Approximating Description Logic Classi?cation for
Semantic Web Reasoning
Perry Groot1,Heiner Stuckenschmidt2,and Holger Wache2
1Radboud University Nijmegen,Toernooiveld1,
6500GL Nijmegen,The Netherlands
Perry.Groot@science.ru.nl,
2Vrije Universiteit Amsterdam,de Boelelaan1081a,
1081HV Amsterdam,The Netherlands
{holger,heiner}@cs.vu.nl.
Abstract.In many application scenarios,the use of the Web ontology language
OWL is hampered by the complexity of the underlying logic that makes reason-
ing in OWL intractable in the worst case.In this paper,we address the question
whether approximation techniques known from the knowledge representation lit-
erature can help to simplify OWL reasoning.In particular,we carry out experi-
ments with approximate deduction techniques on the problem of classifying new
concept expressions into an existing OWL ontology using existing Ontologies on
the web.Our experiments show that a direct application of approximate deduc-
tion techniques as proposed in the literature in most cases does not lead to an
improvement and that these methods also suffer from some fundamental prob-
lems.
1Introduction and Motivation
A strength of the current proposals for the foundational languages of the Semantic Web is that they are all based on formal logic.This makes it possible to formally reason about information and derive implicit knowledge.However,this reliance on logics is not only a strength but also a weakness.Traditionally,logic has always aimed at modelling idealised forms of reasoning under idealised circumstances.Clearly,this is not what is required under the practical circumstances of the Semantic Web.Instead,the following are all needed:
–reasoning under time-pressure
–reasoning with other limited resources besides time
–reasoning that is not‘perfect’but instead‘good enough’for given tasks under given circumstances
It is tempting to conclude that symbolic,formal logic fails on all these counts,and to abandon that paradigm.Our aim is to keep the advantages of formal logic in terms of de?nitional rigour and reasoning possibilities,but at the same time address the needs of the Semantic Web.
Research in the past few years has developed methods with the above properties while staying within the framework of symbolic,formal logic.However,many of those
previously developed methods have never been considered in the context of the Seman-tic Web.Some of them have only been considered for some very simple underlying description languages [1].As the languages proposed for modelling Ontologies in the Semantic Web are becoming more and more complex,it is an open question whether those approximation methods are able to meet the practical demands of the Semantic Web.In this article,we look at approximation methods for Description Logics (DLs),which are closely related to some of the currently proposed Semantic Web languages,e.g.,OWL.
The remainder of the article is structured as follows.Section 2gives an overview of various approximation approaches and techniques useful in the context of the Semantic Web.Thereafter,the article focuses on the investigation of a particular approximation technique in the context of a particular reasoning task.Section 3describes the approx-imation approach used.Section 4describes the reasoning task focused on.Section 5gives experimental results of the approximation method applied to the classi?cation of concepts in a number of Ontologies.Section 6gives conclusions,discusses the results and the applicability of the analysed approximation approach to the Semantic Web.2Approximation Approaches
Fig.1.Architecture of a KR system based on Description Logic together with possi-ble approximation approaches.A typical architecture for a KR system based on DLs can be sketched as in Fig-ure 1[2],which contains three compo-
nents that can be approximated to obtain a simpli?ed system that is more robust and
more scalable.These components are:(1)
the underlying description language,(2)the knowledge base,and (3)the reasoner.The knowledge base itself comprises two
components (TBox and ABox),which
can also be approximated as a whole or separately.Some general approximation techniques that can be applied to one or more of these components are the following:
Language Weakening:The idea of language weakening is based on the well-known trade-off between the expressiveness and the reasoning complexity of a logical lan-guage.By weakening the logical language in which a theory is encoded,we are able to trade the completeness of reasoning against run-time.For example,[3]shows how hierarchical knowledge bases can be used to reason approximately with disjunctive in-formation.The logic that underlies OWL Full for example is known to be intractable,reasoners can use a slightly weaker logic (e.g.,OWL Lite)that still allows the com-putation of certain consequences.This idea can be further extended by starting with a very simple language and iterating over logics of increasing strength supplementing previously derived facts.
Knowledge Compilation:In order to avoid complexity at run-time,knowledge compi-lation aims at pre-processing the ontology off-line such that on-line reasoning becomes faster.For example,this can be achieved by explicating hidden knowledge.Derived facts are added to the original theory as axioms,avoiding the need to deduce them again.In the case of ontological reasoning,implicit subsumption and membership rela-tions are good candidates for compilation.For example,implicit subsumption relations in an OWL ontology could be identi?ed using a DL reasoner,the resulting more com-plete hierarchy could be encoded e.g.,in RDF schema and used by systems that do not have the ability to perform complex reasoning.This example can be considered to be a transformation of the DL language.When one transforms an ontology into a less expressive DL language[4,5],this often results in an approximation of the original ontology.
Approximate Deduction:Instead of modifying the logical language,approximations can also be achieved by weakening the notion of logical consequence[1,6].The approximated consequences are usually characterised as sound but incomplete,or complete but unsound.Only[1]has made some effort in the context of DLs.
The approximation method focused on in this article belongs to the last category, however there is not always a clear classi?cation of one method to the three categories de?ned above.In the following section we discuss in more detail what is meant by approximating DLs in the remainder of this paper.
3Approximating Description Logics
The elements of a DL are concept expressions and determining their satis?ability is the most basic task.Other reasoning services(e.g.,subsumption,classi?cation,instance re-trieval)can often be restated in terms of satis?ability checking[2].With approximation in DLs,we mean determining the satis?ability of a concept expression through some other means than computing the satis?ability of the concept expression itself.This use of approximation differs with other work on approximating DLs[4,5]in which a con-cept expression is translated to another concept expression,de?ned in a second typically less expressive DL.
In our approach(originally proposed in[1]),in a DL only other,somehow‘related’, concept expressions can be used that are in some way‘simpler’when determining their satis?ability.For example,a concept expression can be related to another concept ex-pression through its subsumption relation,and a concept expression can be made sim-pler by either forgetting some of its subconcepts or by replacing some of its subconcepts with simpler concepts.In particular,there are two ways that a concept expression C can be approximated by a related simpler concept expression D.Either the concept expres-sion C is approximated by a weaker concept expression D(i.e.,less speci?c,C D) or by a stronger concept expression D(i.e.,more speci?c,D C).When C D,un-satis?ability of D implies unsatis?ability of C.When D C,satis?ability of D implies satis?ability of C.Note that this is similar to set theory.For two sets C,D,when C?D holds,emptiness of D implies emptiness of C,and when D?C holds,non-emptiness of D implies non-emptiness of C.
In[1]Cadoli and Schaerf propose a syntactic manipulation of concept expressions that simpli?es the task of checking their satis?ability.The method generates two se-quences of approximations,one sequence containing weaker concepts and one sequence containing stronger concepts.The sequences of approximations are obtained by substi-tuting a substring D in a concept expression C by a simpler concept.
More precisely,for every substring D they de?ne the depth of D to be‘the number of universal quanti?ers occurring in C and having D in its scope’[1].The scope of ?R.φisφwhich can be any concept term containing https://www.doczj.com/doc/5c9667435.html,ing the de?nition of depth a sequence of weaker approximated concepts can be de?ned,denoted by C i,by replacing every existentially quanti?ed subconcept,i.e.,?R.φwhereφis any concept term,of depth greater or equal than i by .Analogously,a sequence of stronger approximated concepts can be de?ned,denoted by C⊥i,by replacing every existentially quanti?ed subconcept of depth greater or equal than i by⊥.The concept expressions are assumed to be in negated normal form(NNF)before approximating them.These de?nitions lead to the following result:
Theorem1.For each i,if C i is unsatis?able then C j is unsatis?able for all j≥i, hence C is unsatis?able.For each i,if C⊥i is satis?able then C⊥j is satis?able for all j≥i,hence C is satis?able.
These de?nitions are illustrated by the following concept expression in NNF taken from the Wine ontology3
Merlot≡Wine ≤1madeFromGrape. ?madeFromGrape.{MerlotGrape},
which states that a Merlot wine is a wine that is made from the Merlot grape and no other grape.This concept expression contains no?-quanti?ers.Therefore the depth of the only existentially quanti?ed subconcept‘?madeFromGrape.{MerlotGrape}’is0. Substituting either or⊥leads to the following approximations for level0:
Merlot 0≡Wine (≤1madeFromGrape. ) ,
Merlot⊥0≡Wine (≤1madeFromGrape. ) ⊥.
No subconcepts of level1appear in the concept expression for Merlot.Therefore, Merlot 1and Merlot⊥1are equivalent to Merlot.The nesting of existential and uni-versal quanti?ers is an important measure of the complexity of satis?ability checking when considered from a worst case complexity perspective[8].This is a motivation for Cadoli and Schaef to make their speci?c substitution choices.Furthermore,they are able to show a relation between C i-and C⊥i-approximation and their multi-valued logic based on S-1-and S-3-interpretations[1].Therefore,properties obtained for S-1-and S-3-approximation also hold for C i-and C⊥i-approximation.These properties include the following:(1)Semantically well founded,i.e.,there is a relation with a logic that can be used to give meaning to approximate answers;(2)Computationally attractive, i.e.,approximate answers are cheaper to compute than the original problem;(3)Dual-ity,i.e.,both sound but incomplete and complete but unsound approximations can be 3A wine and food ontology which forms part of the OWL test suite[7].
constructed;(4)Improvable,i.e.,approximate answers can be improved while reusing previous computations;(5)Flexible,i.e.,the method can be applied to various problem domains.These properties were identi?ed by Cadoli and Schaerf to be necessary for any approximation method.
