Wyniki wyszukiwania - Biblioteka Nauki

1

Research on Online Condition monitoring for Complex System based on Modified Broad Learning Systems

100%

Wang C. , Liu J.

Eksploatacja i Niezawodność

|

2024

|

tom Vol. 26, no. 2

art. no. 183600

EN

Complex systems contain numerous interacting components, thus deep learning methods with powerful performance and complex structure are often used to achieve condition monitoring. However, the deep learning methods are always too time-consuming and hardware-demanding to be loaded into complex systems for online training and updates. To achieve accurate and timely monitoring of complex system state, based on broad learning system (BLS), an online condition monitoring method is proposed in this paper. GeneralBLSs are based on a randomly generated hidden-layer, usually perform poorly in high-dimensional data classification tasks. In this work, based on correlation and causality, two modified BLSs are proposed and mixed to establish the online monitoring system. Specifically, logistic regression (LR) and structural causal model (SCM) are considered to form rough predictions of the system state, thus to replace the randomly generated ones with no practical significance. The effectiveness of the proposed online monitoring method is verified by both simulation data and real data.

2

Improvement of Cerebral Microbleeds Detection Based on Discriminative Feature Learning

80%

Hong J. , Cheng H. , Wang S.-H. , Liu J.

Fundamenta Informaticae

|

2019

|

tom Vol. 168, nr 2-4

231--248

EN

The existence and distribution pattern of cerebral microbleeds (CMBs) are associated with some underlying aetiologies caused by intra-cerebral hemorrhage (ICH). CMBs as a kind of subclinical sign can be recognized via magnetic resonance (MR) imaging technique in a few years before the onset of the disease. Hence, detecting CMBs accurately is important for treating and preventing related cerebral disease. In this study, we employed convolution neural network (CNN) for CMBs detection because of its powerful ability in image recognition. In view of too many efforts on optimizing the structure of CNN for achieving a better performance, we introduced center loss, which can greatly enhance the discriminative power of the deeply learned features, to CMBs detection for the first time. It is found that the performances of convolution neural network (CNN) trained under the joint supervision of softmax loss and center loss were significantly better than that under the supervision of softmax loss, even if there are few mislabelled samples in training data. With this trick, we achieved a high performance with a sensitivity of 98.869 ± 1.026%, a specificity of 96.491 ± 0.367%, and an accuracy of 97.681 ± 0.497%, which is better than four state-of-the-art methods.

3

An atomic scale investigation of the adsorption of sodium oleate on Ca2+ activated quartz surface

80%

Gong G. , Liu J. , Han Y. , Zhu Y.

|

2019

|

tom Vol. 55, iss. 2

426--436

EN

In this study, the surface properties and flotation behavior of quartz with NaOl as a collector in the presence of Ca2+ ions were investigated using density functional theory (DFT) calculations in conjunction with flotation tests, adsorption experiments, zeta potential measurements, and solution chemistry calculations. The results of the flotation and adsorption tests proved that Ca2+ promoted the flotation recovery and the adsorption density of sodium oleate on quartz at pH > 8. Zeta potential analyses and solution chemistry calculations demonstrated that Ca(OH)+ was the functional species which activated quartz. DFT calculations indicated that O atoms dominated the quartz (101) surface, and great electrostatic repulsion and space resistance existed between the surface and oleate anion.The spontaneous adsorption of H2O and OH- on the (101) surface made quartz surfaces hydrated and hydroxylated, and resulted in the hydrophilicity of quartz. The adsorption of Ca(OH)+ on quartz (101) surface was more favorable and able to repulse the water film, which decreased the electrostatic repulsion and space resistance, and further facilitated the adsorption of oleate anion. During the activating and collecting adsorption processes, electron transition occurred along the O1—Ca—O2 path, implying Ca(OH)+ acted as an intermediary and electron donator in the activation process.

4

Stress evolution of rock breakages by a disc cutter assisted by pre-cuts

80%

Liu J. , Chen W. , Liu T. , Peng D.

