检索结果-南通市图书馆

i.ternal Combusti.n Engi.e as a New Source for Enhanci.g Di.tri.uti.n System Resi.i.nce

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2021年第5期9卷 1130-1136页

作者： Ahad Abessi.Shahram jadid Department of Electri.al Engi.eeri.g Iran University of Science and TechnologyTehranIran

Enhanci.g di.tri.uti.n system resi.i.nce i. a new challenge for *** di.tri.uti.n loads,especi.lly the resi.enti.l customers and hi.h-pri.ri.y loads after di.asters,i. vi.al for thi. *** thi. paper,the i.ternal combusti.n engi.e(i.E)vehi.les are fi.stly i.troduced as valuable backup energy sources i. the aftermath of di.asters and the use of thi. technology i. ***,the i.provement of di.tri.uti.n system resi.i.nce i. i.vesti.ated through supplyi.g smart resi.enti.l customers and i.jecti.g extra power to the mai. *** thi. method,i. i. assumed that the i.frastructure of di.tri.uti.n system i. parti.lly damaged(common cases)and i. can be restored i. less than one *** extra power of resi.enti.l customer can be deli.ered to other loads.A novel formulati.n for i.creasi.g the i.jected power of the smart home to the mai. gri. usi.g i.E vehi.les i. ***,the maxi.um backup durati.n i. case of extensi.e damages i. the di.tri.uti.n system i. calculated for some commerci.l i.E *** thi. case,the smart home cannot deli.er extra energy to the mai. gri. because of i.s *** results demonstrate the effecti.eness of the proposed method for i.creasi.g backup power duri.g power *** i. also shown that i.E vehi.les can supply resi.enti.l customers for a reasonable amount of ti.e duri.g a power outage.

关键词： i.ternal combusti.n engi.e(i.E) smart home bui.di.g energy management system(BEMS) di.tri.uti.n system resi.i.nce

Opti.al Day-ahead Dynami. Pri.i.g of Gri.-connected Resi.enti.l Renewable Energy Resources Under Di.ferent Meteri.g Mechani.ms

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Journal of Modern Power Systems and Clean Energy 2023年第1期11卷 168-178页

作者： Ki.i. Parandeh Abed Bagheri.Shahram jadid Center of Excellence for Power System Automati.n and Operati.n School of Electrical EngineeringIran University of Science and Technology(IUST)TehranIran

Nowadays,gri.-connected renewable energy resources have wi.espread appli.ati.ns i. the electri.i.y ***,provi.i.g household consumers wi.h photovoltai.(PV)systems requi.es bi.ateral i.terfaces to exchange energy and *** addi.i.n,resi.enti.l consumers’contri.uti.n requi.es guaranteed pri.acy and secured data *** dynami. pri.i.g i. one of the i.centi.e-based demand response methods that has substanti.l effects on the i.tegrati.n of renewable energy resources wi.h smart gri.s and soci.l *** meteri.g mechani.ms of renewable energy resources such as feed-i. tari.fs,net meteri.g,and net purchase and sale are i.portant i.sues i. power gri. operati.n *** thi. paper,opti.al condi.i.n decomposi.i.n method i. used for dayahead dynami. pri.i.g of gri.-connected resi.enti.l renewable energy resources under di.ferent meteri.g mechani.ms:feed-i.tari.fs,net meteri.g,and net purchase and sale i. conjuncti.n wi.h carbon emi.si.n *** to the stochasti. nature of consumers’load and PV system products,uncertai.ti.s are consi.ered i. a two-stage deci.i.n-maki.g *** results demonstrate that the net meteri.g wi.h the sati.facti.n average of 68%for consumers and 32%for the i.vesti.ated electri. company leads to 28%total load *** the case of net purchase and sale mechani.m,a sati.facti.n average of 15%for consumers and 85%for the electri. company results i. 11%total load *** feed-i.-tari.f mechani.m,i. spi.e of i.creased soci.l welfare,load reducti.n does not take place.

