CN108960673A - Sewage treatment method for diagnosing faults and device - Google Patents
Sewage treatment method for diagnosing faults and device Download PDFInfo
- Publication number
- CN108960673A CN108960673A CN201810821381.8A CN201810821381A CN108960673A CN 108960673 A CN108960673 A CN 108960673A CN 201810821381 A CN201810821381 A CN 201810821381A CN 108960673 A CN108960673 A CN 108960673A
- Authority
- CN
- China
- Prior art keywords
- node
- sewage treatment
- semantic
- clause
- failure
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The present invention provides a kind of sewage treatment method for diagnosing faults and device, and the sewage treatment method for diagnosing faults includes: at least one fault message received about sewage treatment;At least one described fault message is sent to expert module, and receives fault diagnosis result corresponding with each fault message from expert module;Or, sewage treatment knowledge base is called to determine fault diagnosis result corresponding with each fault message based on each fault message, wherein, the sewage treatment knowledge base is the sewage treatment knowledge base established based on an at least semantic tree corresponding with sewage treatment process.Sewage treatment method for diagnosing faults of the invention and device, the accuracy of fault message description can be improved, the knowledge in sewage treatment knowledge base is dynamically updated, interaction can also be carried out with online technique expert, to obtain appropriate effective fault diagnosis result.
Description
Technical field
All things considered of the present invention is related to technical field of sewage, more particularly, is related to a kind of sewage treatment failure and examines
Disconnected method and apparatus.
Background technique
Sewage treatment is a complicated process, and common tertiary treatment process includes physical method, biological method and change
Method.It is influenced by resident living rule, industrial wastewater discharge, weather and geographical conditions etc., the water of sewage treatment plant inflow
Certain fluctuation can be presented in matter and water, this makes sewage disposal process have multivariable, non-linear, time variation and not know
The characteristics of property.Therefore, in the operational process of sewage treatment plant, it is often necessary to which supervisory engineering staff implement intervention appropriate,
The normal operation that can guarantee sewage treatment process makes to be discharged qualified discharge.
Currently, usually by expert system application in the fault diagnosis of sewage treatment process, to be technical management people
Member provides the scheme or measure that can solve practical problem, to make up the deficiency of sewage treatment practical experience, improves operation pipe
Reason level and efficiency, save operating cost.Expert system is as a kind of general aid decision-making system, by knowledge base, inference machine
And accessory is constituted, and is focused primarily upon the expression of knowledge in knowledge base in application and the expansion of sewage treatment field and is known
Know the building in library.
However, existing sewage treatment expert system knowledge base is often directed to the FAQs in sewage treatment process
It is constructed, is difficult to be provided according to the failure problems occurred in sewage treatment process sometimes satisfactory as a result, reason
It essentially consists in following three aspects: (1) inputting information inaccuracy, it is excessively single to provide result.For example, due to sewage treatment expert
Not perfect, the phenomenon of the failure description of system does not know and influences that the condition for the result that sewage treatment expert system provides is more etc.
Input information inaccuracy, the result for providing sewage treatment expert system caused by reason are difficult to obtain good desired effect.
(2) knowledge in knowledge base neatly cannot be changed dynamically.For example, the knowledge base foundation of some sewage treatment expert systems finishes
Do not allow to increase new knowledge later, or every for quite a long time, knowledge base is re-established once again, seriously affected
To the practicability and validity of sewage treatment expert system.(3) lack the interaction with online technique expert.
Summary of the invention
The purpose of the present invention is to provide a kind of sewage treatment method for diagnosing faults and devices, and fault message can be improved and retouch
The accuracy stated dynamically updates the knowledge in sewage treatment knowledge base, can also carry out interaction with online technique expert,
To obtain appropriate effective fault diagnosis result.
An aspect of of the present present invention provides a kind of sewage treatment method for diagnosing faults, the sewage treatment method for diagnosing faults packet
It includes: receiving at least one fault message about sewage treatment;At least one described fault message is sent to expert module, and
And fault diagnosis result corresponding with each fault message is received from expert module;Alternatively, being called based on each fault message dirty
Water process knowledge base determines fault diagnosis result corresponding with each fault message, wherein the sewage treatment knowledge base is base
In the sewage treatment knowledge base that an at least semantic tree corresponding with sewage treatment process is established.
Optionally, the step of sending expert module at least one described fault message includes: that will assist in solicited message
The expert module is sent jointly to at least one described fault message.
Optionally, the sewage treatment method for diagnosing faults further include: receive the selection to expert module, wherein by institute
It states at least one fault message and is sent to expert module, and receive failure corresponding with each fault message from expert module and examine
The step of disconnected result includes: that will assist in solicited message to send jointly to expert's mould of selection at least one described fault message
Block;Fault diagnosis result corresponding with each fault message is received from the expert module of selection, wherein from the expert module of selection
Received fault diagnosis result corresponding with each fault message is based on empirically determined fault diagnosis result or to pass through tune
The fault diagnosis result determined with sewage treatment knowledge base.
Optionally, sewage treatment knowledge base is called to determine failure corresponding with each fault message based on each fault message
The step of diagnostic result includes: to convert failure clause corresponding with each fault message for each fault message;Identification is each
The matching degree of failure clause and the semantic node in sewage treatment knowledge base;Judge whether each matching degree reaches predetermined threshold;
It deletes and is not up to failure clause corresponding to the matching degree of predetermined threshold, and based on reaching corresponding to the matching degree of predetermined threshold
Failure clause inquires corresponding fault diagnosis result from sewage treatment knowledge base.
Optionally, the sewage treatment method for diagnosing faults further include: the matching of predetermined threshold is not up in response to deleting
The corresponding failure clause of degree prompts adjustment fault message corresponding with the failure clause deleted.
Optionally, the sewage treatment knowledge base stores every semantic tree in the form of relevant database.
Optionally, every semantic tree includes being packaged by the predetermined information of each process of sewage treatment process in not
At least two semantic nodes of same layer, wherein each entry in the relevant database is corresponding with each semantic node,
The content of each program recording is to indicate the node content of each semantic node and the domain of hierarchical relationship.
Optionally, it is inquired from sewage treatment knowledge base based on failure clause corresponding to the matching degree for reaching predetermined threshold
The step of corresponding fault diagnosis result includes: that the matching degree institute for each reaching predetermined threshold is searched in sewage treatment knowledge base
The number of plies of the corresponding failure clause on semantic tree, and sorted based on the number of plies searched and determine failure sequence;Determine the event
Hinder the first semantic node and the maximum second failure clause of the number of plies corresponding to the smallest Fisrt fault clause of the number of plies in sequence
The second corresponding semantic node;When detecting third semantic node corresponding to the third failure clause in the failure sequence
When being the child node of first semantic node, judge whether the third semantic node is identical as second semantic node;
When the third semantic node is identical as second semantic node, by detecting the out-degree of the third semantic node and entering
Degree is to determine whether the third semantic node is leaf node;It, will be described when the third semantic node is leaf node
The node content of third semantic node is determined as fault diagnosis result, wherein the third failure clause is the failure sequence
In the next failure clause adjacent with the Fisrt fault clause.
Optionally, when the third semantic node and second semantic node be not identical, more by third failure clause
Newly it is the next failure clause adjacent with the third failure clause, and third semantic node is updated to and updated the
The corresponding semantic node of three failure clauses.
It optionally, will be described when detecting the third semantic node is not the child node of first semantic node
Third semantic node is determined as semantic node corresponding to the 4th failure clause, then executes and determines that the third semantic node is
It is no be leaf node the step of, wherein the 4th failure clause be the failure sequence in third failure clause's phase
An adjacent upper failure clause.
Optionally, it when the third semantic node is not leaf node, searches for all under the third semantic node
Leaf node, and the node content of all leaf nodes searched is determined as fault diagnosis result.
Optionally, the sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node;
The sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node, wherein when original node
Scoring be deleted when failing to reach predetermined threshold within a predetermined period of time.
Optionally, the step of executing the growth course of semantic tree includes: when the child node of root node or internal node includes
Clause when constituting logical complement relationship or being leaf node, will be patrolled comprising the clause that newly inputs with the clause's composition newly inputted
The node for collecting the clause of complementary relationship is added between former father and son's node as new node;When the child node of root node or internal node
It will include that the node of the clause newly inputted is made for one or multiple child nodes when the clause for including does not constitute logical complement relationship
For the new child node of root node or internal node.
Optionally, the step of executing the beta pruning process of semantic tree includes: father's section when the leaf node for executing beta pruning process
When point has the brotgher of node, the leaf node, father node and the brotgher of node are deleted, and the child node of sworn brothers' node is as the brother
The child node of the father node of younger brother's node;When executing the father node of leaf node of beta pruning process does not have the brotgher of node, institute is deleted
State leaf node and father node.
