CN107317866A - A kind of intelligent communication server and its construction method based on finite-state automata framework - Google Patents
A kind of intelligent communication server and its construction method based on finite-state automata framework Download PDFInfo
- Publication number
- CN107317866A CN107317866A CN201710524565.3A CN201710524565A CN107317866A CN 107317866 A CN107317866 A CN 107317866A CN 201710524565 A CN201710524565 A CN 201710524565A CN 107317866 A CN107317866 A CN 107317866A
- Authority
- CN
- China
- Prior art keywords
- state
- node
- frames
- communication server
- condition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/30—Definitions, standards or architectural aspects of layered protocol stacks
- H04L69/32—Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
- H04L69/322—Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions
- H04L69/329—Intralayer communication protocols among peer entities or protocol data unit [PDU] definitions in the application layer [OSI layer 7]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/2205—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing using arrangements specific to the hardware being tested
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/03—Protocol definition or specification
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
- H04L69/26—Special purpose or proprietary protocols or architectures
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of intelligent communication server and its construction method based on finite-state automata framework.The intelligent communication server includes finite-state automata system, real-time data base and diagnostic expert system;The software architecture and analytical algorithm of communication protocol are parsed in the apparatus, the algorithm judged than the more conventional substantial amounts of conditional logic of use, it is possible to increase tackle the complexity of stipulations and reduce the probability that malfunctions.The invention further relates to realize and set up " expert knowledge library " and " causal reasoning process ", the knowledge of the data resource in communication server and the brainstrust of device service object can be made full use of, the exception and event of the device in itself can be not only diagnosed, and the exception and event of its service object can be diagnosed.
Description
Technical field
The invention belongs to electric intelligent communication technique field, it is related to a kind of intelligence based on finite-state automata framework and leads to
Inquiry server and its construction method.
Background technology
Existing communication server species is various, and function respectively has feature, but topmost function is " communication ".Communication is needed
Corresponding agreement is observed jointly, and famous ISO/OSI (International Organization for standardization) seven layer reference model is almost widely known
." application layer " protocol contents in model are the most complicated and with industry characteristic, strongly professional.The application layer of power industry
With the IEC61850 of IEC (International Electrotechnical Commission) successively issue, the serial protocols such as IEC61870-5 are representative, and version of an agreement is patrolled
Collect property strong.The especially foundation of intelligent grid and digital transformer substation, the popularization communication server of IEC61850 international standards needs
Access various IED (Intelligent electronic device) device, it is however generally that, new IED should be supported
IEC61850 standards, its communication protocol includes MMS (Manufacturing Message Specification) and GOOSE again
(Generic Object Oriented Substation Event), its communication protocol is very complicated.Existing communication server
Mainly focus on and solve the logical relation of message when parsing communication packet, include the judgement of substantial amounts of logical condition, software system
The combination of various logic relation is slipped unavoidably in system design, so as to cause packet parsing to malfunction.
The input of " intelligent communication server " be input on the passage from communication, normal condition lower channel be according to
What the rule described by stipulations was sent, stipulations conversion typically will not be problematic;But if passage it is of low quality, just have unavoidably
" error code " occurs, and common logic judgment can only handle mistake set in advance, and other randomness mistakes will produce unpredictable
Consequence.
Because communication server needs to handle various application layer communication protocols, " restructuring " of message is can hardly be avoided,
So communication server can all have " real-time data base ".General real-time data base is mainly used in depositing the input of message, organizes defeated
Data item of outgoing packet etc.;Obviously the utilization of resources for being somebody's turn to do " real-time data base " is insufficient.
The content of the invention
Present invention solves the problem in that provide a kind of intelligent communication server based on finite-state automata framework and
Its construction method, the careless omission of logic judgment in programming is avoided using finite-state automata, is effectively filtered out random
The error code of property, so as to improve the stability and fault-tolerance of system;And with Implementation of Expert System, it is intelligent further combined with diagnosis.
The present invention is to be achieved through the following technical solutions:
A kind of intelligent communication server based on finite-state automata framework, including finite-state automata system, reality
When database and diagnostic expert system;
Described finite-state automata system includes:
State matrix, constructs the static models and dynamic model of finite-state automata, and place is needed according to communication server
The communication protocol text of reason, the data structure of structural regime matrix calculates internal machine representation, and the resolving of stipulations is resolved into one
The function of series, includes the state transition rules and its function of movement of each state matrix element;
Circulate thread, be the thread of Infinite Cyclic, its usually in a dormant state, when finding have from incoming message
Message is inputted, or has system event generation, and circulation thread is activated, and state automata is called after activation;State automata is read
Incoming message is taken, according to state transition matrix transition status, and the function of movement of corresponding state element is called;
Incoming message, is the application layer data of message, including the incoming message on passage passes through physical layer, data link
Layer, Internet, transport layer, session layer and expression layer, are finally sent to application layer from bottom to top;
" action " collection of functions of state elements, the analysis result of reception state automatic machine is divided packet parsing function
Solve and constitute the set of all polygamma functions;
Outgoing message, is the application layer data of message, top-down to pass through expression layer, session layer, transport layer, Internet
And data link layer, it is sent to passage up finally by physical layer;The co-ordination of the action function of state automata
Real time data write into Databasce needed for being formed, or outgoing message output is formed according to communication protocol;
Described real-time data base, is the computer with the real time data used in memory database realizes communication server
Internal representation, deposits the real time data of power system;
Described diagnostic expert system includes:
Knowledge base, knowledge rule set constitutes knowledge base;
Organon of causation, causal reasoning is carried out using knowledge base:Startup and parsing is called when needing to carry out fault diagnosis
The function of knowledge rule, rule parsing is carried out by performing deductive logic;
Person on duty's answer, in Organon of causation resolving, real time data is read if data dot values are included in condition
Storehouse, man-machine interaction is carried out if comprising person on duty's answer and reads person on duty's answer;
Inference conclusion feedback form, collects the conclusion of Organon of causation parsing.
Described state matrix (A) is the matrix being made up of state elements, and it is expressed as:
Eij={ xj, σj, aij, wherein:I is the i-th row of matrix, and j arranges for the jth of matrix, xjFor the defeated of the state elements
Enter, σjFor next state of the element;aijFor the action of the element, a function is called in expression, if this yuan have it is " defeated
Go out " also realized in the function;
Also state matrix (A) is represented with an one-dimension array, and increases an array of indexes p to indicate every a line
Original position.
