Search

Volumetric efficiency (VE) is an extremely important index to evaluate the performance of pump and motor. Efforts have been concentrated on the causes and character of leakage loss, but few quantitative analyses of pump kinematic clearances (KCs) identification and nonlinear compression loss caused by air content are discussed. This paper clarifies the nonlinear compression loss resulted from the air content of the oil in detail and proposes an improved flow loss model of the swash-plate axial piston pump (SAPP). Then, combining the improved model with experimental results, the KCs of SAPP are identified. It has been verified that a reasonable sampling path is useful for reducing sampling numbers and guaranteeing identification accuracy. Finally, the identification and quantization of the wear fault of SAPP are realized by analyzing the estimated value of KCs. The proposed method is simple and easy to implement and shows high predictive accuracy. It provides a possible and promoting technology support for the autonomous calibration of the system parameters and the active compensation of the system control.

Pumps are vital components in hydraulic equipment, and their malfunction is directly related to the normal operation of the whole system. The hydraulic system is accompanied by the conversion of multi-domain energy during operation. In particular, it exhibits non-stationarity and nonlinearity under variable operating conditions, which brings difficulties to fault diagnosis. In this paper, the instantaneous angular speed (IAS) signal obtained by the equal-angle measurement is studied to diagnose pump faults under non-stationary conditions. For this purpose, a synchroextracting normal S transform method is proposed to extract the time-frequency feature components properly. Then, the extracted signal is reconstructed into a high-dimensional phase space based on the improved C-C method. On this basis, the quantitative analysis indicators of pump state are obtained by the G-P algorithm and small-data method, including correlation dimension, Kolmogorov entropy, and largest Lyapunov exponent. Pump faults can be classified in a chaotic space by using these sensitive features. The results show that the proposed method is capable of diagnosing different states of pumps and robust to the variation of load values.

The axial piston pump is a core component for power output conversion in the hydraulic system. Monitoring pump status and diagnosing faults in an optimal way are of profound significance for ensuring system reliability. Currently, considerable studies concentrate primarily on the development of vibration-monitoring technologies. However, due to strong fluid shock and noise, vibration analysis has low signal-to-noise ratio and difficulty in fault location. Therefore, an approach of instantaneous angular speed-based fault detection is introduced in this paper since it has the advantages of short transfer path and non-intrusive measurement. Instantaneous angular speed (IAS) is obtained by cross-period linear interpolation (CLI) algorithm for the voltage square wave induced by a magneto-electric tachometer transducer. Compared with the elapsed time method, CLI has the capacity to precisely capture the speed fluctuation. Weighted angle synchronous averaging (WASA) is then utilized to realize the extraction of abnormal wave components in IAS as a vital signal preprocessing method. Instantaneous angular speed fluctuation (IASF) characteristics of the angular domain are analyzed to study the fault diagnosis under varying working conditions. Moreover, it is proven that after processing, the fault features are extracted in the order spectrum where the deterministic shaft order and its harmonics corresponding to wear characteristics are displayed clearly. Experimental results indicate that IAS has demonstrated more effective and sensitive than vibration signals, thus providing a promising tool for the health monitoring of a pump.

An indium tin oxide (ITO) and silicon nitride (Si3N4) assisted compact TE-pass waveguide polarizer based on lithium-niobate-on-insulator is proposed and numerically analyzed. By properly designing the ITO and Si3N4 assisted structure and utilizing the epsilon-near-zero effect of ITO, the TM mode is strongly confined in the ITO layer with extremely high loss, while the TE mode is hardly affected and passes through the waveguide with low loss. The simulation results show that the polarizer has an extinction ratio of 22.5 dB and an insertion loss of 0.8 dB at the wavelength of 1.55 μm, and has an operating bandwidth of about 125 nm (from 1540 nm to 1665 nm) for an extinction ratio of > 20 dB and an insertion loss of < 0.95 dB. Moreover, the proposed device exhibits large fabrication tolerances. More notably, the device is compact, with a length of only 7.5 μm, and is appropriate for on-chip applications.