关键词： Dynami. pri.i.g renewable energy resources feed-i.-tari.fs net meteri.g net purchase and sale stochasti. deci.i.n-maki.g

Subspace Clusteri.g i. Hi.h-Di.ensi.nal Data Streams:A Systemati. Li.erature Revi.w

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Computers, Materi.ls & Conti.ua 2023年第5期75卷 4649-4668页

作者： Nur Lai.a Ab Ghani.i.zatdi. Abdul Azi. Sai. jadid AbdulKadi. Department of Computer and i.formati.n Sci.nces Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Centre for Research i. Data Sci.nce(CeRDAS) Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia

Clusteri.g hi.h di.ensi.nal data i. challengi.g as data di.ensi.nali.y i.creases the di.tance between data poi.ts,resulti.g i. sparse regi.ns that degrade clusteri.g *** clusteri.g i. a common approach for processi.g hi.h-di.ensi.nal data by fi.di.g relevant features for each cluster i. the data *** clusteri.g methods extend tradi.i.nal clusteri.g to account for the constrai.ts i.posed by data *** streams are not only hi.h-di.ensi.nal,but also unbounded and *** necessi.ates the development of subspace clusteri.g algori.hms that can handle hi.h di.ensi.nali.y and adapt to the uni.ue characteri.ti.s of data *** many arti.les have contri.uted to the li.erature revi.w on data stream clusteri.g,there i. currently no speci.i. revi.w on subspace clusteri.g algori.hms i. hi.h-di.ensi.nal data ***,thi. arti.le ai.s to systemati.ally revi.w the exi.ti.g li.erature on subspace clusteri.g of data streams i. hi.h-di.ensi.nal streami.g *** revi.w follows a systemati. methodologi.al approach and i.cludes 18 arti.les for the fi.al *** analysi. focused on two research questi.ns related to the general clusteri.g process and deali.g wi.h the unbounded and evolvi.g characteri.ti.s of data *** mai. fi.di.gs relate to si. elements:clusteri.g process,cluster search,subspace search,synopsi. structure,cluster mai.tenance,and evaluati.n *** algori.hms use a two-phase clusteri.g approach consi.ti.g of an i.i.i.li.ati.n stage,a refi.ement stage,a cluster mai.tenance stage,and a fi.al clusteri.g *** densi.y-based top-down subspace clusteri.g approach i. more wi.ely used than the others because i. i. able to di.ti.gui.h true clusters and outli.rs usi.g projected *** algori.hms i.pli.i.ly adapt to the evolvi.g nature of the data stream by usi.g a ti.e fadi.g functi.n that i. sensi.i.e to *** work can focus on the clusteri.g framework,parameter opti.i.a

关键词： Clusteri.g subspace clusteri.g projected clusteri.g data stream stream clusteri.g hi.h di.ensi.nali.y evolvi.g data stream concept dri.t

A Novel Wrapper-Based Opti.i.ati.n Algori.hm for the Feature Selecti.n and Classi.i.ati.n

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materi.ls & Conti.ua 2023年第3期74卷 5799-5820页

作者： Noureen Talpur Sai. jadid Abdulkadi. Mohd Hi.mi.Hasan Hi.ham Alhussi.n Ayed Alwadai. Computer and i.formati.n Sci.nces Department Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Computer Sci.nce Department Community CollegeKing Saud UniversityRiyadh145111Saudi Arabia

Machi.e learni.g(ML)practi.es such as classi.i.ati.n have played a very i.portant role i. classi.yi.g di.eases i. medi.al *** medi.al sci.nce i. a sensi.i.e fi.ld,the pre-processi.g of medi.al data requi.es careful handli.g to make quali.y cli.i.al ***,medi.al data i. consi.ered hi.h-di.ensi.nal and complex data that contai.s many i.relevant and redundant *** factors i.di.ectly upset the di.ease predi.ti.n and classi.i.ati.n accuracy of any ML *** address thi. i.sue,vari.us data pre-processi.g methods called Feature Selecti.n(FS)techni.ues have been presented i. the ***,the majori.y of such techni.ues frequently suffer from local mi.i.a i.sues due to large soluti.n ***,thi. study has proposed a novel wrapper-based Sand Cat SwarmOpti.i.ati.n(SCSO)techni.ue as an FS approach to fi.d opti.um features from ten benchmark medi.al *** SCSO algori.hm repli.ates the hunti.g and searchi.g strategi.s of the sand cat whi.e havi.g the advantage of avoi.i.g local opti.a and fi.di.g the i.eal soluti.n wi.h mi.i.al control ***,K-Nearest Nei.hbor(KNN)classi.i.r was used to evaluate the effecti.eness of the features i.enti.i.d by the proposed SCSO *** performance of the proposed SCSO algori.hm was compared wi.h si. state-of-the-art and recent wrapper-based opti.i.ati.n algori.hms usi.g the vali.ati.n metri.s of classi.i.ati.n accuracy,opti.um feature si.e,and computati.nal cost i. *** si.ulati.n results on the benchmark medi.al datasets revealed that the proposed SCSO-KNN approach has outperformed comparati.e algori.hms wi.h an average classi.i.ati.n accuracy of 93.96%by selecti.g 14.2 features wi.hi. 1.91 ***,the Wi.coxon rank test was used to perform the si.ni.i.ance analysi. between the proposed SCSOKNN method and si. other algori.hms for a p-value less than *** fi.di.gs revealed that the proposed algori.hm produces better outcomes wi.h an average p-value o

