Intelligent Management of Electrical Systems in Industries

Intelligent Management Of Electrical Systems in Industries Industries

S.R.M.S.C.E.T Bareilly

SEMINAR REPORT ON

Intelligent Management of Electrical Systems in Industries

Submitted to Mr. Mr. Vinee ineett Sr Sriv ivas asta tava va

Submitted by Sadd Saddam am Hu Hussa ssain in 1101!10"

1


A#stra$t

The automation of public electricity distribution has developed very rapidly in the past few years. The same basis can be used to develop new intelligent applications for electricity distribution networks in industrial plants. Many Many new appli applica cati tion onss have have to be intr introd oduc uced ed beca becaus usee of the the diff differ eren entt envi enviro ronm nmen entt and needs needs in indu indust stri rial al sect sector or.. The The paper paper incl include udess a syst system em description of industrial electric system management. The paper discusses on the requi require reme ment ntss of new new appl applic icat atio ions ns and and meth methods ods that that can be used used to solv solvee problems in the areas of distribution management and condition monitoring of industrial networks.

2


CONTENTS

1 Introduction …………………..…………………….………………... ! 2 "pplications "pplications for supporting the public distribution network management ................................................ ................................................ # $ %escription of the system environment …………………………….….& ………………………… ….….& ! "pplication "pplication functions for distribution management in industrial plants ………………………………............ 11 # "dvanced %istribution "utomation "utomation ………………...............................………………....……..1! #.1 %istribution 'ystem of (uture with "%" ………………………………………….….1) * %istribution Management (unctions …...............……………………………….…....1& )"pplication (unctions of %ata Management 'ystems ………………………………........................…......…...…..21 ).1+ ,oad modeling modeling ………..……….…................................ ………..……….….........................................21 .........21 ).2+ -eliability management………………………….........……..2$ ).$+ oltage dip analyses.........……………........……......... analyses.........……………........……...............…2# ......…2# ).!+ /ower quality analyses……………………………................2* ).#+ 0ondition monitoring…………………………………..........2* & 0onclusion…...................................... 0onclusion….......................................……………..…………….…...2 .……………..…………….…...2  ibliography.……………….............. ibliography.……………….................……………………................$ ...……………………................$

$


Intr%du$ti%n

Indus Industr tria iall plan plants ts have have put put cont contin inuo uous us pres pressu sure re on the the advan advance ced d proce process ss automa automatio tion. n. 3owever 3owever44 there there has not been been so much much focus focus on the automati automation on of the electricity distribution networks. "lthough4 the uninterrupted electricity distribution is one basic basic requir requireme ement nt for the proces process. s. " disturba disturbance nce in electr electrici icity ty supply supply causin causing g the5downrun6 of the process may cost huge amount of money. Thus the intelligent management management of electricit electricity y distributi distribution on including4 including4 for e7ample4 e7ample4 preventive preventive condition condition monitoring monitoring and on8line on8line reliabili reliability ty analysis analysis has a great importance. importance. 9owadays the above needs have aroused the increased interest in the electricity distribution automation of industrial plants. The automation of public electricity distribution has developed very rapidly in the past few years. ery promising results has been gained4 for e7ample4 in decreasing outage times of customers. 3owever4 the same concept as such cannot be applied in the field of industrial electricity distribution4 although the bases of automation systems are common. The infrastructures of different industry plants vary more from each other as compared to the public electricity distribution4 which is more homogeneous domain. The automation devices4 computer systems4 and databases are not in the same level and the integration of them is more complicated.

!


