoid the edge points missing. The adaptive method is used to compute the threshold of edge-detection, the image is represented by histogram. Then use the ratio of the number of pixels in the bar and the total numbers of pixels to set the initial threshold. The regions on both sides of the initial threshold are used to calculate the high threshold and low threshold until the reasonable error between the current threshold and the previous threshold is very small iteratively. The acquired threshold makes the self-adaptability more reasonable and stronger, it also avoids the detection errors, the connection errors and the pseudo-edges which are caused by setting threshold artificially. The experimental results show that the proposed algorithm of edge detection has a good effect of preserving edge detail and filtering noise of image.

Adaptive and ADRC information fusion method for high speed train braking system

Aiming at the problem of poor adaptability and lag of traditional braking control methods of high-speed train, a high-speed train braking information fusion method based on adaptive linear auto disturbance rejection is proposed to arrange the transition process for

Design of active disturbance rejection controller for compass-like biped walking

This article proposes an active disturbance rejection controller design scheme to stabilize the unstable limit cycle of a compass-like biped robot. The idea of transverse coordinate transformation is applied to form the control system based on angular momentum. With the linearization approximation, the limit cycle stabilization problem is simplified into the stabilization of an linear time-invariant system, which is known as transverse coordinate control. In order to solve the problem of poor adaptability caused by linearization approximation, we design an active disturbance rejection controller in the form of a serial system. With the active disturbance rejection controller, the system error can be estimated by extended state observer and compensated by nonlinear state error feedback, and the unstable limit cycle can be stabilized. The numerical simulations show that the control law enhances the performance of transverse coordinate control.

Fault Diagnostic Method for Micro-grid Based on Wavelet SOM Neural Network and Multi Agent System

Fault diagnosis using traditional neural network method requires training samples to train the neural network. The micro grid system has the characteristics of flexible operation mode and variety of topology structure, the fault diagnosis method of it by using neural network has the problem of poor adaptability and requires a large number of training samples. A fault diagnosis method combining multi agent system with wavelet som neural network is proposed. The wavelet som neural network can judge the reason of fault and the multi agent system judge the location of the fault. A micro grid simulation system is established based on PSCAD. The simulation results prove the feasibility of the fault diagnosis method based on multi agent system and wavelet som neural network. the wavelet som neural network only need typical fault training samples, this method will not affected by fault location, fault time and other factors, it has good adaptability to the change of the topology structure of micro grid system.

A Method of Road-Edge Detection Based on Perception Dominance

Aiming at the problem of poor adaptability to environment when using a single sensor to detect road-edge, this paper proposed a road-edge detection method based on perception dominance. Firstly, the confidence model of 2D laser radar and vision sensor was established, and then the Dempster-Shafer Theory was used to fuse the detection results of 2D laser radar and vision sensor. Finally, the effectiveness of the method in harsh environment was verified by experiments.

Implementation and Analysis of the File System Based on Metadata Dynamic Hashing

Metadata management algorithm plays an important role in file system performance and file system is the most common way of data accessing in mass storage systems, so metadata management algorithm is very important for the performance of mass storage system. In this paper we analyze current metadata management algorithms and bring in the metadata dynamic hashing algorithm to solve the problem of poor adaptability of them. We realize the prototype on Lustre. After testing system performance with common tools by adjusting the several parameters’ value of metadata dynamic hashing algorithm, we find that performance of the prototype I/O is superior to Lustre. Furthermore, we analyze impacts of the parameters of dynamic hashing function on I/O performance of prototype