关键词： Machi.e learni.g opti.i.ati.n feature selecti.n classi.i.ati.n medi.al data

Hyper-Parameter Opti.i.ati.n of Semi.Supervi.ed GANs Based-Si.e Cosi.e Algori.hm for Multi.edi. Datasets

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materi.ls & Conti.ua 2022年第10期73卷 2169-2186页

作者： Anas Al-Ragehi.Sai. jadid Abdulkadi. Amgad Muneer Safwan Sadeq Qasem Al-Tashi Computer and i.formati.n Sci.nces Department Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Centre for Research i. Data Sci.nce(CERDAS) Universiti Teknologi PETRONASSeri Iskandar32610Malaysia Mechani.al Engi.eeri.g Department Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Department of i.agi.g Physi.s The University of Texas MD Anderson Cancer CenterHoustonTXUSA Uni.ersi.y of Albaydha AlbaydhaYemen

Generati.e Adversari.l Networks(GANs)are neural networks that allow models to learn deep representati.ns wi.hout requi.i.g a large amount of trai.i.g ***-Supervi.ed GAN Classi.i.rs are a recent i.novati.n i. GANs,where GANs are used to classi.y generated i.ages i.to real and fake and multi.le classes,si.i.ar to a general multi.class ***,GANs have a sophi.ti.ated desi.n that can be challengi.g to *** i. because obtai.i.g the proper set of parameters for all models-generator,di.cri.i.ator,and classi.i.r i. *** a result,trai.i.g a si.gle GAN model for di.ferent datasets may not produce sati.factory ***,thi. study proposes an SGAN model(Semi.Supervi.ed GAN Classi.i.r).Fi.st,a baseli.e model was *** model was then enhanced by leveragi.g the Si.e-Cosi.e Algori.hm and Syntheti. Mi.ori.y Oversampli.g Techni.ue(SMOTE).SMOTE was used to address class i.balances i. the dataset,whi.e Si.e Cosi.e Algori.hm(SCA)was used to opti.i.e the wei.hts of the classi.i.r *** opti.al set of hyperparameters(learni.g rate and batch si.e)were obtai.ed usi.g gri. manual *** well-known benchmark datasets and a set of evaluati.n measures were used to vali.ate the proposed *** proposed method was then compared agai.st exi.ti.g models,and the results on each dataset were recorded and demonstrated the effecti.eness of the proposed *** proposed model successfully showed i.proved test accuracy scores of 1%,2%,15%,and 5%on benchmarki.g multi.edi. datasets;Modi.i.d Nati.nal i.sti.ute of Standards and Technology(MNi.T)di.i.s,Fashi.n MNi.T,Pneumoni. Chest X-ray,and Faci.l Emoti.n Detecti.n Dataset,respecti.ely.

关键词： Generati.e adversari.l networks semi.supervi.ed generati.e adversari.l network si.e-cosi.e algori.hm SMOTE pri.ci.al component analysi. gri. search

Energy Pri.e Forecasti.g Through Novel Fuzzy Type-1 Membershi. Functi.ns

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materi.ls & Conti.ua 2022年第10期73卷 1799-1815页

作者： Muhammad Hamza Azam Mohd Hi.mi.Hasan Azli.da A Mali. Sai.a Hassan Sai. jadid Abdulkadi. Centre of Research i. Data Sci.nce Computerand Information Sciences DepartmentUniversiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Petroleum Engi.eeri.g Department Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia i.sti.ute of Computi.g Kohat University of Science and TechnologyKohat26000K.P.KPakistan

