Working condition recognition of sucker rod pumping system based on 4-segment time-frequency signature matrix and deep learning

High-precision and real-time diagnosis of sucker rod pumping system(SRPS)is important for quickly mastering oil well operations.Deep learning-based method for classifying the dynamometer card(DC)of oil wells is an efficient diagnosis method.However,the input of the DC as a two-dimensional image into the deep learning framework suffers from low feature utilization and high computational effort.Additionally,different SRPSs in an oil field have various system parameters,and the same SRPS generates different DCs at different moments.Thus,there is heterogeneity in field data,which can dramatically impair the diagnostic accuracy.To solve the above problems,a working condition recognition method based on 4-segment time-frequency signature matrix(4S-TFSM)and deep learning is presented in this paper.First,the 4-segment time-frequency signature(4S-TFS)method that can reduce the computing power requirements is proposed for feature extraction of DC data.Subsequently,the 4S-TFSM is constructed by relative normalization and matrix calculation to synthesize the features of multiple data and solve the problem of data heterogeneity.Finally,a convolutional neural network(CNN),one of the deep learning frameworks,is used to determine the functioning conditions based on the 4S-TFSM.Experiments on field data verify that the proposed diagnostic method based on 4S-TFSM and CNN(4S-TFSM-CNN)can significantly improve the accuracy of working condition recognition with lower computational cost.To the best of our knowledge,this is the first work to discuss the effect of data heterogeneity on the working condition recognition performance of SRPS.

Few-shot working condition recognition of a sucker-rod pumping system based on a 4-dimensional time-frequency signature and meta-learning convolutional shrinkage neural network

The accurate and intelligent identification of the working conditions of a sucker-rod pumping system is necessary. As onshore oil extraction gradually enters its mid-to late-stage, the cost required to train a deep learning working condition recognition model for pumping wells by obtaining enough new working condition samples is expensive. For the few-shot problem and large calculation issues of new working conditions of oil wells, a working condition recognition method for pumping unit wells based on a 4-dimensional time-frequency signature (4D-TFS) and meta-learning convolutional shrinkage neural network (ML-CSNN) is proposed. First, the measured pumping unit well workup data are converted into 4D-TFS data, and the initial feature extraction task is performed while compressing the data. Subsequently, a convolutional shrinkage neural network (CSNN) with a specific structure that can ablate low-frequency features is designed to extract working conditions features. Finally, a meta-learning fine-tuning framework for learning the network parameters that are susceptible to task changes is merged into the CSNN to solve the few-shot issue. The results of the experiments demonstrate that the trained ML-CSNN has good recognition accuracy and generalization ability for few-shot working condition recognition. More specifically, in the case of lower computational complexity, only few-shot samples are needed to fine-tune the network parameters, and the model can be quickly adapted to new classes of well conditions.

A Novel Deep Learning Representation for Industrial Control System Data

Feature extraction plays an important role in constructing artificial intel-ligence(AI)models of industrial control systems(ICSs).Three challenges in this field are learning effective representation from high-dimensional features,data heterogeneity,and data noise due to the diversity of data dimensions,formats and noise of sensors,controllers and actuators.Hence,a novel unsupervised learn-ing autoencoder model is proposed for ICS data in this paper.Although traditional methods only capture the linear correlations of ICS features,our deep industrial representation learning model(DIRL)based on a convolutional neural network can mine high-order features,thus solving the problem of high-dimensional and heterogeneous ICS data.In addition,an unsupervised denoising autoencoder is introduced for noisy ICS data in DIRL.Training the denoising autoencoder allows the model to better mitigate the sensor noise problem.In this way,the represen-tative features learned by DIRL could help to evaluate the safety state of ICSs more effectively.We tested our model with absolute and relative accuracy experi-ments on two large-scale ICS datasets.Compared with other popular methods,DIRL showed advantages in four common indicators of AI algorithms:accuracy,precision,recall,and F1-score.This study contributes to the effective analysis of large-scale ICS data,which promotes the stable operation of ICSs.