Another aspect of the present invention also provides a kind of sewage treatment trouble-shooter, the sewage treatment fault diagnosis dress
Setting includes: fault message receiving unit, is configured as receiving at least one fault message about sewage treatment;At fault message
Unit is managed, is configured as at least one described fault message being sent to expert module, and from expert module reception and each
The corresponding fault diagnosis result of fault message;Alternatively, determining and each based on each fault message calling sewage treatment knowledge base
The corresponding fault diagnosis result of fault message, wherein the sewage treatment knowledge base be based on sewage treatment process pair
The sewage treatment knowledge base that at least semantic tree answered is established.
Optionally, the faulted-phase judgment unit is also configured to will assist in solicited message and at least one described event
Barrier information sends jointly to the expert module.
Optionally, the sewage treatment trouble-shooter further include: selection receiving unit is configured as receiving to expert
The selection of module, wherein the faulted-phase judgment unit is also configured to will assist in solicited message and at least one described event
Barrier information sends jointly to the expert module of selection;Failure corresponding with each fault message is received from the expert module of selection to examine
Disconnected result, wherein from the received fault diagnosis result corresponding with each fault message of the expert module selected for based on experience
Determining fault diagnosis result or the fault diagnosis result by calling sewage treatment knowledge base to determine.
Optionally, the faulted-phase judgment unit is also configured to convert each fault message to and each failure
The corresponding failure clause of information;Identify the matching degree of the semantic node in each failure clause and sewage treatment knowledge base;Judgement
Whether each matching degree reaches predetermined threshold;It deletes and is not up to failure clause corresponding to the matching degree of predetermined threshold, and be based on
Reach failure clause corresponding to the matching degree of predetermined threshold and inquires corresponding fault diagnosis result from sewage treatment knowledge base.
Optionally, the faulted-phase judgment unit is also configured to be not up to the matching of predetermined threshold in response to deleting
The corresponding failure clause of degree prompts adjustment fault message corresponding with the failure clause deleted.
Optionally, the sewage treatment knowledge base stores every semantic tree in the form of relevant database.
Optionally, every semantic tree includes being packaged by the predetermined information of each process of sewage treatment process in not
At least two semantic nodes of same layer, wherein each entry in the relevant database is corresponding with each semantic node,
The content of each program recording is to indicate the node content of each semantic node and the domain of hierarchical relationship.
Optionally, the faulted-phase judgment unit is also configured to search in sewage treatment knowledge base and each reach
The number of plies of the failure clause on semantic tree corresponding to the matching degree of predetermined threshold, and event is determined based on the number of plies sequence searched
Hinder sequence;Determine the first semantic node corresponding to the smallest Fisrt fault clause of the number of plies in the failure sequence and the number of plies most
Second semantic node corresponding to the second big failure clause;When detecting that the third failure clause institute in the failure sequence is right
When the third semantic node answered is the child node of first semantic node, judge the third semantic node whether with described
Two semantic nodes are identical;When the third semantic node is identical as second semantic node, by detecting the third language
The out-degree and in-degree of adopted node determines whether the third semantic node is leaf node;When the third semantic node is leaf
When child node, the node content of the third semantic node is determined as fault diagnosis result, wherein the third failure clause
For next failure clause adjacent with the Fisrt fault clause in the failure sequence.
Optionally, the faulted-phase judgment unit is also configured to when the third semantic node and second language
When adopted node is not identical, third failure clause is updated to the next failure clause adjacent with the third failure clause, and
Third semantic node is updated to semantic node corresponding with updated third failure clause.
Optionally, the faulted-phase judgment unit, which is also configured to work as, detects that the third semantic node is not institute
When stating the child node of the first semantic node, the third semantic node is determined as the section of semanteme corresponding to the 4th failure clause
Then point executes the step of whether the third semantic node is leaf node determined, wherein the 4th failure clause is institute
State a upper failure clause adjacent with the third failure clause in failure sequence.
Optionally, the faulted-phase judgment unit is also configured to when the third semantic node not be leaf node
When, all leaf nodes under the third semantic node are searched for, and the node content of all leaf nodes searched is true
It is set to fault diagnosis result.
Optionally, the sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node;
The sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node, wherein when original node
Scoring be deleted when failing to reach predetermined threshold within a predetermined period of time.
Optionally, the faulted-phase judgment unit is also configured to work as root node or the child node of internal node includes
Clause when constituting logical complement relationship or being leaf node, will be patrolled comprising the clause that newly inputs with the clause's composition newly inputted
The node for collecting the clause of complementary relationship is added between former father and son's node as new node;When the child node of root node or internal node
It will include that the node of the clause newly inputted is made for one or multiple child nodes when the clause for including does not constitute logical complement relationship
For the new child node of root node or internal node.
Optionally, the faulted-phase judgment unit is also configured to father's section when the leaf node for executing beta pruning process
When point has the brotgher of node, the leaf node, father node and the brotgher of node are deleted, and the child node of sworn brothers' node is as the brother
The child node of the father node of younger brother's node;When executing the father node of leaf node of beta pruning process does not have the brotgher of node, institute is deleted
State leaf node and father node.
Another aspect of the present invention also provides a kind of computer readable storage medium, and be stored with makes when being executed by a processor
Obtain the computer program that processor executes sewage treatment method for diagnosing faults as described above.
Another aspect of the present invention also provides a kind of computing device, comprising: processor;Memory is located for storing to work as
Device is managed to execute so that processor executes the computer program of sewage treatment method for diagnosing faults as described above.
Sewage treatment method for diagnosing faults of the invention and device improve fault message description in a manner of feedback information
Accuracy, and combine sewage treatment process fault diagnosis actual conditions, dynamically increased based on semantic tree or
The knowledge (that is, semantic node) in sewage treatment knowledge base is reduced, to improve sewage treatment knowledge base;It can also be with online technique
Expert carries out interaction, to obtain the help of online technique expert, to obtain sewage treatment specialized guidance as much as possible.
In addition, also constantly matching with the knowledge in sewage treatment knowledge base using fault message, inference chain is formed, to obtain just
When effective fault diagnosis result.
Detailed description of the invention
By the detailed description carried out below in conjunction with the accompanying drawings, above and other objects of the present invention, features and advantages will
It becomes more fully apparent, in which:
Fig. 1 shows sewage treatment process figure in the prior art;
Fig. 2 shows the flow charts of the sewage treatment method for diagnosing faults of embodiment according to the present invention;
Fig. 3 shows the logic signal of the semantic tree of the sludge bulking in the secondary settling tank failure of embodiment according to the present invention
Figure;
Fig. 4 shows the flow chart of the inquiry fault diagnosis result of embodiment according to the present invention;
Fig. 5 shows the flow chart of the fault diagnosis of embodiment according to the present invention;
Fig. 6 shows the example of the growth course of the semantic tree of embodiment according to the present invention;
Fig. 7 shows the user interface schematic diagram of embodiment according to the present invention;
Fig. 8 shows the overall structure of the sewage treatment process fault diagnosis expert system of embodiment according to the present invention
Schematic diagram;
Fig. 9 shows the block diagram of the sewage treatment trouble-shooter of embodiment according to the present invention.
Specific embodiment
Now, different example embodiments is more fully described with reference to the accompanying drawings, wherein some exemplary embodiments are attached
It is shown in figure.
Sewage disposal process is the complex process flow including physics, chemistry and biological respinse, is usually contained multiple
Processing unit realizes the removal and conversion of multiple pollutant in the different disposal stage.Fig. 1 shows sewage treatment in the prior art
Process flow chart, including water inlet desilting unit, hydrolysis acidification unit, biochemical treatment unit, dosing disinfection unit and sludge dewatering
Unit.
Sewage into sewage treatment plant is first passed around into water desilting unit, and sewage flows through coarse rack, larger in the water that shuts off
The impurity etc. of particle, then the water pump of lift pump room, which is taken out, rises sewage into the pond of certain altitude, guarantees that sewage can lean on gravity
Subsequent each processing links are flowed through in effect.Next, sewage further removes the impurity of smaller particle by fine fack.Rotation
Setting pot is funnel-shaped, built-in paddle board, and sewage rotation is driven under motor driven, carries out girt-water separation.
Next, sewage enters hydrolysis acidification unit.The effect of hydrolysis acidification pool is that the non-solubilised state in sewage is organic
Object retains and is gradually changed into dissolved organic matter, some to be difficult to biodegradable macromolecular substances and be converted into be easy to the small of degradation
Molecular substance such as organic acid etc., so that the biodegradability and degradation speed of sewage are improved, in favor of subsequent Aerobic biological process.