Although the proposed method by Cadoli and Schaerf[1]satis?es the needs of the Semantic Web identi?ed in Section1in theory,little is known about the applicability of their method to practical problem solving.Few results have been obtained for S-1-and S-3-approximation when applied to propositional logic[9–11],but no results are currently known to the authors when their proposed method is applied to DLs.Current work focuses on empirical validation of their proposed method.Furthermore,DLs have changed considerably in the last decade.Cadoli and Schaerf proposed their method for approximating the language ALE(they also give an extension for ALC),but ALE has a much weaker expressivity then the languages that are currently proposed for ontology modeling on the Semantic Web such as OWL.The applicability of their method to a more expressive language like OWL is an open question.Current work takes the method of Cadoli and Schaerf as a basis and focuses on extending it to more expressive DLs. 4Approximating Classi?cation
The problem of classi?cation is to arrange a complex concept expression into the sub-sumption hierarchy of a given TBox.We choose this task for two reasons.First,the worst-case complexity of a classi?cation algorithm for expressive representation lan-guages like OWL-DL is known to be intractable.Ef?cient alternatives have only been proposed for subsets of DLs[12].
Second,classi?cation is a very important part of many other reasoning services and applications.For example,classi?cation is used to generate the subsumption hierarchy of the concept descriptions in an ontology.Furthermore,classi?cation is used in the task of retrieving instances.From a theoretical point of view,checking whether an instance i is member of a concept Q can be done by proving the unsatis?ability of?Q(i).Doing this for all existing instances,however,is intractable.Therefore,most DL systems use a process that reduces the number of instance checks.It is assumed that the ontology is classi?ed and all instances are assigned to the most speci?c concept they belong to. Instance retrieval is then done by?rst classifying the query concept Q in the subsump-tion hierarchy and then selecting the instances of all successors of Q and of all direct predecessors of Q that pass the membership test in Q.We conclude that there is a lot of potential for approximating the classi?cation task.
In the following,we?rst describe the process of classi?cation in DL systems.Af-terwards we explain how the approximation technique introduced in Section3can be used to approximate(part of)this problem.
For classifying a concept expression Q into the concept hierarchy(Algorithm1)a number of subsumption tests are required for comparing the query concept with other concepts C i in the hierarchy.As the classi?cation hierarchy is assumed to be known, the number of subsumption tests can be reduced by starting at the highest level of the hierarchy and to move down to the children of a concept only if the subsumption test is positive.The most speci?c concepts w.r.t.the subsumption hierarchy which passed the
Algorithm1classi?cation
Require:A query concept Q
V ISITED:=?
R ESULT:=?
G OALS:={ }
while Goals=?do
C∈Goals where{direct parents of C}?Visited
G OALS:=Goals\{C}
V ISITED:=Visited∪{C}
if subsumed-by(Q,C)then
G OALS:=Goals∪{direct children of C}
R ESULT:=(Result∪{C})\{all ancestors of C}
end if
end while
if|Result|=1∧subsumed-by(C,Q)then
E QUAL:=‘yes’
else
E QUAL:=‘no’
end if
return Equal,Result
subsumption test are collected for the results.At the end of the algorithm,we check if the result is subsumed by Q as this implies that both are equal.
Algorithm1contains more than one step that can be approximated.For example, the subsumption tests,represented by subsumed-by(X,Y)in the algorithm,can be approximated using the method of Cadoli and Schaerf.
The subsumption test Q C can be reformulated into the unsatis?ability test of Q ?C(Algorithm2).The idea is to replace standard subsumption checks by a series of approximate checks of increasing exactness.In particular,we use weaker approxima-tions C i in the C -subsumption algorithm(Algorithm3)and stronger approxima-tions C⊥i in the C⊥-subsumption algorithm(Algorithm4).(Note the difference be-tween Algorithm2and Algorithms3and4.)The approximations are easily constructed in a linear way.When the approximation at a certain level I does not lead to a conclusion (based on Theorem1)the level I is increased by one.This is repeated until the original concept expression is obtained,i.e.,the exact subsumption test has to be performed.Al-gorithm5integrates both approximations in one procedure.The approximate versions, Algorithm2subsumption
Require:A complex concept expression C
Require:A Query Q
C URRENT:=Q ?C
R ESULT:=unsatis?able(Current)
return Result
Algorithm3C -subsumption Require:A complex concept expression C Require:A Query Q
I:=0
repeat
C URRENT:=(Q ?C) I
R ESULT:=unsatis?able(Current)
if Result=‘true’then
break
end if
I:=I+1
until Current=Q ?C
return Result
Algorithm4C⊥-subsumption Require:A Query Q
I:=0
repeat
C URRENT:=(Q ?C)⊥I
R ESULT:=unsatis?able(Current)
if Result=‘false’then
break
end if
I:=I+1
until Current=Q ?C
return Result
Algorithm5C⊥I-C I-subsumption Require:A complex concept expression C Require:A Query Q
I:=0
repeat
C URRENT:=(Q ?C)⊥I
R ESULT:=unsatis?able(Current)
if Result=‘false’then
break
end if
C URRENT:=(Q ?C) I
R ESULT:=unsatis?able(Current)
if Result=‘true’then
break
end if
I:=I+1
until Current=Q ?C
return Result
i.e.,C -subsumption ,C ⊥-subsumption ,and C ⊥I -C I -subsumption will re-place the method subsumed-by in Algorithm 1in the forthcoming experiments.Fact
Query
Racer Fig.2.Architecture of experimental setup.While these approximate versions can in principle be applied to all occurrences
of subsumption tests,we restricted the use
of approximations to the ?rst part of the
algorithm where the query concept is clas-si?ed into the hierarchy.
Each DL reasoner (e.g.,Fact [13],
Racer [14,15])implements the classi?-
cation functionality internally.In order
to obtain comparable statements about
approximate classi?cation,independently
from the implementation of a particular
DL reasoner,which may use highly opti-
mised heuristics,we implement our own
and independent classi?cation method.The classi?cation procedure was built on
top of an arbitrary DL reasoner accord-
ing to Algorithm 1,which can call the various approximation forms stated in Al-
gorithms 3,4,and 5The satis?ability
tests are propagated to the DL reasoner
through the DIG interface [16]as depicted in Figure 2.
5Experiments
The main question focused on in the experiments is which form of approximation,i.e.,C i ,C ⊥i ,or their combination,can be used to reduce the complexity of the classi?cation task.The focus of the experiments will not be on the overall computation time,but on the number of operations needed.The goal of approximation is to replace costly reasoning operations by a (small)number of cheaper approximate reasoning operations.The suitability of the method of Cadoli and Schaerf therefore depends on the number of classical reasoning operations that can be replaced by their approximate counterparts without changing the result of the computation.
In the experiments queries are generated automatically.The system randomly se-lects a number of concept descriptions from the loaded ontology.These de?nitions are used as queries and are reclassi?ed into the subsumption hierarchy.Note that the queries are ?rst randomly selected,then they are used in the experiments with all forms of ap-proximation.
The ?rst experiments were made with the TAMBIS ontology in which we (re)classi?ed 16unfoldable concept de?nitions.4Only the approximation method orig-inally suggested by Cadoli and Schaerf [1]for ALE (described in Section 3)was used.4A biochemistry ontology developed in the TAMBIS project [17].
The results of the ?rst experiments are shown in Table 1,which is divided into four columns.Each column reports the number of subsumption tests when using a certain form of approximation.The ?rst column reports results for the experiment with normal classi?cation (i.e.,without approximation),the second column for C ⊥i -approximation,the third column for C i -approximation,and the fourth column for a combination of C ⊥i -and C i -approximation.
Each column of Table 1is divided into a number of smaller rows and columns.The rows represent the level of the approximation used,where N denotes normal subsump-tion testing,i.e.,without approximation.The columns represent whether the subsump-tion test resulted in true or false.5This distinction is important,because Theorem 1tells us that only one of those two results will immediately lead to a reduction in complexity,while for the other result approximation has to continue at the next level.This continues until no more approximation steps can be done.
