CN103745261B - Method for dynamic fault diagnosis by constructing three-dimensional DUCG intelligent system - Google Patents

Abstract

On the basis of the technical scheme of the existing authorized patents for invention with the patent number of ZL200680055266 and US8255353B2, the invention brings forward a novel method in which a computer is used to carry out processing on fault diagnosis uncertain causality information so as to generate a three-dimensional dynamic uncertain causality graph, and dynamic fault diagnosis is carried out on an object system based on the graph. Therefore, value new information including health state monitoring, fault forecasting, diagnosis, prediction and decision support and the like of the object system can be provided for users, thereby improving the health state of the object system and enhancing the security and availability.

Description

A kind of constructing stereo DUCG intelligent system is used for the method for dynamic fault diagnosis

Technical field

The present invention relates to intelligent information processing technology, particularly relate to a kind of dynamic fault diagnosis technology of uncertain cause-effect relationship class.

Background technology

Fault diagnosis is an important applied field of artificial intelligence technology, particularly to the monitoring of the health status of large-scale complicated system, failure prediction, diagnosis, prediction and decision support.For nuclear power station, the major parameter that power station is run and component status signal are pooled to master control room by real time data acquisition control system, and are presented in various instrument or electronic display.The task of operating personnel monitors and checks these data, judges that whether it is normal.When noting abnormalities or occur reporting to the police, diagnosing its reason, and taking measures in time, eliminate or control fault.But because important number of parameters is many, change frequency is fast, and situation is complicated, and master control room operating personnel have a great responsibility, cause operating personnel to be nervous, be usually difficult to correctly diagnose the reason of exception and adopt right measures in time, thus may heavy losses be caused.

U.S.'s Three Mile Island Nuclear Station accident in 1979 is exactly a typical case.This accident is caused by a conventional unit failure.But because operating personnel have made wrong comprehensive descision to numerous exceptions and normal signal, take wrong measure, not only fail to eliminate or control accident, expand accident on the contrary, a conventional accident is caused to become a major accident: reactor core burns, and bed rearrangement nuclear power station is scrapped.

Similarly, conventional power plant, electrical power transmission system, chemical system, aerospace system, Ship System and other relate to the system of fault, all there is similar problem.

Artificial intelligence technology is the important method solved the problem.Artificial intelligence method for diagnosing faults popular at present can be divided three classes substantially: the first kind is the method for logic-based knowledge, such as rule-based method (IF-THEN Based), main expression domain expert is about the logic knowledge of objective system.But the method there is no the technical scheme of rigorous process uncertainty propagation so far, the management of knowledge base also becomes difficulty along with the increase of knowledge, how to process contradiction or inconsistent knowledge also there is no unified strict theoretical foundation; Equations of The Second Kind is the method for object-based mathematical physics feature, such as noise diagnostics (extract the noise data of objective system vibration, extracted the fault characteristic value of object, and diagnose accordingly by means such as Fourier transforms).This method is generally only applicable to special object, does not have general using value.3rd class is data learning methods, such as neural network (Neural Networks), Bayesian network (Bayesian Networks) etc., can from a large amount of fault sample data discovered knowledge carry out fault diagnosis.When objective system exists a large amount of usable samples data, this method is applicable.But for large-scale complicated systems such as nuclear power stations, be characterized in that fault mode is numerous, but the probability occurred is low, the probability repeated in the same manner is lower, thus available fault sample data are few, and portable very poor.On the contrary, the Dynamic Signal of objective system that arrives of on-site collection is a lot.Require to carry out Multi-parameter Combined Tool reasoning based on dynamic evidence to the fault diagnosis of these systems, speed wants fast, accuracy wants high, the reasoning results will be easy to user and understand.For this kind of troubleshooting issue, above-mentioned three class methods are all difficult to be competent at.Given this, the patented technology that Zhang Qin etc. propose " a kind of building method processing the intelligent system of uncertain cause-effect relationship category information ", and obtain the Chinese invention patent mandate (patent No.: ZL 2,006 8 0055266.X in August, 2010.Hereinafter referred to as " above-mentioned patent "), and obtain U.S. patent Nos mandate (patent No.: US 8255353B2) for inscribing in April, 2012 with " A METHOD FOR CONSTRUCTING AN INTELLIGENT SYSTEM PROCESSINGUNCERTAIN CAUSAL RELATIONSHIP INFORMATION ".This technical scheme can referred to as DUCG (Dynamic UncertainCausality Graph), can effectively utilize domain expert to the causal knowledge of concrete object, and causal logic knowledge is combined with uncertainty, modular knowledge representation and synthesis is carried out by patterned mode, patterned explanation is carried out to the reasoning results, inference speed is fast, accuracy is high, interpretation is strong.

But such scheme carries out in the process of Multi-parameter Combined Tool reasoning at utilization dynamic evidence, and when objective system exists negative feedback, must meet two conditions could obtain correct result:

If t _mfor receiving the moment of one group of new evidence, m=1,2, Establish E again _j(t _m) be t _mthe label that moment receives is the evidence of j, and this evidence determines the state of X categorical variable in the original DUCG figure recorded in above-mentioned patent documentation, such as, observe variable X _nstate be X _nk, k=0,1 ..., wherein 0 represent normal condition, then

Condition 1: to any time t _mthe one group of evidence E (t collected _m) ≡ ∏ _je _j(t _m), identical possible breakdown hypothesis event H all can be found in original DUCG _kjexplain these evidences, and H _kjin comprise true fault event;

Condition 2: given H _kj, not t in the same time _mevidence E (t _m) independent of one another.

In essence, this method dynamic problem is decomposed into one group of static problem in succession occurred to process, thus must meet above-mentioned two conditions.But under objective system exists degenerative situation, condition 1 is difficult to meet.So in actual process, often only consider the initial offset direction of signal, do not consider the changing inversely in signal later stage.Although but such disposal route effectively can reduce the difficulty of building original DUCG, can not effectively utilize and explain the signal of change in oscillation.

The technical scheme that the present invention proposes is based on above-mentioned patented technology, but no longer rely on above-mentioned two conditions in practical application, do not need to adopt process (Treatment) method only considering signal initial change direction yet, but based on original DUCG, according to the signal E (t of each time point dynamically received _m), automatically and dynamically grow one through abbreviation by computing machine and add the three-dimensional DUCG:Cubic_DG (t of time dimension _m), and according to Cubic_DG (t _m) carry out fault diagnosis.

Summary of the invention

For solving the deficiency that existing intelligent system exists, the present invention proposes a kind of new technical scheme, that is: three-dimensional Dynamic Uncertain cause-and-effect diagram (Cubic Dynamic Uncertain Causality Graph is called for short Cubic_DG) method for diagnosing faults.The technical scheme using the present invention to propose, by computer disposal, dynamically can generate three-dimensional DUCG (Dynamic Uncertain cause-and-effect diagram) online, and utilize this figure to carry out the fault diagnosis of objective system.Cubic_DG method proposed by the invention had both proposed a kind of based on original DUCG and dynamic evidence E (t _m) by computing machine automatically and dynamically generate Cubic_DG (t _m) method, it is also proposed based on Cubic_DG (t _m) the method for automatically being carried out fault diagnosis by computing machine.The present invention can be performed by the computing machine that at least one processor is housed, and this computing machine comprises further: input media, for the signal that input fault diagnosis is relevant; Output unit, for exporting generated three-dimensional DUCG figure, to carry out fault diagnosis.The Cubic_DG technology that application the present invention proposes, can process the dynamic fault diagnosis problem with degenerative objective system more rationally and effectively.Present patent application is the later patents application of having authorized Chinese invention patent ZL 2,006 8 0055266.X and U.S. patent Nos US 8255353 B2, is further developing above-mentioned granted patent technology.

Except term explained above, other relational language of explained later.

Similar with above-mentioned patent, in the present invention, so-called cause-effect relationship (causality), refers to the relation of logical relation between reason (cause) between anything and result (consequence) or effect (effect) or formal available causal logic relationship expression.Such as jet chimney leaks and vapor pressure will be caused to reduce and steam generator water level reduction, and leakage is reason, and it is result that pressure and water level reduce.

So-called objective system, comprises the industrial systems such as power plant, chemical plant, naval vessel, aerospace vehicle and other are various once occur abnormal symptom, need computer to diagnose its reason thus to take to successfully manage the system of measure.