Next, sewage enters biochemical unit.Integrated oxidation ditch uses inversion A2O technique (anoxic-anaerobic-aerobic),
Most of pollutant in air and sewage that activated sludge is exposed using air blower carries out biochemical reaction, plays denitrogenation dephosphorizing
Effect.A part of activated sludge is continuously circulated in integrated oxidation ditch, and another part then flow to secondary settling tank with sewage together.
The effect of secondary settling tank is to make mixed liquor clarification, be concentrated to realize that sludge separates, and providing and flow back into integrated oxidation ditch inlet
Activated sludge.
Next, the water outlet of secondary settling tank enters dosing disinfection unit.The water outlet of secondary settling tank enters efficient sedimentation tank, by adding
Add PAC and PAM medicament, make particulate matter in sewage further obtains flocculation sedimentation.After efficient sedimentation tank plus filtering cloth filtering pool, be for
The suspended matter after flocculation sedimentation in sewage is stopped, can play the role of removing total solid suspended matter, total phosphorus and heavy metal.Contact
Sterilization pool is gone out by chlorination removes the viral germ of influence environment in sewage, makes processed sewage may finally row up to standard
It puts.
In addition, the activated sludge of secondary settling tank discharge also passes through sludge dewatering unit.The activated sludge of secondary settling tank discharge removes one
It is most of to need to be discarded in the form of excess sludge other than partial reflux to integrated oxidation ditch.Then, surplus due at this time
Remaining sludge has mobility, and water content is still very high, needs further thickening, is converted into the lower mud cake of water content,
It is transported outward again by sludge vehicle.
Exactly pass through above-mentioned processing links, polluter contained in sewage is separated from the water removal, turns harmful substance
Harmless or even utility is turned to, sewage is purified.
However, in sewage disposal process, in fact it could happen that various failures.Therefore, it is necessary to pass through sewage treatment fault diagnosis
The scheme or measure that can solve practical problem are provided for supervisory engineering staff, to enable supervisory engineering staff in face
It, can be by the help of sewage treatment knowledge base and/or online technique expert, as early as possible when to complicated sewage treatment process problem
Finding out most appropriate effective treatment measures reduces the adverse effect to environment to guarantee sewage qualified discharge.
The sewage treatment method for diagnosing faults and device of embodiment according to the present invention are described referring to Fig. 2 to Fig. 9.
Fig. 2 shows the flow charts of the sewage treatment method for diagnosing faults of embodiment according to the present invention.
In step S10, at least one fault message about sewage treatment is received.
As an example, fault message can be the event for describing to occur in sewage treatment process of natural language form
The information of barrier.For example, fault message can be " secondary settling tank sludge bulking ", " having no a large amount of der Pilz wadding bodies " etc..
Preferably, comma is used to accord between multiple fault messages as isolation, that is, with comma point between different fault messages
Every.
In one embodiment of step S10, can receive by user (that is, user) input about sewage treatment extremely
A few fault message.
In step S20, at least one fault message is sent to expert module, and receive and each event from expert module
Hinder the corresponding fault diagnosis result of information.
Preferably, it will assist in solicited message and at least one fault message send jointly to expert module, and from expert
Module receives and shows fault diagnosis result corresponding with each fault message.
As an example, expert module can access online technique expert, which can be current online to sewage
Processing technological flow has personnel that are extensive and understanding deeply, however, the present invention is not limited thereto.
In the case, sewage treatment method for diagnosing faults can further include: receives the selection to expert module.
In one embodiment of step S20, it will assist in solicited message and send jointly to select at least one fault message
The expert module selected;Fault diagnosis result corresponding with each fault message is received from the expert module of selection.
Preferably, it may also display received fault diagnosis result corresponding with each fault message.
It can from the received fault diagnosis result corresponding with each fault message of the expert module of selection are as follows: true based on experience
Fixed fault diagnosis result or the fault diagnosis result by calling sewage treatment knowledge base to determine.
It is appreciated that can be provided by following two mode when the expert module of selection obtains assistance solicited message
Fault diagnosis result: if (1) the online technique expert in the expert module thinks the experience for only relying on oneself, event can be made
Judgement and the treatment measures etc. for hindering reason, then can directly give fault diagnosis result;(2) if it is online in the expert module
Technical specialist thinks to need by sewage treatment process fault diagnosis expert system, can according to assist solicited message one
The fault message sent is acted, failure-description is rearranged, sewage treatment knowledge base is called to execute fault inquiry, to provide event
Hinder diagnostic result.In the case, the online technique expert in the expert module can apply when rearranging failure-description
The more accurate of failure-description is increased and decreased some conditions or limitation perhaps to change the range of inquiry or reasoning by specialized vocabulary.
Selectively, it in step S30, calls sewage treatment knowledge base determining based on each fault message and each failure is believed
Cease corresponding fault diagnosis result.
Preferably, it may also display determining fault diagnosis result.
As an example, sewage treatment knowledge base may be based on an at least semantic tree corresponding with sewage treatment process
The sewage treatment knowledge base of foundation.
That is, sewage treatment process fault diagnosis expert system based on semantic tree, establish knowledge can be with
Increase or the dynamic sewage of reduction handles knowledge base.In this way, the inquiry of fault message and reasoning can be converted on semantic tree
Search process, and the increase and decrease of knowledge can be realized by executing growth and/or cut operator to semantic tree.
As an example, the characteristics of according to sewage treatment process, water inlet desilting, hydrolysis acidification, oxidation ditch, clarification filter
The processing links such as cloth, dosing disinfection, sludge dewatering are relatively independent, and the reason of leading to failure is different with treatment measures, point
Not Wei each processing links establish different semantic trees, that is, a semantic tree can correspond to one in sewage treatment process
Processing links.Whole semantic trees corresponding to sewage treatment process are stored in the form of semantic forest.
Every semantic tree may include being packaged by the predetermined information of each process of sewage treatment process in different layers
At least two semantic nodes.
For example, it is assumed that ss is the clause for describing certain processing ring-segment-like state information in sewage treatment knowledge base, as is failure letter
The corresponding failure clause of breath, then, state set SS is the set of ss, i.e. SS={ ss1,ss2,...,ssp, wherein p is to retouch
State the quantity of the clause of certain sewage treatment link status information;Fault setWherein, q
For the quantity of failure clause corresponding to fault message.
Semantic tree T about AS is the directed tree grown downwards, T=<V, E>it is a binary group, V={ r, ni,
lj, 0≤i < ∞, 1≤j < ∞.Here, r is root node, and it includes limited height sentence ss or clause that in-degree 0, out-degree, which are greater than 1,
The negative of ss;niFor internal node (i-th of internal node), it includes limited height sentence ss or clause that in-degree 1, out-degree, which are greater than 1,
The negative of ss;ljFor leaf node (j-th of internal node), in-degree 1, out-degree 0 include failure cause and treatment measures;E
For by r, niOr ljThe set of the orderly binary group constituted;V is the node collection of semantic tree.
Semantic forest F is the set of semantic tree, then F={ T1,T2,...,Tz, wherein z is the quantity of semantic tree.
It is appreciated that semantic tree T is a multiway tree, root node and internal node can have multiple child nodes, but a certain
A leaf node or internal node can only have unique father node.At least one root node of minimum semantic tree and a leaf section
Point.Semantic forest F can only have a semantic tree T.
Fig. 3 shows the logic signal of the semantic tree of the sludge bulking in the secondary settling tank failure of embodiment according to the present invention
Figure.
Referring to Fig. 3, the characteristics of according to sewage treatment process, the fault message of secondary settling tank, failure cause and processing
The relevant informations such as measure are packaged into semantic node, and the hierarchical relationship of relevant information is embodied with the structure of tree, establishes secondary settling tank failure
Semantic tree, and emphasis describes the phenomenon of the failure, failure cause and treatment measures of sludge bulking, that is to say, that the semantic tree
Integrate phenomenon of the failure, failure cause and treatment measures.In Fig. 3, DO is dissolved oxygen, SVI is sludge volume index, pH is
PH value, F/M are sludge loading.
Preferably, sewage treatment knowledge base stores every language in the form of relevant database (that is, relational database)
Justice tree.Each entry in relevant database is corresponding with each semantic node, and the content of each program recording is that instruction is every
The node content of a semantic node and the domain of hierarchical relationship.In this way, in sewage treatment process fault diagnosis expert system
The inquiry and reasoning of fault message, the data access form that can be provided by relevant database execute the state space search of tree
To realize.