Archives of Civil and Mechanical Engineering

|

2023

|

tom Vol. 23, no. 2

art. no. e115, 2023

EN

To study the rock breakage mechanism by constant cross-section cutters assisted by pre-cuts, the present study first performed small-scaled linear cutting tests on sandstone specimens containing pre-cuts. The laboratory tests indicate that the sufficiently large penetration, causing successful internal and surface crack incisions, is essential for large chip formation. In addition, the small pre-cut depth may fail to form large chips. The numerical results agree well with laboratory tests in fracture patterns. More importantly, the numerical analysis indicates that the increases in rolling force frequently result in stress concentrations. When the stresses concentrate to critical values, fracture propagation occurs. The fracture propagation causes stress dissipation and the decrease in rolling force. Thus, the relation between cutting load fluctuations and crack propagation is revealed. Moreover, the influence of penetration on crack propagation is analyzed. By analyzing the stress fields at typical peak points of the rolling force, the crack propagation direction is predicted, and the influence of pre-cut depth on fracture propagation is studied.

5

A machine learning method for soil conditioning automated decision-making of EPBM : hybrid GBDT and Random Forest Algorithm

61%

Lin L. , Guo H. , Lv Y. , Liu J. , Tong C. , Yang S.

Eksploatacja i Niezawodność

|

2022

|

tom Vol. 24, no. 2

237--247

EN

There lacks an automated decision-making method for soil conditioning of EPBM with high accuracy and efficiency that is applicable to changeable geological conditions and takes drive parameters into consideration. A hybrid method of Gradient Boosting Decision Tree (GBDT) and random forest algorithm to make decisions on soil conditioning using foam is proposed in this paper to realize automated decision-making. Relevant parameters include decision parameters (geological parameters and drive parameters) and target parameters (dosage of foam). GBDT, an efficient algorithm based on decision tree, is used to determine the weights of geological parameters, forming 3 parameters sets. Then 3 decision-making models are established using random forest, an algorithm with high accuracy based on decision tree. The optimal model is obtained by Bayesian optimization. It proves that the model has obvious advantages in accuracy compared with other methods. The model can realize real-time decision-making with high accuracy under changeable geological conditions and reduce the experiment cost.

6

Antitumor chemical constituents of Toddalia asiatica (Linn) Lam root bark and its rational alternative medicinal parts by multivariate statistical analysis

61%

Luo C. , Liu J. , Liang Y. , Shen X. , Zhang X. , Zhou W.

Acta Chromatographica

|

2021

|

tom Vol. 33, no. 2

153--161

EN

Toddalia asiatica (Linn) Lam (T. asiatica) as a traditional Miao medicine was investigated to find rational alternative medicinal parts for T. asiatica root bark and its antitumor chemical constituents by quantitative pharmacognostic microscopy, high performance liquid chromatography (HPLC) fingerprint and multivariate statistical analysis. A bivariate correlation analysis method based on microscopic characteristics and content of chemical constituents was established for the first time, there were some regular discoveries between powder microscopic characteristics and common chromatographic peaks of T. asiatica through quantitative pharmacognostic microscopy, cork cells, calcium oxalate square crystal, brown clump, starch granule and phloem fiber, as powder microscopic characteristics may be placed where the main chemical constitutes were enriched. Scores plot of principal component analysis (PCA) and dendrogram of hierarchical clustering analysis (HCA) showed that 18 T. asiatica samples were distinguished correctly, clustered clearly into two main groups as follows: S01∼S03 (root bark) and S07∼S09 (stem bark) in cluster 1, S04∼S06 and S10∼S18 in cluster 2. Nineteen common peaks were obtained in HPLC fingerprint of T. asiatica, loadings plot of PCA indicated seven compounds played important roles in different part of samples (P10 > P08 > P07 > P14 > P16 > P17 > P19), peaks 04, 06, 07, 08, 10 were identified as hesperidin, 4-methoxycinnamic acid, toddalolactone, isopimpinlline and pimpinellin. MTT assay was used to determine the inhibitory activity of different medicinal parts of T. asiatica on human breast cancer MCF-7 cells, all parts of T. asiatica had different inhibitory effects on MCF-7 cell lines, root and stem barks of T. asiatica showed the best inhibitory activity. The relationship between chemical constituents and the inhibitions on MCF-7 cell had been established, significant antitumor constituents of T. asiatica were identified by correlation analysis, the order of the antitumor effect of the main compounds was P07 (toddalolactone) > P16 > P06 (4-methoxycinnamic acid), P11 > P18 > P10 (pimpinellin) > P08 (isopimpinellin) > P01 > P19 > P14 > P04 (hesperidin) > P17, which were antitumor chemical constituents of T. asiatica root bark. T. asiatica stem bark was the most rational alternative medicinal part for T. asiatica root bark.