So-called variable, expresses the event variable of proposition (proposition), exactly as temperature, pressure, flow, speed, frequency, content, proportion etc., its each state, as normal, higher, on the low side etc., be a value of variable, be also referred to as a state of variable.Causal variable V _irepresent, outcome variable X _nrepresent, no matter V _ior X _nwhether state itself discrete, be all discrete by its Unified Expression or fuzzy discrete state to adopt unified method to process, by V _iand X _ndifferent conditions Unified Expression be V _ijand X _nk, wherein i with n is the label distinguishing different variable, j with k is the label distinguishing different discrete or fuzzy discrete state, works as V _ior X _nduring for continuous variable, its any value e _ior e _nbe under the jurisdiction of V respectively _ijor X _nkdegree of membership m _ij(e _i) or m _nk(e _n) meet or v _ijand X _nktreated by as event, i.e. V _ijrepresent variable V _ibe in this event of j state, X _nkrepresent variable X _nbe in this event of k state, and if j ≠ j ', k ≠ k ', then V _ijwith V _{ij '}mutual exclusion, X _nkwith X _{nk '}mutual exclusion, if i ≠ i ', then B _ijand B _{i ' j '}separate random occurrence, its probability of happening b _ij≡ Pr{B _ijbe known parameters, meet

Above-mentioned V ∈ { X, B, D, G}.X _nbeing outcome variable, also can be causal variable, Ke Yiyong or or other graphical symbol represents; B _ican only be causal variable, Ke Yiyong or or other graphical symbol represents; D _ix _idefault causal variable (or unknown cause variable), only have one for genuine state, Ke Yiyong or other graphical symbol represents; G _ibe logic gate variable, express the cause-effect relationship between the various logic combination of its input variable and sub-variable, Ke Yiyong or other graphical symbol represents, and use or input variable is connected with logic gate by other graphical symbol, the particular content of logic gate is by logic gate instruction card LGS _i(Logic Gate Specification) illustrates:

j	G ij
		0	Logical expression 1
1	Logical expression 2
		……	……
m	Logical expression m

Such as, " logical expression 1 "=X _1,1+ X _1,2b _4,2.Comma is for distinguishing variable label and state number.

As patent documentation (ZL 2,006 8 0055266.X) is recorded, actuation variable (Functional Variable) F _{n; i}express causal variable V _iwith outcome variable X _nbetween cause-effect relationship, F _{nk; i}express causal variable V _iwith result event X _nkbetween cause-effect relationship, F _{n; Ij}express reason event V _ijwith outcome variable X _nbetween cause-effect relationship, F _{nk; Ij}express reason event V _ijwith result event X _nkbetween cause-effect relationship, and can to represent with directed arc → or other graphical symbol, to point to result from reason.F _nk；ij≡(r _n；i/r _n)A _nk；ij。Wherein r _{n; i}>0 is causal variable V _iwith outcome variable X _nbetween strength of association (relationship intensity), r _n≡ ∑ _ir _{n; i}, A _{nk; Ij}for the V when not considering other causal variable _ijcause X _nkthere is this random occurrence, a _{nk; Ij}≡ Pr{A _{nk; Ij}, r _{n; i}and a _{nk; Ij}can be the function of time, meet f _{nk; Ij}≡ Pr{F _{nk; Ij}}=(r _{n; i}/ r _n) a _{nk; Ij}.F _{nk; Ij}v _ijto X _nkthe contribution margin of probability, meet

Above-mentioned actuation variable F _{n; i}can be conditioning variable, use dotted line directed arc or other graphical symbol represents.Conditioning variable expresses its causal variable V _iwith outcome variable X _nbetween be conditioning relation, namely according to conditional event Z _{n; i}whether meet to judge actuation variable F _{n; i}whether set up.Such as Z _{n; i}=X _1,2, work as X _1,2for true time, Z _{n; i}satisfied, F _{n; i}establishment, this conditioning variable become actuation variable →; Work as X _1,2during non-real, Z _{n; i}dissatisfied, F _{n; i}be false, this conditioning variable deleted.

An illustration of the DUCG built in the manner described above as shown in Figure 1.This figure corresponds to the Fig. 7 in above-mentioned patent specification, and be the DUCG graphic knowledge storehouse about nuclear power station secondary loop fault diagnosis, its variable description is see this patent specification, and two-way arc wherein represents two rightabout directed arcs.

Record according to the embodiment in above-mentioned patent documentation, t ₁moment receives evidence wherein E ' (t) represents abnormal evidence, E ' (t in this example ₁)=X _6,2, E " and (t) represent normal evidence, in this example in this instructions, state 0 is normal condition, state 1 is on the low side or the abnormality of two-value state variable, state 2 for higher, state 3 be low state, state 4 is high high state.Variable states can also be low amplitude vibrations, middle width vibrates and significantly vibrate, steadily rise, accelerate rising, steadily decline, accelerate lower degradation.According to the method that above-mentioned patent documentation is recorded, original DUCG figure (Fig. 1) can be that Fig. 2 (corresponding to the Figure 25-1 in above-mentioned patent documentation, has filled the X missed by abbreviation _3,0).Except subscript, can also represent state by color, such as green representative is normal, and yellow represents higher, and blueness represents on the low side, and grey represents the abnormality etc. of two-valued variable.

It is important to note that when objective system is the procedures system such as power plant, chemical plant, category-B type event exist just because of and non-ly just therefore to divide.So-called initial event, refers to that event that will directly cause objective system state to change occurs for it; So-called non-initial event, when referring to that certain initial event occurs, need certain parts or subsystem responses, and these parts or subsystem correctly can not respond, causes system to occur the event of special state.Non-initial event may occur already, but does not show before initial event occurs.The two or more independently simultaneous probability of initial event is high-order a small amount of, negligible.Have just because of and non-initial event exists time, non-initial event variable or event can be used or or other graphical symbol represents.At this moment the graphical symbol of single frame represents initial event variable or initial event.

For simplicity, below the state (that is event) of variable and variable is referred to as node.

Technical scheme of the present invention is as follows:

The structure that § 1, a kind of computing machine by being equipped with at least one processor perform is called as three-dimensional Dynamic Uncertain cause-and-effect diagram Cubic_DG (t _m) intelligent system and method in order to carry out objective system dynamic fault diagnosis, the method comprises the steps: that computing machine continues to scan the objective system signal by batch dynamically input, when detecting that the 1st batch exists abnormal signal, is t ₁in the moment, detecting that the 2nd batch when being different from the abnormal signal of the 1st batch, is t ₂moment, etc.Usually, when detect m criticize be different from m-1 criticize abnormal signal (being the normal signal be different from before this during m=1) time, be t _mmoment, t _mthe evidence in moment is E (t _m)=∏ _je _j(t _m), wherein E _j(t _m) be the jth evidence that m criticizes in evidence, (1) works as t _m=t ₁time, based on t ₁evidence E (the t that moment obtains ₁)=∏ _je _j(t ₁), make m=1, adopt the original DUCG figure of following rules reduction, the DUCG seal after the abbreviation obtained is Slice_DG (t ₁), be also designated as Cubic_DG (t ₁), i.e. Slice_DG (t ₁)=Cubic_DG (t ₁):

Rule 1: as E (t _m) display Z _{n; i}do not meet, by F _{n; i}delete from DUCG, as E (t _m) display Z _{n; i}meet, condition directed arc F _{n; i}become common directed arc F _{n; i}, figure is presented as become →.

Rule 2: as E (t _m) display V _ij(V ∈ B, X}) be true, but V _ijbut not X _nreason event, by directed arc F _{n; Ij}delete from DUCG.

Rule 3: as E (t _m) display X _nkbe true, but X _nkcan not by V _iany state of (V ∈ { B, X, G, D}) causes, by directed arc F _{nk; i}delete from DUCG.

Rule 4: as E (t _m) { this node and input directed arc thereof are deleted by B, X} type node Status unknown and no-output directed arc from DUCG in display.

Rule 5: as E (t _m) display X _n0be true, and X _n0with abnormal evidence E ' (t) without any connected relation, by X _n0delete from DUCG.

Rule 6: as E (t _m) show one group of Status unknown node except by X _n0, otherwise not with X _nkbe connected (k ≠ 0), this group Status unknown node and the directed arc that is attached thereto are deleted from DUCG.

Rule 7: because any reason causes G _ido not export, by G _iand input directed arc delete from DUCG; As m=1, because any reason causes G _ido not input directed arc, by G _iand output directed arc is deleted from DUCG.

Rule 8: when directed arc does not have reason node or node of coming to nothing, this directed arc is deleted from DUCG.

Rule 9: as existence one group node and directed arc and E (t _m) in the node that relates to without connected relation, this group node and directed arc are deleted from DUCG.

Rule 10: if 1. E (t _m) show or cause some V _ij, but V _nstatus unknown, { B, X, G}, n ≠ i, 2. according to DUCG, V for V ∈ _nbe only V _ijfather node or child node, and F _{nk; Ij}≠ 0 or F _{ij; Nk}≠ 0, F _{nk '; Ij}=0 or F _{ij; Nk '}=0, k ≠ k ', then by the V in DUCG _nchange V into _nk.

Rule 11: above-mentioned rule can be used alone according to any order, conbined usage, reuse.

It should be noted that, above-mentioned rule is equally applicable to the technical scheme described in above-mentioned patent documentation.

(2) as m>1, based on t _mevidence E (the t that moment obtains _m)=∏ _je _j(t _m), the original DUCG figure of abbreviation, is designated as Slice_DG (t _m), comprise the steps: that (a) uses above-mentioned rule to carry out abbreviation to original DUCG figure, but except following (b) and (c) described situation; (b) in execution step (a), if Slice_DG (t _m-1) in there is X _n0, and E (t _m) in also comprise X _n0, retain nodes X _n0; (c) in execution step (a), if X _nkdo not input directed arc, and k ≠ 0, retain nodes X _nk.