Described knowledge rule is defined as:If<Condition>Then<Consequent>;
If<Condition>It is true, then produces<Consequent>And the adjacent rule of next stage is communicated information to, otherwise information is not
One-level is transmitted still further below;
The definition of condition:<Condition>::=^<Condition>|<Data dot values>|<Person on duty's answer>|<Finish condition>;Wherein:
" ^ " symbol represents " non-" that non-real is false, and non-vacation is true;<Data dot values>It is expressed as the value of real-time data base data point;<It is on duty
Member's answer>Expression is provided the condition of answer by person on duty;<Finish condition>The node not transmitted still further below is represented, for termination knot
Point;
<Data dot values>::=<Analog quantity>|<Digital quantity>|<Amount of calculation>;Wherein:<Analog quantity>Floating number table is used in expression
The real time data shown;<Digital quantity>That is switching value real time data;<Amount of calculation>Calculated by arithmetic and logical expression
The real time data obtained afterwards;
The tree structure that correlation is formed between the set of knowledge rule, and expression rule, is stored in database
In be knowledge base;Knowledge base reads in the requirement that real-time data base meets the real-time calculated by tree structure.
The tree structure deposited in knowledge base, its node and its attribute are deposited with internal storage data block, tree structure
Branch road index of reference is represented;
The node parameters of tree structure include this node condition, higher level's node condition, the expert info and priority of association;
The index of tree structure includes:This grade of node is indexed forward, this grade of node is indexed backward, point to higher level's node and point to subordinate's knot
Point;
The intelligent communication server is playing background music always:Phase is played according to the diagnostic result of diagnostic expert system
Corresponding music, does not have event occurs just to play easily music;According to the serious journey for occurring event if event occurs
Degree plays the music that can be come into the picture;
Operator on duty can be according to the background music of device come the working condition of identifying device.
The causal reasoning that described Organon of causation is carried out is the decision tree that traversal and parsing are made up of knowledge rule,
It is deductive procedure to decision tree traversal and resolving, the forward chaining of its path composition rule passed through:A → B, B → C, C
→D...;
The function parsing of Organon of causation is triggered by event, and the corresponding root node searched for afterwards in knowledge base is solved step by step
In analysis rule<Condition>And according to condition transmission downwards;
On the one hand the conclusion of reasoning reports person on duty by dialog box, on the other hand then plays corresponding background music and reminds
Operator on duty;
Organon of causation also receives person on duty's answer by diagnosing dialog box.
A kind of construction method of the intelligent communication server based on finite-state automata framework, including following operation:
1) finite-state automata static models are constructed in embedded systems, including computer data structure and corresponding
Function of movement:Communication protocol text to be processed, the data structure computer of structural regime matrix are needed according to communication server
It is interior to represent, the resolving of stipulations is resolved into a series of function, includes the state transition rules of each state matrix element
And its function of movement;
2) real-time data base is constructed:With in the computer of the real time data used in memory database realizes communication server
Portion is represented;
3) dynamic model of finite-state automata is constructed, so as to the operationally processing communication immediately of intelligent communication server
The input of message, output and system event;
4) knowledge base of diagnostic expert system is constructed:Invite expert to provide the knowledge rule collection of diagnostic expert system, make
With configuration instrument typing intelligent communication server, and it is automatically converted to computer-internal and represents, so as to forms knowledge base;
5) Organon of causation is realized, so as to the event activation that intelligent communication server can be operationally preset, root
According to the answer of the data of real-time data base, and person on duty, logical deduction is carried out;The inference conclusion drawn passes through man-machine interface exhibition
Show to operator on duty, and play corresponding background music to point out the relevant personnel.
Compared with prior art, the present invention has following beneficial technique effect:
The intelligent communication server and its construction method based on finite-state automata framework that the present invention is provided, in " intelligence
The system model of " finite-state automata " is constructed in energy communication server ", is parsed using finite-state automata
MMS and GOOSE communication protocols in the stipulations of the IEC61850 worlds, " finite-state automata " can then effectively filter out randomness
" error code ", so as to improve the stability and " fault-tolerant " property of system, and due to the description clear layer to stipulations, debugging is very
It is convenient;" expert system " model (including knowledge base and inference machine) is also constructed in " intelligent communication server " simultaneously, for examining
This intelligent communication server apparatus break in itself and relevant equipment is with the presence or absence of failure and exception, to realize equipment and power system
Fault diagnosis functions;
The intelligent communication server and its construction method based on finite-state automata framework that the present invention is provided, also exist
Background music is played in " intelligent communication server ", makes device more hommization, i.e., according to the diagnostic result of diagnostic expert system,
Corresponding background music is played to point out the relevant personnel.So diagnostic result melody different by playing points out person on duty,
Make work on duty very happy, it is not necessary to 8 hours inflexibly keep a close watch on equipment or screen, it appears very hommization.
Brief description of the drawings
Fig. 1 is configuration diagram of the invention.The flow direction of arrows show data in figure, it is clear that have the lines of unilateral arrow
Represent that data flow is unidirectional, being represented by the lines of four-headed arrow can be with the data of two-way flow.
Fig. 2 represents for the one-dimension array of state transition matrix, to save memory space in computer, by a two-dimemsional number
Group is compressed to the expression way of an one-dimension array.
Fig. 3 is the flow chart of the finite-state automata of the present invention.
Fig. 4 is the node and its index of internal memory tree structure.
Fig. 5 is knowledge base recursion resolution function block diagram representation of the invention.
Fig. 6 is the block diagram representation of decision tree.
Embodiment
With reference to specific embodiment, the present invention is described in further detail, it is described be explanation of the invention and
It is not to limit.
The intelligent communication server and its construction method based on finite-state automata framework that the present invention is provided, are related to
Intelligent communication server construction finite-state automata and diagnostic expert system, are described separately below.
On finite-state automata:One system can be described with a "black box", and it has some end points, " input "
End represents " exciting variable " and " output " end represents " response variable ".Finite state machine model assume to have one it is independent same
Step source controls it, is just submitted by synchronisation source at the time of only discrete associated with particular event, is referred to as " during sampling
Between " tv(v=1,2 ...).