"pplications for supporting the public distribution network management

It was seen already in the end of &:s that the conventional automation system ;i.e. '0"%"+ cannot solve all the problems regarding to network operation. I'+ include vast amount of data which is useful in network operation. The operators had considerable heuristic knowledge to be utili?ed4 too. Thus new tools for practical problems were called for4 to which "I8 based methods ;e.g. ob@ect8oriented approach4 rule8based technique4 uncertainty modeling and fu??y sets4 hyperte7t technique4 neural networks and genetic algorithms+ offers new problem solving methods. 'o far a computer system entity4 entity4 called as a distribution management system ;%M'+4 has been developed. The %M' is a part of an integrated environment composed of the '0"%"4 distribution automation ;e.g. microprocessor8 based protection relays+4 the network database ;i.e. "M=(M=>I'+4 the geographical database4 the customer database4 and the automatic telephone answering machine system. The %M' includ includes es many many intell intellige igent nt applica applicatio tions ns needed needed in networ network k operati operation. on. 'uch 'uch applica applicatio tions ns are4 are4 for e7ampl e7ample4 e4 norma normall state8 state8mon monito itorin ring g and optimi optimi?at ?ation ion44 real8t real8tim imee netwo network rk calc calcul ulat atio ions ns44 shor shortt term term load load fore foreca cast stin ing4 g4 swit switch chin ing g plan planni ning ng44 and and faul faultt management.

#


The core of the whole %M' is the dynamic ob@ect8oriented network model. The distribution network is modeled as dynamic ob@ects which are generated based on the network data read from the network database. The network model includes the real8t real8time ime state state of the networ network k ;e.g. ;e.g. topolo topology gy and loads+ loads+.. %iffe %ifferen rentt networ network k operat operation ion tasks tasks call call for differ different ent kinds kinds of proble problem m solvin solving g methods methods.. arious rious modules can operate interactively with each other through the network model4 which works as a blackboard ;e.g. the results of load flow calculations are stored in the network model4 where they are available in all other modules for different purposes+.The present %M' is a Aindows Aindows 9T 8program implemented by isual isual 0BB. 0BB. The The prot prototy otypi ping ng mean meantt the the iter iterat atio ion n loop loop of know knowle ledg dgee acqui acquisi siti tion on44 modeling4 implementation4 and testing. /rototype versions were tested in a real enviro environme nment nt from from the very beginning beginning.. Thus Thus the feedba feedback ck on new infere inference nce models4 e7ternal connections4 and the user8interface was obtained at a very early stage. The aim of a real application in the technical sense was thus been achieved. The %M' entity was tested in the pilot company4 Coillis8'atakunnan 'DhkE
*


The e7periences as a whole were so encouraging that the %M' was modified as a commercial product. The vendor was first a small (innish software company. 'ince 1) the %M' has been a worldw worldwide ide software software product product of " Trans Transmit mit
)


&es$ri'ti%n &es$ri'ti%n %( t)e system envir%nment " big industrial plant differs differs from public distribution distribution company by organi?atory organi?atory structure and by system environment. " production is divided into many departments or many companies. These units have the responsibility of production and maintenance. ery ery often the maintenance is maintained by a service company. "n "n energy department or company is in charge of local energy production and of the distribution network. "bove organi?ations may have some control systems that serve for their needs only4 but usually information systems are closely connected together. " process automation system is the most important system in an industrial plant4 sometimes including other systems4 as illustrated in (ig. 1. (or e7ample4 all energy production and distribution network control tasks can be done in a process automation system. 9ormally4 because of the reliability reason reasons4 s4 vital vital parts parts of distri distribut bution ion networ network k control control is indepe independen ndentt on the proces processs automation. The independency of process automation system vendor has been one reason for separate systems4 too.

(igure1G "utomation and information systems of an industrial plant.

&

Intelligent Management Of Electrical Systems in Industries Industries The systems in (ig. 1 utili?e many databases4 which contain data that can be used in new applications. /rocess automation systems collect data for process monitoring and optimi?ation tools. The databases contain co ntain information of material flow4 energy flow and control data of production machines. Maintenance databases include technical specifications and condition data of production machine components. 'imilar information of electricity network components is supported by network database. /roduction programs are stored in the databases of administrative systems. systems. Intelligent applications are needed toG 8 3andle large amount of information available. This includes filtering of data and producing new information by collecting data. 8 Illustrate comple7 dependencies of electricity distribution and production processes in abnormal situations. 8 >ive instructions for operators in fault situations. " ri risk sk of misoperation in unusual fault situation is obvious and prevents or delay operatorsH decision making. 8 "utomi?e analysis tasks. 0ontinuous information analysis is not possible manually.

