Hierarchical multihead self-attention for time-series-based fault diagnosis

Fault diagnosis is important for maintaining the safety and effectiveness of chemical process.Considering the multivariate,nonlinear,and dynamic characteristic of chemical process,many time-series-based data-driven fault diagnosis methods have been developed in recent years.However,the existing methods have the problem of long-term dependency and are difficult to train due to the sequential way of training.To overcome these problems,a novel fault diagnosis method based on time-series and the hierarchical multihead self-attention(HMSAN)is proposed for chemical process.First,a sliding window strategy is adopted to construct the normalized time-series dataset.Second,the HMSAN is developed to extract the time-relevant features from the time-series process data.It improves the basic self-attention model in both width and depth.With the multihead structure,the HMSAN can pay attention to different aspects of the complicated chemical process and obtain the global dynamic features.However,the multiple heads in parallel lead to redundant information,which cannot improve the diagnosis performance.With the hierarchical structure,the redundant information is reduced and the deep local time-related features are further extracted.Besides,a novel many-to-one training strategy is introduced for HMSAN to simplify the training procedure and capture the long-term dependency.Finally,the effectiveness of the proposed method is demonstrated by two chemical cases.The experimental results show that the proposed method achieves a great performance on time-series industrial data and outperforms the state-of-the-art approaches.

An intelligent control method based on artificial neural network for numerical flight simulation of the basic finner projectile with pitching maneuver

In this paper,an intelligent control method applying on numerical virtual flight is proposed.The proposed algorithm is verified and evaluated by combining with the case of the basic finner projectile model and shows a good application prospect.Firstly,a numerical virtual flight simulation model based on overlapping dynamic mesh technology is constructed.In order to verify the accuracy of the dynamic grid technology and the calculation of unsteady flow,a numerical simulation of the basic finner projectile without control is carried out.The simulation results are in good agreement with the experiment data which shows that the algorithm used in this paper can also be used in the design and evaluation of the intelligent controller in the numerical virtual flight simulation.Secondly,combined with the real-time control requirements of aerodynamic,attitude and displacement parameters of the projectile during the flight process,the numerical simulations of the basic finner projectile’s pitch channel are carried out under the traditional PID(Proportional-Integral-Derivative)control strategy and the intelligent PID control strategy respectively.The intelligent PID controller based on BP(Back Propagation)neural network can realize online learning and self-optimization of control parameters according to the acquired real-time flight parameters.Compared with the traditional PID controller,the concerned control variable overshoot,rise time,transition time and steady state error and other performance indicators have been greatly improved,and the higher the learning efficiency or the inertia coefficient,the faster the system,the larger the overshoot,and the smaller the stability error.The intelligent control method applying on numerical virtual flight is capable of solving the complicated unsteady motion and flow with the intelligent PID control strategy and has a strong promotion to engineering application.

Kinematic deformation and intensity assessment of the 2021 Maduo M_(S)7.4 earthquake in Qinghai revealed by high-frequency GNSS

Rapid acquisition of the kinematic deformation field and seismic intensity distribution of large earthquakes is crucial for postseismic emergency rescue,disaster assessment,and future seismic risk research.The advancement of GNSS observation and data processing makes it play an important role in this field,especially the high-frequency GNSS.We used the differential positioning method to calculate the 1 HZ GNSS data from 98 sites within 1000 km of the M_(S)7.4 Maduo earthquake epicenter.The kinematic deformation field and the distribution of the seismic intensity by using the peak ground velocity derived from displacement waveforms were obtained.The results show that:1)Horizontal coseismic response deformation levels ranging from 25 mm to 301 mm can be observed within a 1000 km radius from the epicenter.Coseismic response deformation on the east and west sides shows bilateral asymmetry,which markedly differs from the symmetry presented by surface rupture.2)The seismic intensity obtained through high-frequency GNSS and field investigations exhibits good consistency of the scope and orientation in the high seismic intensity area,although the former is generally slightly smaller than the latter.3)There may exist obstacles on the eastern side of the seismogenic fault.The Maduo earthquake induced a certain tectonic stress loading effect on the western Kunlun Pass-Jiangcuo fault(KPJF)and Maqin-Maqu segment,resulting in higher seismic risk in the future.