(3) by Slice_DG (t _m) and Cubic_DG (t _m-1) carry out the connection of crossing time sheet, generate three-dimensional DUCG and scheme Cubic_DG (t _m), comprise the steps:

A () is by Slice_DG (t _m) and Cubic_DG (t _m-1) category-B type node that interior joint label is identical overlaps as Cubic_DG (t after getting and occuring simultaneously _m-1) in node.

B () is by Cubic_DG (t _m-1) in node as being different from Slice_DG (t _m) in node treat, scheme based on original DUCG, find Cubic_DG (t _m-1) in whether there is Slice_DG (t _m) in result node X _n(t _m) or X _nk(t _m) reason node V _i(t _{m '}) or V _ij(t _{m '}), { B, X, G}, if existed, couple together result node and reason node with the corresponding F type directed arc in original DUCG figure, and figure is presented as use directed arc → by V for m ' <m and V ∈ _i(t _{m '}) or V _ij(t _{m '}) and X _n(t _m) or X _nk(t _m) couple together, point to X from V.

C () is by Slice_DG (t _m) and Cubic_DG (t _m-1) middle t _m-1the undirected arc of X type node----that nodal scheme in timeslice and state are all identical or other symbol couples together, deletes Slice_DG (t simultaneously _m) in this X type node with undirected arc and Cubic_DG (t _m-1) in identical and its upstream also identical input directed arc of coupling together of identical node.So-called upstream is exactly reason direction.

If d () Cubic_DG is (t _m-1) in there is X _n0(t _{m '}), m ' <m, simultaneously Slice_DG (t _m) in there is X _nk(t _m), k ≠ 0, deletes Cubic_DG (t _m-1) in X _n0(t _{m '}) and the Status unknown X of upstream and G type node and the directed arc that is attached thereto.

If e () Cubic_DG is (t _m-1) in the X of existence the unknown _n(t _{m '}), m ' <m, simultaneously Slice_DG (t _m) in there is X _nk(t _m), delete Cubic_DG (t _m-1) in X _n(t _{m '}) with the directed arc be attached thereto, to delete with deleted output directed arc be the node uniquely inputting directed arc and the node be connected with undirected arc with this node thereof simultaneously.

If f () Slice_DG is (t _m) in exist not have input G type node, at Cubic_DG (t _m-1) in searching whether there is the input node relevant with its output directed arc of this G type node, if found, using the input of input node as this G type node, figure is presented as use input node is connected to G type node.

(g) according to above-mentioned steps obtain by Slice_DG (t _m) and Cubic_DG (t _m-1) figure that links together is called as t _mthe three-dimensional DUCG figure in moment, is designated as Cubic_DG (t _m), which includes from t ₁to t _meach timeslice (time slice) in node and cross over and do not cross over the F type directed arc of timeslice.Export Cubic_DG (t _m), to provide current possible fault, and these faults cause objective system that current abnormal causal logic relation occurs, thus carry out fault diagnosis, to take reasonable measure elimination fault.

§ 2, one are based on Cubic_DG (t _m), carried out the method for objective system fault diagnosis by the computing machine that at least one processor is housed, comprise the steps:

(1) to Cubic_DG (t _m) in node and F type directed arc mark the timestamp of its place timeslice, wherein, V _nor V _nk, { B, X, D, G} are labeled as V to V ∈ _i(t _p) or V _ij(t _p), F _{nk; Ij}be labeled as F _{nk; Ij}(t _p, t _q), wherein p ∈ 1,2 ..., m} marks X _nkplace timeslice, q ∈ 1,2 ..., m} marks V _ijplace timeslice, q≤p.

(2) to t _m≤ t≤t _m+1, Cubic_DG (t will be contained in _m) in Status unknown event and combination be defined as assumed fault event H _kj∈ S _h(t), wherein, for Cubic_DG (t _m) in evidence, abnormality evidence E ' (t) and normal condition evidence E can be decomposed into " (t); it can be used as the abnormality evidence E ' in above-mentioned patent documentation claim 10 and normal condition evidence E respectively " treat, according to the method described in above-mentioned patent documentation, logic expansion is carried out to E ' (t), obtain all possible breakdown hypothesis events under E (t) or event sets H _kj∈ S _h(t).E ' can be called as common abnormal evidence, does not namely consider the abnormal evidence of time attribute, is made up of the logical and of several abnormal signal events, such as E '=X _1,2x _2,3x _3,1, etc.; E " common normal evidence can be called as, namely do not consider the normal evidence of time attribute, such as E "=X _4,0x _5,0x _7,0, etc.So-called E ' (t) be used as E ' treat, exactly by E ' (t) not evidence in the same time regard different abnormal evidences as.Such as E ' (t)=X _1,2(t ₁) X _2,3(t ₂) X _3,1(t ₂) X _1,2(t ₃) X _3,3(t ₄) be exactly be made up of, although X 5 different evidences _1,2(t ₁) and X _1,2(t ₃) be that identical variable states repeats, X _3,1(t ₂) and X _3,3(t ₄) be identical variable states contradiction, but because they are in different time points respectively, therefore treated by as different evidences.That is, time point is also used for distinguishing different evidences by one of attribute as evidence, and in effect, these different evidences are relations of logical and; Similarly, if E " (t)=X _4,0(t ₁) X _5,0(t ₂) X _4,0(t ₃), X _4,0(t ₁) and X _4,0(t ₃) be also two different evidences.

(3) by Cubic_DG (t _m) as the DUCG in above-mentioned patent documentation, will as the E in above-mentioned patent documentation, the method recorded according to above-mentioned patent documentation calculates H _kj∈ S _hthe conditional probability Pr{H of (t) _kj| E (t) }=Pr{H _kje (t) }/Pr{E (t) }, wherein relate to and launch set event E (t) and H _kje (t), no longer comprises X and G type node in the expression formula after expansion.So-called common evidence E is exactly E=E ' E ", wherein E ' and E " implication as above-mentioned.Obviously, E (t)=E ' (t) E " (t).

(4) in the deployment step of above-mentioned (3), the different conditions of identical variable is in not mutual exclusion in the same time, but in mutual exclusion in the same time mutually, the equal state of identical variable not independent in the same time, but is absorbing mutually in the same time mutually.Such as, as k ≠ k ', X _nk(t ₁) X _{nk '}(t ₂) ≠ 0, X _nk(t ₂) X _{nk '}(t ₂)=0, F _{nk; Ij}(t ₂, t ₁) F _{nk '; I ' j '}(t ₃, t ₂) ≠ 0, F _{nk; Ij}(t ₂, t ₁) F _{nk '; I ' j '}(t ₂, t ₂)=0, X _nk(t ₂) X _nk(t ₃) ≠ X _nk(t ₂) or ≠ X _nk(t ₃), X _nk(t ₂) X _nk(t ₂)=X _nk(t ₂).

(5) in the process performing (3), the undirected arc of the connection same node point of different time sheet is crossed over according to the process of extension mode.So-called extend, be exactly the node node of comparatively early timeslice and input directed arc thereof being substituted into more late timeslice, the timestamp of coupled F type directed arc also does corresponding conversion, remerges identical directed arc.

(6) based on the logical expression that above-mentioned steps is launched, the event in formula or event variable probability of occurrence corresponding in original DUCG or event variable probability matrix are substituted, calculate H _kj∈ S _hthe state probability of (t) with sequence probability $h_{\mathrm{kj}}^r\&equiv;h_{\mathrm{kj}}^s\left(t\right)/{\mathrm{\&Sigma;}}_{H_{\mathrm{kj}}\&Element;S_H}{H}_{\mathrm{kj}}^s\left(t\right).$

(7) export above-mentioned (6) described state probability and sequence probability, so that user understands probability of happening and the sequence thereof of current possible breakdown, thus take reasonable measure and eliminate fault as early as possible.

§ 3, when application is industrial process systems (process system), the technical scheme that can realize described in § 1 and § 2 according to following step:

(1) original DUCG can be decomposed according to initial event variable, namely retain the initial event variable B in original DUCG one by one _i, delete remaining initial event variable B _{i '}, i ' ≠ i, then make m=1 abbreviation only comprise B according to the described rule of § 1 (1) _ioriginal DUCG figure, obtain carrying out the multiple DUCG figure after decomposing abbreviation according to initial event variable, be designated as DUCG (B _i).

(2) DUCG (B is used one by one _i) as original DUCG, according to step described in § 1, build and B _icorresponding Slice_DG (B _i, t _m) and Cubic_DG (B _i, t _m).

(3) step when (2) can not realize constructing complete Cubic_DG (B _i, t _m) time, namely cannot construct Cubic_DG (B _i, t _m) explain all abnormal nodes in E (t) time, by Cubic_DG (B _i, t _m) delete.

(4) remaining Cubic_DG (B is exported _i, t _m), so that user understands and currently has which possible breakdown, and these faults cause industrial process systems to occur abnormal causal logic relation.