Characteristics of laser-induced breakdown spectroscopy of liquid slag

Rapid online analysis of liquid slag is essential for optimizing the quality and energy efficiency of steel production. To investigate the key factors that affect the online measurement of refined slag using laser-induced breakdown spectroscopy(LIBS), this study examined the effects of slag composition and temperature on the intensity and stability of the LIBS spectra. The experimental temperature was controlled at three levels: 1350℃, 1400℃, and 1450℃. The results showed that slag composition and temperature significantly affected the intensity and stability of the LIBS spectra. Increasing the Fe content and temperature in the slag reduces its viscosity, resulting in an enhanced intensity and stability of the LIBS spectra. Additionally, 42 refined slag samples were quantitatively analyzed for Fe, Si, Ca, Mg, Al, and Mn at 1350℃, 1400℃, and 1450℃.The normalized full spectrum combined with partial least squares(PLS) quantification modeling was used, using the Ca Ⅱ 317.91 nm spectral line as an internal standard. The results show that using the internal standard normalization method can significantly reduce the influence of spectral fluctuations. Meanwhile, a temperature of 1450℃ has been found to yield superior results compared to both 1350℃ and 1400℃, and it is advantageous to conduct a quantitative analysis of the slag when it is in a “water-like” state with low viscosity.

A method of laser focusing control in microlaser-induced breakdown spectroscopy

This paper presents a method for the automatic adjustment of the laser defocusing amount in micro-laser-induced breakdown spectroscopy. A microscopic optical imaging system consisting of a CCD camera and a 20× objective lens was adopted to realize the method. The real-time auto-focusing of the system was achieved by detecting the effective pixels of the light spot generated by the laser pointer. The focusing accuracy of the method could achieve 3 μm. The element concentrations of Mn and Ni in low-alloy steels were analyzed at a crater diameter of about 35 μm using the presented method. After using the presented method, the determination coefficients of Mn and Ni both exceeded 0.997, with the root-mean-square errors being 0.0133 and 0.0395, respectively. Scanning analysis was performed on the inclined plane and the curved surface by means of focusing control and non-focusing control. Ten characteristic spectral lines of Fe were selected as the analysis lines. With the focusing control, the average relative standard deviations obtained on the inclined plane and curved surface were both less than 5%, and much less than the values without focusing control, 14.6% and 40.39%.

A Priority-aware Frequency Domain Polling MAC Protocol for OFDMA-based Networks in Cyber-physical Systems

Wireless networking in cyber-physical systems(CPSs) is characteristically different from traditional wireless systems due to the harsh radio frequency environment and applications that impose high real-time and reliability constraints.One of the fundamental considerations for enabling CPS networks is the medium access control protocol. To this end, this paper proposes a novel priority-aware frequency domain polling medium access control(MAC) protocol, which takes advantage of an orthogonal frequency-division multiple access(OFDMA)physical layer to achieve instantaneous priority-aware polling.Based on the polling result, the proposed work then optimizes the resource allocation of the OFDMA network to further improve the data reliability. Due to the non-polynomial-complete nature of the OFDMA resource allocation, we propose two heuristic rules,based on which an efficient solution algorithm to the OFDMA resource allocation problem is designed. Simulation results show that the reliability performance of CPS networks is significantly improved because of this work.

Joint Algorithm of Message Fragmentation and No-Wait Scheduling for Time-Sensitive Networks

Time-sensitive networks(TSNs)support not only traditional best-effort communications but also deterministic communications,which send each packet at a deterministic time so that the data transmissions of networked control systems can be precisely scheduled to guarantee hard real-time constraints.No-wait scheduling is suitable for such TSNs and generates the schedules of deterministic communications with the minimal network resources so that all of the remaining resources can be used to improve the throughput of best-effort communications.However,due to inappropriate message fragmentation,the realtime performance of no-wait scheduling algorithms is reduced.Therefore,in this paper,joint algorithms of message fragmentation and no-wait scheduling are proposed.First,a specification for the joint problem based on optimization modulo theories is proposed so that off-the-shelf solvers can be used to find optimal solutions.Second,to improve the scalability of our algorithm,the worst-case delay of messages is analyzed,and then,based on the analysis,a heuristic algorithm is proposed to construct low-delay schedules.Finally,we conduct extensive test cases to evaluate our proposed algorithms.The evaluation results indicate that,compared to existing algorithms,the proposed joint algorithm improves schedulability by up to 50%.