Table 1 shows the example in the domain of semantic node.As shown in table 1, the data structure of semantic node includes four domains, respectively
For the domain NodeID, the domain FatherID, the domain ChildID and the domain Content.Here, the domain NodeID, the domain FatherID and ChildID
Domain can embody hierarchical relationship, and the domain Content can embody node content.
Table 1: the example in the domain of semantic node
Field name | Data type | Default value | Explanation |
NodeID | Text | Node serial number (unique) | |
FatherID | Text | Father node number | |
ChildID | Text | Child node number | |
Content | Text | null | Node content |
For example, the value that the value in the domain root node NodeID is the domain 0, FatherID is -1, the domain the Content storage of internal node
The clause set of sewage treatment link status information, the domain the Content storage failure cause and treatment measures of leaf node are described.
It is described below in detail and " calls the determination of sewage treatment knowledge base corresponding with each fault message based on each fault message
Fault diagnosis result " process.
Preferably, in step s 30, failure clause corresponding with each fault message is converted by each fault message;
Fault diagnosis result corresponding with each fault message is inquired from sewage treatment knowledge base based on each failure clause.
It is used in sewage treatment knowledge base inquire for example, converting the fault message of each natural language form to
Or the failure clause of reasoning.
It is described in detail and " is inquired from sewage treatment knowledge base and each event based on each failure clause below with reference to Fig. 4
The process of the corresponding fault diagnosis result of barrier information ".
Fig. 4 shows the flow chart of the inquiry fault diagnosis result of embodiment according to the present invention.
Preferably, referring to Fig. 4, in step S301, the semantic section in each failure clause and sewage treatment knowledge base is identified
The matching degree of point.
That is, by the clause in each failure clause and sewage treatment knowledge base (for example, in the node of semantic node
Hold) it is compared, obtain matching degree.
It should be appreciated that the various matching degrees that can obtain failure clause and the clause in sewage treatment knowledge base can be used
Method calculates matching degree, and this is not limited by the present invention.
In step S302, judge whether each matching degree reaches predetermined threshold.
It is appreciated that predetermined threshold can be set in advance as needed, this is not limited by the present invention.
In step S303, failure clause corresponding to the matching degree for being not up to predetermined threshold is deleted, and in step S304, base
The failure clause corresponding to the matching degree for reaching predetermined threshold inquires corresponding fault diagnosis knot from sewage treatment knowledge base
Fruit.
In addition, sewage treatment method for diagnosing faults can further include: be not up to the matching degree institute of predetermined threshold in response to deleting
Corresponding failure clause prompts adjustment fault message corresponding with the failure clause deleted.
That is, feedback information can be provided according to the judging result of matching degree, to receive event as accurate as possible
Hinder information, to improve the accuracy of description fault message, making fault message as far as possible is that sewage treatment process failure is examined
Disconnected expert system is understood, the difficulty that fault message obtains and understands is reduced.If the accuracy of some failure clause is too low, and
It is still not up to the degree that sewage treatment process fault diagnosis expert system requires after modification, failure clause can be deleted, and
Only the higher failure clause of accuracy is conducive to obtain appropriate effective fault diagnosis result for inquiring reasoning.
The characteristics of according to the fault message and semantic tree of sewage treatment process, the intuitive forward reasoning of comparison can be used
To realize semantic traversal of tree.Below with reference to Fig. 5 detailed description " based on of failure corresponding to the matching degree for reaching predetermined threshold
Sentence corresponding fault diagnosis result is inquired from sewage treatment knowledge base " process.
Fig. 5 shows the flow chart of the fault diagnosis of embodiment according to the present invention.
Referring to Fig. 5, in step S401, it is right that each matching degree institute for reaching predetermined threshold is searched in sewage treatment knowledge base
The number of plies of the failure clause answered on semantic tree, and sorted based on the number of plies searched and determine failure sequence.
Preferably, it is searched in sewage treatment knowledge base and each reaches failure clause corresponding to the matching degree of predetermined threshold
The number of plies on semantic tree, and using the number of plies searched as failure clause corresponding to the matching degree for each reaching predetermined threshold
The number of plies;Ascending order row is carried out to failure clause corresponding to all matching degrees for reaching predetermined threshold according to the number of plies searched
Column, obtain failure sequence.
It is appreciated that the number of plies of the clause on semantic tree can be leading to from semantic node where semantic root vertex to clause
Road length.For example, could dictate that the number of plies of the root node comprising clause is 0.
In step S402, the first semantic node corresponding to the smallest Fisrt fault clause of the number of plies in failure sequence is determined
With the second semantic node corresponding to the maximum second failure clause of the number of plies.
In step S403, third corresponding to the third failure clause in failure sequence is detected in sewage treatment knowledge base
Semantic node whether be the first semantic node child node.
That is, judging that third semantic node whether there is in sewage treatment knowledge base.When judging third language for the first time
When adopted node whether there is, third failure clause is next failure clause adjacent with Fisrt fault clause in failure sequence.
When detecting third semantic node is the child node of the first semantic node, in step S404, third semanteme is judged
Whether node is identical as the second semantic node.
That is, judging whether the third semantic node is number of plies maximum in the presence of detecting third semantic node
Failure clause corresponding to semantic node.
When third semantic node and not identical the second semantic node, in step S405, third failure clause is updated to
The next failure clause adjacent with third failure clause, and third semantic node is updated to and updated third failure
The corresponding semantic node of sentence, and return step S403.
When third semantic node is identical as the second semantic node, in step S406, pass through detection third semantic node
Out-degree and in-degree determines whether third semantic node is leaf node.
When detecting third semantic node not is the child node of the first semantic node, in step S407, by third semanteme
Node is determined as semantic node corresponding to the 4th failure clause, then executes step S406.
4th failure clause can be a upper failure clause adjacent with third failure clause in failure sequence.It can manage
Solution, when third failure clause updates, the 4th failure clause can be failure sequence in updated third failure clause
An adjacent upper failure clause.
When third semantic node is leaf node, in step S408, the node content of third semantic node is determined as
Fault diagnosis result.
When third semantic node is not leaf node, in step S409, all leaves under third semantic node are searched for
Node, and the node content of all leaf nodes searched is determined as fault diagnosis result.
As an example, specific step is as follows for the fault diagnosis of the embodiment of the present invention:
(1) fault set AS={ as is searched for1,as2,...,asqThe number of plies on syntax tree T, by each failure clause as1、
as2、……、asqAccording to the carry out ascending order arrangement of its number of plies, failure sequence AS'={ as' is obtained1,as'2,...,as'q}。
(2) the smallest failure clause as of current layer number in failure sequence AS' is searched for1' where the first semantic node and layer
The maximum failure clause as' of numberqSecond semantic node at place.
(3) next clause as' in depth-first search failure sequence AS'2The third semantic node at place.
(4) judge that third semantic node whether there is, if third semantic node exists, then follow the steps (5);Otherwise,
Third semantic node is chosen as as' in failure sequence AS'2Previous failure clause (that is, as'1) where semantic node,
And execute step (6).
(5) judge whether third semantic node is the maximum failure clause as' of the number of pliesqThe semantic node at place is (that is, third
Whether semantic node and the second semantic node are identical), if so, thening follow the steps (6);Otherwise, (3) are returned to step,
In this case, returning in the step of executing (3), depth-first search clause as'2Next clause as'3The semantic section at place
Point, and third semantic node is updated to the clause as' searched3The semantic node at place, and so on, it is executed when returning
When step (3), next clause is determined in the manner described above, and third semantic node is updated to next clause of the determination
The node at place.
(6) judge whether third semantic node is leaf node, if so, thening follow the steps (8);Otherwise, step is executed
(7)。
(7) all leaf nodes under depth-first search third semantic node.
(8) node content (that is, failure cause and treatment measures) in leaf node is exported.
In sewage treatment process fault diagnosis expert system, the knowledge in sewage treatment knowledge base can increase or
Person is reduced, to dynamically update the knowledge in sewage treatment knowledge base.
Preferably, sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node, thus
Increase the knowledge in sewage treatment knowledge base.
Fig. 6 shows the example of the growth course of the semantic tree of embodiment according to the present invention.
With clause ssnewIncrease to root node r or internal node niUnder for, following two situation can be divided into:
(1) if root node r or internal node niThe child node clause that includes constitute logical complement relationship or for leaf
Node then will include ssnewClause andClause is (that is, and ssnewClause composition logical complement relationship clause) node
As root node r or internal node niChild node, and include ssnewClause orThe node of clause as root node r or
Internal node niThe originally father node of child node.ssnewClause withThe selection of clause can be determined by actual conditions.Ginseng
It include ss according to Fig. 6newClause withFormer father and son's node (ss is added as new node in the node of clausecur、ssson、
) between.