(5) if remaining Cubic_DG (B _i, t _m) more than one, calculate its Pr{E (t) respectively }, be designated as ζ (B _i, t _m), wherein relate to the calculating launching E (t) identical with step described in claim 2.

(6) if remaining Cubic_DG (B _i, t _m) more than one, calculate Cubic_DG (B respectively _i, t _m) weight factor $\mathrm{\&xi;}(B_i,t_m)\&equiv;\mathrm{\&zeta;}(B_i,t_m)/\underset{B_i}{\mathrm{\&Sigma;}}(B_i,t_m).$

(7) Cubic_DG (B is calculated according to the step of claim 2 _i, t _m) in comprise by initial event variable B _iin initial event and possible non-initial event form possible breakdown hypothesis H _kjstate probability and revise by weight factor: $h_{\mathrm{ij}}^s\left(t\right)=h_{\mathrm{ij}}^s\left(t\right)\mathrm{\&xi;}(B_i,t_m),$ Calculate sequence probability again $h_{\mathrm{kj}}^r\left(t\right)\&equiv;h_{\mathrm{kj}}^s\left(t\right)/{\mathrm{\&Sigma;}}_{H_{\mathrm{kj}}\&Element;S_H}{H}_{\mathrm{kj}}^s\left(t\right).$

(8) export (7) gained state probability and sequence probability, so that user understands probability of happening and the sequence thereof of current possible breakdown, instruct user to take reasonable measure as early as possible and fix a breakdown.

The technical scheme of § 3 as shown in Figure 3.

Accompanying drawing explanation

The original DUCG figure of Fig. 1 nuclear power station secondary loop fault diagnosis;

T in Fig. 2 Fig. 1 ₁the DUCG abbreviation figure in moment;

Technical scheme process flow diagram in Fig. 3 procedures system situation;

DUCG (the B after abbreviation is decomposed according to initial event variable in Fig. 4 example 1 ₁₀₂) figure;

T in Fig. 5 example 1 ₂slice_DG (the B in moment ₁₀₂, t ₂) figure;

T in Fig. 6 example 1 ₂moment completes the result after § 1 (3) (a);

T in Fig. 7 example 1 ₂moment completes the result after § 1 (3) (c);

T in Fig. 8 example 1 ₂the three-dimensional DUCG that moment obtains schemes Cubic_DG (B ₁₀₂, t ₂);

T in Fig. 9 example 1 ₃slice_DG (the B in moment ₁₀₂, t ₃);

T in Figure 10 example 1 ₃moment completes the result after § 1 (3) (a);

T in Figure 11 example 1 ₃moment completes the result after § 1 (3) (c);

T in Figure 12 example 1 ₃the three-dimensional DUCG that moment obtains schemes Cubic_DG (B ₁₀₂, t ₃);

T in Figure 13 example 1 ₄slice_DG (the B in moment ₁₀₂, t ₄);

T in Figure 14 example 1 ₄moment completes the result after § 1 (3) (a);

T in Figure 15 example 1 ₄the three-dimensional DUCG that moment obtains schemes Cubic_DG (B ₁₀₂, t ₄);

T in Figure 16 example 1 ₅slice_DG (the B in moment ₁₀₂, t ₅);

T in Figure 17 example 1 ₅moment completes the result after § 1 (3) (a);

T in Figure 18 example 1 ₅the three-dimensional DUCG that moment obtains schemes Cubic_DG (B ₁₀₂, t ₅);

The original DUCG figure of Figure 19 example 2;

Cubic_DG (t in Figure 20 example 2 ₁);

Slice_DG (t in Figure 21 example 2 ₂);

Cubic_DG (t in Figure 22 example 2 ₂);

Slice_DG (t in Figure 23 example 2 ₃);

Cubic_DG (t in Figure 24 example 2 ₃);

Slice_DG (t in Figure 25 example 2 ₄);

Cubic_DG (B in Figure 26 example 2 ₁, t ₄);

Cubic_DG (B in Figure 27 example 2 ₂, t ₄).

Embodiment

Method of the present invention is described in detail below in conjunction with accompanying drawing.

Example 1 nuclear power station secondary loop dynamic fault diagnosis

The original DUCG figure of nuclear power station secondary loop as shown in Figure 1, its variable declaration is see patent documentation (ZL 2006 80055266.X) table 1.Wherein X _n0both without input, also no-output.All r _{n; i}=1.Because parameter is too many, when not affecting explanation technical solution of the present invention, only provide part b and a type parameter herein:

$b_{102}=\left(\begin{array}{c}{b}_{\mathrm{102,0}}\\ b_{\mathrm{102,1}}\end{array}\right)=\left(\begin{array}{c}0.999\\ 0.001\end{array}\right)$

$a_{23;26}=\left(\begin{array}{ccc}{a}_{\mathrm{23,0};\mathrm{26,0}}& a_{\mathrm{23,0};\mathrm{26,1}}& a_{\mathrm{23,0};\mathrm{26,2}}\\ a_{\mathrm{23,1};\mathrm{26,0}}& a_{\mathrm{23,1};\mathrm{26,1}}& a_{\mathrm{23,1};\mathrm{26,2}}\\ a_{\mathrm{23,2};\mathrm{26,0}}& a_{\mathrm{23,2};\mathrm{26,1}}& a_{\mathrm{23,2};\mathrm{26,2}}\end{array}\right)=\left(\begin{array}{ccc}-& -& -\\ -& 0.6& 0\\ -& 0& 0.6\end{array}\right)$

$a_{23;26}=\left(\begin{array}{ccc}{a}_{\mathrm{26,0};\mathrm{23,0}}& a_{\mathrm{26,0};\mathrm{23,1}}& a_{\mathrm{26,0};\mathrm{23,2}}\\ a_{\mathrm{26,1};\mathrm{23,0}}& a_{\mathrm{26,1};\mathrm{23,1}}& a_{\mathrm{26,1};\mathrm{23,2}}\\ a_{\mathrm{26,2};\mathrm{23,0}}& a_{\mathrm{26,2};\mathrm{23,1}}& a_{\mathrm{26,2};\mathrm{23,2}}\end{array}\right)=\left(\begin{array}{ccc}-& -& -\\ -& 0& 0.8\\ -& 0.8& 0\end{array}\right)$

Wherein "-" expression is not paid close attention to, and is equivalent to " 0 ".Every parameter is "-" or " 0 " place, and cause-effect relationship does not exist, i.e. corresponding F _{nk; Ij}=0 or A _{nk; Ij}=0.

Because this example is procedures system, according to § 3 (1), decompose the DUCG (B after abbreviation ₁₀₂) as shown in Figure 4 (other decomposes abbreviation figure slightly).Below by DUCG (B ₁₀₂) treat as the original DUCG figure in patent documentation (ZL 2,006 8 0055266.X).

1. t ₁moment receives evidence according to § 1 (1) described step, abbreviation is carried out to Fig. 4, the Slice_DG (t after abbreviation ₁)=Cubic_DG (B ₁₀₂, t ₁) as shown in Figure 2.

Based on Cubic_DG (t ₁), according to § 2 (1), (2) described step, launch E ' (t)=X _6,2(t ₁), E ' (t)=F can be obtained _{6,2; 102,1}(t ₁, t ₁) B _102,1.After removing F categorical variable, S can be obtained _h(t)={ B _102,1.

According to the described step of § 1 (1) to other DUCG (B _i), i ≠ 102, carry out abbreviation, and launch E ' (t)=X according to § 2 (1), (2) described step _6,2(t ₁), acquired results is S _h(t)=0 (empty set).According to § 3 (3), these Cubic_DG (B _i, t _m), i ≠ 102, are deleted.

According to § 2 (3), (6) described step, and notice § 2 (4), the situation of (5) do not exist herein, calculates H _kj=B _102,1state probability and sequence probability: launch E (t)=X _6,2x _3,0x _12,0x _15,0x _23,0in X _6,2, obtain

E(t)＝F _6,2；102,1(t ₁,t ₁)B _102,1X _3,0X _12,0X _15,0X _23,0

B _102,1E(t)＝B _102,1F _6,2；102,1(t ₁,t ₁)B _102,1X _3,0X _12,0X _15,0X _23,0

＝F _6,2；102,1(t ₁,t ₁)B _102,1X _3,0X _12,0X _15,0X _23,0＝E(t)

Wherein X _3,0, X _12,0, X _15,0, X _23,0should according to X _n0=(1-∑ _{k ≠ 0}x _nk) mode launch.But in this example, due to B _102,1e (t)=E (t), no longer launches for the sake of simplicity.So have

$h_{\mathrm{102,1}}^s\left(t\right)=\mathrm{Pr}\left\{B_{\mathrm{102,1}}\right|E\left(t\right)\}=\mathrm{Pr}\{B_{\mathrm{102,1}}E\left(t\right)\}/\mathrm{Pr}\{E\left(t\right)\}=1$

$h_{\mathrm{102,1}}^r\left(t\right)=1$

2. t ₂the evidence that moment receives is

$\begin{array}{c}E\left(t_2\right)=E_1\left(t_2\right)E_2\left(t_2\right)E_3\left(t_2\right)E_6\left(t_2\right)E_{12}\left(t_2\right)E_{13}\left(t_2\right)E_{23}\left(t_2\right)E_{26}\left(t_2\right)\underset{j\&NotElement;\left\{\mathrm{1,2,3,6,12,13,23,26}\right\}}{\mathrm{\&Pi;}}{E}_j\left(t_2\right)\\ =X_{\mathrm{1,2}}{X}_{\mathrm{2,2}}{X}_{\mathrm{3,2}}{X}_{\mathrm{6,2}}{X}_{\mathrm{12,2}}{X}_{\mathrm{13,2}}{X}_{\mathrm{23,2}}{X}_{\mathrm{26,1}}\underset{j\&NotElement;\left\{\mathrm{1,2,3,6,12,13,23,26}\right\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

According to § 1 (2), based on the Slice_DG (B after Fig. 4 abbreviation ₁₀₂, t ₂) as shown in Figure 5.