Other important definition:
Input alphabet:X={ ξ 1, ξ 2 ..., ξ p }
Output alphabet:Z={ ζ 1, ζ 2 ..., ζ q }
Finite state set:S=σ 1, σ 2 ... σ n }
Paired characteristic function fz and fs
The system that finite-state automata is described, can use form, figure, and matrix to represent it.It is given below one
The example of individual state transition matrix:
Obviously, state transition matrix is a matrix for including many empty elements, and we are referred to as " sparse matrix ".The present invention
According to the operation rule of the specification of various communication protocols and expression formula in communication server, corresponding state transition matrix is built,
And set up with running status automatic machine to parse the complex calculation such as communication protocol and parsing arithmetic expression.
On expert system:Using a large amount of expertises to solve the problem of only expert could solve;The present invention is related to
And " knowledge base " and " inference machine " two main contents in expert system.
The rule of knowledge is described in knowledge base:If<Condition>Then<Consequent>If,<Condition>For "true", then produce<
Consequent>And next stage adjacent " node " is communicated information to, otherwise information not transmit still further below by one-level.Above-listed regular set
Constitute " knowledge base ", " rule " is to be provided by brainstrust, then regular collection is stored in " knowledge base ".
The process made inferences using a large amount of rules in knowledge base is referred to as " inference machine ", is referred to as judging in expert system
Tree.Actually perform " deductive logic " (Deductive Logic);It is triggered by " event ", is searched for afterwards in knowledge base
Corresponding root node, and step by step in resolution rules<Condition>, passed on downwards if " condition " is "true", i.e. causal reasoning.
Because the quantity of rule is very big, it is understood that there may be potential uncertain and conflict, the conclusion of reasoning allows that generation is multiple, and expert is also
" priority " attribute can be endowed, inference machine is queued up parsing according to expert's priority.
Referring to Fig. 1, a kind of intelligent communication server based on finite-state automata framework, including finite-state automata
System, real-time data base F and diagnostic expert system;
Described finite-state automata system includes:
State matrix A, constructs the static models and dynamic model of finite-state automata, and place is needed according to communication server
The communication protocol text of reason, the data structure of structural regime matrix calculates internal machine representation, and the resolving of stipulations is resolved into one
The function of series, includes the state transition rules and its function of movement of each state matrix element;
Circulate thread B, be the thread of Infinite Cyclic, its usually in a dormant state, when finding have from incoming message C
Message input, or have a system event generation, circulation thread is activated, and state automata is called after activation;State automata
Incoming message is read, according to state transition matrix transition status, and the function of movement of corresponding state element is called;
Incoming message C, is the application layer data of message, including the incoming message on passage passes through physical layer, data link
Layer, Internet, transport layer, session layer and expression layer, are finally sent to application layer from bottom to top;
" action " collection of functions E of state elements, the analysis result of reception state automatic machine is divided packet parsing function
Solve and constitute the set of all polygamma functions;
Outgoing message D, is the application layer data of message, top-down to pass through expression layer, session layer, transport layer, network
Layer and data link layer, passage is sent to up finally by physical layer;The coordination work of the action function of state automata
Real time data write into Databasce needed for being formed, or outgoing message output is formed according to communication protocol;
Described real-time data base, is the computer with the real time data used in memory database realizes communication server
Internal representation, deposits the real time data of power system;
Described diagnostic expert system includes:
Knowledge base G, knowledge rule set constitutes knowledge base;
Organon of causation I, causal reasoning is carried out using knowledge base:Startup and parsing is called when needing to carry out fault diagnosis
The function of knowledge rule, rule parsing is carried out by performing deductive logic;
Person on duty answer J, in Organon of causation I resolvings, real-time number is read if data dot values are included in condition
According to storehouse, man-machine interaction is carried out if comprising person on duty's answer and reads person on duty's answer;
Inference conclusion feedback form H, collects the conclusion of Organon of causation I parsings.
A kind of construction method of the intelligent communication server based on finite-state automata framework, including following operation:
Step 1:In embedded systems construct finite-state automata static models, including computer data structure and
Accordingly " function of movement ":
Table in communication protocol text to be processed, the data structure computer of structural regime matrix is needed according to communication server
Show, including each state matrix element state transition rules and its " function of movement ", its essence is by the resolving of stipulations
A series of function is resolved into complete.
It has to be noted here that the input of state automata not only include incoming message, also including it is other " event input ", for example
Command event of time out event and person on duty etc..
Step 2:Construct real-time data base:
Realize that the computer-internal of the real time data used in communication server is represented, usually use memory database.Remove
Outside the data item of message input, in addition to by arithmetic or logical expression calculate " amount of calculation ", it is and on duty
Data of member's setting etc..
Step 3:The dynamic model of finite-state automata is constructed, so that intelligent communication server " " is being located immediately during operation
Manage the input of communication packet, output and system event.
Step 4:" knowledge base " of " diagnostic expert system " is constructed, that is, invites expert to provide " knowing for diagnostic expert system
Know rule " collection, use " configuration instrument " typing intelligent communication server, and be automatically converted to computer-internal and represent, so that shape
Into " knowledge base ".
Step 5:Organon of causation is realized, so that intelligent communication server is in " " " event " that can be preset during operation
Activation, according to the answer of the data of real-time data base, and person on duty, carries out logical deduction.The inference conclusion drawn passes through people
Machine showing interface plays corresponding background music to point out the relevant personnel to operator on duty.
It is specifically described respectively with constructing finity state machine and diagnostic expert system separately below.
1) on finite state machine model
In three kinds of mathematical descriptions of finite-state automata, it is best suitable for being accomplished that State Transferring square on computers
Battle array:
(1) " state elements ", are firstly introduced into, it is defined as a structure:
Eij={ xj, σj, aij} (3.1)
Wherein:I-th row of i- representing matrixs, i.e. original state are σ i;
J- represents that jth is arranged, and it is σ to represent next statej;
Xj- represent the inputs of the state elements;
σj- represent next state of the element;
aij- represent " action " of the element, typically represent call a function in a computer, if this yuan have it is " defeated
Go out " also realize it in the function.