Table 1.Subsumption tests for the reclassi?cation of 16concepts in TAMBIS.normal
C ⊥i C i C ⊥i &C i true false
true false true false true false C ⊥015732C 08181C ⊥015732Tambis (16)
C ⊥100C 1
00C 0
8149N 24279N 24247N 16279N 16247
The ?rst column shows that for the reclassi?cation of 16concepts in the TAMBIS ontology,24true subsumption tests and 279false subsumption tests were needed.The second column shows that C ⊥i -approximation leads to a change in normal sub-sumption https://www.doczj.com/doc/5c9667435.html,pared to the normal case,the number of false subsumption tests are reduced from 279to 247.However,the 24true subsumption tests are not reduced.Note
that 32(279-247)false subsumption tests are replaced by 157true C ⊥0-subsumption tests and 32false C ⊥0-subsumption tests.6The third column shows that C i -approximation also leads to a change in normal
subsumption tests,but quite different when compared to C ⊥i -approximation.With C i -approximation we reduce the true subsumption tests from 24to 16.However,the 279false subsumption tests are not reduced.Note that 8(24-16)true subsumption tests are
replaced by 8true C 0-subsumption tests and 181false C 0-subsumption tests.Analo-gously to C ⊥i -approximation,no C i -approximation was used when this would not lead
to a change in the subsumption expression.
The fourth column shows the combination of C ⊥i -and C i -approximation by using
the approximation sequence C ⊥0,C 0,C ⊥1,C 1,...,C ⊥n ?1,C n ?1
,normal .This combination 5
We will use the shorthand ‘true subsumption test’and ‘false subsumption test’to indicate these two distinct results.6Note that the numbers do not add up.The reason for this is that approximation is not used when there is no change in the subsumption expression after approximation,i.e.,when C ⊥i =C the DL reasoner is not called and no subsumption check for C ⊥i is performed.
leads to a reduction of normal subsumption tests,which is the combination of the re-ductions found when using C⊥i-or C i-approximation by itself.The true subsumption tests are reduced from24to16and the false subsumption tests are reduced from279 to247.Note that the reduction of8(24-16)true subsumption tests and32(279-247) are now replaced by157true C⊥0-subsumption tests,32false C⊥0-subsumption tests,8 true C 0-subsumption tests,and149false C 0-subsumption tests.
5.1Analysis of C⊥i-/C i-Approximation
The approximation of concept classi?cation in the TAMBIS ontology using the method of Cadoli and Schaerf reveals at least four points of interest.First,Table1shows that using C⊥i-approximation can only lead to a reduction of the false subsumption tests and C i-approximation can only lead to a reduction of the true subsumption tests.These results could be expected as they follow from Theorem1and are re?ected by Algo-rithm3,4,https://www.doczj.com/doc/5c9667435.html,ing Theorem1we have the following reasoning steps for C⊥i-approximation:
Query Concept?(Query ?Concept)is satis?able
?(Query ?Concept)⊥i is satis?able.
Hence,when(Query ?Concept)⊥i is not satis?able,nothing can be concluded and approximation can not lead to any gain.
Using Theorem1we have the following reasoning steps for C i-approximation: Query Concept?(Query ?Concept)is not satis?able
?(Query ?Concept) i is not satis?able. Hence,when(Query ?Concept) i is satis?able,nothing can be concluded and approximation can not lead to any gain.
Second,no approximations are used on a level higher than zero.This is a direct consequence of the TAMBIS ontology containing no nested concept de?nitions.Further on,we show this to be the case for most ontologies found in practice.
Third,both C⊥i-and C i-approximation are not applied in all subsumption tests that are theoretically possible.With normal classi?cation303(24+279)subsump-tion tests are needed.However,with C⊥i-approximation in only189(157+32)cases approximation was actually used.In the remaining114(303-189)cases approxima-tion had no effect on the concept de?nitions,i.e.,C⊥i=C,and no test was therefore performed.Hence,in38%of the subsumption tests,approximation was not used.Sim-ilar observations hold for C i-approximation.This observation indicates that C⊥i-/C i-approximation is not very useful(at least for the TAMBIS ontology)for approximating classi?cation in an ontology.
Fourth,apart from the successful reduction of normal subsumption tests,we must also consider the cost for obtaining the reduction.For example,with C⊥i-approximation we obtained a reduction in32false subsumption tests,i.e.,32normal false subsump-tion tests could be replaced by32cheaper false C⊥0-subsumption tests,however it also
cost an extra157true C⊥0-subsumption tests that did not lead to any reduction.As noth-ing can be deduced from these157true C⊥0-subsumption tests,these computations are wasted and reduce the gain obtained with the32reduced false subsumption tests consid-erably.Obviously,these unnecessary true C⊥0-subsumption tests should be minimised. No?nal verdict can be made however,because it all depends on the computation time needed to compute the normal subsumption tests and C⊥0-subsumption tests,but157 seems rather high.Similar observations hold for C i-approximation.
Analysing the high amount of unnecessary subsumption tests,we discovered a phenomenon,which we call term collapsing.We illustrate term collapsing through an example taken from the Wine ontology.Suppose that during a classi?cation the subsumption test Query WhiteNonSweetWine is generated.The de?nition for WhiteNonSweetWine is:
Wine ?hasColor.{White} ?hasSugar.{OffDry,Dry}.
The subsumption query is?rst transformed into a satis?ability test,i.e.,Query WhiteNonSweetWine?Query ?WhiteNonSweetWine is unsatis?able,be-cause C⊥i-/C i-approximation is de?ned in terms of satis?ability checking.
The de?nition of?WhiteNonSweetWine is
≡?(Wine ?hasColor.{White} ?hasSugar.{OffDry,Dry})
≡?Wine ?hasColor.?{White} ?hasSugar.?{OffDry,Dry}.
and therefore the approximation(?WhiteNonSweetWine) 0is
≡(?Wine ?hasColor.?{White} ?hasSugar.?{OffDry,Dry}) 0
≡(?Wine) 0 (?hasColor.?{White}) 0 (?hasSugar.?{OffDry,Dry}) 0
≡?Wine ?hasColor.?{White}
≡ .
Therefore,approximating the expression Query ?WhiteNonSweetWine results in checking unsatis?ability of Query 0,i.e.,(Query ?WhiteNonSweetWine) 0?Query 0 ?Query 0is unsatis?able.This test most likely fails,because in a consistent ontology Query will be satis?able and as Query is more speci?c than Query 0,i.e.,Query Query 0,the latter will be satis?able.
Analogously,applying C⊥i-approximation may result in a collapse of the Query to ⊥.This occurs whenever Query contains a conjunction with at least one?-quanti?er. In this case,the entire subsumption test is collapsed into checking the satis?ability of ⊥.As⊥can never be satis?ed,this results in an unnecessary subsumption test.
For the TAMBIS ontology we counted the numbers of occurrences of term collaps-ing in approximated concept expressions.With C i-approximation65terms out of181 collapsed.In other words,35.9%of the approximated false subsumption tests are obvi-ously not needed and should be avoided.With C⊥i-approximation it is even more severe drastic:157terms out of157collapsed.With C⊥i-and C i-approximation190terms out of306collapsed.In other words62.1%of the approximated subsumption tests are not needed.
An additional linear time test could be added to the approximation algorithm to detect term collapsing.However,an optimised DL reasoner with lazy evaluation would perform this simple test in a similar way.Experiments indeed show that the DL reasoner quickly detects term collapsing.
Summarising,using the proposed approximation method by Cadoli and Schaerf [1]on query classi?cation in the TAMBIS ontology leads to many collapsing terms. Furthermore,in only a few cases expensive subsumption tests are replaced by cheaper approximated subsumption tests.These results indicate that their approximation method does not?t practical situations well.A different approximation method may provide better approximation.
5.2Further Experiments
Although practical results of C⊥i-/C i-approximation are somewhat disappointing for the TAMBIS ontology,similar experiments were made with other ontologies.Table2 summarises the results of C⊥i-/C i-approximation applied to the reclassi?cation of10 unfoldable concepts in?ve other Ontologies.
Table2.Number of subsumption tests for reclassi?cation in?ve ontologies.
normal C⊥
C i C⊥i&C i
i
true false true false true false true false
C⊥0--00--00
----0000 Dolce(10)C
normal10113101131011310113
C⊥0--00--00
----0000 Galen(10)C
normal1012190101219010121901012190
C⊥0--00--00
----0000 Monet(10)C
normal20656206562065620656
C⊥0--1450--1450
----51405140 MadCow(10)C
normal66152661526115261152
C⊥0--2281--2281
----62236222 Wine(10)C
normal33252332512725227251
For the?rst three Ontologies in Table2,the DOLCE7,Galen8,and Monet on-tology9,C⊥i-or C i-approximation has no effect.In these three Ontologies,C⊥i-/C i-7An ontology for linguistic and cognitive engineering[18].
8A medical terminology developed in the Galen project[19].
9An ontology for mathematical web services[20].
approximation does not change any concept expression and therefore no reduction in normal subsumption tests can be obtained.An analysis of these three Ontologies shows that the Ontologies use some roles and/or attributes,but the?-and/or?-quanti?ers are very rarely used.For example,the Monet ontology contains2037concepts,34roles, and10attributes.The?-constructor is only used in13de?nitions(0.64%of all concept de?nitions).The?-constructor is only used in11cases(0.54%of all concept de?ni-tions).Therefore no quanti?ers were present in the ten randomly selected queries.As quanti?ers are so rare,C⊥i-/C i-approximation seems to be useless for those Ontologies.