According to § 1 (3) (a), by the Slice_DG (B in Fig. 5 ₁₀₂, t ₂) with Fig. 2 in Slice_DG (B ₁₀₂, t ₁) identical category-B type variable get occur simultaneously after overlap, obtain result shown in Fig. 6.

The situation of § 1 (3) (b) does not exist in this example.

According to § 1 (3) (c), by Slice_DG (B ₁₀₂, t ₂) and Slice_DG (B ₁₀₂, t ₁) in X _6,2use undirected arc----to couple together, delete Slice_DG (B simultaneously ₁₀₂, t ₂) in X _6,2node with Slice_DG (B ₁₀₂, t ₁) in X _6,2the input directed arc F that node is identical _{6,2; 102,1}, obtain the result of Fig. 7.

According to § 1 (3) (d), (g), by t in Fig. 5 ₁the X in moment _12,0, X _15,0, X _23,0and input directed arc F _{12,0; 102,1}, F _{15,0; 102,1}and F _{23,0; 102,1}leave out, obtain the result shown in Fig. 8, be Cubic_DG (B ₁₀₂, t ₂).

The situation of § 1 (3) (e), (f) does not exist in this example.

Based on Cubic_DG (B ₁₀₂, t ₂), can obtain the evidence wherein comprised is:

E(t)＝E ₆(t ₁)E ₁(t ₂)E ₂(t ₂)E ₃(t ₂)E ₁₂(t ₂)E ₁₃(t ₂)E ₁₅(t ₂)E ₂₃(t ₂)E ₂₆(t ₂)＝X _1,2X _2,2X _3,2X _6,2X _12,2X _13,2X _15,0X _23,2X _26,1

According to step described in § 2, each evidence carries out logic expansion according to Fig. 8 and substitutes into above-mentioned evidence expression formula carrying out logical and respectively, and the logic expansion result of E (t) is as follows, wherein for the sake of simplicity, timestamp is omitted (will embody in example 2), and according to F _{nk; Ij}=(r _{n; i}/ r _n) A _{nk; Ij}launch, calculate weight coefficient.

E(t)＝(1/24)A _13,2；1,2A _1,2；23,2A _2,2；23,2A _3,2；23,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/48)A _13,2；1,2A _1,2；23,2A _2,2；23,2A _3,2；13,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/48)A _13,2；1,2A _1,2；23,2A _2,2；23,2A _3,2；13,2A _13,2；2,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/48)A _13,2；1,2A _1,2；23,2A _2,2；13,2A _3,2；23,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/96)A _13,2；1,2A _1,2；23,2A _2,2；13,2A _3,2；13,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/96)A _13,2；1,2A _1,2；23,2A _2,2；13,2A _3,2；13,2A _13,2；2,2A _2,2；23,2X _23,2A _23,2；102,1A _26,1；23,2A _6,2；102,1

A _12,2；102,1A _15,0；102,1B _102,1

+(1/24)A _13,2；2,2A _1,2；23,2A _2,2；23,2A _3,2；23,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+(1/48)A _13,2；2,2A _1,2；23,2A _2,2；23,2A _3,2；13,2A _23,2；102,1A _26,1；23,2A _6,2；102,1A _12,2；102,1A _15,0；102,1B _102,1

+ ... (totally 81)

Carry out logic expansion to E (t) in Fig. 8, delete after category-A type event and weight coefficient remerge similar events, each affair logic remaining in expression formula and item are possible hypothesis event.In this example, this possible hypothesis event is B _102,1, i.e. S _h(t)={ B _102,1, because above-mentioned every middle category-B type event all only has B _102,1.

Due to S _han event B is only had in (t) _102,1, according to the described step of § 2 (3)-(6), B _102,1state and sequence probability must be

$h_{\mathrm{102,1}}^s\left(t\right)=h_{\mathrm{102,1}}^r\left(t\right)=1.$

3. t ₃the evidence that moment receives is:

$E\left(t_3\right)=E_6\left(t_3\right)E_{15}\left(t_3\right)E_{26}\left(t_3\right)\underset{j\&NotElement;\left\{\mathrm{6,15,26}\right\}}{\mathrm{\&Pi;}}{E}_j\left(t_3\right)=X_{\mathrm{6,2}}{X}_{\mathrm{15,2}}{X}_{\mathrm{26,1}}\underset{j\&NotElement;\left\{\mathrm{6,15,26}\right\}}{\mathrm{\&Pi;}}{X}_{j0}.$

According to § 1 (2), the Slice_DG (B after abbreviation ₁₀₂, t ₃) as shown in Figure 9.

According to § 1 (3) (a), by the Slice_DG (B in Fig. 9 ₁₀₂, t ₃) with Fig. 8 in Cubic_DG (B ₁₀₂, t ₂) identical category-B type variable get occur simultaneously after overlap, obtain result shown in Figure 10.

The situation of § 1 (3) (b) does not exist herein.

According to § 1 (3) (c), by Slice_DG (B ₁₀₂, t ₃) and Cubic_DG (B ₁₀₂, t ₂) in t ₂x in timeslice _6,2and X _26,1use undirected arc----to couple together respectively, delete Slice_DG (B simultaneously ₁₀₂, t ₃) in X _6,2with Cubic_DG (B ₁₀₂, t ₂) in t ₂the X in moment _6,2the input directed arc F of the same node point be connected by undirected arc _{6,2; 102,1}, obtain the result of Figure 11.

According to § 1 (3) (d), (g), by t in Figure 11 ₂the X in moment _15,0and input directed arc F _{15,0; 102,1}delete, obtain the result shown in Figure 12, be Cubic_DG (B _102,t ₃).

The situation of § 1 (3) (e), (f) does not exist in this example.

Based on Cubic_DG (B _102,t ₃), be similar to the calculating 2. of this example, still can obtain S _h(t)={ B _102,1, and the result of calculation of its state probability and sequence probability constant (process slightly).

4. t ₄the evidence that moment receives is:

$\begin{array}{c}E\left(t_4\right)=E_1\left(t_4\right)E_2\left(t_4\right)E_3\left(t_4\right)E_6\left(t_4\right)E_{15}\left(t_4\right)E_{26}\left(t_4\right)E_{15}\left(t_4\right)E_{26}\left(t_4\right)\underset{j\&NotElement;\left\{\mathrm{1,2,3,6,13,15,23,26}\right\}}{\mathrm{\&Pi;}}{E}_j\left(t_4\right)\\ =X_{1,1}{X}_{2,1}{X}_{3,1}{X}_{\mathrm{6,2}}{X}_{13,1}{X}_{\mathrm{15,2}}{X}_{\mathrm{23,2}}{X}_{\mathrm{26,1}}\underset{j\&NotElement;\left\{\mathrm{1,2,3,6,13,15,23,26}\right\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}.$

According to § 1 (2), the Slice_DG (B after abbreviation _102,t ₃) as shown in figure 13.

According to § 1 (3) (a), by the Slice_DG (B in Figure 13 _102,t ₄) with Figure 12 in Cubic_DG (B _102,t ₃) identical category-B type variable get occur simultaneously after overlap, obtain result shown in Figure 14.

The situation of § 1 (3) (b) does not exist herein.

According to § 1 (3) (c), by Slice_DG (B _102,t ₄) and Cubic_DG (B _102,t ₃) in t ₃x in timeslice _6,2, X _15,2and X _26,1use undirected arc----to couple together respectively, delete Slice_DG (B simultaneously _102,t ₄) in X _6,2and X _15,2with Cubic_DG (B _102,t ₃) in t ₃the X in moment _6,2and X _15,2the input directed arc F of the same node point be connected by undirected arc _{6,2; 102,1}and F _{15,2; 102,1}obtain the result of Figure 15.

The situation of § 1 (3) (e), (f) does not exist in this instance, and thus result shown in Figure 15 is Cubic_DG (B _102,t ₄).

Based on Cubic_DG (B _102,t ₄), be similar to the calculating 2. of this example, can S be obtained _h(t)={ B _102,1, and the result of calculation of its state probability and sequence probability constant (process slightly).