(2) we are still referred to as " state transition matrix " to the matrix, being made up of state elements, and it and original state turn
It is equally also " sparse " matrix to change matrix, and empty element, which need not be stored, need not also carry out computing.To save memory space,
We can neglect these sky elements, and state transition matrix is represented with an one-dimension array;But need increase
One array of indexes p indicates the original position of every a line, as shown in Figure 2.
(3) finite-state automata, is realized on " intelligent communication server ", that is, realizes the State Transferring (one shown in Fig. 2
Dimension) matrix, expressed to be expressed with flow chart, it is necessary to be converted to programming language:
State elements are defined as a structure:
." state elements " (such as E that the same current state (such as σ i) of correspondence includesi0, Ei1...), order is deposited;Again
Arranged according to the sequencing of state, be formed structural array tran [];
.The position of an element of foremost in every group of state elements, index of reference array p [] is indicated;
.In one special state elements of last increase of every group of state elements, its " input " is " other ", i.e., with it is upper
The element that the element that face is enumerated all is not matched." other inputs " is represented with 0 in structural array, notices that this 0 is not
Represent that the element is " empty element " but represents " other inputs ".
.Variable State represents current state, and variable pp represents current transition matrix index, and variable i nEle represents defeated
Enter symbol.
The logic diagram of " state automata " is as shown in Figure 3.
The workflow of explained later once state automata:
1. P represents the array of pointers of state elements table, and current state is stored in State variables, and pp ← p [State] is represented
Variable pp is sent by first position of the corresponding state elements of current state;
2. judge " input " (tran [pp] .in) in the signified state elements structure of pp variables whether with incoming symbol
InEle matches;Gone to 5. if matching;Gone to 3. if mismatching;
3. determine whether whether pp pointers have reached the afterbody of the corresponding state elements of the state, be if reaching afterbody
Last element, then tran [pp] .in is necessarily equal to<Other inputs>(being represented with 0);If "Yes" is so namely
The element is have matched, is gone to 5.;Otherwise continue to 4.;
4. 2. pp variables go to afterwards to moving down one;
5. " next transition stage " in the state elements structure matched with incoming symbol is tran [pp] .next, such as
Fruit is less than 0 and represents that next transition stage is identical with current state, goes to frame 7.;Otherwise go to downwards 6.;
6. " the next transition stage " of the state elements is sent to current state variable State ← tran [pp] .next;
7. " function of movement " tran [pp] .action of the state elements is called;
8. return
2) on Expert System Model
The present invention will realize " knowledge base " and " Organon of causation " on " intelligent communication server ", with reference to sentencing for knowledge base
It is fixed to set to illustrate the specific method realized.
(1) rule, is defined first
If<Condition>Then<Consequent>
If<Condition>For "true", then produce<Consequent>And next stage adjacent " rule " is communicated information to, otherwise believe
Breath not transmit still further below by one-level.
The definition of " condition ":<Condition>::=^<Condition>|<Data dot values>|<Person on duty's answer>|<Finish condition>
Wherein:" ^ " symbol represents " non-" that non-real is false, and non-vacation is true;
Wherein:<Analog quantity>The real time data that expression is represented with floating number;
<Digital quantity>That is switching value real time data;
<Amount of calculation>The real time data obtained after being calculated by arithmetic and logical expression.
(2) knowledge base, being made up of substantial amounts of rule, for representing correlation is formed between " rule " tree-like knot
Structure, being stored in realizes the definition (configuration) of knowledge base in database.
Use for sake of convenience<Cond>To represent " condition ";Quoting the method for " knowledge node " is:Information is from original node
" condition " for "true" that current " rule " will pass through is reached, form is:
.<Cond1>.<Cond2>....<Condn>.
Wherein:
<Cond1>……<Condn>It in the path that the rule passes through is genuine condition to be.
" " of foremost represents also to have one " " to represent information transmission behind root node, each condition.
<Condn>It is this grade of condition.Part before this grade of condition is " higher level's knowledge node ".
Illustrate this knowledge node with an example below:Assuming that there is three below condition, i.e. XL1_ in knowledge base
HW represents that the position of the switch of circuit 1 is being closed a floodgate, and XL1_TW represents the position of the switch relay of circuit 1 in trip position, XL1_JX
It is " circuit 1, which is switched, to be overhauled " to represent operator's answer.So " .XL1_HW.XL1_JX.XL1_TW. " is a knowledge knot
Point, entitled " switch of circuit 1 is tested ";Similar " .XL1_HW.^XL1_JX.XL1_TW. " is also a knowledge
Node, entitled " secondary circuit failure that circuit 1 is switched ".
In fact, can there be accordingly " action " each knowledge in knowledge base, " the thing for example reported to person on duty
Part " content;What etc. the background music that device is played be.
In order to create more humane working environment to operator on duty, intelligent communication server is playing background sound always
Happy, system plays corresponding music according to the diagnostic result of diagnostic expert system, does not have event occurs just to play easily sound
It is happy, the music that can be come into the picture is played according to the order of severity for occurring event if event occurs.Operator on duty is not
Screen may be monitor constantly, but can be according to the background music of device come the working condition of identifying device.
(3) tree structure, deposited in database is, it is necessary to read in real-time data base (i.e. memory database), Cai Nengman
The requirement for the real-time that foot is calculated, " node " and its attribute of tree structure are deposited with internal storage data block, and tree structure
" branch road " index of reference represents, as shown in Figure 4:
Wherein:" node parameters " include this node " condition ", higher level's node " condition ", the expert info and priority of association
Deng;
chainBackIdx:This grade of node is indexed forward;
chainIdx:This grade of node is indexed backward;
fatherIdx:Point to higher level's node (father node);
childIdx:Point to subordinate's node (child's node).
(4) inference machine, is realized on " intelligent communication server ", is substantially to travel through and parse knowledge as shown in Figure 4
The decision tree that rule is constituted;The decision tree that example described above is constituted is as shown in Figure 6.Wherein the branch on the left side can reduce
For knowledge:
Root node
Closing a floodgate the position of the switch of If circuits 1
THEN switches are in work
Switch is in work
IF^ circuits 1, which are switched, to be overhauled
THEN switchs charging operation
Switch charging operation
The position of the switch relay of IF circuits 1 is in trip position
The secondary circuit failure that THEN circuits 1 are switched
In other words, it is deductive procedure to decision tree traversal and resolving, its path passed through is constituting " rule " just
To inference chain:Root node → switch is switched in work, switch in work → switch charging operation, switch charging operation → circuit 1
Secondary circuit failure.