The next two Ontologies in Table2,MadCow10and Wine,are somewhat arti?cial because they are developed for demonstrating the expressive power of DLs.C⊥i-/C i-approximation was applied to classi?cation in both Ontologies,but this leads to almost no reduction of normal subsumption tests.In the Madcow ontology only5true sub-sumption tests are reduced and in the Wine ontology only7subsumption tests are re-duced(6true subsumption tests+1false subsumption test).Many more subsumption tests are not reduced.In many cases approximating subsumption tests leads to term collapsing and useless subsumption tests.
6Conclusions
We argued that the idea of approximate logical reasoning matches the requirements of the Semantic Web in terms of robustness against errors and the ability to cope with limited resources better than conventional reasoning methods.At the same time,ap-proximate logical inference avoids the problems of many numerical approaches for approximate reasoning like the proper interpretation of the numeric values assigned to statements and the problem of acquiring these numbers.We tested a concrete method for approximate logical reasoning in DLs against these claims by applying it to the classi?cation problem on a number of Ontologies.In particular,subsumptions were approximated by sequences of weaker and stronger subsumptions.We showed that in principle both approximations can contribute to the ef?ciency of query classi?cation.
The main result,however,is that the use of the approximation method for DLs proposed by Cadoli and Schaerf is problematic for two reasons:
–A problematic side effect of using the approximation method is the collapsing of concept expressions leading to many unnecessary approximation steps.This hap-pens either when terms of a disjunction are replaced by or terms of a conjunction are replaced by⊥.The former case happens when C i is used on a concept that contains a universal quanti?er at the top level of the de?nition.The latter happens when C⊥i is used on a concept with an existential quanti?er at the top level of the de?nition.This feature of the approach is quite problematic as it excludes an impor-tant class of query concepts from the method,namely translations of conjunctive queries which are mostly translated using nested existential quanti?cations[22].–The experiments show that only in some cases the method is able to successfully replace subsumption tests by cheaper approximations.In many cases like DOLCE, Galen,and Monet no test could successfully be approximated.This observation 10Ontology about mad cows,part of the OWL Reasoning Examples[21].
can be explained by the fact that the approximation method only works on nested expressions that are existentially quanti?ed.Many existing Ontologies,however, do not contain concept expressions with nested expressions.The average ontology on the Semantic Web uses quite simple concept expressions that,if at all,are of depth one.The approximation method by Cadoli and Schaerf was designed based on theoretical considerations to reduce worst case complexity of the subsumption problem,but it does not take practical considerations like the nature of de?nitions that are likely to be found in Ontologies into account.
We conclude that the use of this speci?c method of approximating subsumption is often not suited for Semantic Web reasoning.Nevertheless,we believe in the general idea of approximate logical reasoning.The goal is to?nd an approximation strategy that takes the speci?cs of Ontologies into account.A particular problem with the current approach is the reliance on alternations of?-and?-quanti?ers.A straightforward way to modify this approach is to?nd alternative strategies for selecting subexpressions that are to be replaced by ,⊥,or other simpler subconcepts.A good candidate,that will be explored in future work,can use domain knowledge to determine the subset of the vocabulary to be replaced.We could for example?rst exclude very speci?c terms and then gradually add more speci?c ones.This and other options for approximating Semantic Web reasoning will be studied in future research.
References
1.Schaerf,M.,Cadoli,M.:Tractable reasoning via approximation.Arti?cial Intelligence74
(1995)249–310
2.Baader,F.,Calvanese,D.,McGuinness,D.L.,Nardi,D.,Patel-Schneider,P.F.:The Descrip-
tion Logic Handbook-Theory,Implementation and Applications.Cambridge University Press(2003)
3.Borgida,A.,Etherington,D.W.:Hierarchical knowledge bases and ef?cient disjunctive rea-
soning.In Brachman,R.J.,Levesque,H.J.,Reiter,R.,eds.:KR’89:Principles of Knowledge Representation and Reasoning.Morgan Kaufmann,San Mateo,California(1989)33–43 4.Baader,F.,K¨u sters,R.,Molitor,R.:Rewriting concepts using terminologies.In Cohn,A.G.,
Giunchiglia,F.,Selman,B.,eds.:Proceedings of the Seventh International Conference on Principles of Knowledge Representation and Reasoning(KR2000),San Francisco,Morgan Kaufman(2000)297–308
5.Brandt,S.,K¨u sters,R.,Turhan,A.Y.:Approximation and difference in description log-
ics.In Fensel,D.,Giunchiglia,F.,McGuiness,D.,Williams,M.A.,eds.:Proceedings of the Eighth International Conference on Principles of Knowledge Representation and Reasoning (KR2002),San Francisco,CA,Morgan Kaufman(2002)203–214
6.McAllester,D.:Truth maintenance.In:Proceedings of AAAI’90,Morgan Kaufmann(1990)
1109–1116
7.OWL test suite.https://www.doczj.com/doc/5c9667435.html,/TR/owl-test/.
8.Donini,F.,Hollunder,B.,Lenzerini,M.,Spaccamela,A.M.,Nardi,D.,Nutt,W.:The com-
plexity of existential quanti?cation in concept languages.Arti?cial Intelligence53(1992) 309–327
9.Groot,P.,ten Teije,A.,van Harmelen,F.:Towards a Structured Analysis of Approximate
Problem Solving:a Case Study in Classi?cation.In Dubois,D.,Welty,C.,Williams,M., eds.:Principles of Knowledge Representation and Reasoning:Proceedings of the Ninth In-ternational Conference(KR2004),Whistler,BC,Canada,AAAI Press(2004)399–406
10.ten Teije,A.,van Harmelen,F.:Exploiting domain knowledge for approximate diagnosis.
In Pollack,M.,ed.:Proceedings of the Fifteenth International Joint Conference on Arti?cial Intelligence(IJCAI-97).V olume1.,Nagoya,Japan,Morgan Kaufmann(1997)454–459 11.ten Teije,A.,van Harmelen,F.:Computing approximate diagnoses by using approximate
entailment.In Aiello,G.,Doyle,J.,eds.:Proceedings of the Fifth International Conference on Principles of Knowledge Representation and Reasoning(KR-96),Boston,Massachusetts, Morgan Kaufman(1996)
12.Grosof,B.N.,Horrocks,I.,V olz,R.,Decker,S.:Description logic programs:Combining
logic programs with description logic.In:Proc.of the Twelfth International World Wide Web Conference(WWW2003),ACM(2003)48–57
13.Horrocks,I.:The FaCT system.In:Proceedings of the second International Conference
on Analytic Tableaux and Related Methods.V olume1397of Lecture Notes in Arti?cial Intelligence.,Springer(1998)307–312
14.Haarslev,V.,M¨o ller,R.:Race system description.In:Proceedings of the1999Description
Logic Workshop(DL’99).CEUR Electronic Workshop Proceedings(1999)130–132
15.Haarslev,V.,M¨o ller,R.:Racer system description.In:Proceedings of the International
Joint Conference on Automated Reasoning(IJCAR2001).V olume2083of Lecture Notes in Arti?cial Intelligence.,Springer(2001)701–705
16.Bechhofer,S.,M¨o ller,R.,Crowther,P.:The dig description logic interface.In:Proceedings
of DL2003International Workshop on Description Logics,Rome(2003)
17.Baker,P.,Brass,A.,Bechhofer,S.,Goble,C.,Paton,N.,Stevens,R.:TAMBIS:Transparent
Access to Multiple Bioinformatics Information Sources.An Overview.In:Proceedings of the Sixth International Conference on Intelligent Systems for Molecular Biology(ISMB’98), Menlow Park,California,AAAI Press(1998)25–34
18.Masolo,C.,Borgo,S.,Gangemi,A.,Guarino,N.,Oltramari,A.:Ontology Library.Wonder-
Web Deliverable https://www.doczj.com/doc/5c9667435.html,boratory For Applied Ontology-ISTC-CNR.(20003)
19.Rector,A.L.,Nowlan,W.A.,Glowinski,A.:Goals for concept representation in the galen
project.In:Proceedings of the Seventeenth Annual Symposium on Computer Applications in Medical Care(SCAMC-93),Washington DC,USA(1993)414–418
20.Caprotti,O.,Dewar,M.,Turi,D.:Mathematical service matching using description logic
and owl.In:To appear in Proceedings3rd Int’l Conference on Mathematical Knowledge Management(MKM’04).V olume3119of Lecture Notes in Computer Science.,Springer-Verlag(2004)
21.Bechhofer,S.:OWL Reasoning Examples.University of Manchester(2003)
https://www.doczj.com/doc/5c9667435.html,/2003/why/latest/.