5. t ₅the evidence that moment receives is:

$\begin{array}{c}E\left(t_5\right)=E_1\left(t_5\right)E_2\left(t_5\right)E_3\left(t_5\right)E_6\left(t_5\right)E_{15}\left(t_5\right)E_{26}\left(t_4\right)E_{15}\left(t_5\right)E_{26}\left(t_5\right)\underset{j\&NotElement;\{\mathrm{1,2,3,6},13,\mathrm{15,23,26}\}}{\mathrm{\&Pi;}}{E}_j\left(t_5\right)\\ =X_{1,1}{X}_{2,1}{X}_{3,1}{X}_{\mathrm{6,2}}{X}_{13,1}{X}_{15,2}{X}_{23,1}{X}_{\mathrm{26,1}}\underset{j\&NotElement;\left\{\mathrm{1,2,3,6,13,15,23,26}\right\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

According to § 1 (2), the Slice_DG (t after abbreviation ₅) as shown in figure 16.

According to § 1 (3) (a), by the Slice_DG (B in Figure 16 _102,t ₅) with Figure 15 in Cubic_DG (B _102,t ₄) identical category-B type variable get occur simultaneously after overlap, obtain result shown in Figure 17.

The situation of § 1 (3) (b) does not exist herein.

According to § 1 (3) (c), by Slice_DG (B _102,t ₅) and Cubic_DG (B _102,t ₄) in t ₄x in timeslice _1,2, X _2,1, X _3,1, X _6,2, X _13,1, X _15,2and X _26,1use undirected arc----to couple together respectively, delete Slice_DG (B simultaneously _102,t ₅) in X _1,2, X _2,1, X _3,1, X _6,2, X _13,1, X _15,2and X _26,1with Cubic_DG (B _102,t ₄) in t ₄the X in moment _1,2, X _2,1, X _3,1, X _6,2, X _13,1, X _15,2and X _26,1the input directed arc F of the same node point be connected by undirected arc _{1,1; 13,1}, F _{2,1; 13,1}, F _{3,1; 13,1}, F _{13,1; 1,1}, F _{13,1; 2,1}, F _{13,1; 3,1}, F _{23,1; 1,1}, F _{23,1; 2,1}, F _{23,1; 3,1}, F _{6,2; 102,1}and F _{15,2; 102,1}obtain the result of Figure 18.

The situation of § 1 (3) (e), (f) does not exist in this instance, and thus result shown in Figure 18 is Cubic_DG (B _102,t ₅).

Based on Cubic_DG (B _102,t ₅), be similar to the calculating 2. of this example, can S be obtained _h(t)={ B _102,1, and the result of calculation of its state probability and sequence probability constant (process slightly).The fault that Figure 18 reflects 5 timeslices occurs and evolution, clearly presents vapor (steam) temperature X ₂₃with reactor capability X ₂₆between positive feedback and negative-feedback signal oscillatory condition, original DUCG figure and reasoning process do not use aforementioned condition 1 and condition 2, do not use Treatment yet.

Example 2

Figure 19 is an original DUCG figure for fault diagnosis, and described object is a procedures system, wherein all r _{n; i}=1, conditioning event Z _{4; 2}=X _3,2.Interrelated logic door and parameter values as follows:

In this example, we are first according to the three-dimensional DUCG figure of method construct described in § 1, below again according to procedures system first because of and non-initial event split, observe multiple the DUCG figure split according to initial event are put in a figure, essence be equivalent to application § 3 described in step.

Below be described in four moment in succession receive new abnormal evidence successively after three-dimensional DUCG figure growth course, mainly with t ₃moment is example, describes the technical scheme how applying the present invention's proposition in detail and carries out fault diagnosis.

1. t ₁moment

T ₁the evidence that moment receives is

$\begin{array}{c}E\left(t_1\right)=E_1\left(t_1\right)E_2\left(t_1\right)E_6\left(t_1\right)E_8\left(t_1\right)\underset{j\&NotElement;\left\{\mathrm{1,2,6,8,14}\right\}}{\mathrm{\&Pi;}}{E}_j\left(t_1\right)\\ =X_{\mathrm{1,1}}{X}_{\mathrm{2,1}}{X}_{\mathrm{6,2}}{X}_{\mathrm{8,1}}\underset{j\&NotElement;\left\{\mathrm{1,2,6,8,14}\right\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

Slice_DG (t is obtained according to § 1 (1) abbreviation ₁) (i.e. Cubic_DG (t ₁)) as shown in figure 20.Wherein B _1,1and B _1,2all X can be caused _1,1and X _2,1.

2. t ₂moment

T ₂the evidence that moment receives is

$\begin{array}{c}E\left(t_2\right)=E_1\left(t_2\right)E_2\left(t_2\right)E_3\left(t_2\right)E_4\left(t_2\right)E_6\left(t_2\right)E_8\left(t_2\right)E_9\left(t_2\right)\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9,14}\}}{\mathrm{\&Pi;}}{E}_j\left(t_2\right)\\ =X_{1,1}{X}_{2,1}{X}_{\mathrm{3,2}}{X}_{\mathrm{4,2}}{X}_{\mathrm{6,2}}{X}_{8,1}{X}_{\mathrm{9,1}}\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9},14\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

According to the Slice_DG (t that § 1 (2) abbreviation obtains ₂) as shown in figure 21.

According to § 1 (3), Slice_DG (t ₂) and Cubic_DG (t ₁) connect, obtain Cubic_DG (t ₂) as shown in figure 22.3. t ₃moment

T ₃the evidence that moment receives is

$\begin{array}{c}E\left(t_3\right)=E_1\left(t_3\right)E_2\left(t_3\right)E_3\left(t_3\right)E_4\left(t_3\right)E_6\left(t_3\right)E_8\left(t_3\right)E_9\left(t_3\right)E_{15}\left(t_3\right)\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9,15}\}}{\mathrm{\&Pi;}}{E}_j\left(t_3\right)\\ =X_{1,1}{X}_{2,1}{X}_{\mathrm{3,2}}{X}_{4,1}{X}_{6,4}{X}_{8,1}{X}_{\mathrm{9,1}}{X}_{\mathrm{15,1}}\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9},15\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

Slice_DG (t is obtained according to § 1 (2) abbreviation ₃) as shown in figure 23.

According to § 1 (3), by the Slice_DG (t in Figure 23 ₃) with Figure 22 in Cubic_DG (t ₂) connect, obtain Cubic_DG (t ₃) as shown in figure 24.Wherein, application § 1 (3) (e), the X in Figure 22 ₁₄and X _5,0deleted.X is embodied in Figure 24 _4,2(t ₂) event initiation X _15,1(t ₃), X _15,1(t ₃) merge again non-initial event B _4,1acting in conjunction causes X ₄state fall back to signal X on the low side _4,1(t ₃) concussion process.

According to the described initial event decomposing scheme of § 3 (1), and according to step described in § 2, Cubic_DG (B can be obtained ₁, t ₃) proof logic launch result be:

E(t)＝{(1/3)[A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,1(t ₁；t ₁)

A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₂)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,1(t ₁；t ₁)

+A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,1(t ₁；t ₁)A _8,1；6,2(t ₁；t ₁)

A _6,2；9,1(t ₂；t ₂)A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₁)A _6,2；2,1(t ₁；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,1(t ₁；t ₁)

+A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,1(t ₁；t ₁)A _9,1；8,1(t ₂；t ₂)

A _8,1；6,2(t ₂；t ₁)A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,1(t ₁；t ₁)]B _1,1B _4,1

+(1/3)[A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,2(t ₁；t ₁)

A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₂)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,2(t ₁；t ₁)

+A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,2(t ₁；t ₁)A _8,1；6,2(t ₁；t ₁)

A _6,2；9,1(t ₂；t ₂)A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₁)A _6,2；2,1(t ₁；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,2(t ₁；t ₁)

+A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；1,2(t ₁；t ₁)A _9,1；8,1(t ₂；t ₂)

A _8,1；6,2(t ₂；t ₁)A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；1,2(t ₁；t ₁)]B _1,2B _4,1}

A _13,0；13D(t _2；t ₂)

Substitute the A in above formula with a, substitute the B in above formula with b, and bring the value of a and b type parameter into above formula and calculate, can obtain

Pr{E(t)}＝0.00000008535；

Cubic_DG (B ₂, t ₃) in evidence logic launch result be:

E(t)＝(1/6)[(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))

(A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)(A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₂)A _6,2；2,1(t ₂；t ₁))A _6,4；2,1(t ₃；t ₁)A _2,1；2,2(t ₁；t ₁))

(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

+(1/6)[(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))

(A _8,1；6,2(t ₁；t ₁)(A _6,2；9,1(t ₂；t ₂)A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₁))A _6,2；2,1(t ₁；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；2,2(t ₁；t ₁))

(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

+(1/6)[(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；3,2(t ₂；t ₂)A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))

((A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₁)A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁))A _2,1；2,2(t ₁；t ₁))

(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

+(1/6)[(A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))(A _8,1；6,2(t ₁；t ₁)A _6,2；2,1(t ₁；t ₁)(A _9,1；8,1(t ₂；t ₂)

A _8,1；6,2(t ₂；t ₂)A _6,2；2,1(t ₂；t ₁))A _6,4；2,1(t ₃；t ₁)A _2,1；2,2(t ₁；t ₁))(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)