If we parse any rule with a function, then this function must be " recurrence ", so its core
The heart is to call the function of one " recurrence ".
" inference machine " of the present invention, it is triggered by " event ", the corresponding root node searched for afterwards in knowledge base, is solved step by step
In analysis rule<Condition>And according to condition transmission, i.e. causal reasoning downwards.Because the quantity of rule is very big, it is understood that there may be potential
Uncertain and conflict, even if inference machine can be according to the priority queueing of " rule ", the conclusion of reasoning still has multiple;Push away
On the one hand the conclusion of reason reports person on duty by dialog box, on the other hand then plays corresponding background music and reminds operator on duty.
In addition, this inference machine can also receive " person on duty's answer " by diagnosing dialog box, because they can not be represented by " data point "
" condition ", can only artificially be answered by person on duty.
Fig. 5 is the block diagram for the recursive function for parsing " rule ", and the recursive function of parsing knowledge rule is the core of inference machine
The heart:
.In order to be shown to the inference conclusion of operations staff " knowledge rule tree ", and show that request person on duty provides answer
Devised in entry, program one " feedback form ", its entry is probably dynamic increase and decrease.Two functions are refer in Fig. 5:
CopyOneCon (tIdx, cNodeIdx) function representations copy node parameters " feedback form " to, and AddOneCon
(cNodeIdx) one feedback table clause of increase is represented.
.Node in block diagram in decision tree is represented with structural array eleNodes [];
.Recursive function ScanKnLib has two parameters, and nodeIdx represents current father node, tIdx represent in feedback form when
The call number of preceding available items;
.Initialization block carries out variable initializer show, and the meaning of primary variables is as follows:
The entry number that the NextFlag- nodes occupy in feedback form;
SucFlag--- indicates that 0 represents that " condition " is "No", and 1 represents that " condition " is "Yes", and 2 represent that " condition " is " eventually
Knot " is " unknown ".
CNodeIdx- current nodes, childIdx indexes during initialization according to its father node are obtained, i.e.,:
cNodeIdx←eleNodes[nodeIdx].childIdx
CtIdx and tIdx--- are the call numbers of feedback form, but the latter is that the former is to move back into index during function
Go out index during function.
.From block diagram, for " data point " condition, parsing can be worth to according to the current of real-time data base at once,
This function it can continue to parse its downstream site with recursive call if " condition " is "Yes".
It is " termination " (SucFlag=2) for condition, only adds " feedback form " display, be no longer transmitted to subordinate.
It is " person on duty's answer " person for " condition ", if may determine that its true and false (example according to " condition " that has parsed
As in the past parse the condition " ^ ", it is clear that can it be negated till now again " condition " value), also can just solve immediately
Separate out;Otherwise just should " condition " addition " feedback form " wait person on duty provide answer.
.The nethermost square frame cNodeIdx of this block diagram ← eleNodes [cNodeIdx] .chainIdx represents to continue to parse
Next node at the same level.Just returned until node at the same level parsing finishes function.
Flow-interpret is as follows:
A recursive function entrances:ScanKnLib (nodeIdx, tIdx).Wherein nodeIdx is father node, and tIdx is represented
The current call number of " feedback form ".Due to starting to parse always since root node, so when calling the recursive function first
The two parameters are " 0 ".
B function variables are initialized:Including mark NextFlag ← SecFlag ← 0;Current node index cNodeIdx ←
EleNodes [nodeIdx] .childIdx (be equal to father node son, if father node does not have son the call number for "-
1”);Feedback form working cable quotation marks:ctIdx←tIdx.
C decision blocks:Whether inference chain terminates.CNodeIdx < 0 represent that node has been parsed and finished, and go to returns frame W.It is no
Then order performs D.
D decision blocks:If the species (i.e. eleNodes [cNodeIdx] .CondTypeId) of the node is " data point ",
Then go to J frames.Otherwise order performs E frames.
E decision blocks:If the species (i.e. eleNodes [cNodeIdx] .CondTypeId) of the node is " person on duty answers
Case ",
Then go to G frames.Otherwise order performs F frames.
F puts mark SucFlag ← 2.K frames are gone to afterwards.
G decision blocks:Judge person on duty to this " rule " whether to answer (be included in decision tree to the rule
" no rule " gave answer, also correspond to this rule to answer), I frames are gone to if giving.Otherwise sequentially go to
H frames.
H puts mark SucFlag ← 2.K frames are gone to afterwards.
I indicates SucFlag according to original set, and put 1 if affirmative acknowledgement (ACK) is done originally, and otherwise set to 0.K is gone to afterwards
Frame.
J decision blocks:Whether the instantaneous value for judging the data point is " 0 ", and mark SucFlag ← 0 is then put if " 0 ", no
Then put mark SucFlag ← 1.Put and K frames are gone to after mark.
K decision blocks:Judgement symbol SucFlag > 0, M frames are then gone to if greater than " 0 ".Otherwise order performs L frames.
Younger brother's (i.e. next node at the same level) of current node is changed into current node, i.e. cNodeIdx ← eleNodes by L
[cNodeIdx].chainIdx.C frames are gone to afterwards, continue to parse.
M decision blocks:Judgement symbol NextFlag > 0, then turn R frames, and otherwise order performs N frames.
The N node parameters are copied to " feedback form ", i.e. call function CopyOneCon (tIdx, cNodeIdx), Zhi Houshun
Sequence goes to O.
O decision blocks:Whether judgement symbol SucFlag is equal to 1, and Q frames are gone to if being not equal to 1;Equal to 1 order is performed
P frames.
P calls the recursive function with " son " (i.e. subordinate's node) ScanKnLib of the recursion resolution node
(cNodeIdx, tIdx).Q frames are sequentially gone to afterwards.
Q mark NextFlag increases by 1, go to L frames afterwards.
R increases a feedback table clause, and send ctIdx, i.e. ctIdx ← AddOneCon by increased entry index
(cNodeIdx).S frames are sequentially gone to afterwards.
S decision blocks:Whether judgement symbol SucFlag is equal to 1, and Q frames are gone to if being not equal to 1;Equal to 1 order is performed
T frames.