22.Horrocks,I.,Tessaris,S.:A conjunctive query language for description logic aboxes.In:
National conference on arti?cial intelligence(AAAI2000).(2000)399–404
Multisim12 常用元件库分类 图1 做到熟悉常用元件的放置即可；对于不同功能的元器件的使用不做详细讲解。 1.点击“放置源”按钮，弹出对话框中的“系列”栏如图2所示。 图2 (1). 选中“电源(POWER_SOURCES)”，其“元件”栏下内容如图3所示： 图3 (2). 选中“信号电压源(SIGNAL_VOLTAGE_SOURCES)”，其“元件”栏下内容如图4所示： 图4 (3). 选中“信号电流源(SIGNAL_CURRENT_SOURCES)”，其“元件”栏下内容如图5所示：
图5 (4). 选中“控制函数块(CONTROL_FUNCTION_BLOCKS)”，其“元件”栏下内容如图6所示： 图6 (5). 选中“电压控源(CONTROLLED_VOLTAGE_SOURCES)”，其“元件”栏下内容如图7所示： 图7 (6). 选中“电流控源(CONTROLLED_CURRENT_SOURCES)”，其“元件”栏下内容如图8所示： 图8 2. 点击“放置模拟元件”按钮，弹出对话框中“系列”栏如图9 所示。
图9 (1). 选中“模拟虚拟元件(ANALOG_VIRTUAL)”，其“元件”栏中仅有虚拟比较器、三端虚拟运放和五端虚拟运放3个品种可供调用。 (2). 选中“运算放大器(OPAMP)”。其“元件”栏中包括了国外许多公司提供的多达4243种各种规格运放可供调用。 (3). 选中“诺顿运算放大器(OPAMP_NORTON)”，其“元件”栏中有16种规格诺顿运放可供调用。 (4). 选中“比较器(COMPARATOR)”，其“元件”栏中有341种规格比较器可供调用。 (5). 选中“宽带运放(WIDEBAND_AMPS)”其“元件”栏中有144种规格宽带运放可供调用，宽带运放典型值达100MHz，主要用于视频放大电路。 (6). 选中“特殊功能运放(SPECIAL_FUNCTION)”，其“元件”栏中有165种规格特殊功能运放可供调用，主要包括测试运放、视频运放、乘法器/除法器、前置放大器和有源滤波器等。 3．点击“放置基础元件”按钮，弹出对话框中“系列”栏如图10所示。 图10
方案一： 1.进入C\WINDOWS下找到ADV PCB99SE.INI和ADV SCH99SE.INI两个文件； 2.用写字板打开ADVSCH99SE.INI文件,在[Change Library File List]下找到File0,大家可以发现,等号后面的的内容就是默认已经添加的库，如果要添加多个怎么办呢?简单，在File0后面添File1,File2..依次类推,但注意最后修改File0上面的Count属性,如果你添了两个,就把它的值改为2。 我用的是windows7系统，我如下改可行：配置SCH库 TypeCount=2 Count=5 File0=C:\Program Files\Design Explorer 99 SE\Library\Sch\Miscellaneous Devices.ddb File1=C:\Program Files\Design Explorer 99 SE\Library\Sch\Protel DOS Schematic Libraries.ddb File2=C:\Program Files\Design Explorer 99 SE\Library\Sch\Intel Databooks.ddb File3=C:\Program Files\Design Explorer 99 SE\Examples\Backup of AT89C2051.Lib File4=C:\Program Files\Design Explorer 99 SE\Examples\Backup of SHUMAGUAN.Lib 其中File3和File4是我自己画的两个元件。 3.同样对ADVPCB99SE.INI更改以配置PCB库。 同样是windows7系统改后可行。我做了如下更改可行： TypeCount=2 Count=5 File0=D>MSACCESS:$RP>C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>Advpcb.ddb$OP>$ON>PCB Footprints.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File1=D>MSACCESS:$RP>C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>Miscellaneous.ddb$OP>$ON>Miscellaneous.lib$ID>-1$ATTR>0$E>PCBLIB$S TF> File2=D>MSACCESS:$RP>C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>General IC.ddb$OP>$ON>General IC.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File3=D>MSACCESS:$RP>C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>InternationalRectifiers.ddb$OP>$ON>InternationalRectifiers.lib$ID>-1$A TTR>0 $E>PCBLIB$STF> File4=D>MSACCESS:$RP>C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>Transistors.ddb$OP>$ON>Transistors.lib$ID>-1$ATTR>0$E>PCBLIB 4.修改后,再打开PROTEL 99SE，OK。（注意每次修改时必须保证protel99se程序是关闭着的，不然你就白改了，因为protel在退出时会修改这两人文件。）
Multisim 元件库分类介绍 电子仿真软件“Mumsim8.3.30特殊版”的元件库中把元件分门别类地分成13个类别，每个类别中又有许多种具体的元器件，为便于读者在创建仿真电路时寻找元器件，现将电子仿真软件“Mumsim8.3.30特殊版”元件库和元器件的中文译意整理如下，供读者参考。 电子仿真软件Mumsim8.3.30特殊版的元件工具条如图1 所示。 图1 1.点击“放置信号源”按钮，弹出对话框中的“系列”栏如图2 所示。 图2 (1). 选中“电源(POWER_SOURCES)”，其“元件”栏下内容如图3所示： (2). 选中“信号电压源如图4所示： 图4 (3).选中“信号电流源(SIGNAL_CURRENT_SOURCES)”，其“元件”栏下内容如图5 所示： 图5 (4)选中“控制函数块 (CONTROL_FUNCTION_BLOCKS)”，其“元件”栏下内容如图6 所示： 图6 (5). 选中“电压控源 (CONTROLLED_VOLTAGE_SOURCES)”，其“元件”栏下内容如图7所示：
图 7 (6). 选中“电流控源 (CONTROLLED_CURRENT_SOURCES)”，其“元件”栏下内容如图8 所示： 图8 2. 点击“放置模拟元件”按钮，弹出对话框中“系列”栏如图9 所示。 图 9 (1). 选中“模拟虚拟元件(ANALOG_VIRTUAL)”，其“元件”栏中仅有虚拟比较器、三端虚拟运放和五端虚拟运放3个品种可供调用。 (2). 选中“运算放大器(OPAMP)”。其“元件”栏中包括了国外许多公司提供的多达4243种各种规格运放可供调用。 (3). 选中“诺顿运算放大器(OPAMP_NORTON)”，其“元件”栏中有16种规格诺顿运放可供调用。 (4). 选中“比较器(COMPARATOR)”，其“元件”栏中有341种规格比较器可供调用。 (5). 选中“宽带运放(WIDEBAND_AMPS)”其“元件”栏中有144种规格宽带运放可供调用，宽带运放典型值达100MHz，主要用于视频放大电路。 (6). 选中“特殊功能运放(SPECIAL_FUNCTION)”，其“元件”栏中有165种规格特殊功能运放可供调用，主要包括测试运放、视频运放、乘法器/除法器、前置放大器和有源滤波器等。 3．点击“放置基础元件”按钮，弹出对话框中“系列”栏如图10所示。 图10 (1). 选中“基本虚拟元件库(BASIC_VIRTUAL)”，其“元件”栏中如图 11
Multisim14使用multisim12元件库的方法 如题，步骤如下： 1、下载multisim12，multisim14，multisim12库文件。 2、安装multisim14，安装multisim12，安装方法及安装包自己百度 3、打开multisim12，导入multisim12库文件。工具----数据库----数据库管理器 ---导入-----选择下载好的数据库，按照提示操作。 4、导入成功后，打开数据库管理器（打开顺序：工具----数据库---数据库管理器）， 点击右下角的关于，查找已导入数据库的存放位置。如导入到用户数据，则复制用户数据库地址，如下图，我的存放地址为：C:\Users\Administrator\AppData\Roaming\National Instruments\Circuit Design Suite\12.0\database
5、打开数据库存放位置，可看到当前数据库，usr文件为数据库文件。 6、关闭multisim12，运行multisim14，执行工具----数据库----转换数据库---选择 v12→v14-----选择源数据库名称
7、打开到multisim12中usr库文件存放位置，即第四步所示地址，右下角选择 所有文件，这是可看到第三步导入的库文件存放文件，选择该文件，点击打开，点击开始，选择自动重命名或覆盖、忽略，点击确定。 8、等待导入结束后，即可使用。 该方法可用于其他版本数据库导入，如multisim10数据库导入multisim12或14等。 另外，也可以下载别人转换好的数据库文件，但是是否可行，有待验证。