(2) if root node r or internal node niChild node only have one or clause that multiple child nodes include not structure
It then will include ss at logical complement relationshipnewThe node of clause is as root node r or internal node niNew child node.
When certain knowledge in sewage treatment knowledge base all obtain too low evaluation in one section of longer time, at sewage
These can be evaluated too low knowledge and deleted by reason knowledge base.Increase with knowledge on the contrary, knowledge deletes the beta pruning for corresponding to semantic tree
Process.
Preferably, sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node, thus
Reduce the knowledge in sewage treatment knowledge base.For example, the scoring when original node fails to reach predetermined threshold within a predetermined period of time
It is deleted when value.
As an example, whether having the brotgher of node according to the father node of deleted leaf node, two kinds of situations can be divided into:
(1) if being deleted leaf node ldeleteFather node have the brotgher of node, illustrate clause that father node includes and its
Logical complement relationship is constituted between the clause that the brotgher of node includes, then deletes leaf node ldelete, father node and the brotgher of node,
And child node of the child node of sworn brothers' node as the father node of the brotgher of node.
(2) if being deleted leaf node ldeleteFather node there is no the brotgher of node, then directly delete leaf node
ldeleteAnd father node.
The example of " user interface " is described in detail below with reference to Fig. 7.
Fig. 7 shows the user interface schematic diagram of embodiment according to the present invention.
Referring to Fig. 7, sewage treatment process may include following processing links: water inlet desilting, hydrolysis acidification, at biochemistry
Reason, dosing disinfection, sludge dewatering.For the text box of " failure-description " for receiving fault message, fault message is defeated with natural language
Enter, respectively " secondary settling tank sludge bulking " and " having no a large amount of der Pilz wadding bodies ", is separated between the two fault messages with comma.
After receiving a fault message to end up with comma, the clause in the fault message and sewage treatment knowledge base is searched for automatically
Matching degree, be expressed as a percentage.Specifically, about fault message " secondary settling tank sludge bulking ", matching degree 100%;About
Fault message " has no a large amount of der Pilz wadding bodies ", matching degree 95%.If user has found that matching degree is too low, explanation may be sent out
The case where raw error understanding fault message, can prompt user to adjust the verbal description of " fault message ", correctly be looked into executing
Ask reasoning;Otherwise, when the matching degree that search obtains is as low as certain matching degree threshold value, the too low event of matching degree can be automatically deleted
Hinder information.This matching degree evaluation mechanism can reduce fault message retrieval and matched difficulty, also reduce retouching for fault message
State adverse effect of the inaccuracy to fault diagnosis result.
User can choose online expert from specialist list, to seek the support of online technique expert, obtain
Specialized guidance as much as possible.As shown in fig. 7, expert 1 and expert 2 are online, expert 3 is not online, user selected expert 1 for
It provides help.
" failure cause and suggestion and measure " can provide multiple fault diagnosis results of inquiry reasoning, for user's selection.
User can score to fault diagnosis result, for example, five kinds of evaluation results can be provided with five-pointed star.If certain failures are examined
Disconnected result long-term assessment is too low, can will delete it.
The example of " sewage treatment process fault diagnosis expert system " is described in detail below with reference to Fig. 8.
Fig. 8 shows the overall structure of the sewage treatment process fault diagnosis expert system of embodiment according to the present invention
Schematic diagram.
It can be the expert of user (for example, supervisory engineering staff of sewage treatment plant) and sewage treatment field referring to Fig. 8
User interface and expert interface are separately designed, by assigning different operating rights to user and expert, is realized at sewage
The dynamic increase and decrease and the interaction of user and online technique expert for managing knowledge in knowledge base, to provide needle for user
To multiple treatment measures of current failure information or suggestion.
As an example, expert can possess the permission for increasing sewage treatment knowledge in knowledge base, but have no right to delete at sewage
The existing knowledge in knowledge base is managed, to reduce personal limitation and prejudice as far as possible.
As an example, fault diagnosis result may there are three source, be respectively as follows: user be based on sewage treatment knowledge base into
Diagnostic result caused by row inquiry reasoning, technical specialist are based on sewage treatment knowledge base and carry out diagnosis caused by inquiry reasoning
As a result with technical specialist by diagnostic result given by oneself experience.If there is technical specialist participates in, then separate sources is examined
There may be the case where repetition or conflict between disconnected result, in the case, duplicate diagnostic result can be deleted, and enumerate conflict
Diagnostic result.
The sewage treatment trouble-shooter of embodiment according to the present invention is described referring to Fig. 9.
Fig. 9 shows the block diagram of the sewage treatment trouble-shooter of embodiment according to the present invention.
Referring to Fig. 9, the sewage treatment trouble-shooter of embodiment according to the present invention includes: fault message receiving unit
100 and faulted-phase judgment unit 200.
Fault message receiving unit 100 receives at least one fault message about sewage treatment;
As an example, fault message can be the event for describing to occur in sewage treatment process of natural language form
The information of barrier.Comma is used to accord between multiple fault messages as isolation, that is, to be separated between different fault messages with comma.
As an example, fault message receiving unit 100 can receive at least one about sewage treatment inputted by user
A fault message.
At least one fault message is sent to expert module (not shown) by faulted-phase judgment unit 200, and from special
Family's module receives fault diagnosis result corresponding with each fault message.
Preferably, faulted-phase judgment unit 200 will assist in solicited message and send jointly to at least one fault message
Expert module, and received from expert module and show fault diagnosis result corresponding with each fault message.
As an example, expert module can have people that is extensive and understanding deeply to sewage treatment process to be currently online
Member, however, the present invention is not limited thereto.
In the case, sewage treatment trouble-shooter can further include: selection receiving unit (not shown).Selection receives
Unit also receives the selection to expert module.
In one embodiment of faulted-phase judgment unit 200, faulted-phase judgment unit 200 will assist in solicited message
The expert module of selection is sent jointly to at least one fault message.
Faulted-phase judgment unit 200 receives fault diagnosis knot corresponding with each fault message from the expert module of selection
Fruit.
Preferably, faulted-phase judgment unit 200 can also show received fault diagnosis corresponding with each fault message
As a result.
From the received fault diagnosis result corresponding with each fault message of the expert module selected for based on empirically determined
Fault diagnosis result or by call sewage treatment knowledge base determine fault diagnosis result.
Selectively, faulted-phase judgment unit 200 based on each fault message call sewage treatment knowledge base determine with
The corresponding fault diagnosis result of each fault message, and show determining fault diagnosis result,
As an example, sewage treatment knowledge base is to be built based on an at least semantic tree corresponding with sewage treatment process
Vertical sewage treatment knowledge base.
The characteristics of according to sewage treatment process, water inlet desilting, hydrolysis acidification, oxidation ditch, clarification filter cloth, dosing disappear
The processing links such as poison, sludge dewatering are relatively independent, and the reason of leading to failure is different with treatment measures, may respectively be each
Processing links establish different semantic trees, that is, a semantic tree can correspond to a processing links in sewage treatment process.
Whole semantic trees corresponding to sewage treatment process are stored in the form of semantic forest.
Every semantic tree may include being packaged by the predetermined information of each process of sewage treatment process in different layers
At least two semantic nodes.
It is appreciated that semantic tree is a multiway tree, root node and internal node can have multiple child nodes, but some
Leaf node or internal node can only have unique father node.At least one root node of minimum semantic tree and a leaf section
Point.Semantic forest can only have a semantic tree.
Preferably, sewage treatment knowledge base stores every language in the form of relevant database (that is, relational database)
Justice tree.Each entry in relevant database is corresponding with each semantic node, and the content of each program recording is that instruction is every
The node content of a semantic node and the domain of hierarchical relationship.In this way, in sewage treatment process fault diagnosis expert system
The inquiry and reasoning of fault message, the data access form that can be provided by relevant database execute the state space search of tree
To realize.
As an example, faulted-phase judgment unit 200 can convert each fault message to it is corresponding with each fault message
Failure clause, identify the matching degree of the semantic node in each failure clause and sewage treatment knowledge base, judge each matching
Whether degree reaches predetermined threshold, deletes failure clause corresponding to the matching degree for being not up to predetermined threshold, and is based on reaching predetermined
Failure clause corresponding to the matching degree of threshold value inquires corresponding fault diagnosis result from sewage treatment knowledge base.
It is used for for example, faulted-phase judgment unit 200 converts the fault message of each natural language form in sewage
The failure clause of inquiry or reasoning is carried out in processing knowledge base.
It is appreciated that predetermined threshold can be set in advance as needed, this is not limited by the present invention.