Electri.i.y pri.e forecasti.g i. a subset of energy and power forecasti.g that focuses on projecti.g commerci.l electri.i.y market present and future *** pri.e forecasti.g have been a cri.i.al i.put to energy corporati.ns’strategi. deci.i.n-maki.g systems over the last 15 *** strategi.s have been uti.i.ed for pri.e forecasti.g i. the past,however Arti.i.i.l i.telli.ence Techni.ues(Fuzzy Logi. and ANN)have proven to be more effi.i.nt than tradi.i.nal techni.ues(Regressi.n and Ti.e Seri.s).Fuzzy logi. i. an approach that uses membershi. functi.ns(MF)and fuzzy i.ference model to forecast future electri.i.y *** c-means(FCM)i. one of the popular clusteri.g approach for generati.g fuzzy membershi. ***,the fuzzy c-means algori.hm i. li.i.ed to produci.g only one type of MFs,Gaussi.n *** generati.n of vari.us fuzzy membershi. functi.ns i. cri.i.al si.ce i. allows for more effi.i.nt and opti.al problem *** a result,for the best and most i.proved results for electri.i.y pri.e forecasti.g,an approach to generate multi.le type-1 fuzzy MFs usi.g FCM algori.hm i. ***,the objecti.e of thi. paper i. to propose an approach for generati.g type-1 fuzzy tri.ngular and trapezoi.al MFs usi.g FCM algori.hm to overcome the li.i.ati.ns of the FCM *** approach i. used to compute and i.prove forecasti.g accuracy for electri.i.y pri.es,where Australi.n Energy Market Operator(AEMO)data i. *** results show that the proposed approach of usi.g FCM to generate type-1 fuzzy MFs i. effecti.e and can be adopted.

关键词： Fuzzy logi. fuzzy C-means type-1 fuzzy membershi. functi.n electri.i.y pri.e forecasti.g

Supplementi.g Conventi.nal Treatment wi.h Pycnogenol(R) May i.prove Hepati.i. C Vi.us-Associ.ted Type 2 Di.betes:A Mi.i.Revi.w

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Cli.i.al and Translati.nal Hepatology 2016年第3期4卷 228-233页

作者： Sayeh Ezzi.ouri.Fati.a Zahra jadid Salsabi. Hamdi.Lahcen Wakri. Kyoko Tsuki.ama-Kohara Soumaya Benjelloun Vi.ology Uni. Viral Hepatitis Laboratory Institut Pasteur du Maroc Casablanca Morocco Vi.ology Uni. Immunovirology Laboratory Institut Pasteur du Maroc Casablanca Morocco Laboratory of Ani.al Hygi.ne Joint Faculty of Veterinary MedicineKagoshima University Kagoshima Japan Transboundary Ani.al Di.eases Centre Joint Faculty of Veterinary MedicineKagoshima University Kagoshima Japan

Hepati.i. C vi.us (HCV) i.fecti.n and type 2 di.betes melli.us (T2DM) present a si.ni.i.ant health burden,wi.h i.creasi.g compli.ati.ns and mortali.y rates ***(R) (PYC),a natural product,possesses anti.i.beti. and anti.i.al properti.s that may i.prove HCV-associ.ted *** thi. revi.w,we present previ.usly publi.hed data on the effecti.eness of PYC agai.st HCV repli.ati.n and *** beli.ve that supplementi.g conventi.nal treatment wi.h PYC may i.prove the current HCV therapy,attenuate HCV-associ.ted T2DM,and reduce the ri.k of compli.ati.ns such as ci.rhosi. or hepatocellular carci.oma and cardi.vascular di.ease.

关键词： T2DM Hepati.i. C Pycnogenol(R) Complementary medi.i.e

Type 2 Di.betes Ri.k Predi.ti.n Usi.g Deep Convoluti.nal Neural Network Based-Bayesi.n Opti.i.ati.n

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materi.ls & Conti.ua 2023年第5期75卷 3223-3238页

作者： Alawi.Alqushai.i.Mohd Hi.mi.Hasan Sai. jadid Abdulkadi. Amgad Muneer Mohammed Gamal Qasem Al-Tashi.Shaki.ah Mohd Tai. Hi.ham Alhussi.n Computer&i.formati.n Sci.nces Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Centre for Research i. Data Sci.nce(CERDAS) Universiti Teknologi PETRONASSeri Iskandar32610PerakMalaysia Department of i.agi.g Physi.s University of Texas MD Anderson CancerHoustonTXUSA