T calls the recursive function with " son " (i.e. subordinate's node) ScanKnLib of the recursion resolution node
(cNodeIdx, tIdx).Q frames are sequentially gone to afterwards.
W function is exported, and returns to ctIdx.
Example given above is to realize the present invention preferably example, and the invention is not restricted to above-described embodiment.This area
Technical staff any nonessential addition, the replacement made according to the technical characteristic of technical solution of the present invention, belong to this
The protection domain of invention.
Claims (10)
1. a kind of intelligent communication server based on finite-state automata framework, it is characterised in that including finity state machine
Machine system, real-time data base (F) and diagnostic expert system;
Described finite-state automata system includes:
State matrix (A), constructs the static models and dynamic model of finite-state automata, and processing is needed according to communication server
Communication protocol text, the data structure of structural regime matrix calculates internal machine representation, and the resolving of stipulations is resolved into one is
The function of row, includes the state transition rules and its function of movement of each state matrix element;
Circulate thread (B), be the thread of Infinite Cyclic, its usually in a dormant state, when finding have from incoming message (C)
Message is inputted, or has system event generation, and circulation thread is activated, and state automata is called after activation;State automata is read
Incoming message is taken, according to state transition matrix transition status, and the function of movement of corresponding state element is called;
Incoming message (C), is the application layer data of message, including the incoming message on passage is by physical layer, data link layer,
Internet, transport layer, session layer and expression layer, are finally sent to application layer from bottom to top;
" action " collection of functions (E) of state elements, the analysis result of reception state automatic machine is decomposed packet parsing function
And constitute the set of all polygamma functions;
Outgoing message (D), is the application layer data of message, top-down to pass through expression layer, session layer, transport layer, Internet
And data link layer, it is sent to passage up finally by physical layer;The co-ordination of the action function of state automata
Real time data write into Databasce needed for being formed, or outgoing message output is formed according to communication protocol;
Described real-time data base (F), is the computer with the real time data used in memory database realizes communication server
Internal representation, the real time data of storage system;
Described diagnostic expert system includes:
Knowledge base (G), knowledge rule set constitutes knowledge base;
Organon of causation (I), causal reasoning is carried out using knowledge base:Startup and parsing is called to know when needing to carry out fault diagnosis
Know the function of rule, rule parsing is carried out by performing deductive logic;
Person on duty's answer (J), in Organon of causation (I) resolving, real-time number is read if data dot values are included in condition
According to storehouse, man-machine interaction is carried out if comprising person on duty's answer and reads person on duty's answer;
Inference conclusion feedback form (H), collects the conclusion of Organon of causation (I) parsing.
2. the intelligent communication server as claimed in claim 1 based on finite-state automata framework, it is characterised in that described
State matrix (A) be the matrix being made up of state elements, it is expressed as:Eij={ xj, σj, aij, wherein:I is matrix
I-th row, j arranges for the jth of matrix, xjFor the input of the state elements, σjFor next state of the element;aijFor the element
A function is called in action, expression, is also realized if this yuan have " output " in the function;
Also state matrix (A) is represented with an one-dimension array, and increases an array of indexes p to indicate the starting of every a line
Position.
3. the intelligent communication server as claimed in claim 2 based on finite-state automata framework, it is characterised in that will also
State matrix (A) carries out following change:
Wherein, struct Tran correspondences Eij, short in correspondences xj, short next correspondences σj, Action action correspondences
aij;
The state elements order of the same current state of correspondence is deposited, and is arranged according to the sequencing of state, forms structure number
Group tran [];The position of an element of foremost in every group of state elements, index of reference array p [] is indicated;In every group of state
One special state elements of last increase of element, its input is " other ", i.e., all do not matched with the element enumerated
Element, is represented " other inputs " in structural array with 0.
4. the intelligent communication server as claimed in claim 3 based on finite-state automata framework, it is characterised in that state
Automatic machine it is as follows to calling for state elements:
1. the array of pointers of state elements table is represented with P, current state is stored in State variables, by the corresponding shape of current state
Send variable pp in first position of state element:pp←p[State];Wherein, variable State represents current state, and variable pp is represented
Current transition matrix index;
2. judge input tran [pp] .in in the signified state elements structure of pp variables whether with incoming symbol inEle phases
Match somebody with somebody;Gone to 5. if matching;Gone to 3. if mismatching;Wherein, variable i nEle represents incoming symbol;
3. determine whether whether pp pointers have reached the afterbody of the corresponding state elements of the state, if it is last to reach afterbody
One element, then tran [pp] .in is necessarily equal to<Other inputs>, it is represented with 0;If "Yes" is so namely matched
5. the element, go to;Otherwise continue to 4.;
4. 2. pp variables go to afterwards to moving down one;
5. next transition stage in the state elements structure matched with incoming symbol is tran [pp] .next, if less than 0
Represent that next transition stage is identical with current state, go to frame 7.;Otherwise go to downwards 6.;
6. " the next transition stage " of the state elements is sent to current state variable State ← tran [pp] .next;
7. " function of movement " tran [pp] .action of the state elements is called;
8. return.
5. the intelligent communication server as claimed in claim 1 based on finite-state automata framework, it is characterised in that described
Knowledge base in knowledge rule be defined as:If<Condition>Then<Consequent>;
If<Condition>It is true, then produces<Consequent>And communicate information to the adjacent rule of next stage, otherwise information no longer to
Next stage is transmitted;
The definition of condition:<Condition>::=∧<Condition>|<Data dot values>|<Person on duty's answer>|<Finish condition>;Wherein:
" ∧ " symbol represents " non-" that non-real is false, and non-vacation is true;<Data dot values>It is expressed as the value of real-time data base data point;<It is on duty
Member's answer>Expression is provided the condition of answer by person on duty;<Finish condition>The node not transmitted still further below is represented, for termination knot
Point;
<Data dot values>::=<Analog quantity>|<Digital quantity>|<Amount of calculation>;Wherein:<Analog quantity>Expression is represented with floating number
Real time data;<Digital quantity>That is switching value real time data;<Amount of calculation>After being calculated by arithmetic and logical expression
The real time data arrived;
The tree structure that correlation is formed between the set of knowledge rule, and expression rule, is stored in database i.e.
For knowledge base;Knowledge base reads in the requirement that real-time data base meets the real-time calculated by tree structure.