A _4,2；2,2(t ₂；t ₁))(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

+(1/6)[(A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))(A _8,1；6,2(t ₁；t ₁)(A _6,2；9,1(t ₂；t ₂)A _9,1；8,1(t ₂；t ₂)

A _8,1；6,2(t ₂；t ₁))A _6,2；2,1(t ₁；t ₁)A _6,4；2,1(t ₃；t ₁)A _2,1；2,2(t ₁；t ₁))(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)

A _4,2；2,2(t ₂；t ₁))(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

+(1/6)[(A _3,2；1,1(t ₂；t ₂)A _13,0；13D(t ₂；t ₂)A _1,1；2,2(t ₁；t ₁))((A _9,1；8,1(t ₂；t ₂)A _8,1；6,2(t ₂；t ₁)A _8,1；6,2(t ₁；t ₁))

A _6,2；2,1(t ₁；t ₁)A _6,2；2,1(t ₂；t ₁)A _6,4；2,1(t ₃；t ₁))A _2,1；2,2(t ₁；t ₁)(A _4,1；3,1(t ₃；t ₃)A _15,1；4,2(t ₃；t ₂)A _4,2；2,2(t ₂；t ₁))

(A _5,0；14(t _1；t ₁)A _14；2,2(t _1；t ₁))A _14,0；2,2(t _3；t ₁)]B _2,2B _4,1

Substitute the A in above formula with a, substitute the B in above formula with b, and bring the value of a and b type parameter into above formula and calculate, can obtain

Pr{E(t)}＝0.000000004344。

According to step shown in § 3 (4), (5), Cubic_DG (B can be obtained ₁, t ₃) and Cubic_DG (B ₂, t ₃) corresponding weight coefficient is respectively: ξ (B ₁, t ₂)=0.9516, ξ (B ₂, t ₂)=0.04844.

Finally calculate to obtain t ₃moment fault hypothesis space is S _h(t)={ B _1,1b _4,1, B _1,2b _4,1, B _2,2b _4,1.According to step shown in § 2, each hypothesis event H can be obtained _kj∈ S _ht the state probability of () is:

$h_{\mathrm{1,1}}^s\left(t\right)=\mathrm{Pr}\left\{B_{\mathrm{1,1}}{B}_{\mathrm{4,1}}E\left(t\right)\right\}/\mathrm{Pr}\left\{E\left(t\right)\right\}=0.8157$

$h_{1,2}^s\left(t\right)=\mathrm{Pr}\left\{B_{1,2}{B}_{\mathrm{4,1}}E\left(t\right)\right\}/\mathrm{Pr}\left\{E\left(t\right)\right\}=0.1359$

$h_{\mathrm{2,2}}^s\left(t\right)=\mathrm{Pr}\left\{B_{\mathrm{2,2}}{B}_{\mathrm{4,1}}E\left(t\right)\right\}/\mathrm{Pr}\left\{E\left(t\right)\right\}=0.0484$

Sequence probability is:

$h_{\mathrm{1,1}}^r\left(t\right)=h_{\mathrm{1,1}}^s\left(t\right)/\underset{H_{\mathrm{lj}}\&Element;S_H\left(t\right)}{\mathrm{\&Sigma;}}{h}_{\mathrm{kj}}^s\left(t\right)=0.8157$

$h_{\mathrm{1,2}}^r\left(t\right)=0.1359$

$h_{\mathrm{2,2}}^r\left(t\right)=0.0484$

Wherein, due to B _1,1with B _1,2mutual exclusion, must have B _1,1b _4,1with B _1,2b _4,1state probability sum equal 1.Again due to ξ (B ₁, t ₂) and ξ (B ₂, t ₂) be normalizing, so state probability equals the probability that sorts in this example.

4. t ₄moment

T ₄the evidence that moment receives is

$\begin{array}{c}E\left(t_4\right)=E_1\left(t_4\right)E_2\left(t_4\right)E_3\left(t_4\right)E_4\left(t_4\right)E_6\left(t_4\right)E_8\left(t_4\right)E_9\left(t_4\right)E_{14}\left(t_4\right)E_{15}\left(t_4\right)\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9,14,15}\}}{\mathrm{\&Pi;}}{E}_j\left(t_4\right)\\ =X_{1,1}{X}_{2,1}{X}_{\mathrm{3,2}}{X}_{4,1}{X}_{6,4}{X}_{8,1}{X}_{\mathrm{9,1}}{X}_{\mathrm{15,1}}\underset{j\&NotElement;\{\mathrm{1,2,3},\mathrm{4,6},\mathrm{8,9},\mathrm{14,15}\}}{\mathrm{\&Pi;}}{X}_{j0}\end{array}$

Abbreviation obtains Slice_DG (t ₄) as shown in figure 25.

Be similar to this example 3., Slice_DG (t ₄) and Cubic_DG (t ₃) connect obtain Cubic_DG (t ₄) and obtain Figure 26 (Cubic_DG (B respectively after decomposing by initial event ₁, t ₄)) and Figure 27 (Cubic_DG (B ₂, t ₄)).

As seen from Figure 26, at Cubic_DG (B ₁, t ₄) in, abnormal evidence X _14,1(t ₄) can not get any explanation, therefore, according to § 3 (3), B ₁corresponding three-dimensional DUCG schemes Cubic_DG (B ₁, t ₄) deleted.

T ₄the effective three-dimensional DUCG figure that moment finally obtains only has one, as shown in figure 27.Calculate according to § 2 (4), (5) described step, ξ (B must be had ₂, t ₄)=1.S can be obtained according to the described step of § 2 (2) _h(t)={ B _2,2b _4,1.

Due to S _ht () only comprises an assumed fault, calculate, must have according to the described step in § 2 (3)-(6) namely diagnostic result is diagnosed as by initial event B _2,2with non-initial event B _4,1concurrent, jointly form fault rootstock.

industrial applicability

The present invention can industrially apply.Its effect provides the valuable fresh informations such as objective system health monitoring, failure prediction, diagnosis, prediction and decision support for people, to improve security and the availability of objective system.

Claims (3)

1. the structure performed by the computing machine that at least one processor is housed is called as three-dimensional Dynamic Uncertain cause-and-effect diagram Cubic_DG (t _m) intelligent system and method in order to carry out objective system dynamic fault diagnosis, the method comprises the steps: that computing machine continues to scan the objective system signal by batch dynamically input, when detecting that the 1st batch exists abnormal signal, is t ₁in the moment, detecting that the 2nd batch when being different from the abnormal signal of the 1st batch, is t ₂moment, usually, when detect m criticize be different from m-1 criticize abnormal signal time, being wherein the normal signal be different from before this as m=1, is t _mmoment, t _mthe evidence in moment is E (t _m)=∏ _je _j(t _m), E _j(t _m) be the jth evidence that m criticizes in evidence,

(1) t is worked as _m=t ₁time, based on t ₁evidence E (the t that moment obtains ₁)=∏ _je _j(t ₁), make m=1, adopt the original DUCG figure of following rules reduction, obtain the DUCG after abbreviation and scheme, be designated as Slice_DG (t ₁), be also designated as Cubic_DG (t ₁), i.e. Slice_DG (t ₁)=Cubic_DG (t ₁),

Rule 1: as E (t _m) display Z _{n; i}do not meet, by F _{n; i}delete from DUCG, as E (t _m) display Z _{n; i}meet, condition directed arc F _{n; i}become common directed arc F _{n; i},

Rule 2: as E (t _m) display V _ij(V ∈ B, X}) be true, but V _ijbut not X _nreason event, by directed arc F _{n; Ij}delete from DUCG,

Rule 3: as E (t _m) display X _nkbe true, but X _nkcan not by V _iany state of (V ∈ { B, X, G}) causes, by directed arc F _{nk; i}delete from DUCG,

Rule 4: as E (t _m) display B, X} type node Status unknown and no-output directed arc, delete this node and input directed arc thereof from DUCG,

Rule 5: as E (t _m) display X _n0be true, and X _n0with abnormal evidence E ' (t) without any connected relation, by X _n0delete from DUCG,

Rule 6: as E (t _m) show one group of Status unknown node except by X _n0, otherwise not with X _nkbe connected (k ≠ 0), this group Status unknown node and the directed arc that is attached thereto deleted from DUCG,

Rule 7: because any reason causes G _ido not export, by G _iand input directed arc is deleted from DUCG; As m=1, because any reason causes G _ido not input directed arc, by G _iand output directed arc is deleted from DUCG,

Rule 8: when directed arc does not have reason node or node of coming to nothing, this directed arc is deleted from DUCG,

Rule 9: as existence one group node and directed arc and E (t _m) in the node that relates to without connected relation, this group node and directed arc are deleted from DUCG,

Rule 10: if 1. E (t _m) show or cause some V _ij, but V _nstatus unknown, { B, X, G}, n ≠ i, 2. according to DUCG, V for V ∈ _nbe only V _ijfather node or child node, and F _{nk; Ij}≠ 0 or F _{ij; Nk}≠ 0, F _{nk '; Ij}=0 or F _{ij; Nk '}=0, k ≠ k ', then by the V in DUCG _nchange V into _nk,

Rule 11: above-mentioned rule can be used alone according to any order, conbined usage, reuse;

(2) as m>1, based on t _mevidence E (the t that moment obtains _m)=∏ _je _j(t _m), the original DUCG figure of abbreviation, is designated as Slice_DG (t _m), comprise the steps: that (a) uses above-mentioned rule to carry out abbreviation to original DUCG figure, but except following (b) and (c) described situation, (b) in execution step (a), if Slice_DG (t _m-1) in there is X _n0, and E (t _m) in also comprise X _n0, retain nodes X _n0, (c) in execution step (a), if X _nkdo not input directed arc, and k ≠ 0, retain nodes X _nk;

(3) by Slice_DG (t _m) and Cubic_DG (t _m-1), carry out the connection of crossing time sheet, generate three-dimensional DUCG and scheme Cubic_DG (t _m), comprise the steps:

A () is by Slice_DG (t _m) and Cubic_DG (t _m-1) category-B type node that interior joint label is identical overlaps as Cubic_DG (t after getting and occuring simultaneously _m-1) in node,

B () is by Cubic_DG (t _m-1) in node as being different from Slice_DG (t _m) in node treat, scheme based on original DUCG, find Cubic_DG (t _m-1) in whether there is Slice_DG (t _m) in result node X _n(t _m) or X _nk(t _m) reason node V _i(t _{m '}) or V _ij(t _{m '}), m ' <m and V ∈ B, X, G}, if existed, couple together result node and reason node with the corresponding F type directed arc in original DUCG figure,

C () is by Slice_DG (t _m) and Cubic_DG (t _m-1) middle t _m-1the X type node that nodal scheme in timeslice and state are all identical couples together with undirected arc, deletes Slice_DG (t simultaneously _m) in this X type node with undirected arc and Cubic_DG (t _m-1) in identical and its upstream also identical input directed arc of coupling together of identical node,

If d () Cubic_DG is (t _m-1) in there is X _n0(t _{m '}), m ' <m, simultaneously Slice_DG (t _m) in there is X _nk(t _m), k ≠ 0, deletes Cubic_DG (t _m-1) in X _n0(t _{m '}) and the Status unknown X of upstream and G type node and the directed arc that is attached thereto,

If e () Cubic_DG is (t _m-1) in the X of existence the unknown _n(t _{m '}), m ' <m, simultaneously Slice_DG (t _m) in there is X _nk(t _m), delete Cubic_DG (t _m-1) in X _n(t _{m '}) with the directed arc be attached thereto, to delete with deleted output directed arc be the node uniquely inputting directed arc and the node be connected with undirected arc with this node thereof simultaneously,

If f () Slice_DG is (t _m) in exist not have input G type node, at Cubic_DG (t _m-1) in searching whether there is the input node relevant with its output directed arc of this G type node, if found, using the input of input node as this G type node,

(g) according to above-mentioned steps obtain by Slice_DG (t _m) and Cubic_DG (t _m-1) figure that links together is called as t _mthe three-dimensional DUCG figure in moment, is designated as Cubic_DG (t _m), export Cubic_DG (t _m), to provide current possible fault, and these faults cause objective system that current abnormal causal logic relation occurs, thus carry out fault diagnosis, to take reasonable measure elimination fault.

2. method according to claim 1, comprises the steps:

(1) to Cubic_DG (t _m) in node and F type directed arc mark the timestamp of its place timeslice, wherein, V _nor V _nk, { B, X, D, G} are labeled as V to V ∈ _i(t _p) or V _ij(t _p), F _{nk; Ij}be labeled as F _{nk; Ij}(t _p, t _q), wherein p ∈ 1,2 ..., m} marks X _nkplace timeslice, q ∈ 1,2 ..., m} marks V _ijplace timeslice, q≤p;

(2) to t _m≤ t≤t _m+1, Cubic_DG (t will be contained in _m) in Status unknown event and combination be defined as assumed fault event H _kj∈ S _h(t), wherein, for Cubic_DG (t _m) in evidence, abnormality evidence E ' (t) and normal condition evidence E can be decomposed into " (t); i.e. E (t)=E ' (t) E " (t), respectively E ' (t) is treated as common abnormality evidence E ', using E " (t) as common normal condition evidence E " treat, namely the time attribute of evidence is increased, identical or the contradiction evidence of different time is different evidence, logic expansion is carried out to E ' (t), obtains all possible breakdown hypothesis events under E (t) or event sets H _kj∈ S _h(t);

(3) by Cubic_DG (t _m) treat as original DUCG, will treat as common evidence E, calculate H _kj∈ S _hthe conditional probability Pr{H of (t) _kj| E (t) }=Pr{H _kje (t) }/Pr{E (t) }, wherein relate to and launch set event E (t) and H _kje (t), no longer comprises X and G type node in the expression formula after expansion;

(4) in the deployment step of above-mentioned (3), the different conditions of identical variable is in not mutual exclusion in the same time, but in mutual exclusion in the same time mutually, the equal state of identical variable not independent in the same time, but is absorbing mutually in the same time mutually; Such as, as k ≠ k ', X _nk(t ₁) X _{nk '}(t ₂) ≠ 0, X _nk(t ₂) X _{nk '}(t ₂)=0, F _{nk; Ij}(t ₂, t ₁) F _{nk '; I ' j '}(t ₃, t ₂) ≠ 0, F _{nk; Ij}(t ₂, t ₁) F _{nk '; I ' j '}(t ₂, t ₂)=0, X _nk(t ₂) X _nk(t ₃) ≠ X _nk(t ₂) or X _nk(t ₃), X _nk(t ₂) X _nk(t ₂)=X _nk(t ₂);

(5) in the process performing (3), cross over the undirected arc of the connection same node point of different time sheet according to the process of extension mode, so-called extension, it is exactly the node node of comparatively early timeslice and input directed arc thereof being substituted into more late timeslice, the timestamp of coupled F type directed arc also does corresponding conversion, remerges identical directed arc;

(6) based on the logical expression that above-mentioned steps is launched, the event in formula or event variable probability of occurrence corresponding in original DUCG or event variable probability matrix are substituted, calculate H _kj∈ S _hthe state probability of (t) with sequence probability $h_{\mathrm{kj}}^r\&equiv;h_{\mathrm{kj}}^s\left(t\right)/{\mathrm{\&Sigma;}}_{H_{\mathrm{kj}}\&Element;S_H}{h}_{\mathrm{kj}}^s\left(t\right);$

(7) state probability described in above-mentioned (6) and sequence probability is exported.

3. method according to claim 2, when application is industrial process systems, comprises the steps:

(1) original DUCG can be decomposed according to initial event variable, namely retain the initial event variable B in original DUCG one by one _i, delete remaining initial event variable B _{i '}, i ' ≠ i, then make m=1 abbreviation only comprise B according to the described rule of claim 1 (1) _ioriginal DUCG figure, obtain carrying out the multiple DUCG figure after decomposing abbreviation according to initial event variable, be designated as DUCG (B _i);

(2) DUCG (B is used one by one _i) substitute original DUCG, according to step described in claim 1, build and B _icorresponding Slice_DG (B _i, t _m) and Cubic_DG (B _i, t _m);

(3) step when (2) can not realize constructing complete Cubic_DG (B _i, t _m) time, namely cannot construct the Cubic_DG (B that can explain E (t) _i, t _m) time, by Cubic_DG (B _i, t _m) delete;

(4) remaining Cubic_DG (B is exported _i, t _m), to provide current possible fault, and these faults cause industrial process systems to occur abnormal causal logic relation;

(5) if remaining Cubic_DG (B _i, t _m) more than one, calculate its Pr{E (t) respectively }, be designated as ζ (B _i, t _m), wherein relate to the calculating launching E (t) identical with step described in claim 2;

(6) if remaining Cubic_DG (B _i, t _m) more than one, calculate Cubic_DG (B respectively _i, t _m) weight factor $\mathrm{\&xi;}(B_i,t_m)\&equiv;\mathrm{\&zeta;}(B_i,t_m)/\underset{B_i}{\mathrm{\&Sigma;}}\mathrm{\&zeta;}(B_i,t_m);$

(7) Cubic_DG (B is calculated according to the step of claim 2 _i, t _m) in comprise by initial event variable B _iin initial event and possible non-initial event form possible breakdown hypothesis H _kjstate probability and revise by weight factor: $h_{\mathrm{ij}}^s\left(t\right)=h_{\mathrm{ij}}^s\left(t\right)\mathrm{\&xi;}(B_i,t_m),$ Calculate sequence probability again $h_{\mathrm{kj}}^r\&equiv;h_{\mathrm{kj}}^s\left(t\right)/{\mathrm{\&Sigma;}}_{H_{\mathrm{kj}}\&Element;S_H}{h}_{\mathrm{kj}}^s\left(t\right);$

(8) export (7) gained state probability and sequence probability, to provide probability of happening and the sequence thereof of current possible breakdown, thus take reasonable measure and eliminate fault as early as possible.