A feature selection method combined with ridge regression and recursive feature elimination in quantitative analysis of laser induced breakdown spectroscopy

In the spectral analysis of laser-induced breakdown spectroscopy,abundant characteristic spectral lines and severe interference information exist simultaneously in the original spectral data.Here,a feature selection method called recursive feature elimination based on ridge regression(Ridge-RFE)for the original spectral data is recommended to make full use of the valid information of spectra.In the Ridge-RFE method,the absolute value of the ridge regression coefficient was used as a criterion to screen spectral characteristic,the feature with the absolute value of minimum weight in the input subset features was removed by recursive feature elimination(RFE),and the selected features were used as inputs of the partial least squares regression(PLS)model.The Ridge-RFE method based PLS model was used to measure the Fe,Si,Mg,Cu,Zn and Mn for 51 aluminum alloy samples,and the results showed that the root mean square error of prediction decreased greatly compared to the PLS model with full spectrum as input.The overall results demonstrate that the Ridge-RFE method is more efficient to extract the redundant features,make PLS model for better quantitative analysis results and improve model generalization ability.

Selection of Spectral Data for Classification of Steels Using Laser-Induced Breakdown Spectroscopy

Principal component analysis （PCA） combined with artificial neural networks was used to classify the spectra of 27 steel samples acquired using laser-induced breakdown spectroscopy. Three methods of spectral data selection, selecting all the peak lines of the spectra, selecting intensive spectral partitions and the whole spectra, were utilized to compare the infiuence of different inputs of PCA on the classification of steels. Three intensive partitions were selected based on experience and prior knowledge to compare the classification, as the partitions can obtain the best results compared to all peak lines and the whole spectra. We also used two test data sets, mean spectra after being averaged and raw spectra without any pretreatment, to verify the results of the classification. The results of this comprehensive comparison show that a back propagation network trained using the principal components of appropriate, carefully selecred spectral partitions can obtain the best results accuracy can be achieved using the intensive spectral A perfect result with 100% classification partitions ranging of 357-367 nm.

Application of Stand-off Double-Pulse Laser-Induced Breakdown Spectroscopy in Elemental Analysis of Magnesium Alloy

In this study, a stand-off and collinear double pulse laser-induced breakdown spectroscopy （DP LIBS） system was designed, and the magnesium alloy samples at a distance of 2.5 m away from the LIBS system were measured. The effect of inter-pulse delay on spectra was studied, and the signal enhancement was observed compared to the single pulse LIBS （SP LIBS）. The morphology of the ablated crater on the sample indicated a higher efficiency of surface pretreatment in DP LIBS. The calibration curves of Ytterbium （Y） and Zirconium （Zr） were investigated. The square of the correlation coefficient of the calibration curve of element Y reached up to 0.9998.

Horseshoe and entropy in a fractional-order unified system

This paper studies chaotic dynamics in a fractional-order unified system by means of topological horseshoe theory and numerical computation. First it finds four quadrilaterals in a carefully-chosen Poincare section, then shows that the corresponding map is semiconjugate to a shift map with four symbols. By estimating the topological entropy of the map and the original time-continuous system, it provides a computer assisted verification on existence of chaos in this system, which is much more convincible than the common method of Lyapunov exponents. This new method can potentially be used in rigorous studies of chaos in such a kind of system. This paper may be a start for proving a given fractional-order differential equation to be chaotic.

Quantitative analysis of the main components in ceramic raw materials based on the desktop LIBS analyzer

The concentrations of SiO,Al2O,KO,NaO,CaO,MgO,Fe2Oand TiO,and loss on ignition(L.O.I.) are the main inorganic components of geological samples.Concentrations of the eight oxides and L.O.I.are also the main indicators of concern in the production of building ceramics.Quantitative analysis of the eight oxides and L.O.I.was performed using fiber-laserbased laser-induced breakdown spectroscopy(LIBS).A combination of continuous background deduction,full width at half maximum(FWHM) intensity integral and spectral sum normalization was proposed for data processing.After the data processing combined the continuous background deduction,FWHM intensity integral and spectral sum normalization,the mean absolute errors(MAEs) of the calibration of L.O.I.,SiO,Al2O,KO,NaO,CaO,MgO,Fe2Oand TiOwas reduced from 2.03%,12.06%,4.84%,1.10%,0.69%,0.31%,0.11%,0.20%and 0.10% to 1.80%,9.48%,2.12%,0.36%,0.58%,0.11%,0.08%,0.19% and 0.05%,respectively.This multivariate method was further introduced and discussed to improve the analysis performance.The MAEs of L.O.I.,SiO,Al2O,KO and NaO were further reduced to1.12%,2.07%,1.38%,0.35% and 0.43%,respectively.The results show that the overall prediction error can meet the requirements for the production of building ceramics.The LIBS desktop analyzer has great potential in detection applications on geological samples.