Multisim 2001的器件库 Multisim 2001含有4个种类的器件库，执行View\Component Bars命令即可显示如图2-1所示的下拉菜单。 图2-1 View\Component Bars命令的下拉菜单 图2-1中的Multisim Database也称为Multisim Master，用来存放软件自带的元件模型。随着版本的不同，该数据库中包含的仿真元件的数量也不一样。 Corporate Database 仅专业版有效，为用于多人协同开发项目时建立的共用器件库。 User Database 用来存放用户使用Multisim编辑器自行创建的元器件模型。 EDAParts Bar 为用户提供通过因特网进入https://www.doczj.com/doc/5c9667435.html,网站，下载有关元器件的信息和资料。 Multisim 2001的Multisim Database中含有14个器件库(即Component Toolbar)，每个器件库中又含有数量不等的元件箱(又称之为Farmily)，共有6000多个元器件，各种元器件分门别类地放在这些器件箱中供用户调用。User Database在开始使用时是空的，只有在用户创建或修改了元件并存放于该库后才能有元件供调用。 本章将分别对Multisim Database中的14个器件库中的元器件加以介绍。 第一节电源库 一、电源库组成 电源库(Sources)如图2-2所示，其中共有30个电源器件，有为电路提供电能的电源，也有作为输入信号的信号源及产生电信号转变的控制电源，还有两个接地端。电源库中的器件全部为虚拟器件。
图2-2 电源库 二、电源库中的器件箱 1.接地端（Ground） 在电路中，“地”是一个公共参考点，电路中所有的电压都是相对于该点而言的电势差。在Multisim电路图上可以同时调用多个接地端，它们的电位都是OV。 2．数字接地端(Digital Ground) 在实际数字电路中，许多数字器件需要接上直流电源才能正常工作，而在原理图中并不直接表示出来。为更接近于现实，Multisim在进行数字电路的“Real”仿真时，电路中的数字元件要接上示意性的电源，数字接地端是该电源的参考点。 3．Vcc电压源(Vcc Voltage Source) 直流电压源的简化符号，常用于为数字元件提供电能或逻辑高电平。 4. V DD电压源(V DD Voltage Source) 与Vcc基本相同。当为CMOS器件提供直流电源进行“Real”仿真时，只能用V DD。 5．直流电压源(DC Voltage Source(Battery)) 这是一个理想直流电压源，使用时允许短路，但电压值将降为0。 6．直流电流源(DC Current Source) 这是一个理想直流电流源，使用时允许开路，但电流值将降为0。 7．交流电压源(AC Voltage Source) 这是一个正弦交流电压源，电压显示的数值是其有效值(均方根值)。 8．正弦交流电流源(AC Current Source) 这是一个正弦交流电流源，电流显示的数值是其有效值(均方根值)。
1.电阻 固定电阻：RES 半导体电阻：RESSEMT 电位计；POT 变电阻；RV AR 可调电阻;res1 2.电容 定值无极性电容；CAP 定值有极性电容;CAP 半导体电容：CAPSEMI 可调电容：CAPV AR 3.电感：INDUCTOR 4.二极管：DIODE.LIB 发光二极管：LED 5.三极管:NPN1 6.结型场效应管：JFET.lib 7.MOS场效应管 8.MES场效应管 9.继电器：PELAY. LIB 10.灯泡:LAMP 11.运放:OPAMP 12.数码管：DPY_7-SEG_DP (MISCELLANEOUS DEVICES.LIB) 13.开关;sw_pb 原理图常用库文件： Miscellaneous Devices.ddb Dallas Microprocessor.ddb Intel Databooks.ddb Protel DOS Schematic Libraries.ddb PCB元件常用库： Advpcb.ddb General IC.ddb Miscellaneous.ddb 部分分立元件库元件名称及中英对照 AND 与门 ANTENNA 天线 BATTERY 直流电源 BELL 铃,钟 BVC 同轴电缆接插件 BRIDEG 1 整流桥(二极管) BRIDEG 2 整流桥(集成块) BUFFER 缓冲器 BUZZER 蜂鸣器 CAP 电容 CAPACITOR 电容
CAPACITOR POL 有极性电容CAPV AR 可调电容 CIRCUIT BREAKER 熔断丝 COAX 同轴电缆 CON 插口 CRYSTAL 晶体整荡器 DB 并行插口 DIODE 二极管 DIODE SCHOTTKY 稳压二极管DIODE V ARACTOR 变容二极管DPY_3-SEG 3段LED DPY_7-SEG 7段LED DPY_7-SEG_DP 7段LED(带小数点) ELECTRO 电解电容 FUSE 熔断器 INDUCTOR 电感 INDUCTOR IRON 带铁芯电感INDUCTOR3 可调电感 JFET N N沟道场效应管 JFET P P沟道场效应管 LAMP 灯泡 LAMP NEDN 起辉器 LED 发光二极管 METER 仪表 MICROPHONE 麦克风 MOSFET MOS管 MOTOR AC 交流电机 MOTOR SERVO 伺服电机 NAND 与非门 NOR 或非门 NOT 非门 NPN NPN三极管 NPN-PHOTO 感光三极管 OPAMP 运放 OR 或门 PHOTO 感光二极管 PNP 三极管 NPN DAR NPN三极管 PNP DAR PNP三极管 POT 滑线变阻器 PELAY-DPDT 双刀双掷继电器 RES1.2 电阻 RES3.4 可变电阻 RESISTOR BRIDGE ? 桥式电阻
网上找的，比较齐全，归纳较好，分享之~ Multisim元件库介绍 1.Source库：包括电源、信号电压源、信号电流源、可控电压源、可控电流源、函数控制器件6个类。 2.BASIC库：包含基础元件，如电阻、电容、电感、二极管、三极管、开关等； 3.Diodes：二极管库，包含普通二极管、齐纳二极管、二极管桥、变容二极管、PIN二极管、发光二极管等。 4.Transisitor库：三极管库，包含NPN、PNP、达林顿管、IGBT、MOS管、场效应管、可控硅等； 5.Analog库：模拟器件库，包括运放、滤波器、比较器、模拟开关等模拟器件 6.TTL库：包含TTL型数字电路如7400 7404等门BJT电路。 https://www.doczj.com/doc/5c9667435.html,S库：COMS型数字电路如74HC00 74HC04等MOS管电路。 8.MCU Model：MCU模型，Multisim的单片机模型比较少，只有8051 PIC16的少数模型和一些ROM RAM等 9.Advance Periphearls库：外围器件库，包含键盘、LCD、和一个显示终端的模型。 10.MIXC Digital：混合数字电路库，包含DSP、CPLD、FPGA、PLD、单片机-微控制器、存储器件、一些接口电路等数字器件。 11.Mixed：混合库，包含定时器、AC/DA转换芯片、模拟开关、震荡器等； 12.Indicators：指示器库，包含电压表、电流表、探针、蜂鸣器、灯、数码管等等显示器件。 13.Power：电源库，包含保险丝、稳压器、电压抑制、隔离电源等 14.Misc：混合库，包含晶振、电子管、滤波器、MOS驱动、和其他一些器件等 15.RF：RF库，包含一些RF器件，如高频电容电感、高频三极管等 16.Elector Mechinical：电子机械器件库，包含传感开关、机械开关、继电器、电机等。
分立元件库中英文对照/仿真元件库--protel99se 元件名系表： 原理图常用库文件Miscellaneous Devices.ddb Dallas Microprocessor.ddb Intel Databooks.ddb Protel DOS SchematicLibraries.ddb ：PCB元件常用库Advpcb.ddb General IC.ddb Miscellaneous.ddb 分立元件库分立元件库元件名称及中英对照部分 AND 与门ANTENNA 天线 BATTERY 直流电源钟BELL 铃, 同轴电缆接插件BVC ) 二极管整流桥(BRIDEG 1 ) (整流桥集成块BRIDEG 2 BUFFER 缓冲器 BUZZER 蜂鸣器电容CAP 电容CAPACITOR 有极性电容CAPACITOR POL CAPvar 可调电容熔断丝CIRCUIT BREAKER COAX 同轴电缆插口CON 1 晶体整荡器CRYSTAL 并行插口DB 二极管DIODE 稳压二极管DIODE SCHOTTKY 变容二极管DIODE varACTOR LED 段DPY_3-SEG 3LED 段DPY_7-SEG 7) 带小数点段LED(DPY_7-SEG_DP 7 电解电容ELECTRO FUSE 熔断器 INDUCTOR 电感 INDUCTOR IRON 带铁芯电感 INDUCTOR3 可调电感 JFET N N沟道场效应管沟道场效应管JFET P P 灯泡LAMP 起辉器LAMP NEDN 发光二极管LED 仪表METER MICROPHONE 麦克风 MOSFET MOS管 MOTOR AC 交流电机伺服电机MOTOR SERVO 与非门NAND 或非门NOR NOT 非门三极管NPN NPN NPN-PHOTO 感光三极管运放OPAMP 2 或门OR
二、Multisim10常用元件库分类 图1 1.点击“放置信号源”按钮，弹出对话框中的“系列”栏如图2所示。 图2 (1). 选中“电源(POWER_SOURCES)”，其“元件”栏下内容如图3所示： 图3 (2). 选中“信号电压源(SIGNAL_VOLTAGE_SOURCES)”，其“元件”栏下内容如图4所示：
图4 (3). 选中“信号电流源(SIGNAL_CURRENT_SOURCES)”，其“元件”栏下内容如图5所示： 图5 (4). 选中“控制函数块(CONTROL_FUNCTION_BLOCKS)”，其“元件”栏下内容如图6所示： 图6 (5). 选中“电压控源(CONTROLLED_VOLTAGE_SOURCES)”，其“元件”栏下内容如图7所示：
图7 (6). 选中“电流控源(CONTROLLED_CURRENT_SOURCES)”，其“元件”栏下内容如图8所示： 图8 2. 点击“放置模拟元件”按钮，弹出对话框中“系列”栏如图9 所示。 图9 (1). 选中“模拟虚拟元件(ANALOG_VIRTUAL)”，其“元件”栏中仅有虚拟比较器、三端虚拟运放和五端虚拟运放3个品种可供调用。 (2). 选中“运算放大器(OPAMP)”。其“元件”栏中包括了国外许多公司提供的多达4243种各种规格运放可供调用。 (3). 选中“诺顿运算放大器(OPAMP_NORTON)”，其“元件”栏中有16种规格诺顿运放可供调用。 (4). 选中“比较器(COMPARATOR)”，其“元件”栏中有341种规格比较器可供调用。 (5). 选中“宽带运放(WIDEBAND_AMPS)”其“元件”栏中有144种规格宽带运放可供调用，宽带运放典型值达100MHz，主要用于视频放大电路。 (6). 选中“特殊功能运放(SPECIAL_FUNCTION)”，其“元件”栏中有165种规格特殊功能运放可供调用，主要包括测试运放、视频运放、乘法器/除法器、前置放大器和有源滤波器等。 3．点击“放置基础元件”按钮，弹出对话框中“系列”栏如图10所示。