In addition, faulted-phase judgment unit 200 is additionally in response to delete event corresponding to the matching degree for being not up to predetermined threshold
Hedge sentence, and prompt adjustment fault message corresponding with the failure clause deleted.
That is, faulted-phase judgment unit 200 can provide feedback information according to the judging result of matching degree, so as to event
Barrier information receiving unit 100 receives fault message as accurate as possible, to improve the accuracy of description fault message, makes event
Barrier information understood as far as possible by sewage treatment process fault diagnosis expert system, reduce fault message obtain and understanding
Difficulty.If the accuracy of some failure clause is too low, and it is special to be still not up to sewage treatment process fault diagnosis after modifying
The degree of family's system requirements, faulted-phase judgment unit 200 can delete failure clause, and only higher failure of accuracy
Sentence is conducive to obtain appropriate effective fault diagnosis result for inquiring reasoning.
The characteristics of according to the fault message and semantic tree of sewage treatment process, the intuitive forward reasoning of comparison can be used
To realize semantic traversal of tree.
As an example, faulted-phase judgment unit 200 is searched in sewage treatment knowledge base each reaches predetermined threshold
The number of plies of the failure clause on semantic tree corresponding to matching degree, and sorted based on the number of plies searched and determine failure sequence;Really
Determine the first semantic node corresponding to the smallest Fisrt fault clause of the number of plies in failure sequence and maximum second failure of the number of plies
Second semantic node corresponding to clause;It is detected in sewage treatment knowledge base corresponding to the third failure clause in failure sequence
Third semantic node whether be the first semantic node child node;When detecting that third semantic node is the first semantic node
When child node, judge whether third semantic node is identical as the second semantic node;When third semantic node and the second semantic node
When identical, determine whether third semantic node is leaf node by detecting the out-degree and in-degree of third semantic node;When
When three semantic nodes are leaf node, the node content of third semantic node is determined as fault diagnosis result.
When whether judge third semantic node for the first time is the child node of the first semantic node, third failure clause is failure
The next failure clause adjacent with Fisrt fault clause in sequence.
It is appreciated that the number of plies of the clause on semantic tree can be leading to from semantic node where semantic root vertex to clause
Road length.For example, could dictate that the number of plies of the root node comprising clause is 0.
Preferably, faulted-phase judgment unit 200 can be searched in sewage treatment knowledge base each reaches predetermined threshold
The number of plies of the failure clause on semantic tree corresponding to matching degree, and using the number of plies searched as each reaching predetermined threshold
The number of plies of failure clause corresponding to matching degree;Reach corresponding to the matching degree of predetermined threshold according to the number of plies searched to all
Failure clause carry out ascending order arrangement, obtain failure sequence.
When third semantic node and not identical the second semantic node, faulted-phase judgment unit 200 is sub by third failure
Sentence is updated to the next failure clause adjacent with third failure clause, and third semantic node is updated to and updated the
The corresponding semantic node of three failure clauses returns to execute and " detects the third failure in failure sequence in sewage treatment knowledge base
Third semantic node corresponding to clause whether be the first semantic node child node " processing.
When detecting third semantic node not is the child node of the first semantic node, faulted-phase judgment unit 200 will
Third semantic node is determined as semantic node corresponding to the 4th failure clause, then execute determine third semantic node whether be
The processing of leaf node.
4th failure clause is a upper failure clause adjacent with third failure clause in failure sequence.It is appreciated that
When third failure clause updates, the 4th failure clause can be adjacent with updated third failure clause in failure sequence
A upper failure clause.
When third semantic node is not leaf node, faulted-phase judgment unit 200 is searched under third semantic node
All leaf nodes, and the node content of all leaf nodes searched is determined as fault diagnosis result.
In sewage treatment process fault diagnosis expert system, the knowledge in sewage treatment knowledge base can increase or
Person is reduced, to dynamically update the knowledge in sewage treatment knowledge base.
As an example, sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node, from
And increase the knowledge in sewage treatment knowledge base.
Preferably, when the clause that the child node of root node or internal node includes constitutes logical complement relationship or is leaf section
When point, faulted-phase judgment unit 200 will constitute logical complement relationship with the clause newly inputted comprising the clause newly inputted
The node of clause is added between former father and son's node as new node.
Preferably, when the clause that the child node of root node or internal node is one or multiple child nodes include is not constituted
When logical complement relationship, faulted-phase judgment unit 200 is using the node comprising the clause newly inputted as root node or internal section
The new child node of point.
When certain knowledge in sewage treatment knowledge base all obtain too low evaluation in one section of longer time, at sewage
These can be evaluated too low knowledge and deleted by reason knowledge base.Increase with knowledge on the contrary, knowledge deletes the beta pruning for corresponding to semantic tree
Process.
As an example, sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node, from
And reduce the knowledge in sewage treatment knowledge base, it is preferable that when the scoring of original node fails to reach pre- within a predetermined period of time
Determine to be deleted when threshold value.
Preferably, when executing the father node of leaf node of beta pruning process has the brotgher of node, faulted-phase judgment unit
200 delete leaf nodes, father node and the brotgher of node, and father node of the child node of sworn brothers' node as the brotgher of node
Child node.
Preferably, when executing the father node of leaf node of beta pruning process does not have the brotgher of node, faulted-phase judgment list
Member 200 deletes leaf node and father node.
In addition, the sewage treatment method for diagnosing faults and device of the embodiment of the present invention, are improved in a manner of feedback information
The accuracy of fault message description, and the actual conditions of sewage treatment process fault diagnosis are combined, based on semantic tree
The knowledge (that is, semantic node) in sewage treatment knowledge base is dynamically increased or reduces, to improve sewage treatment knowledge base;
Interaction can also be carried out with online technique expert, to obtain the help of online technique expert, to obtain as much as possible
Sewage treatment specialized guidance.In addition, also constantly being matched with the knowledge in sewage treatment knowledge base using fault message, formed
Inference chain, to obtain appropriate effective fault diagnosis result.
Embodiment according to the present invention also provides a kind of computer readable storage medium.The computer readable storage medium is deposited
Contain the computer program for making processor execute sewage treatment method for diagnosing faults as described above when being executed by a processor.
Embodiment according to the present invention also provides a kind of computing device.The computing device includes processor and memory.It deposits
Reservoir is for storing program instruction.Described program instruction is executed by processor so that processor executes sewage treatment as described above
The computer program of method for diagnosing faults.
Moreover, it should be understood that each list in sewage treatment trouble-shooter according to an exemplary embodiment of the present invention
Member can be implemented hardware component and/or component software.Those skilled in the art's processing according to performed by each unit of restriction,
Can such as use site programmable gate array (FPGA) or specific integrated circuit (ASIC) realize each unit.
Although being particularly shown and describing the present invention, those skilled in the art referring to its exemplary embodiment
It should be understood that in the case where not departing from the spirit and scope of the present invention defined by claim form can be carried out to it
With the various changes in details.
Claims (30)
1. a kind of sewage treatment method for diagnosing faults, which is characterized in that the sewage treatment method for diagnosing faults includes:
Receive at least one fault message about sewage treatment;
At least one described fault message is sent to expert module, and corresponding with each fault message from expert module reception
Fault diagnosis result;Alternatively,
Sewage treatment knowledge base is called to determine fault diagnosis result corresponding with each fault message based on each fault message,
Wherein, the sewage treatment knowledge base is to be established based on an at least semantic tree corresponding with sewage treatment process
Sewage treatment knowledge base.
2. sewage treatment method for diagnosing faults as described in claim 1, which is characterized in that will at least one described fault message
The step of being sent to expert module include:
It will assist in solicited message and at least one described fault message send jointly to the expert module.
3. sewage treatment method for diagnosing faults as claimed in claim 2, which is characterized in that the sewage treatment fault diagnosis side
Method further include:
The selection to expert module is received,
Wherein, at least one described fault message is sent to expert module, and receives from expert module and believes with each failure
The step of ceasing corresponding fault diagnosis result include:
It will assist in solicited message and at least one described fault message send jointly to the expert module of selection;
Fault diagnosis result corresponding with each fault message is received from the expert module of selection,
Wherein, from the received fault diagnosis result corresponding with each fault message of the expert module selected for based on empirically determined
Fault diagnosis result or by call sewage treatment knowledge base determine fault diagnosis result.