Microanalysis of a ductile iron by microchip laser-induced breakdown spectroscopy

Laser beams with ns pulse width are generally employed as an excitation source in the process of detecting inclusions and elemental segregation on a workpiece surface by microanalysis of the laser-induced breakdown spectroscopy.In addition,the ablation crater interval of laser sampling on the sample surface is generally 20μm or more.It is difficult to detect the morphology of inclusions smaller than 50μm in diameter and the micro-segregation of elements.However,in this work,when the laser ablation crater is 10μm and the sampling resolution of the laser on the sample surface is 5μm,the morphology and distribution of spherical inclusions(20–60μm)in ductile iron can be detected according to the difference of the Fe spectrum on the Fe matrix and the spheroidal inclusions.Moreover,the distribution of micro-segregation of Mg and Ti elements in ductile iron was also studied.

Industrial Wireless Control Networks:From WIA to the Future

1.Introduction Industrial automation is undergoing a significant innovation as information,communication,and operation technologies are deeply integrating with each other.Following this trend,industrial wireless control networks(IWCNs)are becoming increasingly attractive to industrial automation since they can help speed up production efficiency,reduce cost,enhance safety,and finally realize intelligent manufacturing[1].

An Anomaly Detection Method of Industrial Data Based on Stacking Integration

With the development of Internet technology,the computing power of data has increased,and the development of machine learning has become faster and faster.In the industrial production of industrial control systems,quality inspection and safety production of process products have always been our concern.Aiming at the low accuracy of anomaly detection in process data in industrial control system,this paper proposes an anomaly detection method based on stacking integration using the machine learning algorithm.Data are collected from the industrial site and processed by feature engineering.Principal component analysis(PCA)and integrated rule tree method are adopted to reduce the dimension of the process data,which can restore the original feature information of the data to the maximum extent.Random forest(RF),Adaboost,XGboost,SVM were selected as the first layer of basic learners.Logistic regression(LR)was used as the secondary learner to build the exception detection model based on stacking integrated method.TE data was used to train the base learner model and the integrated model.By comparing and analyzing the experimental results of between integrated model and each basic learning model.By comparing and analyzing the experimental results of the constructed anomaly detection model and the basic learning model,the accuracy of process data anomaly detection is effectively improved,and the false alarm rate of process data anomaly detection is effectively reduced.

Unsupervised Anomaly Detection Approach Based on Adversarial Memory Autoencoders for Multivariate Time Series

The widespread usage of Cyber Physical Systems(CPSs)generates a vast volume of time series data,and precisely determining anomalies in the data is critical for practical production.Autoencoder is the mainstream method for time series anomaly detection,and the anomaly is judged by reconstruction error.However,due to the strong generalization ability of neural networks,some abnormal samples close to normal samples may be judged as normal,which fails to detect the abnormality.In addition,the dataset rarely provides sufficient anomaly labels.This research proposes an unsupervised anomaly detection approach based on adversarial memory autoencoders for multivariate time series to solve the above problem.Firstly,an encoder encodes the input data into low-dimensional space to acquire a feature vector.Then,a memory module is used to learn the feature vector’s prototype patterns and update the feature vectors.The updating process allows partial forgetting of information to prevent model overgeneralization.After that,two decoders reconstruct the input data.Finally,this research uses the Peak Over Threshold(POT)method to calculate the threshold to determine anomalous samples from normal samples.This research uses a two-stage adversarial training strategy during model training to enlarge the gap between the reconstruction error of normal and abnormal samples.The proposed method achieves significant anomaly detection results on synthetic and real datasets from power systems,water treatment plants,and computer clusters.The F1 score reached an average of 0.9196 on the five datasets,which is 0.0769 higher than the best baseline method.