Di.betes melli.us i. a long-term condi.i.n characteri.ed by *** could lead to plenty of *** to ri.i.g morbi.i.y i. recent years,the world’s di.beti. pati.nts wi.l exceed 642 mi.li.n by 2040,i.plyi.g that one out of every ten persons wi.l be *** i. no doubt that thi. startli.g fi.ure requi.es i.medi.te attenti.n from i.dustry and academi. to promote i.novati.n and growth i. di.betes ri.k predi.ti.n to save i.di.i.uals’*** to i.s rapi. development,deep learni.g(DL)was used to predi.t numerous ***,DLmethods sti.l suffer from thei. li.i.ed predi.ti.n performance due to the hyperparameters selecti.n and parameters ***,the selecti.n of hyper-parameters i. cri.i.al i. i.provi.g classi.i.ati.n *** study presents Convoluti.nal Neural Network(CNN)that has achi.ved remarkable results i. many medi.al domai.s where the Bayesi.n opti.i.ati.n algori.hm(BOA)has been employed for hyperparameters selecti.n and parameters *** i.sues have been i.vesti.ated and solved duri.g the experi.ent to enhance the *** fi.st i. the dataset class i.balance,whi.h i. solved usi.g Syntheti. Mi.ori.y Oversampli.g Techni.ue(SMOTE)*** second i.sue i. the model’s poor performance,whi.h has been solved usi.g the Bayesi.n opti.i.ati.n *** fi.di.gs i.di.ate that the Bayesi.n based-CNN model superbases all the state-of-the-art models i. the li.erature wi.h an accuracy of 89.36%,F1-score of 0.88.6,andMatthews Correlati.n Coeffi.i.nt(MCC)of 0.88.6.

关键词： Type 2 di.betes di.betes melli.us convoluti.nal neural network Bayesi.n opti.i.ati.n SMOTE

A comprehensi.e revi.w of exi.ti.g corpora and methods for creati.g annotated corpora for event extracti.n tasks

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Data and i.formati.n Sci.nce 2024年第4期9卷 196-238页

作者： Mohd Hafi.ul Afi.i.Abdullah Norshaki.ah Azi. Sai. jadid Abdulkadi. Kashi. Hussai. Hi.ham Alhussi.n Noureen Talpur Centre for Cyber-Physi.al Systems(C2PS) Universiti Teknologi PETRONAS32610 Seri IskandarMalaysia Centre for Research i. Data Sci.nce(CeRDaS) Universiti Teknologi PETRONAS32610 Seri IskandarMalaysia Department of Sci.nce and Engi.eeri.g Solent UniversitySO140YNUK

Purpose:The purpose of thi. study i. to serve as a comprehensi.e revi.w of the exi.ti.g annotated *** revi.w study ai.s to provi.e i.formati.n on the exi.ti.g annotated corpora for event extracti.n,whi.h are li.i.ed but essenti.l for trai.i.g and i.provi.g the exi.ti.g event extracti.n *** addi.i.n to the pri.ary goal of thi. study,i. provi.es gui.eli.es for prepari.g an annotated corpus and suggests sui.able tools for the annotati.n ***/methodology/approach:Thi. study employs an analyti.al approach to exami.e avai.able corpus that i. sui.able for event extracti.n *** offers an i.-depth analysi. of exi.ti.g event extracti.n corpora and provi.es systemati. gui.eli.es for researchers to develop accurate,hi.h-quali.y *** ensures the reli.bi.i.y of the created corpus and i.s sui.abi.i.y for trai.i.g machi.e learni.g ***:Our explorati.n reveals a scarci.y of annotated corpora for event extracti.n *** parti.ular,the Engli.h corpora are mai.ly focused on the bi.medi.al and general *** the i.sue of annotated corpora scarci.y,there are several hi.h-quali.y corpora avai.able and wi.ely used as benchmark ***,access to some of these corpora mi.ht be li.i.ed owi.g to closed-access poli.i.s or di.conti.ued mai.tenance after bei.g i.i.i.lly released,renderi.g them i.accessi.le owi.g to broken ***,thi. study documents the avai.able corpora for event extracti.n *** li.i.ati.ns:Our study focuses only on well-known corpora avai.able i. Engli.h and ***,thi. study places a strong emphasi. on the Engli.h corpora due to i.s status as a global li.gua franca,maki.g i. wi.ely understood compared to other *** i.pli.ati.ns:We genui.ely beli.ve that thi. study provi.es valuable knowledge that can serve as a gui.i.g framework for prepari.g and accurately annotati.g events from text *** provi.es comprehensi.e gui.eli.es for researchers to i.prove th

关键词： i.formati.n extracti.n Event extracti.n Text mi.i.g Large language model Natural language processi.g