Optimizing University Faculty Teams: Addressing the Talent Mobility Challenges in the Context of “Double First-Class” Initiative

With the advancement of the“Double First-Class”initiative,universities are increasingly focusing on attracting top-tier talent to enhance faculty quality and strengthen academic disciplines.The blind competition for high-end talent among universities reflects an overshadowing of academic values by economic considerations,neglecting the growth patterns of top talent and the development laws of higher education.This study investigates and reflects on the real-world challenges of faculty mobility in universities,aiming to provide effective guidance for managing university faculty teams.

Research on variety characteristics of China's Mainland troposphere based on CMONOC

In this research, we processed the GPS and meteorological data from about 220 stations of Crustal Movement Observation Network of China(CMONOC) observed in 2014 and derived the Zenith Total Delay(ZTD) map in both spatial and temporal dimension. The results of ZTD have high accurate and reliable as IGS and all sites with varying locations show obvious variety characteristics of Chinese mainland. Meanwhile, the precipitable water vapor(PWV) correlation coefficients between GPS observation and upper air sounding is close to 1, and the comparison of GPS-derived PWV and observed PWV from meteorological sites indicating GPS observation data generated in CMONOC project applied to the weather forecast research is feasible. In addition, based on all stations covered the whole Chinese land area and using interpolation algorithms, we make contour plots of PWV distribution per hour. We observe obvious feature that the precipitable water in north and western area is less than south and east area all over this year. High latitudes area may be dry and low latitudes area is wet.

Orthogonal nonnegative matrix factorization based local hidden Markov model for multimode process monitoring

Traditional data driven fault detection methods assume that the process operates in a single mode so that they cannot perform well in processes with multiple operating modes. To monitor multimode processes effectively,this paper proposes a novel process monitoring scheme based on orthogonal nonnegative matrix factorization(ONMF) and hidden Markov model(HMM). The new clustering technique ONMF is employed to separate data from different process modes. The multiple HMMs for various operating modes lead to higher modeling accuracy.The proposed approach does not presume the distribution of data in each mode because the process uncertainty and dynamics can be well interpreted through the hidden Markov estimation. The HMM-based monitoring indication named negative log likelihood probability is utilized for fault detection. In order to assess the proposed monitoring strategy, a numerical example and the Tennessee Eastman process are used. The results demonstrate that this method provides efficient fault detection performance.

Introduction to China’s Earthquake Emergency Response System

Following the M_(S)6.4 earthquake that occurred on May 21,2021 in Yangbi,Yunnan,China,the earthquake emergency response system(EERS)responded immediately.The real-time software delivered many seismic parameters that provided a preliminary assessment of the earthquake.The 24-hour on-duty staff and scientific researchers revised these parameters and produced more detailed reports to understand the cause of the earthquake and the potential damage,which provided valuable information for emergency rescue operations and earthquake situation assessment.Emergency personnel were dispatched immedia-tely to the earthquake site to observe the aftershocks,investigate the damage,and guide and assist in the relief efforts.This paper describes the EERS response to the Yangbi earthquake to demonstrate the characteristics of the system and discuss the potential for further improvement.

Inversion Precipitable Water Vapor by GPS Observation of CMONOC

In this study, we have processed the GPS （Global Position System） and meteorological data from about 220 stations of CMONOC （Crustal Movement Observation Network of China in short） observed in 2014 by GAMIT software. The comparison result of ZTD （zenith total delay） calculated by GPS data and IGS （International GNSS （Global Navigation Satellite System） Service） ZTD product shows that the tropospheric delay based on calculation of CMONOC project data is accurate and reliable. Meanwhile, the PWV （precipitable water vapor） correlation coefficients between GPS observation and upper air sounding is close to 1, which proves that GPS observation data generated in CMONOC project applied to the weather forecast research is feasible. In addition, we make an isoline image for PWV distribution per hour on all stations covered the whole Chinese land area using interpolation algorithms. We observe obvious feature that the precipitable water in north and western area is less than south and east area all over this year. High latitudes area may be dry and low latitudes area is wet.