In situ analysis of magnesium alloy using a standoff and double-pulse laser-induced breakdown spectroscopy system

To monitor the components of molten magnesium alloy during the smelting process in real time and online, we designed a standoff double-pulse laser-induced breakdown spectroscopy （LIBS） analysis system that can perform focusing, collecting and imaging of long-range samples. First, we tested the system on solid standard magnesium alloy samples in the laboratory to establish a basis for the online monitoring of the components of molten magnesium alloy in the future. The experimental results show that the diameters of the focus spots are approximately 1 mm at a range of 3 m, the ablation depth of the double-pulse mode is much deeper than that of the single-pulse mode, the optimum interpulse delay of the double pulse is inconsistent at different ranges, and the spectral intensity decays rapidly as the range increases. In addition, the enhancement effect of the double pulse at 1.89 m is greater than that at 2.97 m, the maximum enhancement is 7.1-fold for the Y（I）550.35-nm line at 1.89 m, and the calibration results at 1.89 m are better than those at 2.97 m. At 1.89 m, the determination coefficients （R2） of the calibration curves are approximately 99% for Y, Pr, and Zr; the relative standard deviations （RSDs） are less than 10% for Y, Pr, and Zr; the root mean square errors （RMSEs） are less than 0.037% for Pr and Zr; the limits of detection （LODs） are less than 1000 ppm for Y, Pr, and Zr; and the LODs of Y, Pr, and Zr at 2.97 m are higher than those at 1.89 m. Additionally, we tested the system on molten magnesium alloy in a magnesium alloy plant. The calibration results of the liquid magnesium alloy are not as favorable as those of the sampling solid magnesium alloys. In particular, the RSDs of the liquid magnesium alloy are approximately 20% for Pr and La. However, with future improvements in the experimental conditions, the developed system is promising for the in situ analysis of molten magnesium alloy.

High-precision Calibration of Camera and IMU on Manipulator for Bio-inspired Robotic System

Inspired by box jellyfish that has distributed and complementary perceptive system,we seek to equip manipulator with a camera and an Inertial Measurement Unit(IMU)to perceive ego motion and surrounding unstructured environment.Before robot perception,a reliable and high-precision calibration between camera,IMU and manipulator is a critical prerequisite.This paper introduces a novel calibration system.First,we seek to correlate the spatial relationship between the sensing units and manipulator in a joint framework.Second,the manipulator moving trajectory is elaborately designed in a spiral pattern that enables full excitations on yaw-pitch-roll rotations and x-y-z translations in a repeatable and consistent manner.The calibration has been evaluated on our collected visual inertial-manipulator dataset.The systematic comparisons and analysis indicate the consistency,precision and effectiveness of our proposed calibration method.

Monocular Visual-Inertial and Robotic-Arm Calibration in a Unifying Framework

Reliable and accurate calibration for camera,inertial measurement unit(IMU)and robot is a critical prerequisite for visual-inertial based robot pose estimation and surrounding environment perception.However,traditional calibrations suffer inaccuracy and inconsistency.To address these problems,this paper proposes a monocular visual-inertial and robotic-arm calibration in a unifying framework.In our method,the spatial relationship is geometrically correlated between the sensing units and robotic arm.The decoupled estimations on rotation and translation could reduce the coupled errors during the optimization.Additionally,the robotic calibration moving trajectory has been designed in a spiral pattern that enables full excitations on 6 DOF motions repeatably and consistently.The calibration has been evaluated on our developed platform.In the experiments,the calibration achieves the accuracy with rotation and translation RMSEs less than 0.7°and 0.01 m,respectively.The comparisons with state-of-the-art results prove our calibration consistency,accuracy and effectiveness.

Small-invasive determination of iron content in coating of galvanized steel sheets by laser-induced breakdown spectroscopy

A new method was presented to determine the iron content in the coating of galvanized steel sheet based on laser-induced breakdown spectroscopy.The zinc-iron coating was characterized with a series of single laser pulses irradiated on the traversing sheet steel,each on a different steel sheet position.The influences of laser fluence and elemental depth distribution were studied and analyzed.To protect the corrosion performance of the coating and meet requirements for small-invasive measurement,the ablation size of the crater under different laser fluences was studied.Under the optimized experimental parameters,the diameter of ablation craters is about 50μm,and then,the Fe content in the coating was calibrated and analyzed by the linear standard calibration method.The calibration result,however,is not good.Considering that the Zn content in the coating was high and relatively constant,curve calibration was then carried out with the intensity ratio(IFe404.58/Izn468.01)instead of the net line intensity of Fe,and then,the determination coefficient of calibration curve increases from 0.7713 to 0.9511,and the root-mean-square error decreases from 0.4832%to 0.1509%.The results prove that the laser-induced breakdown spectroscopy is an effective way for the analysis of the Fe content in the coating of galvanized steel sheet.