7

CF–PK–PD analysis of natural hemostatic compounds from Toddalia asiatica (Linn) Lam root bark in rats

51%

Zhang X. , Liu J. , Sun W. , Shen X. , Gong X. , Wang C. , Liang Y. , Zhou W.

Acta Chromatographica

|

2021

|

tom Vol. 33, no. 3

261--269

EN

Natural hemostatic compounds from Toddalia asiatica (Linn) Lam (T. asiatica) root bark had been investigated by a novel strategy, chemical fingerprint–pharmacokinetic–pharmacodynamic (CF–PK–PD) for the first time in this study. The extract sample of T. asiatica root bark was subdivided into petroleum ether (PE), ethyl acetate (EA) and n-butanol (n-B) sample by reagent extraction, EA sample showed significant hemostatic activity using prothrombin time (PT), activated partial thromboplastin time (APTT) and fibrinogen (FIB) as evaluation indexes from rat plasma of PK experiment in hemorrhagic rat model. CF analysis was adopted to assist us to discover six natural compounds from T. asiatica root bark in actual rat plasma after sample treatment by Ultra Performance Liquid Chromatography-Electrospray Ionization (UPLC-ESI) MS, there were only lomatin and 5-methoxy-8-hydroxy psoralen showing significant hemostatic effect (P < 0.05) mainly through endogenous coagulation pathway and fibrinolytic system. In PK–PD study, six compounds in EA sample exhibited relatively rapid absorption and slow elimination characteristics. The mean Tmax and t1/2β of isopimpinellin and pimpinellin were 1.74 and 0.59 h, 5.31 and 6.89 h in rats. On the basis of Sigmoid–Emax model, PK–PD related curves of FIB in hemorrhagic rat model after treatment of T. asiatica root bark were obtained. Predicted Emax, EC50 and ke0 of FIB under isopimpinellin were 4.87 mg/mL, 1.39 μg/mL and 0.81 1/h; predicted Emax, EC50 and ke0 of FIB under pimpinellin were 4.29 mg/mL, 2.47 μg/mL and 0.77 1/h. In conclusion, hemostatic compounds from T. asiatica root bark had been materialized, there were lomatin, isopimpinellin, pimpinellin and 5-methoxy-8-hydroxy psoralen at least as its main active substances through coagulation pathways and fibrinolytic system. CF–PK–PD method as a promising method was worthy of follow-up opening, application in pharmaceutical research.

8

Rapid comparison of antitumor chemical constituents and mechanisms between Dendrobium nobile and Dendrobium officinale by UPLC-IT-TOF, network pharmacology and experimental verification

51%

Liu J. , Li Z. , Zeng M. , Liang Y. , Liang B. , Ge Q. , Zhang X. , Zhang S. , Zhou W.

Acta Chromatographica

|

2023

|

tom Vol. 35, no. 2

123--138

EN

Dendrobium nobile and Dendrobium officinale as the main varieties of traditional Chinese medicine Dendrobium are widely used in clinic. The study aimed to systematically explore chemical constituents and their antitumor effect of D. nobile and D. officinale by ultra-performance liquid chromatography coupled with ion trap time-of-flight mass spectrometry (UPLC-IT-TOF), network pharmacology and cancer cell experiments. D. nobile extract and D. officinale extract could significantly inhibit the proliferation of human lung cancer A549 cells, human liver cancer HepG2 cells and human breast cancer MCF-7 cells in the dose-dependent manner (P < 0.05), the antitumor effect of D. officinale extract was stronger than that of D. nobile extract at the same drug concentration. A total of 40 chemical constituents of D. nobile and D. officinale including phenanthrenes, bibenzyls and other types of compounds had been identified by UPLC-IT-TOF, LCMSsolution and MetID software according to retention times, accurate mass, MSⁿ fragmentation, reference compounds and natural product databases. Phenanthrenes with good antitumor activity were mainly present in D. nobile, bibenzyls were the main compounds of D. officinale. Integrated networks of Herb-Compounds-Targets-Cancer revealed that gigantol, moscatilin, tristin, moscatin and densiflorol B were regarded as key antitumor compounds of D. nobile and D. officinale, D. nobile and D. officinale shared 7 targets accounting for 70% of the antitumor core targets, more than half of their antitumor KEGG pathways were similar. The results of molecular docking and western blotting experiments indicated that the antitumor mechanisms of D. nobile and D. officinale may be through inhibiting PI3K-Akt and HIF-1α signaling pathways.