4. sewage treatment method for diagnosing faults as claimed in claim 1 or 3, which is characterized in that be based on each fault message tune
Include: with the step of sewage treatment knowledge base determination fault diagnosis result corresponding with each fault message
Failure clause corresponding with each fault message is converted by each fault message;
Identify the matching degree of the semantic node in each failure clause and sewage treatment knowledge base;
Judge whether each matching degree reaches predetermined threshold;
It deletes and is not up to failure clause corresponding to the matching degree of predetermined threshold, and is right based on the matching degree institute for reaching predetermined threshold
The failure clause answered inquires corresponding fault diagnosis result from sewage treatment knowledge base.
5. sewage treatment method for diagnosing faults as claimed in claim 4, which is characterized in that the sewage treatment fault diagnosis side
Method further include:
It is not up to failure clause corresponding to the matching degree of predetermined threshold in response to deleting, the failure clause for prompting adjustment and deleting
Corresponding fault message.
6. sewage treatment method for diagnosing faults as claimed in claim 4, which is characterized in that the sewage treatment knowledge base is to close
It is the form of type database to store every semantic tree.
7. sewage treatment method for diagnosing faults as claimed in claim 6, which is characterized in that every semantic tree includes by sewage
At least two semantic nodes in different layers that the predetermined information of each process of science and engineering skill is packaged into,
Wherein, each entry in the relevant database is corresponding with each semantic node, the content of each program recording
To indicate the node content of each semantic node and the domain of hierarchical relationship.
8. sewage treatment method for diagnosing faults as claimed in claim 7, which is characterized in that based on the matching for reaching predetermined threshold
Spending the step of corresponding failure clause inquires corresponding fault diagnosis result from sewage treatment knowledge base includes:
Search each reaches failure clause corresponding to the matching degree of predetermined threshold on semantic tree in sewage treatment knowledge base
The number of plies, and sorted based on the number of plies that searches and determine failure sequence;
Determine that the first semantic node corresponding to the smallest Fisrt fault clause of the number of plies in the failure sequence and the number of plies are maximum
The second failure clause corresponding to the second semantic node;
When detecting that third semantic node corresponding to the third failure clause in the failure sequence is the described first semantic section
When the child node of point, judge whether the third semantic node is identical as second semantic node;
When the third semantic node is identical as second semantic node, by the out-degree for detecting the third semantic node
Determine whether the third semantic node is leaf node with in-degree;
When the third semantic node is leaf node, the node content of the third semantic node is determined as fault diagnosis
As a result,
Wherein, the third failure clause is next failure adjacent with the Fisrt fault clause in the failure sequence
Sentence.
9. sewage treatment method for diagnosing faults as claimed in claim 8, which is characterized in that
When the third semantic node and second semantic node be not identical, third failure clause is updated to and described the
The adjacent next failure clause of three failure clauses, and third semantic node is updated to and updated third failure clause couple
The semantic node answered.
10. sewage treatment method for diagnosing faults as claimed in claim 8, which is characterized in that
When detecting the third semantic node is not the child node of first semantic node, by the third semantic node
It is determined as semantic node corresponding to the 4th failure clause, then executes and determine whether the third semantic node is leaf node
The step of,
Wherein, the 4th failure clause is upper failure adjacent with the third failure clause in the failure sequence
Sentence.
11. sewage treatment method for diagnosing faults as claimed in claim 10, which is characterized in that
When the third semantic node is not leaf node, all leaf nodes under the third semantic node are searched for, and
The node content of all leaf nodes searched is determined as fault diagnosis result.
12. sewage treatment method for diagnosing faults as described in claim 1, which is characterized in that
The sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node;
The sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node,
Wherein, it is deleted when the scoring of original node fails to reach predetermined threshold within a predetermined period of time.
13. sewage treatment method for diagnosing faults as claimed in claim 12, which is characterized in that execute the growth course of semantic tree
The step of include:
When the clause that the child node of root node or internal node includes constitutes logical complement relationship or is leaf node, will include
Former father and son is added as new node in the node of the clause newly inputted and the clause that logical complement relationship is constituted with the clause newly inputted
Between node;
When the clause that the child node of root node or internal node is one or multiple child nodes include does not constitute logical complement relationship
When, using the node comprising the clause newly inputted as the new child node of root node or internal node.
14. sewage treatment method for diagnosing faults as claimed in claim 12, which is characterized in that execute the beta pruning process of semantic tree
The step of include:
When executing the father node of leaf node of beta pruning process has the brotgher of node, the leaf node, father node and brother are deleted
Younger brother's node, and child node of the child node of sworn brothers' node as the father node of the brotgher of node;
When executing the father node of leaf node of beta pruning process does not have the brotgher of node, the leaf node and father node are deleted.
15. a kind of sewage treatment trouble-shooter, which is characterized in that the sewage treatment trouble-shooter includes:
Fault message receiving unit is configured as receiving at least one fault message about sewage treatment;
Faulted-phase judgment unit is configured as at least one described fault message being sent to expert module, and from expert
Module receives fault diagnosis result corresponding with each fault message;Alternatively, calling sewage treatment to know based on each fault message
Know library and determine fault diagnosis result corresponding with each fault message,
Wherein, the sewage treatment knowledge base is to be established based on an at least semantic tree corresponding with sewage treatment process
Sewage treatment knowledge base.
16. sewage treatment trouble-shooter as claimed in claim 15, which is characterized in that the faulted-phase judgment unit
It is also configured to
It will assist in solicited message and at least one described fault message send jointly to the expert module.
17. sewage treatment trouble-shooter as claimed in claim 16, which is characterized in that the sewage treatment fault diagnosis
Device further include:
Receiving unit is selected, is configured as receiving the selection to expert module,
Wherein, the faulted-phase judgment unit is also configured to
It will assist in solicited message and at least one described fault message send jointly to the expert module of selection;
Fault diagnosis result corresponding with each fault message is received from the expert module of selection,
Wherein, from the received fault diagnosis result corresponding with each fault message of the expert module selected for based on empirically determined
Fault diagnosis result or by call sewage treatment knowledge base determine fault diagnosis result.
18. the sewage treatment trouble-shooter as described in claim 15 or 17, which is characterized in that the faulted-phase judgment
Unit is also configured to
Failure clause corresponding with each fault message is converted by each fault message;
Identify the matching degree of the semantic node in each failure clause and sewage treatment knowledge base;
Judge whether each matching degree reaches predetermined threshold;
It deletes and is not up to failure clause corresponding to the matching degree of predetermined threshold, and is right based on the matching degree institute for reaching predetermined threshold
The failure clause answered inquires corresponding fault diagnosis result from sewage treatment knowledge base.
19. sewage treatment trouble-shooter as claimed in claim 18, which is characterized in that the faulted-phase judgment unit
It is also configured to
It is not up to failure clause corresponding to the matching degree of predetermined threshold in response to deleting, the failure clause for prompting adjustment and deleting
Corresponding fault message.
20. sewage treatment trouble-shooter as claimed in claim 18, which is characterized in that the sewage treatment knowledge base with
The form of relevant database stores every semantic tree.
21. sewage treatment trouble-shooter as claimed in claim 20, which is characterized in that every semantic tree includes by sewage
At least two semantic nodes in different layers that the predetermined information of each process for the treatment of process is packaged into,
Wherein, each entry in the relevant database is corresponding with each semantic node, the content of each program recording
To indicate the node content of each semantic node and the domain of hierarchical relationship.
22. sewage treatment trouble-shooter as claimed in claim 21, which is characterized in that the faulted-phase judgment unit
It is also configured to
Search each reaches failure clause corresponding to the matching degree of predetermined threshold on semantic tree in sewage treatment knowledge base
The number of plies, and sorted based on the number of plies that searches and determine failure sequence;
Determine that the first semantic node corresponding to the smallest Fisrt fault clause of the number of plies in the failure sequence and the number of plies are maximum
The second failure clause corresponding to the second semantic node;
When detecting that third semantic node corresponding to the third failure clause in the failure sequence is the described first semantic section
When the child node of point, judge whether the third semantic node is identical as second semantic node;
When the third semantic node is identical as second semantic node, by the out-degree for detecting the third semantic node
Determine whether the third semantic node is leaf node with in-degree;
When the third semantic node is leaf node, the node content of the third semantic node is determined as fault diagnosis
As a result,
Wherein, the third failure clause is next failure adjacent with the Fisrt fault clause in the failure sequence
Sentence.
23. sewage treatment trouble-shooter as claimed in claim 22, which is characterized in that the faulted-phase judgment unit
It is also configured to
When the third semantic node and second semantic node be not identical, third failure clause is updated to and described the
The adjacent next failure clause of three failure clauses, and third semantic node is updated to and updated third failure clause couple
The semantic node answered.