In order to introduce new intelligent applications for the management of electric systems in industrial plants4 a basis for implementation is needed. The following requirements should be satisfiedG 8 %ocumentation of electricity distribution network is available for the systems. 9etwork databases can supply this information.



Intelligent Management Of Electrical Systems in Industries Industries 8 9etwork4 process and motor measurements are available for the system. This means4 that data acquisition from multiple sources with capability to use various data transfer methods is needed4 as illustrated in (ig. 2.

1


A''li$ati%n (un$ti%ns (%r distri#uti%n mana*ement in industrial'lants

"s mentioned above the concept of public distribution automation cannot be applied as such in the management of industrial electricity networks. (or e7ample4 fast and accurate fault location has a great importance for reducing the outage time of customers in the public electricity distribution4 while there is no special need of such a function in industrial networks. /redictive condition monitoring4 reliability calculations4 and protection relay coordination to prevent disturbances in advance are more important. 0aused by the features of industrial networks there are needs for methods to model dynamic phenomena and harmonics4 and to calculate load8flow and fault currents in ring connected networks. "n essential need is the load modeling which differs considerable from the public distribution. The basis of the distribution management system ;i.e. the use of network model as the blackboard+ is common in the both domains. The network model includes the real8time topology and network calculation results in the prevailing switching and load conditions. The main functions of system entity for the industrial networks are listed in the followingG  -eal8time network monitoring4 state estimation and optimi?ationG 8 Topology management 8 load flow and fault currents also as dynamic phenomena 8 Monitoring and compensation of reactive power 8 monitoring of harmonics and resonances

11

Intelligent Management Of Electrical Systems in Industries Industries 8 Minimi?ation of power losses  /lanning and simulation of operation o peration actions 8 switching planning 8 "utomatic load shedding and forming a local island 8 switching the network as a part of the national grid 8 fault situations  Management of disturbances 8 Jvent analysis 8 (ault location and network restoration 8 /reventive condition monitoring 8 /rotection relay coordination 8 -eliability calculations 8 reporting

%istribut %istribution ion "utoma "utomation tion which which includes includes feeder feeder automati automation on and distrib distribution ution managem management ent syste systems ms ;%M'+ ;%M'+ is an impor importan tantt techni technique que in distr distribu ibutio tion n networ network. k. The distribution management systems are composed of distribution management functions. The %M( is an entity which incorporates different applications on a single platform over which which superv supervisi ision on is made. made. This This mainly mainly suppor supports ts documen documentat tation ion of networ network k data data planning operation and

reliability management of distribution networks. arious arious

application application functions functions for distributi distribution on management in industrial industrial plants are mainly load modeli modeling ng 4relia 4reliabil bility ity managem management ent 4 power power qualit quality y analys analysis is44 voltag voltagee dip analys analysis is and condition condition monitoring monitoring ."ll this are incorporat incorporated ed in a domain domain of distribut distribution ion management

12

Intelligent Management Of Electrical Systems in Industries Industries functi functions ons.. "dvanc "dvanced ed distri distribut bution ion automa automatio tion n ;"%"+ ;"%"+ modern modern day approach approach toward towardss efficient management of distribution networks.

1$


A&VANCE& &ISTRIB+TION A+TOMATION

Traditional Traditional distribution systems were designed to perform one functionK distributing power to end users. The distribution system of the future will be more versatile and will be multifunctional. 'trategic drivers for "%" are to L Improve system performance L -educe outage times L "llow the efficient use of distributed energy resources L /rovide the customer more choices and L To integrate the customer systems (or "%" to work4 the various v arious intelligent devices must be interoperable both in the electric system architecture and in the communication and control architecture.

. (igure$G "%" architecture

1!

Intelligent Management Of Electrical Systems in Industries Industries "%" will enable the distribution system to be configured in new ways for such things as looped secondaries or intentional islanding to facilitate easy recovery from outages and to deal with other emergencies.

(igG ! The three ma@or components of "%"  (le7ible electrical system architecture  -eal8time state estimation tools  0ommunication and control system based on open architecture standards The intelligent universal transformer is a prime e7ample of a new electronic device that will be a cornerstone of "%". It will provide a variety of functions including  o oltage stepping  o oltage regulation  /ower quality enhancement  9ew customer service options such as %0 power output  /ower electronic replacement for conventional copper and iron transformers

1#


The (le7ible Jlectric "rchitecture and the
Jach of these is made more valuable by its interaction with the other.

"%" will will provide improvements in many areas including  -eliability  'ystem performance  0ondition monitoring 
1*


&istri#uti%n System %( ,uture -it) A&A A&A

1)


&ISTRIB+TION MANAEMENT ,+NCTIONS %istribut %istribution ion management management functions functions form an entity entity of application applicationss supporting supporting documentation of network data4 and planning4 operation and reliability management of distribution network in industrial plants. The functions can be included into different computer computer systems4 systems4 like "M=(M=>I'4 "M=(M=>I'4 %istribut %istribution ion Management Management 'ystem 'ystem ;%M'+4 ;%M'+4 and '0"%" or case specific customi?ed applications. The main functions of distribution management entity for the industrial networks are listed in the followingG L %ocumentation of network data L >raphical user interfaces L -eal8time network monitoring4 state estimation and optimi?ation 8 Topolo opology gy manage managemen ment4 t4 load load flow flow and fault fault curren currentt calcul calculati ation4 on4 monito monitorin ring g and compen compensa sati tion on of reac reacti tive ve powe power4 r4 moni monito tori ring ng of harm harmoni onics cs and and reso resonan nance ce44 and and minimi?ation of power losses L /lanning and simulation of operation ope ration actions 8 switching planning4 fault situations4 automatic load shedding and forming a local island L Management of disturbances and reliability 8 /reventive condition monitoring4 reliability and availability management4 protection relay coordination4 event analysis4 fault location and network restoration4 reporting. 0aused by the features of industrial networks the importance of the distribution management functions are different as in public pub lic electricity networks. There are also needs for new methods. "n essential need is the load modeling which differs considerable from the public distribution. /redictive condition monitoring4 reliability management4 and

1&


protection relay coordination to prevent disturbances in advance adv ance have a great importance. 'ome functions of the %M' for the management of public distribution networks can be applied almost as such also in the management of industrial electricity networks4 e.g. topology management.

1


APP/ICATION ,+NCTIONS O, &ATA &ATA MANAEMENT SSTEMS

1 /%ad m%delin* The essential basis for advanced application functions is the modeling of loads connected to the network. Nsually there are only few measurement points in the network. 3owever4 loading of every load node of the network must be known in the network calculations. (or that purpose the loads are estimated by load models. The essential need for the load models models is that they form a basis for the load8flow calculations. -esults of load8flow calculations are utili?ed different kind of tasks as real8 time network monitoring and optimi?ation4 and switching planning. Information on loads can also be utili?ed in preventive condition monitoring and reliability analyses. "lthough4 the loads ;i.e. the current+ of some nodes can be measured on8line4 models are needful because of the %M' can be used also in simulated state4 when the information of system does not correspond the current real8time state of the distribution network. In the domain of public electricity distribution hourly load curves have been determ determine ined d for each custom customer er group group to be used used in load8f load8flow low calculat calculation ion and load load forecasting. In industrial plants the load modeling should be based mainly on the process

2


itself and its behavior. ,oad models can be determined by making enough measurements in different known process conditions. 3owever4 the industrialplants vary from each o ther quite much4 which means that load models determined in one plant may not be able to used as such in other one.
21


(ig #G 9etwork ,oad Model %etermining

,oad forecasting in the industrial environment cannot be based on any regularity of behavior. -eliable forecasting assumes use of methods which can utili?e production plans in some time distance which also can have a large difference with each other and include inaccurate information. The load forecasting of the network feeding some process bases on the known behavior of the process4 earlier measured values and the planned production. Cal$ulati%n met)%ds (%r mes)ed net-%r2s

The %M' for public distribution management included load flow and fault current calculation calculation procedures4 procedures4 which worked worked only in radial radial networks. networks. The need for calculating calculating meshed networks in industrial distribution networks is anyway obvious ;e.g. there are several fault current sources+. ,oad flow calculation for meshed meshed network leads to a group of non8linear

22


equatio equations. ns. 0lassi 0lassicc 9ewton 9ewton8-a 8-aphs phson on itera iteratio tion n is conside considered red be the most most compet competent ent method for solving load flow equations4 and was selected as the solver. (ault current calculation is performed only in the symmetrical three8phase case. In fact4 the calculation can be done done simp simply ly by inve invert rtin ing g a matr matri7 i7.. To calc calcul ulat atee inve invers rsee of matr matri7 i7 with with conventional methods is now too laborious and therefore discarded. Instead an algorithm called O8bus algorithm is used for calculating inverse effectively. effectively. The load flow and fault current algorithms are implemented as a part of the %M' so that they can utili?e the common network model and topology analysis. The primary information for the load8flow calculation is the loads of the secondary substations and motors connected to the medium voltage network. The loading information is read from the "ccess database including the load models for different situations. The results of load flow and fault current current calculatio calculations ns can be studied studied through the user8interf user8interface ace of the %M' by selecting the desired node.

! Relia#ility mana*ement mana*ement The functions related to reliability have considerable economic significance in industry. The losses of production caused by the disturbances and the inputs into the investment investmentss of the systems systems including including maintenance and operational operational arrangements arrangements @oin here. The reliability can be studied with both qualitative and quantitative methods. Aith a qualitative analysis the possible states of the system and reasons which lead to these are determined with non8numerical methods. The failure modes4 effects and

2$


criticality analyses are adapted generally on the qualitative methods. Nsing failure modes4 effects and criticality analysis it is aimed to identify those faults of the devices or of the subsys subsystem temss which which affec affectt the capabil capabiliti ities es of the syste system m signif significa icantl ntly y. The syste system m is system systemati atical cally ly analy analy?ed ?ed and the effect effectss of the compone component nt fault faultss of the syste system m are evaluated. In a quantitative analysis indicators describing the capabilities of the system are calculated. (or e7ample4 availability4 fault frequencies4 durations of disturbances and indicators which describe the economic appreciation of interruptions can be evaluated. The functions supporting power distribution reliability management can be included in severa severall differ different ent syste systems ms which which are4 are4 among among others others44 "M=(M "M=(M=>I =>I'4 '4 the %istri %istribut bution ion Management 'ystem ;%M'+4 '0"%" system4 maintenance systems4 and documentation systems depending on the total concept.

The load flow calculations and short circuit calculations are applications which have central meaning in reliability analyses. The calculations make it possible to simulate faul faults ts44 to plan plan

rela relayi ying ng arra arrang ngem ement entss and netwo network rk opera operati tion ons. s. 'witc 'witchi hing ng plans plans

operational instructions can furthermore be stored in databases. "n essential function suppor supportin ting g reliab reliabili ility ty managem management ent and analys analyses es is also also the managem management ent of variou variouss instructions and documents. There are many kind of documents which can be used to support support the reliabili reliability ty management. management. The graphical graphical user8int user8interfac erfacee makes available the devel developi oping ng of the the diff differ eren entt soph sophis isti tica cate ted d user user frie friend ndly ly func functi tion ons4 s4 for for e7am e7ampl ple4 e4 determination of the feeding routes of the components or loads to be e7amined e7 amined

2!

Intelligent Management Of Electrical Systems in Industries Industries The estimation of the reliability technical state and capabilities of the distribution system together with real8time condition supervision and maintenance programmes are in a cent centra rall posi positi tion on in the the anti antici cipa pati ting ng and and prev prevent entio ion n of dist distur urban bance cess and in the the minimi?ation of their effects. The analysis analysis of reliabili reliability ty technical state state and capability capability of power distribution distribution network is closely related to the protection coordination4 too. Nsing fault current and load8f load8flow low calcul calculati ations ons person personnel nel can evaluat evaluatee how the distr distribu ibutio tion n and the primar primary y processes will behave in fault situations of the distribution distribution network.

3 V%lta*e di' analyses " voltage dip is a sudden reduction of the supply voltage to a value between  Fand 1 F of the declared voltage4 followed by a voltage recovery after a short period of time. /ossible causes of these dips are typically faults in installations or in feeding public networks and switching of large loads ;e.g. motors+. In rural areas voltage dips are generally caused by short circuit faults in the public M overhead network. The interest in voltage dips is mainly due to the problems they cause on several types of equipment e.g. e.g. trip trippi ping ng of ad@u ad@ust stab able le8s 8spe peed ed driv drives es ;bot ;both h ac and and dc driv drives es+4 +4 proce process ss8c 8cont ontro roll equipment4 computers and contactors in front of some devices. The employment of INT with the support of "%" is a step towards reduction in these voltage dips.

 P%-er 4uality analyses The term term /ower /ower Puali Puality ty ;/P+ ;/P+ is used used with with slight slightly ly diff differe erent nt meani meanings ngs.. More More e7tensive meaning can be associated with any problems in voltage4 current or frequency

2#

Intelligent Management Of Electrical Systems in Industries Industries deviatio deviations ns which which result result in failure4 failure4 malfunc malfunction tion44 disturban disturbances ces or combinat combination ion of voltage voltage quality and current quality. 3owever4 the voltage quality is addressed in most cases. oltage quality is concerned with deviations of the voltage from the ideal and main characteristics

can be described as with regard to frequency4 magnitude4 waveform4 symmetry of the three phase voltages and interruptions. In industrial plants on the other hand increasing amount of disturbing devices ;e.g. ad@ustable drives and power electronics+ and on the other other hand hand increa increasin sing g amount amount of sensit sensitive ive device devicess ;compu ;computer ters4 s4 proces processs automa automati tion on 4electronic devices and ad@ustable drives+ have caused growing concern about power quality. Thus Thus there is also a growing need to manage and monitor power quality. quality. Volts

5  C%nditi%n m%nit%rin* There There e7ist e7ist many many syste systems ms for condit condition ion monit monitori oring ng of indust industria riall proces processes ses44 especially for rotating machines. Monitoring usually covers electric motors that are connected to the monitored processes. There are on8line systems designed mainly for condition monitoring of electric motors4 too. These systems usually include measuring device connected with processing device4 which can be connected permanently to data bus supplying information for analy?ing computer or data can be collected from device occasionally. " selection between continuous data transfer and manually performed data collection is made mainly by the costs of instrumentation and labour. Jlectric motors are often considered to be very reliable4 which means that investment not economically @ustified.

2*

 0omponent data from the network databaseG 8 %ate of installation4 model4 and nominal life time 8 /lan for service and replacement investments 
2)

Intelligent Management Of Electrical Systems in Industries Industries 8 Topology information and estimated reliability of components in a given load situation  "nalysis ;reconstruction+ of actual faultsG 8 'imulated network state using topology4 load and voltage information of previous situation.

2&


C%n$lusi%n

-equirements of intelligent software applications for supporting the operation of industrial distribution networks are different compared to the public distribution. The domain is more segmented segmented and heterogeneou heterogeneous4 s4 and the infrastructu infrastructure re of automation automation and computer systems for electricity networks are not so sophisticated and advanced as other process automation.
2


BIBI/IORAPH

1+ Qero."46 ,oad modeling for distribution distribution management function of industrial industrial medium oltage distribution networks 54 IJJJ Transactions on Industry applications4 ol.$2 9o !4 Qanuary 21. 2+ (rank -. >oodman4 Qr.4 /h.%.6 "dvanced %istribution "utomation64 www.epri.com www.epri.com.. $+ Markku Cauppinen4 Tampere Nniversity of Technology4 Technology4 (inland 5Management of electrical systems in industrial plants64 www.energyline.com www.energyline.com . . !+ ,i@un Pin46" new principle fro system protection in distribution networks64 IJJJ transactions on power delivery4 d elivery4 o ol 14 9o !4 Qune 21. #+ Monclar (.-46 (.-46 Intelligent support system for distribution network management 54 International conference on Intelligent system application to power systems 54 'weden4 Qune 2.

$

Intelligent Management of Electrical Systems in Industries

Recommend Documents