Protel99se在vista和windows7下sch库和pcb库均不能加载，很郁闷。 这是个比较普遍的问题，高手给出解决方案如下： （（注意每次修改时必须保证protel99se程序是关闭着的，不然就白改了，因为protel在退 出时会自动修改这两人文件。）） 方案一： 1.进入C\WINDOWS下找到ADVPCB99SE.INI和ADVSCH99SE.INI两个文件； 2.用写字板打开ADVSCH99SE.INI 文件,在[Change Library File List] 下找到File0,大家可以发现,等号后面的的内容就是默认已经添加的库，如果要添加多个怎么办呢?简单，在File0后 面添File1,File2..依次类推,但注意最后修改File0上面的Count属性,如果你添了一个,就把它的 值改为2；添了两个,就改为3；以此类推。。。 我用的是windows7系统，更改如下可行： TypeCount=2 Count=5 File0=D:\Design Explorer 99 SE\Library\Sch\Miscellaneous Devices.ddb File1=D:\Design Explorer 99 SE\Library\Sch\Sim.ddb File2=D:\Design Explorer 99 SE\Library\Sch\SGS Analog.ddb File3=D:\Design Explorer 99 SE\Library\Sch\Spice.ddb File4=D:\Design Explorer 99 SE\Library\Sch\Analog Devices.ddb 其中，File4是我添加的。 3.同样对ADVPCB99SE.INI 更改以配置PCB库。 在[PCB Libraries] 下找到File0，更改方法跟上面的一样。我更改如下： TypeCount=2 Count=5 File0=D>MSACCESS:$RP>D:\DesignExplorer99 SE\Library\Pcb\Generic Footprints$RN>Advpcb.ddb$OP>$ON>PCB Footprints.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File1=D>MSACCESS:$RP>D:\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>Miscellaneous.ddb$OP>$ON>Miscellaneous.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File2=D>MSACCESS:$RP>D:\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>General IC.ddb$OP>$ON>General IC.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File3=D>MSACCESS:$RP>D:\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>1394 Serial Bus.ddb$OP>$ON>1394 Serial Bus.lib$ID>-1$ATTR>0$E>PCBLIB$STF> File4=D>MSACCESS:$RP>D:\Design Explorer 99 SE\Library\Pcb\Generic Footprints$RN>International Rectifiers.ddb$OP>$ON>InternationalRectifiers.lib$ID>-1$ATTR>0$E >PCBLIB$STF> （注：PCB Footprints.lib包含在Advpcb.ddb文件中；Miscellaneous.lib包含在Miscellaneous.ddb 文件中；其他类推。）修改后,再打开PROTEL 99SE，OK。
protel99se添加SCH元件库 〈1〉用文本文档打开（c:\windows）下的AdvSCH99SE.INI文件 将默认的File0=C:\Program Files\Design Explorer 99 SE\Library\Sch\Miscellaneous Devices.ddb 改为File0=C:\Program Files\Design Explorer 99 SE\Library\Sch\我的常用原理图IC元件 库.ddb，（路径自己定）再打开99,这个常用的IC原理图库就出来了 <2> 加更多的原理图 注意：第一步，一定要照上面改默认的原理图元件库成功后，再如下图添加更多的 File1,File2…… File0=C:\Program Files\Design Explorer 99 SE\Library\Sch\我的常用原理图IC元件库.ddb File1=C:\Program Files\Design Explorer 99 SE\Library\Sch\原理图端子元件库.ddb File2=C:\Program Files\Design Explorer 99 SE\Library\Sch\电阻元件库.ddb …… 第二步， 修改AdvSCH99SE文本中File0=（……）上面的Count属性,如下面有三个File,就把它的值改为3，Count=3；这样就添加成功了 (修改AdvSCH99SE文本时一定要先关闭Protel99SE软件) protel99se添加PCB原件封装 〈1〉找到C:\Program Files\Design Explorer 99 SE\Library\Pcb\Generic Footprints中的Advpcb.ddb 打开Advpcb.ddb，点文件，导入，将自己常用的封装库导入进去，如IC封装库.lib ，电
M u l t i s i m14使用 m u l t i s i m12元件库的 方法 -CAL-FENGHAI.-(YICAI)-Company One1
Multisim14使用multisim12元件库的方法 如题，步骤如下： 1、下载multisim12，multisim14，multisim12库文件。 2、安装multisim14，安装multisim12，安装方法及安装包自己百度 3、打开multisim12，导入multisim12库文件。工具----数据库----数据库管理 器---导入-----选择下载好的数据库，按照提示操作。 4、导入成功后，打开数据库管理器（打开顺序：工具----数据库---数据库管 理器），点击右下角的关于，查找已导入数据库的存放位置。如导入到用户数据，则复制用户数据库地址，如下图，我的存放地址为：C:\Users\Administrator\AppData\Roaming\National Instruments\Circuit Design Suite\\database
5、打开数据库存放位置，可看到当前数据库， usr文件为数据库文件。 6、关闭multisim12，运行multisim14，执行工具----数据库----转换数据库--- 选择v12→v14-----选择源数据库名称
7、打开到multisim12中usr库文件存放位置，即第四步所示地址，右下角 选择所有文件，这是可看到第三步导入的库文件存放文件，选择该文件，点击打开，点击开始，选择自动重命名或覆盖、忽略，点击确定。 8、等待导入结束后，即可使用。 该方法可用于其他版本数据库导入，如multisim10数据库导入multisim12或14等。 另外，也可以下载别人转换好的数据库文件，但是是否可行，有待验证。
Windows7下protel99se加载PCB库文件时出现错误,提示File is not recognized https://www.doczj.com/doc/5c9667435.html,时间:2010-08-17来源:百度空间作者:计算机应用技巧 分享到： 方法一:点击左边find后点击fied now ，在found library里有DDB文件，选中后按add to library list就能添加上 方法二:在protel66SE的开始菜单中的文件夹中有个Xilinx的东西，点一下他，根据提示安装到protel99SE的安装路径下，就行啦，应该可以，我就弄在protel66SE的开始菜单中的文件夹中有个Xilinx 的东西，点一下他，根据提示安装到protel99SE的安装路径下，就行啦，应该可以 方法三:大家都找不到那个所谓的advpcb99se，是因为安装了protel后软件并没生成advpcb99se 这个文件，而是当大家第一次加载任意pcb库时才能自动生成，所以既然加载不了pcb哪能生成这个文件呢? 咱们把重点不放在advpcb99se这个文件，而是要对它自带的默认pcb库做手脚，首先先用protel 打开自己的pcb库然后导出lib文件，再把lib文件导入到protel安装文件的library/pcb/Generic Footprints/Advpcb.ddb里，Advpcb.ddb里就是软件默认的pcb footprints.lib,现在就把pcb footprints.lib随便改个名字，然后把咱们导入进去的自己的库改成“footprints.lib”这样，嘿嘿，打开protel 后默认的就是咱们的库了，这样嘛也只能咱们学生用用，工作用的话还是换xp好点，毕竟所有元件封装都在一个文件里，以后多了会不好找的。 哈哈，这个方法咋样?要是还有更好的，希望各位不吝赐教啊 ，一切都为了让vista用这个老掉牙的protel99se -------netdreamboy 方法四:1.进入C\WINDOWS下找到ADVPCB99SE.INI和ADVSCH99SE.INI两个文件; 2.用记事打开ADVSCH99SE.INI文件,在[Change Library File List]下找到File0,等号后面的的内容就是默认已经添加的库，如果要添加多个怎么办呢?简单，在File0后面添File1,File2..依次类推,但注意最后修改File0上面的Count属性,如果你添了两个,就把它的值改为2。 如下改可行： TypeCount=2 Count=4 File0=d:\Program Files\Design Explorer 99 SE\Library\Sch\Miscellaneous Devices.ddb File1=d:\Program Files\Design Explorer 99 SE\Library\Sch\Protel DOS Schematic Libraries.ddb File2=d:\Program Files\Design Explorer 99 SE\Library\Sch\Intel Databooks.ddb File3=d:\Program Files\Design Explorer 99 SE\Examples\Backup of AT89C2051.Lib 其中File3是自己画的元件。 3.同样对ADVPCB99SE.INI更改以配置PCB库。 同样是windows7系统改后可行。我做了如下更改可行：
方案一： Protel有个默认的sch库（Miscellaneous Devices.ddb）在protel安装时就将 此库添加进去了。 所以，你可以到"...\Design Explorer 99 SE\Library\Sch"里面找到你要添加的 sch库， 即 *.ddb 文件，打开它将其中的 *.lib（你需要的*.lib）文件复制到Miscellaneous Devices.ddb 里。 重新打开Protel 99SE，再次尝试添加sch库（其实没有用，只是象征性地点 一下）， 你会发现还是失败了，只好取消。不过，你会发现在下拉列表里多了很多*.lib，恭喜你，添加成功了！！ 缺点是，随着你添加的越来越多，列表越来越长，找起来越来越麻烦。 方案二： 1.进入C\WINDOWS下找到ADVPCB99SE.INI和ADVSCH99SE.INI两个文件； 2.用写字板打开ADVSCH99SE.INI文件,在[Change Library File List]下找到 File0,大家可以发现,等号后面的的内容就是默认已经添加的库，如果要添加多 个怎么办呢?简单，在File0后面添File1,File2..依次类推,但注意最后修改File0 上面的Count属性,如果你添了两个,就把它的值改为2。 我用的是windows7系统，我如下改可行： TypeCount=2 Count=5 File0=C:\Program Files\Design Explorer 99 SE\Library\Sch\Miscellaneous Devices.ddb File1=C:\Program Files\Design Explorer 99 SE\Library\Sch\Protel DOS Schematic Libraries.ddb File2=C:\Program Files\Design Explorer 99 SE\Library\Sch\Intel Databooks.ddb File3=C:\Program Files\Design Explorer 99 SE\Examples\Backup of AT89C2051.Lib File4=C:\Program Files\Design Explorer 99 SE\Examples\Backup of SHUMAGUAN.Lib 其中File3和File4是我自己画的两个元件。 3.同样对ADVPCB99SE.INI更改以配置PCB库。 同样是windows7系统改后可行。我做了如下更改可行： TypeCount=2 Count=5
Multisim元件库分类介绍 电子仿真软件“Mumsim8.3.30特殊版”的元件库中把元件分门别类地分成13个类别，每个类别中又有许多种具体的元器件，为便于读者在创建仿真电路时寻找元器件，现将电子仿真软件“Mumsim8.3.30特殊版”元件库和元器件的中文译意整理如下，供读者参考。 电子仿真软件Mumsim8.3.30特殊版的元件工具条如图1所示。 图1 1.点击“放置信号源”按钮，弹出对话框中的“系列”栏如图2所示。 图2 (1). 选中“电源(POWER_SOURCES)”，其“元件”栏下内容如图3所示：
图3 (2). 选中“信号电压源(SIGNAL_VOLTAGE_SOURCES)”，其“元件”栏下内容如图4所示： 图4 (3). 选中“信号电流源(SIGNAL_CURRENT_SOURCES)”，其“元件”栏下内容如图5所示： 图5 (4). 选中“控制函数块(CONTROL_FUNCTION_BLOCKS)”，其“元件”栏下内容如图6所示：
图6 (5). 选中“电压控源(CONTROLLED_VOLTAGE_SOURCES)”，其“元件”栏下内容如图7所示： 图7 (6). 选中“电流控源(CONTROLLED_CURRENT_SOURCES)”，其“元件”栏下内容如图8所示： 图8 2. 点击“放置模拟元件”按钮，弹出对话框中“系列”栏如图9 所示。 图9 (1). 选中“模拟虚拟元件(ANALOG_VIRTUAL)”，其“元件”栏中仅有虚拟比较器、三端虚拟运放和五端虚拟运放3个品种可供调用。
(2). 选中“运算放大器(OPAMP)”。其“元件”栏中包括了国外许多公司提供的多达4243种各种规格运放可供调用。 (3). 选中“诺顿运算放大器(OPAMP_NORTON)”，其“元件”栏中有16种规格诺顿运放可供调用。 (4). 选中“比较器(COMPARATOR)”，其“元件”栏中有341种规格比较器可供调用。 (5). 选中“宽带运放(WIDEBAND_AMPS)”其“元件”栏中有144种规格宽带运放可供调用，宽带运放典型值达100MHz，主要用于视频放大电路。 (6). 选中“特殊功能运放(SPECIAL_FUNCTION)”，其“元件”栏中有165种规格特殊功能运放可供调用，主要包括测试运放、视频运放、乘法器/除法器、前置放大器和有源滤波器等。 3．点击“放置基础元件”按钮，弹出对话框中“系列”栏如图10所示。 图10
很多人用Protel99se的时候，在软件打开之后，进入操作界面后，在窗口的左上方有Browse/Sch,点击后会出现Libraries,然后底下是一个空白的窗口，此时，我们如果想在界面上放置元件，就要首先装载入元件库。 然而，现在大部分使用者的电脑都是win7系统，（只有少数年纪比较大的人还在用windows xp 的），点击Add/Remove后，会出现Change library file list,在Sch目录底下有很多后缀为.ddb的文件，选中一个，点击ADD后，系统提示File is not recognized,那么现在应该怎么办呢？本人经过试验之后，找出了如下3种办法，依次由简单到复杂，大家不妨一试。 1、 在软件界面的左下方依次有：Edit Place Find ,点击Find,会出现find schematic component窗口，在此窗口的右下方有Find Now,点击这个按钮后，软件会自动在安装目录下查找已经存在所有的元器件库，稍带几秒后，点击你所需要的元件库后，点击Add to the library list后，点击close关闭窗口即可 2、 把protel99se软件关掉，（此操作必须先执行，否则后面的操作都白做了！！），打开我的电脑/计算机---------》打开本地磁盘C-------------》Windows---------->AdvSch99SE.ini----------->双击打开这个文件，此时你会看到file0……………….、
file1………………..、等，此时我们需要手动将元件库添加进去，将上面的file0……………复制下来，把Sch\后面的删掉，把你需要的元件库名，手动写上去，一个字母都不能错，还要注意英文的大小写区别。此时，我们注意到上面有个Count=**,你数一下有几个File 开头的，那么Count就等于多少。然后，点击文件…………》保存，然后，关掉窗口。重新启动protel99se，进入软件界面后，此时，我们会发现Libraries 底下有很多.lib的文件，那么恭喜你，文件库已经添加成功了。鼠标左键单击元件库，点击Place即可将元件放置在界面上了。 3、 第三种是本人自己钻研出来的。任何添加不进去的元件库都可用此种方法添加进去。首先，找到你安装目录下的后缀名为.ddb的元件库，我们可以现在外面新建一个文件夹 然后用protel99se将其打开，进入后会发现其内部包含了大量的后缀名为.lib的子文件，可以单个选中或将其全部选中后，右击鼠标，选择Export…选项，浏览文件夹选项，找到你安装目录下的Librariy…………>Sch刚刚你新建的文件夹，选中后，点击确定。此时，先关掉protel软件，稍带几秒后，再次打开软件，找到Find ,点击进入后，在窗口中间的位置有一个文件搜索的条件，即File后面的属性框，将里面的*.ddb删掉，留下*.lib,点击Find Now,此时你就可以找到你刚刚新建的文件夹里面被导出的元器件库了，点击Add to the List,此时在Libraries底下的窗口里就会
下载了的朋友如果觉得好用就请顶个贴，以便让更多 有需要的朋友看到。 现在有不少朋友对自己DIY的线路用仿真进行初步的验证，可一般仿真软件自带的元件库里主要是一些欧美型号的管子，我们常用的日系音频专用管几乎没有。本人只会用Multisim，于是用谷歌在老外的网站搜索了一些日系管的Spice模型参数（由于本人非常爱国，不懂英语，所以费了好大的劲），建了几个音频专用管的仿真模型，经试用还行。 8对管分别是： 2SK246/2SJ103 常见的输入级用J-FET管，特点是结电容小； 2SC3423/2SA1360 东芝的音频专用管，主要用于电压放大级； 2SC3601/2SA1407 原本是用于视频放大，但参数非常优异。高频（ft=400MHz）、高压（200V）、电流较大（Ic=140mA）、高功耗（带散热器8W）、结电容小（Cob=2.5pF）。用某些人的话说那是“绝杀”所有的电压放大管。用作预推动也很好； 2SC669/2SA649 推动级用的老牌名管； 2SC5171/2SA1930 性能非常优秀的推动管，东芝的； 2SC5200/2SA1943 输出级用的东芝名管，你懂的； 2SC2922/2SA1216 输出级用的三肯名管，这个你也懂的； 2SK1058/2SJ162 这对管都不知道的人没资格到坛子里来混。 用户数据库.part1.rar (146.48 KB) 用户数据库.part2.rar (92.33 KB) 二个附件都要下载！ 考虑到有些人和我一样菜，说说使用方法： 首先下载上面二个附件，然后把第一个附件命名为1、第二个附件命名为2（不要搞错顺序了，否则不能解压缩！），再解压到你能找到的地方，得到一个名为“用户数据库”的prz