24. sewage treatment trouble-shooter as claimed in claim 22, which is characterized in that the faulted-phase judgment unit
It is also configured to
When detecting the third semantic node is not the child node of first semantic node, by the third semantic node
It is determined as semantic node corresponding to the 4th failure clause, then executes and determine whether the third semantic node is leaf node
The step of,
Wherein, the 4th failure clause is upper failure adjacent with the third failure clause in the failure sequence
Sentence.
25. sewage treatment trouble-shooter as claimed in claim 24, which is characterized in that the faulted-phase judgment unit
It is also configured to
When the third semantic node is not leaf node, all leaf nodes under the third semantic node are searched for, and
The node content of all leaf nodes searched is determined as fault diagnosis result.
26. sewage treatment trouble-shooter as claimed in claim 15, which is characterized in that
The sewage treatment knowledge base is entered and executes the growth course of semantic tree in response to new node;
The sewage treatment knowledge base is deleted and executes the beta pruning process of semantic tree in response to original node, wherein when original
The scoring of node is deleted when failing to reach predetermined threshold within a predetermined period of time.
27. sewage treatment trouble-shooter as claimed in claim 26, which is characterized in that the faulted-phase judgment unit
It is also configured to
When the clause that the child node of root node or internal node includes constitutes logical complement relationship or is leaf node, will include
Former father and son is added as new node in the node of the clause newly inputted and the clause that logical complement relationship is constituted with the clause newly inputted
Between node;
When the clause that the child node of root node or internal node is one or multiple child nodes include does not constitute logical complement relationship
When, using the node comprising the clause newly inputted as the new child node of root node or internal node.
28. sewage treatment trouble-shooter as claimed in claim 26, which is characterized in that the faulted-phase judgment unit
It is also configured to
When executing the father node of leaf node of beta pruning process has the brotgher of node, the leaf node, father node and brother are deleted
Younger brother's node, and child node of the child node of sworn brothers' node as the father node of the brotgher of node;
When executing the father node of leaf node of beta pruning process does not have the brotgher of node, the leaf node and father node are deleted.
29. a kind of computer readable storage medium, which is characterized in that be stored with and execute processor when being executed by a processor
The computer program of sewage treatment method for diagnosing faults as described in any one of claim 1 to 14.
30. a kind of computing device characterized by comprising
Processor;
Memory is executed by processor for storing to work as so that processor is executed as described in any one of claim 1 to 14
Sewage treatment method for diagnosing faults computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810821381.8A CN108960673B (en) | 2018-07-24 | 2018-07-24 | Sewage treatment fault diagnosis method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810821381.8A CN108960673B (en) | 2018-07-24 | 2018-07-24 | Sewage treatment fault diagnosis method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108960673A true CN108960673A (en) | 2018-12-07 |
CN108960673B CN108960673B (en) | 2020-10-30 |
Family
ID=64464702
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810821381.8A Active CN108960673B (en) | 2018-07-24 | 2018-07-24 | Sewage treatment fault diagnosis method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108960673B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115048095A (en) * | 2022-08-12 | 2022-09-13 | 广东粤港澳大湾区硬科技创新研究院 | Expert system fault diagnosis program generation method |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06222922A (en) * | 1993-01-21 | 1994-08-12 | Sankyo Seiki Mfg Co Ltd | Expert system |
CN101470701A (en) * | 2007-12-29 | 2009-07-01 | 日电(中国)有限公司 | Text analyzer supporting semantic rule based on finite state machine and method thereof |
US20100070448A1 (en) * | 2002-06-24 | 2010-03-18 | Nosa Omoigui | System and method for knowledge retrieval, management, delivery and presentation |
CN102184569A (en) * | 2011-06-11 | 2011-09-14 | 福州大学 | Individual plant wood modeling method driven by domain ontology |
CN102254102A (en) * | 2011-07-18 | 2011-11-23 | 华东理工大学 | Ontology-based sewage decision-support system and method |
CN202394263U (en) * | 2012-04-01 | 2012-08-22 | 中国人民解放军总参谋部陆航研究所 | Visual interactive knowledge base editor for fault diagnosis system |
CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
CN104035433A (en) * | 2014-05-30 | 2014-09-10 | 北京必可测科技股份有限公司 | Remote precise diagnosis and maintenance system and method of industrial devices |
CN104376033A (en) * | 2014-08-01 | 2015-02-25 | 中国人民解放军装甲兵工程学院 | Fault diagnosis method based on fault tree and database technology |
CN106155999A (en) * | 2015-04-09 | 2016-11-23 | 科大讯飞股份有限公司 | Semantics comprehension on natural language method and system |
-
2018
- 2018-07-24 CN CN201810821381.8A patent/CN108960673B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06222922A (en) * | 1993-01-21 | 1994-08-12 | Sankyo Seiki Mfg Co Ltd | Expert system |
US20100070448A1 (en) * | 2002-06-24 | 2010-03-18 | Nosa Omoigui | System and method for knowledge retrieval, management, delivery and presentation |
CN101470701A (en) * | 2007-12-29 | 2009-07-01 | 日电(中国)有限公司 | Text analyzer supporting semantic rule based on finite state machine and method thereof |
CN102819239A (en) * | 2011-06-08 | 2012-12-12 | 同济大学 | Intelligent fault diagnosis method of numerical control machine tool |
CN102184569A (en) * | 2011-06-11 | 2011-09-14 | 福州大学 | Individual plant wood modeling method driven by domain ontology |
CN102254102A (en) * | 2011-07-18 | 2011-11-23 | 华东理工大学 | Ontology-based sewage decision-support system and method |
CN202394263U (en) * | 2012-04-01 | 2012-08-22 | 中国人民解放军总参谋部陆航研究所 | Visual interactive knowledge base editor for fault diagnosis system |
CN104035433A (en) * | 2014-05-30 | 2014-09-10 | 北京必可测科技股份有限公司 | Remote precise diagnosis and maintenance system and method of industrial devices |
CN104376033A (en) * | 2014-08-01 | 2015-02-25 | 中国人民解放军装甲兵工程学院 | Fault diagnosis method based on fault tree and database technology |
CN106155999A (en) * | 2015-04-09 | 2016-11-23 | 科大讯飞股份有限公司 | Semantics comprehension on natural language method and system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115048095A (en) * | 2022-08-12 | 2022-09-13 | 广东粤港澳大湾区硬科技创新研究院 | Expert system fault diagnosis program generation method |
Also Published As
Publication number | Publication date |
---|---|
CN108960673B (en) | 2020-10-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102156406B (en) | Expert system and method for preventing and controlling sludge bulking under the diagnosis based on operation condition of sewage disposal plant | |
Madaeni et al. | Fuzzy modeling and hybrid genetic algorithm optimization of virus removal from water using microfiltration membrane | |
CN105836965A (en) | Intelligentized sewage treatment expert system | |
McMorris et al. | A view of some consensus methods for trees | |
Díaz-Madroñero et al. | Analysis of a wastewater treatment plant using fuzzy goal programming as a management tool: A case study | |
CN102063062B (en) | Sludge bulking control expert system for diagnosis based on filamentous bacteria population structure | |
CN108960673A (en) | Sewage treatment method for diagnosing faults and device | |
Han et al. | An intelligent detecting system for permeability prediction of MBR | |
Holland | Robust algorithms for adaptation set in a general formal framework | |
CN109816940A (en) | The fault alarm method and device of sewage treatment plant | |
Sànchez–Marrè et al. | Learning and adaptation in wastewater treatment plants through case–based reasoning | |
KR20140117959A (en) | Apparatus and method for provide of causative factors for stage of outflow water of wastewater treatment plant | |
Shah | Techniques for estimating construction costs of waste treatment plants | |
Liu et al. | Multi-objective optimization of cascade controller in combined biological nitrogen and phosphorus removal wastewater treatment plant | |
Rodriguez‐Roda et al. | Conceptual design of wastewater treatment plants using a design support system | |
CN115329661B (en) | Intelligent dosing model modeling, intelligent dosing system creation and dosing method | |
Schütze et al. | Multi-objective control of urban wastewater systems | |
Comas et al. | Extension of the IWA/COST simulation benchmark to include expert reasoning for system performance evaluation | |
Chen et al. | Structure-based filtering for ontology alignment | |
Lohani et al. | A MULTIOBJECTIVE MODEL FOR REGIONAL WATER QUALITY MANAGEMENT 1 | |
Roda et al. | Development of a case-based system for the supervision of an activated sludge process | |
Roda et al. | Application of a support system to the design of wastewater treatment plants | |
JPH10133737A (en) | Plant operation support system | |
CN110245804B (en) | Method and device for screening wastewater pollutants in area and monomer optimal control characteristics of area | |
Chellaswamy et al. | Cloud assisted waste water management system for smart cities |
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 |