Background: Computerized clinical decision support systems (CDSS) are low-cost, scalable tools with the potential to improve guideline-based antihypertensive treatment in primary care, but their effectiveness needs testing in pragmatic trials. Methods: The Learning Implementation of Guideline-based decision support system for Hypertension Treatment (LIGHT) trial is a pragmatic, four-stage, cluster-randomized trial conducted in 94 primary care practices in China. For each city-based stage, practices are randomly assigned to either implementation of the CDSS for hypertension management (which guides physicians treatment recommendations based on measured blood pressure and patient characteristics), or usual care. Patients are enrolled during the first 3 months after site randomization and followed for 9 months. The primary outcome is the proportion of hypertension management visits at which guideline-based treatment is provided. In a separate sub-study conducted within the CDSS, with the patient as the unit of randomization, the LIGHT-ACD trial, patients are randomized to receive different initial mono- or dual- antihypertensive therapy. The primary outcome of the LIGHT-ACD trial is the change in blood pressure from the first visit after site randomization to 9 months.Discussion: The LIGHT trial will provide evidence on the effectiveness of a CDSS for improving guideline adherence for hypertension management in primary care in China. The sub-study, LIGHT-ACD trial, will provide data on the effect of different initial antihypertensive regimens for blood pressure management in this setting.Trial registration: ClinicalTrials.gov, identifier: LIGHT (NCT03636334) and LIGHT-ACD (NCT03587103). Registered on 3 July 2018.

Two novel compounds based on the non-classic arsenotungstates, [Co(phen)3]2[As3W3O15](OH)(1) and [Ni(phen)3]2[As3W3O15](OH)·2H2O(2), (phen = 1,10- phenanthroline), have been hydrothermally synthesized and characterized by the elemental analysis, TG, IR, XRD and single crystal X-ray diffraction. Compounds 1 and 2 are isostructural, compose of [As3W3O15]3- and {M(phen)3} (M= Co, Ni) as counterions and hold together to form complicated 3D network by the strong supramolecular interactions and the hydrogen bond in ABAB mode, with the aperture of 12.875 × 13.589 A2 (compound 1), 12.802 × 12.338 A2 (compound 2) respectively. Compounds 1–2 showed excellent photocatalytic activities for the degradation of three dyes (MB, RhB and MO) and good electrocatalytic activity for the reduction of H2O2. In addition, the compounds also fluorescence properties in the solid state at room temperature.

We report a mode converter based on a tapered X-cut/Y-propagation ridge-type LiNbO3-on-Insulator (LNOI) photonic wire (PW) with a laterally asymmetric refractive index profile realized by introducing a semi-infinite dielectric cover layer on one side of the LNOI PW. The interplay of material and waveguide birefringence enables to realize mode hybridization, which is a prerequisite for realizing mode conversion. The introduction of the semi-infinite cover layer and the use of a taper structure modify mode index and electric field distribution, enabling to realize mode conversion. Aiming at 1.55 μm wavelength, we have simulated mode hybridization and conversion characteristics of the tapered, laterally asymmetric ridge-type LNOI PW. The results show that a conversion between fundamental TM00 and TE00 modes can be realized. The conversion efficiency can be optimized by choosing an appropriate cover layer material and/or adjusting its thickness. An ultra-compact device with a taper length

Dependences of Judd-Ofelt (JO) intensity parameters Ωt (t = 2,4,6) and their relative ratios (Ω4/Ω2 and Ω6/Ω2) on overall changing factor (q) of integrated absorption bands of Er3+ have been studied theoretically. These dependences can help us to use relevant method to correctly extract Ωt of a powder material. The study was done by exemplifying the widely studied Er3+ system, which is doped into Gd3Ga5O12, LiNbO3 and SrGdGa3O7 bulk single-crystals and β-NaYF4 crystalline powder. The results show that the Ωt changes linearly with the overall changing factor q: Ωt = γtq + ξt. The linear feature changes from one material to another. The magnetic-dipole oscillator of the 1.5 μm transition results in non-zero nodal increment ξt and hence dramatical changes of the relative ratios Ω4/Ω2 and Ω6/Ω2 (with q) in smaller q region. The dramatical feature limits application of the relevant method for Ωt calibration of a luminescent powder. To get the precise Ωt values using the relevant method, the q value must be large enough. This can be realized by adopting a thin powder plate with a low Er3+ concentration for spectral measurement.

Dependences of two factors on Judd-Ofelt (JO) intensity parameters Ωt (t = 2,4,6) have been studied theoretically by exemplifying the widely studied Er3+ ion, which is doped into Gd3Ga5O12, LiNbO3 and SrGdGa3O7 bulk single-crystals and β-NaYF4 powder. The two factors include 1) selection of induced electric-dipole transitions for JO calculation and 2) integrated absorption coefficient (IAC) or integrated absorption cross-section (IACS) of individual electric-dipole transition of Er3+. Absolute and relative root-mean-square methods were considered in the JO fits and consistent results were obtained. Ωt depends mainly on strong and medium-intensity transitions. These transitions must be all included in the fit. The exclusion of an individual transition from the fit leads to deviation of Ωt. From the matrix elements [U(t)]2 values, it is possible to rate how the inclusion or not of a transition affects the Ωt. The study also shows that the Ωt changes almost linearly with the IAC or IACS of an individual transition, and the relevant linear slope may have a positive or negative sign, depending on the [U(t)]2 value. Moreover, the larger the [U(t)]2 value is, the larger the corresponding Ωt value is. In addition, the Ω2, Ω4 and Ω6 are dependent of each other because it is the summation ∑ t Ω t [ U ( t ) ] 2 instead of its component term Ω t [ U ( t ) ] 2 that has a linear relationship to the IAC or IACS.

Judd-Ofelt (J-O) spectroscopic knowledge of a rare-earth-doped luminescent material is crucial to its application. Although a large number of papers with regard to the J-O study of various rare-earth-doped luminescent materials have been reported each year, few papers presented the errors of the J-O intensity parameters Ωi (i = 2, 4, 6) and radiative probabilities evaluated from them. Present study focuses on the error evaluation of the J-O parameters and radiative probabilities. An error theory is established for the J-O analysis and radiative probability. Two error analysis methods based on root mean square of the difference either between measured and calculated oscillator strengths (δfrms) or between measured and calculated line strengths (δSrms) are studied. Explicit error expressions are presented for the J-O parameters and radiative probability. The validity of the theory is verified by applying it to widely studied Er3+, Tm3+, Ho3+ and Nd3+ ions that are doped into four single-crystals (LiNbO3, SrGdGa3O7, LiYF4 and YVO4) and a glass. The two methods are identical in nature and give similar results of errors of Ωi.

In this letter, a variety of simple 6-chloro-4-(4-substituted piperazinyl)quinazoline derivatives was prepared. Preliminary bioassays revealed that these compounds showed good antibacterial activities toward phytopathogens Ralstonia solanacearum and Xanthomonas oryzae pv. oryzae (Xoo). Among these derivatives, compounds 5a, 5d, 5e, 5f, 5p, 5q, 6b, and 6d exhibited potent inhibition effects against R. solanacearum with EC50 within 4.60–9.94 µg/mL, especially, compound 5g exerted the strongest activity with EC50 of 2.72 µg/mL; compound 6b possessed the best inhibitory activity toward Xoo with EC50 of 8.46 µg/mL. Subsequently, a good predictive three-dimensional quantitative structure–activity relationship (3D-QSAR) model was constructed via CoMFA to direct the future structural modification and optimization. Furthermore, the pathogens’ topological studies were performed to explore the possible antibacterial mechanism. Given their simple frameworks and facile synthesis, title compounds can serve as the potential antibacterial leads.

Dilution enthalpies of six α-amino acid enantiomers (l-alanine vs d-alanine, l-serine vs d-serine, and l-proline vs d-proline) in pure water were determined at T = (288.15 to 323.15) K by isothermal titration calorimetry (ITC). Homochiral enthalpic pairwise interaction coefficients (hxx) at each temperature were evaluated according to the McMillan–Mayer theory. The hxx values of α-alanine and α-proline enantiomers are all positive (hLL > hDD > 0), while those of α-serine enantiomers are all negative (hLL < hDD < 0). In both cases, the hxx values increase uniformly and gradually with elevating temperatures. A slight but significant enthalpic discrimination effect is found in the homochiral pairwise interactions of these α-amino acid enantiomers. The positive growth of hxx values indicates that increasing temperature promotes the predominance of hydrophobic interactions in the binary aqueous solutions studied.