A recombineering system for Bacillus subtilis based on the native phage recombinase pair YqaJ/YqaK

Bacillus subtilis plays an important role in fundamental and applied research,and it has been widely used as a cell factory for the production of enzymes,antimicrobial materials,and chemicals for agriculture,medicine,and industry.However,genetic manipulation tools for B.subtilis have low efficiency.In this work,our goal was to develop a simple recombineering system for B.subtilis.We showed that genome editing can be achieved in B.subtiliis 1A751 through co-expression of YqaJ/YqaK,a native phage recombinase pair found in B.sub-tilis 168,and the competence master regulator ComK using a double-stranded DNA substrate with short ho-mology arms(100 bp)and a phosphorothioate modification at the 5′-end.Efficient gene knockouts and large DNA insertions were achieved using this new recombineering system in B.subtilis 1A751.As far as we know,this is the first recombineering system using the native phage recombinase pair YqaJ/YqaK in B.subtilis.In conclusion,this new recombineering system provides a simple and fast tool for genetic manipulation of B.sub-tilis,and it will promote studies of genome function,construction of production strains,and genome mining in B.subtilis.

High-dimensional immune profiling by mass cytometry revealed immunosuppression and dysfunction of immunity in COVID-19 patients

The outbreak of coronavirus disease 2019(COVID-19)caused by the new virus SARS-CoV-2 has been announced as a public health emergency of international concern.1–3 The clinical features of patients with COVID-19 range from common fever and cough to other rare symptoms,such as diarrhea and nausea.This disease can progress quickly,and 2–3%of patients die within a short time,which is generally due to multiple organ failure.4–7 Clinically,COVID-19 patients are classified into mild,moderate,severe,and critical cases.5–7 The immune response against SARS-CoV-2 is probably linked to the severity of disease.Recently,Zheng et al.8 showed that elevated levels of T-cell exhaustion and reduced functional diversity of T cells in peripheral blood may predict severe progression in COVID-19 patients;however,a more comprehensive understanding of the pathology of SARS-CoV-2 infection remains to be delineated.Here,we profiled immune cellular components using mass cytometry(CyTOF)to analyze the peripheral blood mononuclear cells(PBMCs)from patients with differences in disease progression by comparing with the PBMCs from healthy donors(HDs).

Comparison of clinical features and liver histology in liver failure caused by autoimmune hepatitis with different prognosis

To the Editor:Autoimmune hepatitis(AIH)is a type of liver parenchymal inflammation mediated by the autoimmune response of liver cells.AIH was previously considered to be a chronic liver disease,but recent observers have confirmed that AIH patients can have acute onset or even liver failure(AIH-LF).[1]The incidence of AIH-LF has also increased.Once LF occurs,the mortality rate within 90 days could reach>40%.[2]Therefore,the occurrence of LF has a major impact on the prognosis of AIH.Although AIH-LF has a low incidence and a high mortality rate,some cases can still be cured by treatment.The clinical characteristics that may affect the prognosis of patients with AIH-LF are still unclear.To clarify the factors affecting the prognosis of AIH-LF,we compared the clinical characteristics of two groups of AIH-LF patients with different prognosis.

Turing patterns in a reaction-diffusion epidemic model

In this paper, we investigate the spatiotemporal dynamics of a reactio^diffusion epi- demic model with zero-flux boundary conditions. The value of our study lies in two aspects： mathematically, by using maximum principle and the linearized stability theory, a priori estimates of the steady state system and the local asymptotic stability of positive constant solution are given. By using the implicit function theorem, the exis- tence and nonexistence of nonconstant positive steady states are shown. Applying the bifurcation theory, the global bifurcation structure of nonconstant positive steady states is established. Epidemiologically, through numerical simulations, under the conditions of the existence of nonconstant positive steady states, we find that the smaller the space, the easier the pattern formation; the bigger the diffusion, the easier the pattern formation. These results are beneficial to disease control, that is, we must do our best to control the diffusion of the infectious to avoid disease outbreak.