6. the intelligent communication server as claimed in claim 5 based on finite-state automata framework, it is characterised in that if with
<Cond>To represent condition, quoting knowledge node is:Information from original point reach that current " rule " will pass through for "true"
" condition ", form is<Cond1>.<Cond2>.....<Condn>, wherein,<Cond1>....<Condn>It is the regular institute
It is genuine condition in the path passed through;" " of foremost represents also to have one " " to represent letter behind root node, each condition
Breath transmission;<Condn>It is this grade of condition, the part before this grade of condition is " higher level's knowledge node ".
7. the intelligent communication server as claimed in claim 5 based on finite-state automata framework, it is characterised in that knowing
Know the tree structure deposited in storehouse, its node and its attribute are deposited with internal storage data block, the branch road index of reference of tree structure comes
Represent;
The node parameters of tree structure include this node condition, higher level's node condition, the expert info and priority of association;It is tree-like
The index of structure includes:This grade of node is indexed forward, this grade of node is indexed backward, point to higher level's node and point to subordinate's node;
The intelligent communication server is playing background music always:Play corresponding according to the diagnostic result of diagnostic expert system
Music, do not have that event occurs and just play easily music;Broadcast if event occurs according to the order of severity for occurring event
The music that exoergic comes into the picture;
Operator on duty can be according to the background music of device come the working condition of identifying device.
8. the intelligent communication server as claimed in claim 1 based on finite-state automata framework, it is characterised in that described
The causal reasoning that is carried out of Organon of causation be decision tree that traversal and parsing are made up of knowledge rule, decision tree is traveled through
It is deductive procedure with resolving, the forward chaining of its path composition rule passed through:A → B, B → C, C → D...;
The function parsing of Organon of causation is triggered by event, the corresponding root node searched for afterwards in knowledge base, step by step parsing rule
In then<Condition>And according to condition transmission downwards;
On the one hand the conclusion of reasoning reports person on duty by dialog box, on the other hand then plays corresponding background music and reminds on duty
Personnel;
Organon of causation also receives person on duty's answer by diagnosing dialog box.
9. the intelligent communication server as claimed in claim 8 based on finite-state automata framework, it is characterised in that be based on
Feedback form, the process of analysis of the recursive function of described Organon of causation resolution rules is:
A recursive function entrances:ScanKnLib (nodeIdx, tIdx), wherein nodeIdx are father node, and tIdx represents " feedback
The current call number of table ";
B function variables are initialized:Including mark NextFlag ← SecFlag ← 0;Current node index cNodeIdx ←
EleNodes [nodeIdx] .childIdx, equal to the son of father node, if father node does not have son the call number for "-
1”;Feedback form working cable quotation marks:ctIdx←tIdx;
C decision blocks:Whether inference chain terminates, cNodeIdx<0 expression node, which has been parsed, to be finished, and goes to return W frames, otherwise suitable
Sequence performs D;
D decision blocks:If the species of the node is " data point ", J frames are gone to;Otherwise order performs E frames:
E decision blocks:If the species of the node is " person on duty's answer ", G frames are gone to;Otherwise order performs F frames;
F puts mark SucFlag ← 2, and K frames are gone to afterwards;
G decision blocks:Judge person on duty to this " rule " whether to answer, I frames are gone to if giving, otherwise order
Go to H frames;
H puts mark SucFlag ← 2, and K frames are gone to afterwards;
I indicates SucFlag according to original set, and put 1 if affirmative acknowledgement (ACK) is done originally, and otherwise set to 0;K frames are gone to afterwards;
J decision blocks:Whether the instantaneous value for judging the data point is " 0 ", then puts mark SucFlag ← 0 if " 0 ", otherwise puts
Indicate SucFlag ← 1;Put and K frames are gone to after mark;
K decision blocks:Judgement symbol SucFlag > 0, M frames are then gone to if greater than " 0 ", and otherwise order performs L frames;
Next node at the same level of current node is changed into current node, i.e. cNodeIdx ← eleNodes [cNodeIdx] by L
.chainIdx;C frames are gone to afterwards, continue to parse;
M decision blocks:Judgement symbol NextFlag > 0, then turn R frames, and otherwise order performs N frames;
The N node parameters are copied to " feedback form ", i.e. call function CopyOneCon, and O is sequentially gone to afterwards;
O decision blocks:Whether judgement symbol SucFlag is equal to 1, and Q frames are gone to if being not equal to 1;Equal to 1 order performs P frames;
P calls the recursive function with the subordinate node ScanKnLib (cNodeIdx, tIdx) of the recursion resolution node, Zhi Houshun
Sequence goes to Q frames;
Q mark NextFlag increases by 1, go to L frames afterwards;
R increases a feedback table clause, and send ctIdx, i.e. ctIdx ← AddOneCon by increased entry index
(cNodeIdx) S frames, are sequentially gone to afterwards;
S decision blocks:Whether judgement symbol SucFlag is equal to 1, and Q frames are gone to if being not equal to 1;Equal to 1 order performs T frames;
T calls the recursive function with the subordinate node ScanKnLib (cNodeIdx, tIdx) of the recursion resolution node, Zhi Houshun
Sequence goes to Q frames;
W recursive functions are exported.