We propose a new constructive algorithm, called HAPE3D, which is a heuristic algorithm based on the principle of minimum total potential energy for the 3D irregular packing problem, involving packing a set of irregularly shaped polyhedrons into a box-shaped container with fixed width and length but unconstrained height. The objective is to allocate all the polyhedrons in the container, and thus minimize the waste or maximize profit. HAPE3D can deal with arbitrarily shaped polyhedrons, which can be rotated around each coordinate axis at different angles. The most outstanding merit is that HAPE3D does not need to calculate no-fit polyhedron (NFP), which is a huge obstacle for the 3D packing problem. HAPE3D can also be hybridized with a meta-heuristic algorithm such as simulated annealing. Two groups of computational experiments demonstrate the good performance of HAPE3D and prove that it can be hybridized quite well with a meta-heuristic algorithm to further improve the packing quality.

The aim of this study is to investigate the effects of inhibiting Aurora-B on osteosarcoma (OS) cell malignant phenotype, phosphorylation of valosin-containing protein (VCP), and the activity of NF-κB signaling in vitro. The expressions of Aurora-B and p-VCP proteins were detected by immunohistochemistry in 24 OS tissues, and the relationship between Aurora-B and p-VCP was investigated. The results showed that there was a positive correlation between Aurora-B and p-VCP proteins. The expression of Aurora-B in human OS cell lines U2-OS and HOS cells was inhibited by specific short hairpin RNA (shRNA) lentivirus (AURKB-shRNA lentivirus, Lv-shAURKB) which targeted Aurora-B. The results showed that the phosphorylation of VCP, the activity of NF-κB signaling pathway and the malignant phenotype of OS cells were all suppressed by knockdown of Aurora-B. It indicated that the inhibition of Aurora-B alters OS cells malignant phenotype by downregulating phosphorylation of VCP and activating of the NF-κB signaling pathway in vitro.

Three hexa-nuclear substituted sandwich-type arsenotungstates, Na6[{Na(H2O)}3{Ni(Htaz)}3(AsW9O33)2]·4H2O (1), Na6[{Na(H2O)}3{Co(Htaz)}3(AsW9O33)2]·8H2O (2), and Na6[{Na(H2O)0.5}3{Mn(Htaz)}3(AsW9O33)2]·11H2O (3) (taz = 1,2,4-1H-triazole), have been synthesized and structurally characterized. Basic frameworks of 1–3 are constructed from sandwich-type [{Na(H2O)n}3{M(Htaz)}3(AsW9O33)2]6− (n = 1 or 0.5) anions and sodium cations. The transition metal ions M2+ and Na+ in the central belt are coordinated by α-[AsW9O33]9− units, taz, and water molecules to form {MO4(Htaz)} and {NaO4(H2O)m} (m = 1 or 2) segments in which M2+ ions are modified with Htaz ligands. These fragments connect alternately in an edge-sharing mode forming a six-membered ring. Such a triazole coordinated six-nuclear cluster is first observed in a sandwich-type arsenotungstate system. The compounds 1–3 exhibit good electrocatalytic activity to reduce hydrogen peroxide and antiferromagnetic interactions between the trinuclear M(II) clusters. In addition, complex 1 has a potent inhibition effect on cell proliferation of HeLa cells.

A solution based on PLC remote I/O is put forward for the production information collection and management of the automobile assembly line required. SIEMENS S7-200, S7-300PLC and IPC, LED large screen are integrated into an information system which accomplishes various information display of help application on each station, failure alarm for stopping the line, TAKT in the assembly line, periodical production plan and information display of the current completion and production information. Thus the failure can be immediately cleaned and the production status can be dynamically controlled. Through long-term running the system, it has shown that the system can timely acquire data, display them and store complete information.

In the car assembly line made from a mix of ground plate-type conveying machinery and suspension conveying machinery, a control scheme of the car body automatically conveyed from the ground conveying machinery to overhead one is proposed, and the specific implementing methods are presented. The transferring robot adopts the Cartesian coordinate system and achieves the devices of lifting car bodies in linear motion along with four degrees of freedom by using four servo motors. The control system presented in this paper consists of the SIEMENS S7-300 T-CPU PLC, SINAMICS S120 servo driver components.

Two-dimensional bin packing is encountered in various applications where small rectangular items are packed into a minimum number of large rectangular objects (bins). Aiming at an optimal area utilization, the paper presents an effective constructive heuristic approach to two-dimensional bin packing. The heuristic approach integrates ranking, placement and search strategies along with an effective handling method of the remaining areas during the packing process. In order to obtain an optimal arrangement in a given area, all items are examined for possible positions and orientations using the search strategies. In addition, an effective handling method of the remaining areas is applied through appropriate partitioning and merging to minimize waste areas. Tests with a number of standard test and real world instances have shown that the performance of the proposed approach is superior to that of other approaches published.

Dilution enthalpies of two pairs of α-amino acid enantiomers, namely l -proline vs d -proline, and l -hydroxyproline vs d -hydroxyproline, in water-rich regions of dimethyl formamide (DMF) + H2O and dimethylsulfoxide (DMSO) + H2O mixtures (mass fractions of cosolvents wCOS = 0 to 0.30) have been determined respectively at 298.15 K by isothermal titration calorimetry (ITC). The successive values of dilution enthalpy obtained in a single run of ITC determination were used to calculate homochiral enthalpic pairwise interaction coefficients (hxx) at the corresponding composition of mixed solvents according to the McMillan–Mayer’ statistical thermodynamic approach. The sign and magnitude of hxx were interpreted in terms of solute–solute interactions mediated by solvent and cosolvent molecules, and preferential configurations of homochiral pairwise interactions ( l – l or d – d pairs) in aqueous solutions. The variations of hxx with wCOS were considered to be dependent greatly on the competition equilibrium between hydrophobic and hydrophilic interactions, as well as the structural alteration of water caused by the two highly polar aprotic cosolvents (DMF and DMSO). Especially, it was found that when one of the two kinds of interactions (hydrophobic or hydrophilic interactions) preponderates over the other in solutions, enthalpic effect of homochiral pairwise interactions is always remarkable, and is characterized by a large absolute value of hxx, positive or negative, which corresponds respectively to the prevailing interactions of hydrophobicity or hydrophilicity. Interestingly, it was also found that the hxx values of l -enantiomers are generally larger than those of d -enantiomers across the whole studied composition range of mixed solvents, i.e. |hLL| > |hDD|, which is defined as enthalpic discrimination effect and attributed to the influence of preferential configuration on homochiral pairwise interactions in aqueous solutions.

The dilution enthalpies of four 2′-deoxynucleosides, namely 2′-deoxyuridine (dU), 2′-deoxycytidine (dC), 2′-deoxyguanosine (dG) and 2′-deoxythymidine (dT), in dimethylformamide (DMF) + water mixtures have been determined at 298.15 K respectively, using an isothermal titration calorimeter (ITC200 MicroCal). On the basis of the McMillan–Mayer theory, the homotactic enthalpic pairwise interaction coefficients ( h XX ) of the four 2′-deoxynucleosides in DMF + water mixtures of various mass fractions ( w DMF = 0–0.30) have been calculated. It was found that the values of h XX of the four 2′-deoxynucleosides are all large negative and increase gradually across the whole studied composition range of the mixed solvent, though the variation degrees of them are somewhat different. The results are discussed from the point of view of solute–solute and solute–solvent interactions in solutions.

Long-term use of antiplatelet agents, statins, beta-blockers and angiotensin-converting enzyme inhibitors are proven therapies for secondary prevention in acute coronary syndrome. However, little is known of physicians' opinion about their use in China.In 2010, standard questionnaires were posted to chief cardiologists in 1397 geographically diverse and representative sample of tertiary and secondary hospitals in China, collecting information about their opinions on the recommended treatments for acute coronary syndrome.A total of 1009 (72%) cardiologists responded with a valid questionnaire. Of them, 77% reported routine use of all the four main secondary preventive treatments, with little difference between secondary and tertiary hospitals (75% vs. 79%, P = 0.16). Of the cardiologists reporting routine or selective use of aspirin, beta-blockers, statins and angiotensin-converting enzyme inhibitors, 94%, 85%, 73% and 86% would recommend indefinite use respectively. For all the 4 treatments combined, only 63.5% reported indefinite use at the same time, with no significant difference between secondary and tertiary hospitals (62% vs. 65%, P = 0.17), nor between hospitals with or without fast track for primary percutaneous coronary intervention (66% vs. 61%, P = 0.50).Although Chinese cardiologists seem well informed about the value of main secondary preventive treatments for acute coronary syndrome, there is still gap in their understanding of the need for combined and prolonged use of these treatments.

This study investigated current use of ACEI/ARB among high risk patients with coronary heart disease (CHD) in China and factors affecting ACEI/ARB use in these patients.This cross-sectional survey was performed between June to December 2007 and May to November 2009 in 51 hospitals from 14 cities. The characteristics of patients with established CHD were collected by electronic questionnaire.Only 45.8% high risk CHD patients were taking ACEI/ARB and the ACEI/ARB medication decreased significantly with time after initial CHD diagnosis. ACEI/ARB was taken in 46.1% CHD patients complicated with diabetes mellitus and in 56.3% CHD patients complicated with hypertension. Logistic regression analysis showed that comorbid hypertension was the strongest factor associated with ACEI/ARB use. In addition, male gender, history of myocardial infarction (MI), PCI and the time after initial CHD diagnosis were independent factors affecting the use of ACEI/ARB. Captopril was the most commonly prescribed ACEI in this cohort.ACEI/ARB is underused in secondary prevention among high risk CHD patients in China. It remains a major challenge for healthcare professionals and policy makers to make efforts on narrowing the gap between evidence and practice.

Despite considerable improvements in the care of patients with cardiovascular disease in various populations over the last few decades, there are still limited data about long-term treatment patterns among patients with various atherosclerotic vascular conditions in China, especially the use of statin therapy.Between June 2007 and October 2009, 16 860 patients aged 50 - 80 years with established history of atherosclerotic vascular disease (coronary heart disease (CHD), atherosclerotic cerebrovascular disease (CVD), or peripheral arterial disease (PAD)) from 51 hospitals in 14 cities of China were screened for a large randomized trial. Detailed information about current use of statins and various other treatments was recorded and analyzed by prior disease history, adjusting for various baseline characteristics.Among the 16 860 patients, the mean age was 63 years and 74% were male. Overall, 78% of the patients had documented CHD, 40% had CVD, 5% had PAD and 21% reported more than one condition. The median time from initial diagnosis of vascular disease to screening was 18 months. At screening, the proportions who took various treatments were 83% for antiplatelet agents, 49% for beta-blockers, 47% for statins and 28% for angiotensin-converting enzyme inhibitors. The proportion treated with statin was much higher in CHD than in CVD or PAD patients (61% vs. 10% vs. 22% respectively) and decreased significantly with time from initial diagnosis. Simvastatin (mainly 20 mg) and atorvastatin (mainly 10 mg) each accounted for about 40% of total statin use.In urban China, there is still significant underuse of various proven secondary preventive therapies, with particularly low use of statins in patients with ischaemic stroke.

Well designed randomized trials and meta-analyses have clearly shown that statins reduce the risk of major vascular events, including ischemic stroke, in a wide range of patients and their routine use is recommended for patients at an increased cardiovascular risk. Survivors of a first ischemic stroke are at a risk of recurrence and of other vascular events and statins are generally recommended to reduce this risk. This study investigated how widely statins were being used for such patients in China.Totally 6422 patients with a history of ischemic stroke were identified by reviewing medical records at 51 highest-ranking hospitals in 14 cities in China, and invited to attend a screening clinic to assess eligibility for a randomized trial. Their current statin and other drug uses were recorded alongside clinical and demographic characteristics. Univariate chi-square test and multivariate Logistic regression were used to determine the factors associated with treatment.Only 24% of these patients reported currently taking a statin. The most important predictor of statin use among these patients was prior history of coronary heart disease. History of diabetes or hypertension, as well as treated in university affiliated hospitals is related to increased use. The status had improved significantly during a 2-year period. Atorvastatin (40%) and simvastatin (39%) were the most commonly used.In China, statins are still underused for secondary prevention among survivors of ischemic stroke. Reasons for this poor use need to be understood in order to increase use of these evidence based therapies.

Feature selection is a key problem to pattern recognition and machine learning, and it is difficult to get the optimal feature subset for its NP-hard. Currently, the dimensionality of feature set or instance set is very high in many applications, such as information retrieval, so the feature selection from high-dimensional data is also an urgent task for researchers. This paper presents a new approach, which is a two-level filter model system integrating the relief and a newly developed algorithm of feature cluster, to reduce the dimensionality of large-scale feature set via the feature correlation (relevance) including the feature-feature correlation and feature-class correlation. Our major contributions are: (1) to present a system to perform feature selection from high-dimensional data; (2) to analyze the change of system architecture according to the time cost of the parts in the system; (3) to summarize and comment on the calculations of feature correlation; (4) to perform experiments to show the effective of the proposed approach, which has shown that the system can efficiently get a better compromise between dimensionality reduction and accuracy rate of classification than just part of the system. In many cases, it can improve the accuracy rate and dimensionality reduction.