10. a kind of construction method of the intelligent communication server based on finite-state automata framework, it is characterised in that including with
Lower operation:
1) finite-state automata static models are constructed in embedded systems, including computer data structure and corresponding dynamic
Make function:Table in communication protocol text to be processed, the data structure computer of structural regime matrix is needed according to communication server
Show, the resolving of stipulations resolved into a series of function, including each state matrix element state transition rules and its
Function of movement;
2) real-time data base is constructed:With the computer-internal table of the real time data used in memory database realizes communication server
Show;
3) dynamic model of finite-state automata is constructed, so that intelligent communication server operationally handles communication packet immediately
Input, output and system event;
4) knowledge base of diagnostic expert system is constructed:Invite expert to provide the knowledge rule collection of diagnostic expert system, use group
State instrument typing intelligent communication server, and be automatically converted to computer-internal and represent, so as to form knowledge base;
5) Organon of causation is realized, so as to the event activation that intelligent communication server can be operationally preset, according to reality
When database data, and person on duty answer, carry out logical deduction;The inference conclusion drawn is showed by man-machine interface
Operator on duty, and play corresponding background music to point out the relevant personnel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710524565.3A CN107317866B (en) | 2017-06-30 | 2017-06-30 | Intelligent communication server based on finite state automata architecture and construction method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710524565.3A CN107317866B (en) | 2017-06-30 | 2017-06-30 | Intelligent communication server based on finite state automata architecture and construction method thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107317866A true CN107317866A (en) | 2017-11-03 |
CN107317866B CN107317866B (en) | 2020-07-14 |
Family
ID=60179810
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710524565.3A Active CN107317866B (en) | 2017-06-30 | 2017-06-30 | Intelligent communication server based on finite state automata architecture and construction method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107317866B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114179613A (en) * | 2021-12-10 | 2022-03-15 | 常州星宇车灯股份有限公司 | Audio-video touch interactive control method for copilot control panel |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7092858B1 (en) * | 2001-01-17 | 2006-08-15 | Synopsys, Inc. | Method and apparatus for formally constraining random simulation |
CN102630088A (en) * | 2012-03-12 | 2012-08-08 | 北京中创信测科技股份有限公司 | Method for carrying out signaling analysis through customizing state machine |
CN102902664A (en) * | 2012-08-15 | 2013-01-30 | 中山大学 | Artificial intelligence natural language operation system on intelligent terminal |
CN103020219A (en) * | 2012-12-10 | 2013-04-03 | 广东电网公司电力科学研究院 | Network-based expert system tool for power grid fault diagnosis |
CN103426028A (en) * | 2013-08-23 | 2013-12-04 | 北京化工大学 | Fault diagnosis expert system based on inheritance pattern and fault diagnosis method |
CN104506338A (en) * | 2014-11-21 | 2015-04-08 | 河南中烟工业有限责任公司 | Fault diagnosis expert system based on decision tree for industrial Ethernet network |
CN106227584A (en) * | 2016-07-28 | 2016-12-14 | 武汉聚风天下科技有限公司 | A kind of event-driven method based on state machine and system |
CN106502770A (en) * | 2015-09-07 | 2017-03-15 | 上海畅星软件有限公司 | A kind of HMI state transfer methods based on finite state machine |
-
2017
- 2017-06-30 CN CN201710524565.3A patent/CN107317866B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7092858B1 (en) * | 2001-01-17 | 2006-08-15 | Synopsys, Inc. | Method and apparatus for formally constraining random simulation |
CN102630088A (en) * | 2012-03-12 | 2012-08-08 | 北京中创信测科技股份有限公司 | Method for carrying out signaling analysis through customizing state machine |
CN102902664A (en) * | 2012-08-15 | 2013-01-30 | 中山大学 | Artificial intelligence natural language operation system on intelligent terminal |
CN103020219A (en) * | 2012-12-10 | 2013-04-03 | 广东电网公司电力科学研究院 | Network-based expert system tool for power grid fault diagnosis |
CN103426028A (en) * | 2013-08-23 | 2013-12-04 | 北京化工大学 | Fault diagnosis expert system based on inheritance pattern and fault diagnosis method |
CN104506338A (en) * | 2014-11-21 | 2015-04-08 | 河南中烟工业有限责任公司 | Fault diagnosis expert system based on decision tree for industrial Ethernet network |
CN106502770A (en) * | 2015-09-07 | 2017-03-15 | 上海畅星软件有限公司 | A kind of HMI state transfer methods based on finite state machine |
CN106227584A (en) * | 2016-07-28 | 2016-12-14 | 武汉聚风天下科技有限公司 | A kind of event-driven method based on state machine and system |
Non-Patent Citations (1)
Title |
---|
寇红召: "基于云模型的自适应网络安全通信系统", 《计算机工程》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114179613A (en) * | 2021-12-10 | 2022-03-15 | 常州星宇车灯股份有限公司 | Audio-video touch interactive control method for copilot control panel |
CN114179613B (en) * | 2021-12-10 | 2024-03-05 | 常州星宇车灯股份有限公司 | Audio-video touch interactive control method for co-driver control panel |
Also Published As
Publication number | Publication date |
---|---|
CN107317866B (en) | 2020-07-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109902165A (en) | Intelligent interactive answering method, system, device based on Markov Logic Networks | |
Warwick et al. | Artificial intelligence techniques in power systems | |
CN107272646B (en) | Press fault diagnosis system based on expert system | |
CN102982097B (en) | Domain for Knowledge based engineering data quality solution | |
CN106844194A (en) | A kind of construction method of multi-level software fault diagnosis expert system | |
CN104331543A (en) | Fault diagnostic expert system for marine electrical propulsion system and establishing method thereof | |
CN107632590B (en) | A kind of bottom event sort method priority-based | |
CN106874963B (en) | A kind of Fault Diagnosis Method for Distribution Networks and system based on big data technology | |
CN108663581A (en) | A kind of secondary equipment of intelligent converting station test method | |
CN114330138A (en) | Fault diagnosis method and device and electronic equipment | |
CN110535723A (en) | The message method for detecting abnormality of deep learning is used in a kind of SDN | |
CN112217674A (en) | Alarm root cause identification method based on causal network mining and graph attention network | |
CN106291339A (en) | A kind of circuit breaker failure diagnostic expert system based on artificial neural network | |
Sayyari et al. | Automated generation of software testing path based on ant colony | |
He et al. | Fault diagnosis and location based on graph neural network in telecom networks | |
CN115170344A (en) | Intelligent processing method and device, medium and equipment for operation events of regulation and control system | |
CN116771576A (en) | Comprehensive fault diagnosis method for hydroelectric generating set | |
CN107317866A (en) | A kind of intelligent communication server and its construction method based on finite-state automata framework | |
Snel et al. | Multi-task evolutionary shaping without pre-specified representations | |
CN108121796A (en) | Electric energy metering device failure analysis methods and device based on confidence level | |
CN106202805B (en) | A kind of switchgear method for evaluating reliability | |
Grebenyuk et al. | Technological infrastructure management models and methods based on digital twins | |
Marsan et al. | TOPNET: A tool for the visual simulation of communication networks | |
CN113934862B (en) | Community security risk prediction method, device, electronic equipment and medium | |
CN116346392A (en) | Network security situation prediction method and system based on Tranformer-CNN model and application thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |