CN108960673B - Sewage treatment fault diagnosis method and device - Google Patents

Sewage treatment fault diagnosis method and device Download PDF

Info

Publication number
CN108960673B
CN108960673B CN201810821381.8A CN201810821381A CN108960673B CN 108960673 B CN108960673 B CN 108960673B CN 201810821381 A CN201810821381 A CN 201810821381A CN 108960673 B CN108960673 B CN 108960673B
Authority
CN
China
Prior art keywords
node
fault
semantic
sewage treatment
clause
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.)
Active
Application number
CN201810821381.8A
Other languages
Chinese (zh)
Other versions
CN108960673A (en
Inventor
王达
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
DASMART ENVIRONMENTAL TECHNOLOGIES (BEIJING) CO LTD
Original Assignee
Beijing Etechwin Electric Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Etechwin Electric Co Ltd filed Critical Beijing Etechwin Electric Co Ltd
Priority to CN201810821381.8A priority Critical patent/CN108960673B/en
Publication of CN108960673A publication Critical patent/CN108960673A/en
Application granted granted Critical
Publication of CN108960673B publication Critical patent/CN108960673B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Artificial Intelligence (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • Quality & Reliability (AREA)
  • General Engineering & Computer Science (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention provides a sewage treatment fault diagnosis method and a device, wherein the sewage treatment fault diagnosis method comprises the following steps: receiving at least one fault information regarding sewage treatment; transmitting the at least one fault information to an expert module, and receiving a fault diagnosis result corresponding to each fault information from the expert module; or calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information, wherein the sewage treatment knowledge base is established based on at least one semantic tree corresponding to a sewage treatment process flow. The sewage treatment fault diagnosis method and the sewage treatment fault diagnosis device can improve the accuracy of fault information description, dynamically update knowledge in a sewage treatment knowledge base, and can also carry out interactive communication with an online technical expert so as to obtain a proper and effective fault diagnosis result.

Description

Sewage treatment fault diagnosis method and device
Technical Field
The present invention relates generally to the field of sewage treatment technology, and more particularly, to a method and apparatus for diagnosing a fault in sewage treatment.
Background
The sewage treatment is a complex process, and the common three-stage treatment process comprises a physical method, a biological method and a chemical method. Influenced by the life law of residents, industrial wastewater discharge, climate, geographical conditions and the like, the water quality and the water quantity of inlet water of a sewage treatment plant can show certain fluctuation, so that the sewage treatment process has the characteristics of multivariable, nonlinearity, time-varying property and uncertainty. Therefore, in the operation process of the sewage treatment plant, technical management personnel are often required to implement proper intervention to ensure the normal operation of the sewage treatment process flow and ensure that the effluent reaches the standard and is discharged.
At present, an expert system is generally applied to fault diagnosis of a sewage treatment process flow, so that a scheme or measure capable of solving actual problems is provided for technical management personnel, the defect of practical experience of sewage treatment is made up, the operation management level and efficiency are improved, and the operation cost is saved. The expert system is a general aid decision making system, which is composed of a knowledge base, an inference engine and an auxiliary component, and the application and the expansion of the expert system in the sewage treatment field are mainly focused on the representation of knowledge in the knowledge base and the construction of the knowledge base.
However, the existing knowledge base of the expert system for sewage treatment is often constructed for common problems in the sewage treatment process flow, and sometimes it is difficult to give satisfactory results according to fault problems occurring in the sewage treatment process flow, mainly because of the following three aspects: (1) the input information is inaccurate, and the given result is too single. For example, the results from the expert sewage treatment system are difficult to obtain a good expected result due to inaccurate input information caused by the imperfection of the expert sewage treatment system, unclear description of the fault phenomenon, and more conditions affecting the results from the expert sewage treatment system. (2) Knowledge in the knowledge base cannot be dynamically changed flexibly. For example, some expert sewage treatment systems do not allow new knowledge to be added after the knowledge base is built, or the knowledge base is built again every considerable time, which seriously affects the practicability and effectiveness of the expert sewage treatment system. (3) There is a lack of interactive communication with on-line technical experts.
Disclosure of Invention
The invention aims to provide a sewage treatment fault diagnosis method and device, which can improve the accuracy of fault information description, dynamically update knowledge in a sewage treatment knowledge base, and can perform interactive communication with an online technical expert so as to obtain a proper and effective fault diagnosis result.
One aspect of the present invention provides a sewage treatment fault diagnosis method including: receiving at least one fault information regarding sewage treatment; transmitting the at least one fault information to an expert module, and receiving a fault diagnosis result corresponding to each fault information from the expert module; or calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information, wherein the sewage treatment knowledge base is established based on at least one semantic tree corresponding to a sewage treatment process flow.
Optionally, the step of sending the at least one fault information to an expert module comprises: sending assistance request information to the expert module together with the at least one fault information.
Optionally, the sewage treatment fault diagnosis method further includes: receiving a selection of an expert module, wherein the transmitting of the at least one fault information to the expert module and the receiving of the fault diagnosis result corresponding to each fault information from the expert module comprises: sending assistance request information together with the at least one fault information to the selected expert module; and receiving a fault diagnosis result corresponding to each fault information from the selected expert module, wherein the fault diagnosis result corresponding to each fault information received from the selected expert module is a fault diagnosis result determined based on experience or a fault diagnosis result determined by calling a sewage treatment knowledge base.
Optionally, the step of invoking a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information includes: converting each fault information into a fault clause corresponding to each fault information; identifying the matching degree of each fault clause and a semantic node in a sewage treatment knowledge base; judging whether each matching degree reaches a preset threshold value; and deleting the fault clause corresponding to the matching degree which does not reach the preset threshold value, and inquiring a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree which reaches the preset threshold value.
Optionally, the sewage treatment fault diagnosis method further includes: and responding to the deletion of the fault clause corresponding to the matching degree which does not reach the preset threshold value, and prompting to adjust the fault information corresponding to the deleted fault clause.
Optionally, the sewage treatment knowledge base stores each semantic tree in the form of a relational database.
Optionally, each semantic tree includes at least two semantic nodes in different layers encapsulated by predetermined information of each flow of the sewage treatment process, wherein each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and the hierarchical relationship of each semantic node.
Optionally, the step of querying a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree reaching the predetermined threshold includes: searching the number of layers of fault clauses corresponding to each matching degree reaching a preset threshold value on a semantic tree in a sewage treatment knowledge base, and sequencing and determining a fault sequence based on the searched number of layers; determining a first semantic node corresponding to a first fault clause with the minimum layer number and a second semantic node corresponding to a second fault clause with the maximum layer number in the fault sequence; when detecting that a third semantic node corresponding to a third fault clause in the fault sequence is a child node of the first semantic node, judging whether the third semantic node is the same as the second semantic node; when the third semantic node is the same as the second semantic node, determining whether the third semantic node is a leaf node by detecting the out-degree and the in-degree of the third semantic node; and when the third semantic node is a leaf node, determining the node content of the third semantic node as a fault diagnosis result, wherein the third fault clause is the next fault clause adjacent to the first fault clause in the fault sequence.
Optionally, when the third semantic node is different from the second semantic node, the third faulty clause is updated to a next faulty clause adjacent to the third faulty clause, and the third semantic node is updated to a semantic node corresponding to the updated third faulty clause.
Optionally, when it is detected that the third semantic node is not a child node of the first semantic node, determining the third semantic node as a semantic node corresponding to a fourth faulty clause, and then performing a step of determining whether the third semantic node is a leaf node, where the fourth faulty clause is a last faulty clause adjacent to the third faulty clause in the fault sequence.
Optionally, when the third semantic node is not a leaf node, all leaf nodes under the third semantic node are searched, and the node contents of all searched leaf nodes are determined as a fault diagnosis result.
Optionally, the sewage treatment knowledge base performs a semantic tree growing process in response to a new node being input; the wastewater treatment knowledge base performs a pruning process of the semantic tree in response to the predecessor nodes being deleted, wherein the pruning process is deleted when the predecessor nodes fail to reach a predetermined threshold within a predetermined time period.
Optionally, the step of performing a growth process of the semantic tree comprises: when the clauses contained in the child nodes of the root node or the internal node form a logic complementary relationship or are leaf nodes, adding the node containing the newly input clause and the clause forming the logic complementary relationship with the newly input clause between the original parent child nodes as a new node; and when the child nodes of the root node or the internal node are one or the clauses contained in a plurality of child nodes do not form a logical complementary relationship, taking the node containing the newly input clause as a new child node of the root node or the internal node.
Optionally, the step of performing a pruning process of the semantic tree comprises: when the father node of the leaf node executing the pruning process has the brother node, deleting the leaf node, the father node and the brother node, and taking the child node of the brother node as the child node of the father node of the brother node; when the parent node of the leaf node performing the pruning process has no sibling node, the leaf node and the parent node are deleted.
Another aspect of the present invention also provides a sewage treatment failure diagnosis apparatus including: a fault information receiving unit configured to receive at least one fault information regarding sewage treatment; a fault information processing unit configured to transmit the at least one fault information to the expert module and receive a fault diagnosis result corresponding to each fault information from the expert module; or calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information, wherein the sewage treatment knowledge base is established based on at least one semantic tree corresponding to a sewage treatment process flow.
Optionally, the fault information processing unit is further configured to: sending assistance request information to the expert module together with the at least one fault information.
Optionally, the sewage treatment fault diagnosis apparatus further includes: a selection receiving unit configured to receive a selection of an expert module, wherein the fault information processing unit is further configured to: sending assistance request information together with the at least one fault information to the selected expert module; and receiving a fault diagnosis result corresponding to each fault information from the selected expert module, wherein the fault diagnosis result corresponding to each fault information received from the selected expert module is a fault diagnosis result determined based on experience or a fault diagnosis result determined by calling a sewage treatment knowledge base.
Optionally, the fault information processing unit is further configured to: converting each fault information into a fault clause corresponding to each fault information; identifying the matching degree of each fault clause and a semantic node in a sewage treatment knowledge base; judging whether each matching degree reaches a preset threshold value; and deleting the fault clause corresponding to the matching degree which does not reach the preset threshold value, and inquiring a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree which reaches the preset threshold value.
Optionally, the fault information processing unit is further configured to: and responding to the deletion of the fault clause corresponding to the matching degree which does not reach the preset threshold value, and prompting to adjust the fault information corresponding to the deleted fault clause.
Optionally, the sewage treatment knowledge base stores each semantic tree in the form of a relational database.
Optionally, each semantic tree includes at least two semantic nodes in different layers encapsulated by predetermined information of each flow of the sewage treatment process, wherein each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and the hierarchical relationship of each semantic node.
Optionally, the fault information processing unit is further configured to: searching the number of layers of fault clauses corresponding to each matching degree reaching a preset threshold value on a semantic tree in a sewage treatment knowledge base, and sequencing and determining a fault sequence based on the searched number of layers; determining a first semantic node corresponding to a first fault clause with the minimum layer number and a second semantic node corresponding to a second fault clause with the maximum layer number in the fault sequence; when detecting that a third semantic node corresponding to a third fault clause in the fault sequence is a child node of the first semantic node, judging whether the third semantic node is the same as the second semantic node; when the third semantic node is the same as the second semantic node, determining whether the third semantic node is a leaf node by detecting the out-degree and the in-degree of the third semantic node; and when the third semantic node is a leaf node, determining the node content of the third semantic node as a fault diagnosis result, wherein the third fault clause is the next fault clause adjacent to the first fault clause in the fault sequence.
Optionally, the fault information processing unit is further configured to: and when the third semantic node is different from the second semantic node, updating the third fault clause into a next fault clause adjacent to the third fault clause, and updating the third semantic node into a semantic node corresponding to the updated third fault clause.
Optionally, the fault information processing unit is further configured to: when the third semantic node is detected not to be a child node of the first semantic node, determining the third semantic node as a semantic node corresponding to a fourth fault clause, and then executing a step of determining whether the third semantic node is a leaf node, wherein the fourth fault clause is a last fault clause adjacent to the third fault clause in the fault sequence.
Optionally, the fault information processing unit is further configured to: and when the third semantic node is not a leaf node, searching all leaf nodes under the third semantic node, and determining the node contents of all the searched leaf nodes as fault diagnosis results.
Optionally, the sewage treatment knowledge base performs a semantic tree growing process in response to a new node being input; the wastewater treatment knowledge base performs a pruning process of the semantic tree in response to the predecessor nodes being deleted, wherein the pruning process is deleted when the predecessor nodes fail to reach a predetermined threshold within a predetermined time period.
Optionally, the fault information processing unit is further configured to: when the clauses contained in the child nodes of the root node or the internal node form a logic complementary relationship or are leaf nodes, adding the node containing the newly input clause and the clause forming the logic complementary relationship with the newly input clause between the original parent child nodes as a new node; and when the child nodes of the root node or the internal node are one or the clauses contained in a plurality of child nodes do not form a logical complementary relationship, taking the node containing the newly input clause as a new child node of the root node or the internal node.
Optionally, the fault information processing unit is further configured to: when the father node of the leaf node executing the pruning process has the brother node, deleting the leaf node, the father node and the brother node, and taking the child node of the brother node as the child node of the father node of the brother node; when the parent node of the leaf node performing the pruning process has no sibling node, the leaf node and the parent node are deleted.
Another aspect of the present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to execute the sewage treatment failure diagnosis method as described above.
Another aspect of the present invention also provides a computing apparatus, comprising: a processor; a memory for storing a computer program which, when executed by the processor, causes the processor to execute the sewage treatment failure diagnosis method as described above.
According to the sewage treatment fault diagnosis method and device, the accuracy of fault information description is improved in a feedback information mode, and the actual condition of fault diagnosis of a sewage treatment process flow is combined, so that the knowledge (namely, semantic nodes) in a sewage treatment knowledge base is dynamically increased or decreased on the basis of a semantic tree, and the sewage treatment knowledge base is perfected; and the system can also be interactively communicated with an online technical expert to obtain the help of the online technical expert so as to obtain the professional guidance of sewage treatment as much as possible. In addition, fault information is continuously matched with knowledge in a sewage treatment knowledge base to form an inference chain, so that a proper and effective fault diagnosis result is obtained.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which:
FIG. 1 shows a flow diagram of a prior art wastewater treatment process;
FIG. 2 shows a flow chart of a sewage treatment fault diagnosis method according to an embodiment of the present invention;
FIG. 3 shows a logical schematic of a semantic tree of sludge bulking in a secondary sedimentation tank failure, according to an embodiment of the invention;
FIG. 4 illustrates a flow diagram for querying a fault diagnosis result, according to an embodiment of the invention;
FIG. 5 shows a flow diagram of fault diagnosis according to an embodiment of the invention;
FIG. 6 illustrates an example of a process for growing a semantic tree according to an embodiment of the present invention;
FIG. 7 shows a schematic view of a user interface according to an embodiment of the invention;
FIG. 8 shows a schematic diagram of the general architecture of a wastewater treatment process flow troubleshooting expert system according to an embodiment of the invention;
fig. 9 shows a block diagram of a sewage treatment failure diagnosis apparatus according to an embodiment of the present invention.
Detailed Description
Various example embodiments will now be described more fully with reference to the accompanying drawings, in which some example embodiments are shown.
The sewage treatment process is a complex process flow comprising physical, chemical and biological reactions, and usually comprises a plurality of treatment units, so that the removal and conversion of various pollutants are realized at different treatment stages. FIG. 1 shows a flow chart of a sewage treatment process in the prior art, which comprises a water inlet sand settling unit, a hydrolysis acidification unit, a biochemical treatment unit, a dosing disinfection unit and a sludge dehydration unit.
The sewage that gets into sewage treatment plant is at first through the heavy husky unit of intaking, and sewage flows through thick grid, dams the impurity of the great granule in the aquatic etc. and the water pump that promotes the pump house is in pumping the sewage to the pond of a take the altitude after that, guarantees that sewage can flow through follow-up each processing link by the action of gravity. Next, the sewage is passed through a fine grid to further remove smaller particles of impurities. The rotary grit chamber is funnel-shaped, is internally provided with a paddle board, and drives sewage to rotate under the drive of a motor to separate sand from water.
Next, the sewage enters a hydrolysis acidification unit. The hydrolysis acidification pool has the function of intercepting non-dissolved organic matters in the sewage and gradually converting the non-dissolved organic matters into dissolved organic matters, and some macromolecular substances which are difficult to biodegrade are converted into small molecular substances which are easy to degrade, such as organic acid and the like, so that the biodegradability and the degradation speed of the sewage are improved, and the subsequent aerobic biological treatment is facilitated.
Next, the sewage enters a biochemical unit. The integrated oxidation ditch adopts an inverted channel A2In the O process (anoxic-anaerobic-aerobic), the activated sludge performs biochemical reaction with most pollutants in the sewage by utilizing the air aerated by the blower, thereby playing the role of nitrogen and phosphorus removal. One part of the activated sludge continuously circulates in the integrated oxidation ditch, and the other part of the activated sludge and the sewage flow to a secondary sedimentation tank together. The secondary sedimentation tank is used for clarifying and concentrating the mixed liquor to realize sludge separation and providing activated sludge which flows back to the inlet of the integrated oxidation ditch.
And then, the effluent of the secondary sedimentation tank enters a dosing and disinfecting unit. And (3) enabling the effluent of the secondary sedimentation tank to enter a high-efficiency sedimentation tank, and adding PAC and PAM medicaments to further obtain flocculation sedimentation of particles in the sewage. The filter cloth filter tank is added behind the high-efficiency sedimentation tank, which is used for intercepting suspended matters in sewage after flocculation and sedimentation and can play a role in removing total solid suspended matters, total phosphorus and heavy metals. The contact disinfection tank is used for killing virus and germs affecting the environment in the sewage by adding chlorine, so that the treated sewage can be finally discharged up to the standard.
In addition, the activated sludge discharged from the secondary sedimentation tank also passes through a sludge dewatering unit. And (3) returning a part of the activated sludge discharged from the secondary sedimentation tank to the integrated oxidation ditch, and discarding most of the activated sludge in the form of excess sludge. Then, because the residual sludge has fluidity and the water content is still high, the residual sludge needs further concentration and dehydration and is converted into a mud cake with lower water content, and then the mud cake is transported outside the car.
Through the treatment process, pollutants contained in the sewage are separated and removed from the water, so that the harmful substances are converted into harmless and even useful substances, and the sewage is purified.
However, in the sewage treatment process, various malfunctions may occur. Therefore, it is necessary to provide a solution or measure for technical management personnel through sewage treatment fault diagnosis, so that the technical management personnel can find out the most appropriate and effective treatment measure as soon as possible by means of the help of a sewage treatment knowledge base and/or an on-line technical expert when facing a complicated sewage treatment process problem, thereby ensuring that the sewage is discharged up to the standard and reducing the adverse effect on the environment.
A sewage treatment failure diagnosis method and apparatus according to an embodiment of the present invention will be described below with reference to fig. 2 to 9.
Fig. 2 shows a flowchart of a sewage treatment failure diagnosis method according to an embodiment of the present invention.
At step S10, at least one fault information regarding the sewage treatment is received.
As an example, the fault information may be information in a natural language for describing a fault occurring in a wastewater treatment process flow. For example, the failure information may be "sludge bulking in the secondary sedimentation tank", "no filamentous fungus flocs are seen", and the like.
Preferably, commas are used as separators between a plurality of pieces of fault information, i.e., commas separate different pieces of fault information.
In one embodiment of step S10, at least one fault information regarding sewage treatment input by a user (i.e., a user) may be received.
At step S20, at least one fault information is transmitted to the expert module, and a fault diagnosis result corresponding to each fault information is received from the expert module.
Preferably, the assistance request information is transmitted to the expert module together with at least one fault information, and the fault diagnosis result corresponding to each fault information is received and displayed from the expert module.
By way of example, the expert module may have access to an online technical expert, which may be a person currently online having a wide and profound understanding of the wastewater treatment process flow, although the invention is not so limited.
In this case, the sewage treatment failure diagnosis method may further include: a selection of an expert module is received.
In one embodiment of step S20, the assistance request information is sent to the selected expert module together with at least one fault information; and receiving a fault diagnosis result corresponding to each fault information from the selected expert module.
Preferably, the received fault diagnosis result corresponding to each fault information may also be displayed.
The fault diagnosis result corresponding to each fault information received from the selected expert module may be: based on empirically determined fault diagnosis results or fault diagnosis results determined by calling a sewage treatment knowledge base.
It is understood that when the selected expert module obtains the assistance request information, the fault diagnosis result can be given in the following two ways: (1) if the online technical expert in the expert module thinks that the judgment, the treatment measures and the like of the fault reason can be made only by the experience of the online technical expert, the fault diagnosis result can be directly given; (2) if the online technical expert in the expert module considers that a fault diagnosis expert system needs to be used for sewage treatment process flow, fault description can be rearranged according to fault information sent together with the assistance request information, and a sewage treatment knowledge base is called to execute fault inquiry, so that a fault diagnosis result is given. In this case, when the online technical expert in the expert module rearranges the fault description, the fault description can be more accurately described by applying a professional vocabulary, or some conditions or limitations can be increased or decreased, so as to change the range of inquiry or inference.
Alternatively, in step S30, the sewage treatment knowledge base is invoked to determine a fault diagnosis result corresponding to each fault information based on each fault information.
Preferably, the determined fault diagnosis result may also be displayed.
As an example, the wastewater treatment knowledge base may be a wastewater treatment knowledge base established based on at least one semantic tree corresponding to a wastewater treatment process flow.
That is, the expert system for fault diagnosis of wastewater treatment process flow is based on semantic trees to build a dynamic wastewater treatment knowledge base with knowledge that can be increased or decreased. In this way, the query and inference of fault information can be translated into a search process on the semantic tree, and the addition or subtraction of knowledge can be achieved by performing growing and/or pruning operations on the semantic tree.
As an example, according to the characteristics of the sewage treatment process flow, the treatment links of water inlet sand settling, hydrolytic acidification, oxidation ditch, filter cloth clarification, chemical adding disinfection, sludge dewatering and the like are relatively independent, the causes and treatment measures for faults are different, and different semantic trees are respectively established for each treatment link, namely, one semantic tree can correspond to one treatment link in the sewage treatment process flow. And storing all semantic trees corresponding to the sewage treatment process flow in a semantic forest form.
Each semantic tree may include at least two semantic nodes in different layers encapsulated by predetermined information for each flow of the wastewater treatment process.
For example, if SS is a clause describing status information of a certain processing link in a sewage treatment knowledge base, and as is a fault clause corresponding to fault information, then the status set SS is a set of SS, that is, SS ═ SS1,ss2,...,sspP is the number of clauses describing the state information of a certain sewage treatment link; set of faults
Figure BDA0001741476960000101
And q is the number of fault clauses corresponding to the fault information.
The semantic tree T about AS is one growing downDirected tree, T ═<V,E>Is a doublet, V ═ r, ni,lj0 is more than or equal to i and less than infinity, 1 is more than or equal to j and less than infinity. Here, r is a root node, the in-degree is 0, the out-degree is greater than 1, and the negative of a limited number of clauses ss or clauses ss is included; n isiThe internal node is an ith internal node, the in-degree is 1, the out-degree is greater than 1, and the negative result contains a limited number of clauses ss or clauses ss; ljThe leaf node (jth internal node) is a leaf node, the in-degree is 1, the out-degree is 0, and the leaf node comprises fault reasons and processing measures; e is a group consisting of r and niOr ljA set of formed ordered tuples; v is a node set of the semantic tree.
Semantic forest F is a set of semantic trees, then F ═ T1,T2,...,TzAnd z is the number of semantic trees.
It is understood that the semantic tree T is a multi-branch tree, the root node and the internal node may have a plurality of children nodes, but a certain leaf node or internal node only has a unique parent node. The minimum semantic tree has at least one root node and one leaf node. The semantic forest F may have only one semantic tree T.
Fig. 3 shows a logical schematic of a semantic tree of sludge bulking in a secondary sedimentation tank failure according to an embodiment of the invention.
Referring to fig. 3, according to the characteristics of the sewage treatment process flow, the fault information, the fault reason, the treatment measures and other related information of the secondary sedimentation tank are encapsulated into semantic nodes, the hierarchical relation of the related information is embodied by the tree structure, a semantic tree of the fault of the secondary sedimentation tank is established, and the fault phenomenon, the fault reason and the treatment measures of sludge bulking are described in a repeated point mode, namely the semantic tree integrates the fault phenomenon, the fault reason and the treatment measures. In FIG. 3, DO is dissolved oxygen, SVI is sludge volume index, pH is pH, and F/M is sludge load.
Preferably, the sewage treatment knowledge base stores each semantic tree in the form of a relational database (i.e., a relational database). Each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and the hierarchical relationship of each semantic node. Therefore, the inquiry and reasoning of the fault information in the fault diagnosis expert system of the sewage treatment process flow can be realized by searching the state space of the execution tree in a data access form provided by the relational database.
Table 1 shows an example of the domain of the semantic nodes. As shown in table 1, the data structure of the semantic node includes four fields, which are a NodeID field, a farmerid field, a childld field, and a Content field. Here, the NodeID, farmerid and childld fields may represent a hierarchical relationship, and the Content field may represent the Content of a node.
Table 1: example of Domain of semantic nodes
Name of field Data type Default value Description of the invention
NodeID Text Node number (unique)
FatherID Text Parent node numbering
ChildID Text Child node numbering
Content Text null Node content
For example, the value of the NodeID field of the root node is 0, the value of the FatherID field is-1, the Content field of the internal node stores a clause set for describing the state information of the sewage treatment link, and the Content field of the leaf node stores the fault reason and the treatment measure.
The process of "calling the sewage treatment knowledge base to determine the fault diagnosis result corresponding to each fault information based on each fault information" is described in detail below.
Preferably, in step S30, each fault information is converted into a fault clause corresponding to each fault information; and inquiring a fault diagnosis result corresponding to each fault information from the sewage treatment knowledge base based on each fault clause.
For example, each natural language form of fault information is converted into a fault clause for querying or reasoning in the wastewater treatment knowledge base.
The process of "inquiring the fault diagnosis result corresponding to each fault information from the sewage treatment knowledge base on a per fault clause basis" will be described in detail below with reference to fig. 4.
Fig. 4 shows a flowchart for querying a fault diagnosis result according to an embodiment of the present invention.
Preferably, referring to fig. 4, in step S301, the matching degree of each fault clause with semantic nodes in the sewage treatment knowledge base is identified.
That is, each fault clause is compared with clauses (e.g., node contents of semantic nodes) in the sewage treatment knowledge base to obtain a matching degree.
It should be understood that the matching degree can be calculated by various methods capable of obtaining the matching degree of the fault clause and the clause in the sewage treatment knowledge base, and the invention is not limited to this.
In step S302, it is determined whether each matching degree reaches a predetermined threshold value.
It is understood that the predetermined threshold may be preset according to the requirement, and the present invention is not limited thereto.
In step S303, the fault clause corresponding to the matching degree that does not reach the predetermined threshold is deleted, and in step S304, the corresponding fault diagnosis result is queried from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree that reaches the predetermined threshold.
In addition, the sewage treatment failure diagnosis method may further include: and responding to the deletion of the fault clause corresponding to the matching degree which does not reach the preset threshold value, and prompting to adjust the fault information corresponding to the deleted fault clause.
That is to say, feedback information can be provided according to the judgment result of the matching degree so as to receive fault information which is as accurate as possible, so that the accuracy of describing the fault information is improved, the fault information is understood by a fault diagnosis expert system of the sewage treatment process flow as much as possible, and the difficulty in acquiring and understanding the fault information is reduced. If the accuracy of a certain fault clause is too low and the modified fault clause still does not reach the degree required by a fault diagnosis expert system of the sewage treatment process flow, the fault clause can be deleted, and only the fault clause with higher accuracy is used for inquiry and reasoning, so that a proper and effective fault diagnosis result can be obtained.
According to the fault information of the sewage treatment process flow and the characteristics of the semantic tree, the traversal of the semantic tree can be realized by adopting more intuitive forward reasoning. The process of "inquiring the corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree reaching the predetermined threshold" is described in detail below with reference to fig. 5.
Fig. 5 shows a flow diagram of fault diagnosis according to an embodiment of the invention.
Referring to fig. 5, in step S401, the number of layers of the fault clause on the semantic tree corresponding to each matching degree reaching the predetermined threshold is searched in the sewage treatment knowledge base, and the fault sequence is determined based on the searched number of layers in a sorted manner.
Preferably, the number of layers of the fault clause on the semantic tree corresponding to each matching degree reaching the preset threshold is searched in the sewage treatment knowledge base, and the searched number of layers is used as the number of layers of the fault clause corresponding to each matching degree reaching the preset threshold; and according to the searched layer number, performing ascending arrangement on all fault clauses corresponding to the matching degrees reaching the preset threshold value to obtain a fault sequence.
It can be understood that the number of layers of a clause in a semantic tree may be the path length from the root node of the semantic tree to the semantic node where the clause is located. For example, the number of levels at which the root node contains clauses may be specified to be 0.
In step S402, a first semantic node corresponding to a first failure clause with the smallest layer number and a second semantic node corresponding to a second failure clause with the largest layer number in the failure sequence are determined.
In step S403, it is detected in the sewage treatment knowledge base whether a third semantic node corresponding to a third fault clause in the fault sequence is a child node of the first semantic node.
That is, whether the third semantic node exists is judged in the sewage treatment knowledge base. And when judging whether the third semantic node exists or not for the first time, the third fault clause is the next fault clause adjacent to the first fault clause in the fault sequence.
When it is detected that the third semantic node is a child node of the first semantic node, in step S404, it is determined whether the third semantic node is the same as the second semantic node.
That is, when the third semantic node is detected to exist, whether the third semantic node is the semantic node corresponding to the fault clause with the largest layer number is judged.
When the third semantic node is different from the second semantic node, in step S405, the third faulty clause is updated to the next faulty clause adjacent to the third faulty clause, and the third semantic node is updated to the semantic node corresponding to the updated third faulty clause, and the process returns to step S403.
When the third semantic node is the same as the second semantic node, in step S406, it is determined whether the third semantic node is a leaf node by detecting the out-degree and the in-degree of the third semantic node.
When it is detected that the third semantic node is not a child node of the first semantic node, in step S407, the third semantic node is determined as a semantic node corresponding to the fourth faulty clause, and then step S406 is performed.
The fourth faulty clause may be the last faulty clause in the faulty sequence that is adjacent to the third faulty clause. It will be appreciated that when the third faulting clause is updated, the fourth faulting clause may be the last faulting clause in the faulting sequence that is adjacent to the updated third faulting clause.
When the third semantic node is a leaf node, in step S408, the node content of the third semantic node is determined as a fault diagnosis result.
When the third semantic node is not a leaf node, in step S409, all leaf nodes under the third semantic node are searched, and the node contents of all the searched leaf nodes are determined as the fault diagnosis result.
As an example, the specific steps of the fault diagnosis of the embodiment of the present invention are as follows:
(1) search fault set AS ═ AS1,as2,...,asqThe number of layers on the syntax tree T, and each fault clause as1、as2、……、asqThe layers are arranged in ascending order to obtain a fault sequence AS ═ AS'1,as'2,...,as'q}。
(2) Searching for the fault clause AS with the smallest current layer number in the fault sequence AS1'the first semantic node with the largest number of layers and fault clause as'qThe second semantic node.
(3) Depth-first search fault sequence AS 'next clause AS'2The third semantic node where it is located.
(4) Judging whether a third semantic node exists, if so, executing the step (5); otherwise, selecting a third semantic node AS in 'in the fault sequence AS'2Of (a) '(i.e., as'1) The language ofAnd (5) defining the node, and executing the step (6).
(5) Judging whether the third semantic node is a fault clause as with the largest layer number'qThe located semantic node (namely, whether the third semantic node is the same as the second semantic node or not) is determined, if yes, the step (6) is executed; otherwise, returning to the step (3), in which case, in the step (3) of returning to execution, the depth-first search clause as'2Of the next clause as'3The semantic node is located, and the third semantic node is updated to the searched clause as'3And (4) determining the next clause according to the above mode and updating the third semantic node to the node where the determined next clause is located when the step (3) is executed.
(6) Judging whether the third semantic node is a leaf node, if so, executing the step (8); otherwise, step (7) is performed.
(7) And searching all leaf nodes under the third semantic node in a depth-first mode.
(8) The node contents (i.e., fault cause and handling measures) in the leaf nodes are exported.
In a wastewater treatment process flow troubleshooting expert system, the knowledge in the wastewater treatment knowledge base may be increased or decreased to dynamically update the knowledge in the wastewater treatment knowledge base.
Preferably, the wastewater treatment knowledge base performs a semantic tree growing process in response to a new node being entered, thereby increasing knowledge in the wastewater treatment knowledge base.
FIG. 6 illustrates an example of a process for growing a semantic tree according to an embodiment of the present invention.
In clauses ssnewAdded to the root node r or the internal node niFor example, the following two cases can be divided:
(1) if the root node r or the internal node niThe clauses contained in the child nodes form a logical complementary relation or are leaf nodes, the child nodes contain ssnewClauses and
Figure BDA0001741476960000141
clause (I.e. with ssnewClauses constituting clauses of a logical complementary relationship) as a root node r or an internal node niAnd includes ssnewClauses or
Figure BDA0001741476960000142
The node of the clause is used as a root node r or an internal node niParent node of the original child node. ssnewClauses and
Figure BDA0001741476960000143
the choice of clause may be determined by the actual circumstances. Referring to FIG. 6, including ssnewClauses and
Figure BDA0001741476960000144
the nodes of the clause are added to the original parent-child node as new nodes (ss)cur、ssson、
Figure BDA0001741476960000145
) In the meantime.
(2) If the root node r or the internal node niIf only one child node(s) is (are) included, or if the clauses included in a plurality of child nodes do not form a logical complementary relationship, the child nodes will include ssnewThe node of the clause is used as a root node r or an internal node niThe new child node of (2).
When some knowledge in the sewage treatment knowledge base obtains too low evaluation in a longer period of time, the sewage treatment knowledge base can delete the knowledge with too low evaluation. Knowledge deletion corresponds to the pruning process of the semantic tree, as opposed to knowledge addition.
Preferably, the wastewater treatment knowledge base performs a pruning process of the semantic tree in response to the original node being deleted, thereby reducing knowledge in the wastewater treatment knowledge base. For example, the original node is deleted when its score fails to reach a predetermined threshold within a predetermined period of time.
By way of example, depending on whether the parent of the deleted leaf node has a sibling node, two cases can be distinguished:
(1) if deleted leaf node ldeleteOf parent nodeIf there is brother node, it shows that the clause contained in father node and the clause contained in brother node form logic complementary relation, then delete leaf node ldeleteA parent node and a sibling node, and taking the child node of the sibling node as the child node of the parent node of the sibling node.
(2) If deleted leaf node ldeleteIf the father node has no brother node, the leaf node l is deleted directlydeleteAnd a parent node.
An example of a "user interface" is described in detail below in conjunction with FIG. 7.
FIG. 7 shows a schematic view of a user interface according to an embodiment of the invention.
Referring to fig. 7, the sewage treatment process flow may include the following treatment steps: settling sand in water, hydrolyzing and acidifying, performing biochemical treatment, adding chemicals for disinfection, and dehydrating sludge. The text box of the fault description is used for receiving fault information, the fault information is input in natural language, and the fault information is respectively 'secondary sedimentation tank sludge bulking' and 'large amount of filamentous fungus flocs are not seen', and the two fault information are separated by commas. When fault information ending by comma is received, the matching degree of the fault information and clauses in a sewage treatment knowledge base is automatically searched and is expressed by percentage. Specifically, the fault information "secondary sedimentation tank sludge bulking" is 100% in matching degree; the failure information "no filamentous fungus flocs were found", and the degree of matching was 95%. If the user finds that the matching degree is too low, the situation that the fault information is understood by mistake possibly occurs is shown, and the user can be prompted to adjust the text description of the fault information so as to execute correct inquiry reasoning; otherwise, when the matching degree obtained by searching is lower than a certain matching degree threshold value, the fault information with the lower matching degree can be automatically deleted. The matching degree evaluation mechanism can reduce the difficulty of searching and matching the fault information and also reduce the adverse effect of inaccurate description of the fault information on the fault diagnosis result.
The user can select the on-line expert from the expert list, so that the support of the on-line technical expert is sought, and the professional guidance is obtained as much as possible. As shown in fig. 7, expert 1 and expert 2 are online, expert 3 is not online, and the user has selected expert 1 to provide assistance to them.
The 'fault reason and suggested measure' can provide a plurality of fault diagnosis results of inquiry and inference for users to choose. The user can rate the failure diagnosis result, for example, five stars can be used to give five evaluation results. Certain fault diagnosis results may be deleted if they are evaluated too low for long term.
An example of the "sewage treatment process flow troubleshooting expert system" is described in detail below with reference to fig. 8.
Fig. 8 is a schematic diagram showing the general configuration of a sewage treatment process flow troubleshooting expert system according to an embodiment of the present invention.
Referring to fig. 8, a user interface and an expert interface may be designed for a user (e.g., a technical manager of a sewage treatment plant) and an expert in the sewage treatment field, respectively, and by giving different operation authorities to the user and the expert, dynamic increase and decrease of knowledge in a sewage treatment knowledge base and interactive communication between the user and an online technical expert are realized, so that a plurality of treatment measures or suggestions for current fault information are provided for the user.
As an example, the expert may have the authority to add knowledge to the sewage treatment knowledge base, but not the authority to delete existing knowledge from the sewage treatment knowledge base, thereby minimizing personal limitations and biases.
As an example, the fault diagnosis result may have three sources, respectively: the user inquires and infers the diagnostic result produced based on the sewage treatment knowledge base, and the technical expert inquires and infers the diagnostic result produced based on the sewage treatment knowledge base and the diagnostic result given by the technical expert by virtue of own experience. If a technical expert is involved, there may be instances of duplication or conflict between diagnostic results from different sources, in which case duplicate diagnostic results may be deleted and conflicting diagnostic results enumerated.
A sewage treatment failure diagnosis apparatus according to an embodiment of the present invention is described below with reference to fig. 9.
Fig. 9 shows a block diagram of a sewage treatment failure diagnosis apparatus according to an embodiment of the present invention.
Referring to fig. 9, a sewage treatment failure diagnosis apparatus according to an embodiment of the present invention includes: a failure information receiving unit 100 and a failure information processing unit 200.
The fault information receiving unit 100 receives at least one fault information on sewage treatment;
as an example, the fault information may be information in a natural language for describing a fault occurring in a wastewater treatment process flow. Commas are used as separators between a plurality of pieces of fault information, that is, commas are used to separate different pieces of fault information.
As an example, the fault information receiving unit 100 may receive at least one fault information about sewage treatment input by a user.
The fault information processing unit 200 transmits at least one fault information to an expert module (not shown), and receives a fault diagnosis result corresponding to each fault information from the expert module.
Preferably, the fault information processing unit 200 transmits the assistance request information to the expert module together with at least one fault information, and receives and displays a fault diagnosis result corresponding to each fault information from the expert module.
By way of example, the expert module may be a person currently on-line with a wide and profound understanding of the wastewater treatment process flow, although the invention is not so limited.
In this case, the sewage treatment failure diagnosis apparatus may further include: a selection acceptance unit (not shown). The selection receiving unit also receives a selection of an expert module.
In one embodiment of the fault information processing unit 200, the fault information processing unit 200 transmits the assistance request information together with at least one fault information to the selected expert module.
The fault information processing unit 200 receives a fault diagnosis result corresponding to each fault information from the selected expert module.
Preferably, the fault information processing unit 200 may further display the received fault diagnosis result corresponding to each fault information.
The fault diagnosis result corresponding to each fault information received from the selected expert module is a fault diagnosis result determined based on experience or a fault diagnosis result determined by calling a sewage treatment knowledge base.
Alternatively, the fault information processing unit 200 calls the sewage treatment knowledge base on the basis of each fault information to determine a fault diagnosis result corresponding to each fault information, and displays the determined fault diagnosis result,
as an example, the wastewater treatment knowledge base is a wastewater treatment knowledge base established based on at least one semantic tree corresponding to a wastewater treatment process flow.
According to the characteristics of the sewage treatment process flow, the treatment links of water inlet sand settling, hydrolytic acidification, oxidation ditch, filter cloth clarification, medicine adding disinfection, sludge dewatering and the like are relatively independent, the reasons and treatment measures for causing the faults are different, different semantic trees can be respectively established for each treatment link, namely, one semantic tree can correspond to one treatment link in the sewage treatment process flow. And storing all semantic trees corresponding to the sewage treatment process flow in a semantic forest form.
Each semantic tree may include at least two semantic nodes in different layers encapsulated by predetermined information for each flow of the wastewater treatment process.
It is understood that the semantic tree is a multi-branch tree, the root node and the internal node may have a plurality of child nodes, but a certain leaf node or internal node only has a unique parent node. The minimum semantic tree has at least one root node and one leaf node. A semantic forest may have only one semantic tree.
Preferably, the sewage treatment knowledge base stores each semantic tree in the form of a relational database (i.e., a relational database). Each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and the hierarchical relationship of each semantic node. Therefore, the inquiry and reasoning of the fault information in the fault diagnosis expert system of the sewage treatment process flow can be realized by searching the state space of the execution tree in a data access form provided by the relational database.
As an example, the fault information processing unit 200 may convert each fault information into a fault clause corresponding to each fault information, identify a matching degree of each fault clause with a semantic node in the sewage treatment knowledge base, determine whether each matching degree reaches a predetermined threshold, delete the fault clause corresponding to the matching degree that does not reach the predetermined threshold, and query a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree that reaches the predetermined threshold.
For example, the fault information processing unit 200 converts each fault information in the natural language form into a fault clause for query or inference in the sewage treatment knowledge base.
It is understood that the predetermined threshold may be preset according to the requirement, and the present invention is not limited thereto.
Further, the failure information processing unit 200 also responds to deletion of a failure clause corresponding to a matching degree that does not reach a predetermined threshold value, and prompts adjustment of failure information corresponding to the deleted failure clause.
That is, the fault information processing unit 200 may provide feedback information according to the determination result of the matching degree, so that the fault information receiving unit 100 receives fault information as accurate as possible, thereby improving the accuracy of describing the fault information, making the fault information understood by the sewage treatment process flow fault diagnosis expert system as possible, and reducing the difficulty of obtaining and understanding the fault information. If the accuracy of a certain fault clause is too low and the modified fault clause still does not reach the degree required by the fault diagnosis expert system of the sewage treatment process flow, the fault information processing unit 200 can delete the fault clause, and only the fault clause with higher accuracy is used for query reasoning, so that a proper and effective fault diagnosis result can be obtained.
According to the fault information of the sewage treatment process flow and the characteristics of the semantic tree, the traversal of the semantic tree can be realized by adopting more intuitive forward reasoning.
As an example, the fault information processing unit 200 searches the sewage treatment knowledge base for the number of layers of fault clauses on the semantic tree corresponding to each matching degree reaching a predetermined threshold, and determines a fault sequence based on the searched number of layers in a sorted manner; determining a first semantic node corresponding to a first fault clause with the minimum layer number and a second semantic node corresponding to a second fault clause with the maximum layer number in a fault sequence; detecting whether a third semantic node corresponding to a third fault clause in a fault sequence is a child node of the first semantic node or not in a sewage treatment knowledge base; when detecting that the third semantic node is a child node of the first semantic node, judging whether the third semantic node is the same as the second semantic node; when the third semantic node is the same as the second semantic node, determining whether the third semantic node is a leaf node or not by detecting the out-degree and the in-degree of the third semantic node; and when the third semantic node is a leaf node, determining the node content of the third semantic node as a fault diagnosis result.
And when judging whether the third semantic node is a child node of the first semantic node or not for the first time, the third fault clause is the next fault clause adjacent to the first fault clause in the fault sequence.
It can be understood that the number of layers of a clause in a semantic tree may be the path length from the root node of the semantic tree to the semantic node where the clause is located. For example, the number of levels at which the root node contains clauses may be specified to be 0.
Preferably, the fault information processing unit 200 may search the sewage treatment knowledge base for the number of layers of the fault clause on the semantic tree corresponding to each matching degree reaching the predetermined threshold, and use the searched number of layers as the number of layers of the fault clause corresponding to each matching degree reaching the predetermined threshold; and according to the searched layer number, performing ascending arrangement on all fault clauses corresponding to the matching degrees reaching the preset threshold value to obtain a fault sequence.
When the third semantic node is different from the second semantic node, the fault information processing unit 200 updates the third fault clause to a next fault clause adjacent to the third fault clause, updates the third semantic node to a semantic node corresponding to the updated third fault clause, and returns to perform processing of "detect in the sewage treatment knowledge base whether the third semantic node corresponding to the third fault clause in the fault sequence is a child node of the first semantic node".
When it is detected that the third semantic node is not a child node of the first semantic node, the failure information processing unit 200 determines the third semantic node as a semantic node corresponding to the fourth failed clause, and then performs processing of determining whether the third semantic node is a leaf node.
The fourth faulting clause is the last faulting clause in the faulting sequence that is adjacent to the third faulting clause. It will be appreciated that when the third faulting clause is updated, the fourth faulting clause may be the last faulting clause in the faulting sequence that is adjacent to the updated third faulting clause.
When the third semantic node is not a leaf node, the failure information processing unit 200 searches all leaf nodes under the third semantic node, and determines the node contents of all the searched leaf nodes as a failure diagnosis result.
In a wastewater treatment process flow troubleshooting expert system, the knowledge in the wastewater treatment knowledge base may be increased or decreased to dynamically update the knowledge in the wastewater treatment knowledge base.
As an example, the wastewater treatment knowledge base performs a growth process of the semantic tree in response to a new node being input, thereby increasing knowledge in the wastewater treatment knowledge base.
Preferably, when the clauses included in the child nodes of the root node or the internal node constitute a logical complementary relationship or are leaf nodes, the fault information processing unit 200 adds a node including a newly input clause and a clause constituting a logical complementary relationship with the newly input clause as a new node between the original parent child nodes.
Preferably, when the child nodes of the root node or the internal node are one, or the clauses included in a plurality of child nodes do not form a logical complementary relationship, the fault information processing unit 200 takes the node including the newly input clause as a new child node of the root node or the internal node.
When some knowledge in the sewage treatment knowledge base obtains too low evaluation in a longer period of time, the sewage treatment knowledge base can delete the knowledge with too low evaluation. Knowledge deletion corresponds to the pruning process of the semantic tree, as opposed to knowledge addition.
As an example, the sewage treatment knowledge base performs a pruning process of the semantic tree in response to the predecessor node being deleted, preferably when the predecessor node's score fails to reach a predetermined threshold within a predetermined time period, thereby reducing knowledge in the sewage treatment knowledge base.
Preferably, when the parent node of the leaf node that performs the pruning process has a sibling node, the fault information processing unit 200 deletes the leaf node, the parent node, and the sibling node, and takes the child node of the sibling node as the child node of the parent node of the sibling node.
Preferably, when the parent node of the leaf node that performs the pruning process has no sibling node, the failure information processing unit 200 deletes the leaf node and the parent node.
In addition, the sewage treatment fault diagnosis method and the sewage treatment fault diagnosis device of the embodiment of the invention improve the accuracy of fault information description in a feedback information mode, and dynamically increase or decrease knowledge (namely semantic nodes) in a sewage treatment knowledge base on the basis of a semantic tree by combining the actual situation of fault diagnosis of a sewage treatment process flow, thereby perfecting the sewage treatment knowledge base; and the system can also be interactively communicated with an online technical expert to obtain the help of the online technical expert so as to obtain the sewage treatment professional guidance as much as possible. In addition, fault information is continuously matched with knowledge in a sewage treatment knowledge base to form an inference chain, so that a proper and effective fault diagnosis result is obtained.
There is also provided, in accordance with an embodiment of the present invention, a computer-readable storage medium. The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to execute the sewage treatment failure diagnosis method as described above.
There is also provided, in accordance with an embodiment of the present invention, a computing device. The computing device includes a processor and a memory. The memory is for storing program instructions. The program instructions are executed by the processor to cause the processor to execute the computer program of the sewage treatment fault diagnosis method as described above.
Further, it should be understood that each unit in the sewage treatment malfunction diagnosis apparatus according to the exemplary embodiment of the present invention may be implemented as a hardware component and/or a software component. The individual units may be implemented, for example, using Field Programmable Gate Arrays (FPGAs) or Application Specific Integrated Circuits (ASICs), depending on the processing performed by the individual units as defined by the skilled person.
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims.

Claims (30)

1. A sewage treatment failure diagnosis method characterized by comprising:
receiving at least one fault information regarding sewage treatment;
transmitting the at least one fault information to an expert module, and receiving a fault diagnosis result corresponding to each fault information from the expert module; or,
calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information,
wherein the sewage treatment knowledge base is established based on at least one semantic tree corresponding to the sewage treatment process flow, one semantic tree corresponds to one treatment link in the sewage treatment process flow, each semantic tree comprises at least two semantic nodes in different layers and packaged by preset information of each flow of the sewage treatment process,
the step of calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information comprises the following steps:
and inquiring corresponding fault diagnosis results from the sewage treatment knowledge base based on the matching degree of each fault information and the semantic nodes in the sewage treatment knowledge base.
2. The sewage treatment fault diagnosis method of claim 1, wherein the step of transmitting the at least one fault information to an expert module comprises:
sending assistance request information to the expert module together with the at least one fault information.
3. The sewage treatment failure diagnosis method according to claim 2, further comprising:
a selection of an expert module is received,
wherein the transmitting of the at least one fault information to the expert module and the receiving of the fault diagnosis result corresponding to each fault information from the expert module includes:
sending assistance request information together with the at least one fault information to the selected expert module;
receiving a fault diagnosis result corresponding to each fault information from the selected expert module,
wherein the fault diagnosis result corresponding to each fault information received from the selected expert module is a fault diagnosis result determined based on experience or a fault diagnosis result determined by calling a sewage treatment knowledge base.
4. The sewage treatment fault diagnosis method according to claim 1 or 3, wherein the step of calling a sewage treatment knowledge base based on each fault information to determine a fault diagnosis result corresponding to each fault information comprises:
converting each fault information into a fault clause corresponding to each fault information;
identifying the matching degree of each fault clause and a semantic node in a sewage treatment knowledge base;
judging whether each matching degree reaches a preset threshold value;
and deleting the fault clause corresponding to the matching degree which does not reach the preset threshold value, and inquiring a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree which reaches the preset threshold value.
5. The sewage treatment failure diagnosis method according to claim 4, further comprising:
and responding to the deletion of the fault clause corresponding to the matching degree which does not reach the preset threshold value, and prompting to adjust the fault information corresponding to the deleted fault clause.
6. The wastewater treatment failure diagnostic method of claim 4, wherein the wastewater treatment knowledge base stores each semantic tree in the form of a relational database.
7. The wastewater treatment failure diagnostic method of claim 6, wherein each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and hierarchical relationship of each semantic node.
8. The sewage treatment failure diagnosis method according to claim 7, wherein the step of querying the corresponding failure diagnosis result from the sewage treatment knowledge base based on the failure clause corresponding to the matching degree reaching the predetermined threshold value comprises:
searching the number of layers of fault clauses corresponding to each matching degree reaching a preset threshold value on a semantic tree in a sewage treatment knowledge base, and sequencing and determining a fault sequence based on the searched number of layers;
determining a first semantic node corresponding to a first fault clause with the minimum layer number and a second semantic node corresponding to a second fault clause with the maximum layer number in the fault sequence;
when detecting that a third semantic node corresponding to a third fault clause in the fault sequence is a child node of the first semantic node, judging whether the third semantic node is the same as the second semantic node;
when the third semantic node is the same as the second semantic node, determining whether the third semantic node is a leaf node by detecting the out-degree and the in-degree of the third semantic node;
when the third semantic node is a leaf node, determining the node content of the third semantic node as a fault diagnosis result,
wherein the third faulting clause is the next faulting clause in the faulting sequence that is adjacent to the first faulting clause.
9. The sewage treatment failure diagnosis method according to claim 8,
and when the third semantic node is different from the second semantic node, updating the third fault clause into a next fault clause adjacent to the third fault clause, and updating the third semantic node into a semantic node corresponding to the updated third fault clause.
10. The sewage treatment failure diagnosis method according to claim 8,
determining the third semantic node as a semantic node corresponding to a fourth faulty clause when it is detected that the third semantic node is not a child node of the first semantic node, and then performing a step of determining whether the third semantic node is a leaf node,
wherein the fourth fault clause is a last fault clause adjacent to the third fault clause in the fault sequence.
11. The sewage treatment failure diagnosis method according to claim 10,
and when the third semantic node is not a leaf node, searching all leaf nodes under the third semantic node, and determining the node contents of all the searched leaf nodes as fault diagnosis results.
12. The sewage treatment failure diagnosis method according to claim 1,
the sewage treatment knowledge base executes a semantic tree growing process in response to a new node being input;
the wastewater treatment knowledge base performs a pruning process of the semantic tree in response to the predecessor nodes being deleted,
and deleting the original node when the score of the original node fails to reach a preset threshold value within a preset time period.
13. The sewage treatment failure diagnosis method according to claim 12, wherein the step of performing a semantic tree growing process comprises:
when the clauses contained in the child nodes of the root node or the internal node form a logic complementary relationship or are leaf nodes, adding the node containing the newly input clause and the clause forming the logic complementary relationship with the newly input clause between the original parent child nodes as a new node;
and when the child nodes of the root node or the internal node are one or the clauses contained in a plurality of child nodes do not form a logical complementary relationship, taking the node containing the newly input clause as a new child node of the root node or the internal node.
14. The wastewater treatment fault diagnosis method according to claim 12, wherein the step of performing a pruning process of the semantic tree comprises:
when the father node of the leaf node executing the pruning process has the brother node, deleting the leaf node, the father node and the brother node, and taking the child node of the brother node as the child node of the father node of the brother node;
when the parent node of the leaf node performing the pruning process has no sibling node, the leaf node and the parent node are deleted.
15. A sewage treatment failure diagnosis apparatus characterized by comprising:
a fault information receiving unit configured to receive at least one fault information regarding sewage treatment;
a fault information processing unit configured to transmit the at least one fault information to the expert module and receive a fault diagnosis result corresponding to each fault information from the expert module; or, based on each fault information, calling a sewage treatment knowledge base to determine a fault diagnosis result corresponding to each fault information,
wherein the sewage treatment knowledge base is established based on at least one semantic tree corresponding to the sewage treatment process flow, one semantic tree corresponds to one treatment link in the sewage treatment process flow, each semantic tree comprises at least two semantic nodes in different layers and packaged by preset information of each flow of the sewage treatment process,
the fault information processing unit queries corresponding fault diagnosis results from the sewage treatment knowledge base based on the matching degree of each fault information and the semantic nodes in the sewage treatment knowledge base.
16. The sewage treatment failure diagnosis device according to claim 15, wherein the failure information processing unit is further configured to:
sending assistance request information to the expert module together with the at least one fault information.
17. The sewage treatment failure diagnosis device according to claim 16, further comprising:
a selection receiving unit configured to receive a selection of the expert module,
wherein the fault information processing unit is further configured to:
sending assistance request information together with the at least one fault information to the selected expert module;
receiving a fault diagnosis result corresponding to each fault information from the selected expert module,
wherein the fault diagnosis result corresponding to each fault information received from the selected expert module is a fault diagnosis result determined based on experience or a fault diagnosis result determined by calling a sewage treatment knowledge base.
18. The sewage treatment failure diagnosis device according to claim 15 or 17, wherein the failure information processing unit is further configured to:
converting each fault information into a fault clause corresponding to each fault information;
identifying the matching degree of each fault clause and a semantic node in a sewage treatment knowledge base;
judging whether each matching degree reaches a preset threshold value;
and deleting the fault clause corresponding to the matching degree which does not reach the preset threshold value, and inquiring a corresponding fault diagnosis result from the sewage treatment knowledge base based on the fault clause corresponding to the matching degree which reaches the preset threshold value.
19. The sewage treatment failure diagnosis device according to claim 18, wherein the failure information processing unit is further configured to:
and responding to the deletion of the fault clause corresponding to the matching degree which does not reach the preset threshold value, and prompting to adjust the fault information corresponding to the deleted fault clause.
20. The wastewater treatment failure diagnostic apparatus of claim 18, wherein the wastewater treatment knowledge base stores each semantic tree in the form of a relational database.
21. The wastewater treatment failure diagnostic apparatus of claim 20, wherein each entry in the relational database corresponds to each semantic node, and the content recorded by each entry is a domain indicating the node content and hierarchical relationship of each semantic node.
22. The sewage treatment failure diagnosis device according to claim 21, wherein the failure information processing unit is further configured to:
searching the number of layers of fault clauses corresponding to each matching degree reaching a preset threshold value on a semantic tree in a sewage treatment knowledge base, and sequencing and determining a fault sequence based on the searched number of layers;
determining a first semantic node corresponding to a first fault clause with the minimum layer number and a second semantic node corresponding to a second fault clause with the maximum layer number in the fault sequence;
when detecting that a third semantic node corresponding to a third fault clause in the fault sequence is a child node of the first semantic node, judging whether the third semantic node is the same as the second semantic node;
when the third semantic node is the same as the second semantic node, determining whether the third semantic node is a leaf node by detecting the out-degree and the in-degree of the third semantic node;
when the third semantic node is a leaf node, determining the node content of the third semantic node as a fault diagnosis result,
wherein the third faulting clause is the next faulting clause in the faulting sequence that is adjacent to the first faulting clause.
23. The sewage treatment failure diagnosis device according to claim 22, wherein the failure information processing unit is further configured to:
and when the third semantic node is different from the second semantic node, updating the third fault clause into a next fault clause adjacent to the third fault clause, and updating the third semantic node into a semantic node corresponding to the updated third fault clause.
24. The sewage treatment failure diagnosis device according to claim 22, wherein the failure information processing unit is further configured to:
determining the third semantic node as a semantic node corresponding to a fourth faulty clause when it is detected that the third semantic node is not a child node of the first semantic node, and then performing a step of determining whether the third semantic node is a leaf node,
wherein the fourth fault clause is a last fault clause adjacent to the third fault clause in the fault sequence.
25. The sewage treatment failure diagnosis device according to claim 24, wherein the failure information processing unit is further configured to:
and when the third semantic node is not a leaf node, searching all leaf nodes under the third semantic node, and determining the node contents of all the searched leaf nodes as fault diagnosis results.
26. The sewage treatment failure diagnosis device according to claim 15,
the sewage treatment knowledge base executes a semantic tree growing process in response to a new node being input;
the wastewater treatment knowledge base performs a pruning process of the semantic tree in response to the predecessor nodes being deleted, wherein the pruning process is deleted when the predecessor nodes fail to reach a predetermined threshold within a predetermined time period.
27. The sewage treatment failure diagnosis device according to claim 26, wherein the failure information processing unit is further configured to:
when the clauses contained in the child nodes of the root node or the internal node form a logic complementary relationship or are leaf nodes, adding the node containing the newly input clause and the clause forming the logic complementary relationship with the newly input clause between the original parent child nodes as a new node;
and when the child nodes of the root node or the internal node are one or the clauses contained in a plurality of child nodes do not form a logical complementary relationship, taking the node containing the newly input clause as a new child node of the root node or the internal node.
28. The sewage treatment failure diagnosis device according to claim 26, wherein the failure information processing unit is further configured to:
when the father node of the leaf node executing the pruning process has the brother node, deleting the leaf node, the father node and the brother node, and taking the child node of the brother node as the child node of the father node of the brother node;
when the parent node of the leaf node performing the pruning process has no sibling node, the leaf node and the parent node are deleted.
29. A computer-readable storage medium characterized by storing a computer program which, when executed by a processor, causes the processor to execute the sewage treatment failure diagnosis method according to any one of claims 1 to 14.
30. A computing device, comprising:
a processor;
a memory for storing a computer program that, when executed by the processor, causes the processor to execute the sewage treatment failure diagnosis method according to any one of claims 1 to 14.
CN201810821381.8A 2018-07-24 2018-07-24 Sewage treatment fault diagnosis method and device Active CN108960673B (en)

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 CN108960673A (en) 2018-12-07
CN108960673B true 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)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114428746A (en) * 2022-02-14 2022-05-03 北京三快在线科技有限公司 Bad case identification method, device, equipment and storage medium
CN115048095B (en) * 2022-08-12 2023-01-03 广东粤港澳大湾区硬科技创新研究院 Expert system fault diagnosis program generation method

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH06222922A (en) * 1993-01-21 1994-08-12 Sankyo Seiki Mfg Co Ltd Expert system
US6946715B2 (en) * 2003-02-19 2005-09-20 Micron Technology, Inc. CMOS image sensor and method of fabrication
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
CN102819239B (en) * 2011-06-08 2014-09-17 同济大学 Intelligent fault diagnosis method of numerical control machine tool
CN102184569B (en) * 2011-06-11 2012-12-26 福州大学 Individual plant wood modeling method driven by domain ontology
CN102254102B (en) * 2011-07-18 2013-08-21 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
CN104035433B (en) * 2014-05-30 2017-01-11 北京必可测科技股份有限公司 Remote precise diagnosis and maintenance system and method of industrial devices
CN104376033B (en) * 2014-08-01 2017-10-24 中国人民解放军装甲兵工程学院 A kind of method for diagnosing faults 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

Also Published As

Publication number Publication date
CN108960673A (en) 2018-12-07

Similar Documents

Publication Publication Date Title
Poch et al. Designing and building real environmental decision support systems
Rabosky et al. Density-dependent diversification in North American wood warblers
Han et al. Prediction of activated sludge bulking based on a self-organizing RBF neural network
CN108960673B (en) Sewage treatment fault diagnosis method and device
Ferris et al. Indices developed specifically for analysis of nematode assemblages.
Dawyndt et al. Knowledge accumulation and resolution of data inconsistencies during the integration of microbial information sources
Madaeni et al. Fuzzy modeling and hybrid genetic algorithm optimization of virus removal from water using microfiltration membrane
CN102156406A (en) Expert system and method for preventing and controlling sludge bulking under the diagnosis based on operation condition of sewage disposal plant
US11365140B2 (en) Decision support system and method for water treatment
CN102063062B (en) Sludge bulking control expert system for diagnosis based on filamentous bacteria population structure
Burian et al. Predation increases multiple components of microbial diversity in activated sludge communities
KR20120001116A (en) A system and method for diagnosis algorithm development at wastewater treatment plant operation
Todorovski et al. Modelling and prediction of phytoplankton growth with equation discovery
Jiao et al. Eutrophication in subtropical lakes reinforces the dominance of balanced-variation component in temporal bacterioplankton community heterogeneity by lessening stochastic processes
CN115685899B (en) Biochemical pool operation parameter management system of municipal sewage treatment plant
Chan et al. A pretreatment method of wastewater based on artificial intelligence and fuzzy neural network system
JP2001334253A (en) Water quality simulator
Xie et al. Surface water quality evaluation based on Bayesian network
Comas et al. Extension of the IWA/COST simulation benchmark to include expert reasoning for system performance evaluation
Bhattacharjee et al. A Type‐2 Fuzzy Logic Approach for Forecasting of Effluent Quality Parameters of Wastewater Treatment
Vlad et al. MATHEMATICAL MODELLING OF A BIOLOGICAL WASTEWATER TREATMENT PROCESS. CASE STUDY: THE WASTEWATER TREATMENT STATION OF ROMANOFIR S. A. TRADING CO.-TALMACIU
Babuška et al. Fuzzy modeling of enzymatic Penicillin-G conversion
Roda et al. Development of a case-based system for the supervision of an activated sludge process
Van Valen Why misunderstand the evolutionary half of biology?
CN110245804B (en) Method and device for screening wastewater pollutants in area and monomer optimal control characteristics of area

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
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20240823

Address after: Room 103, No. 780, Tanghekou Street, Tanghekou Town, Huairou District, Beijing

Patentee after: DASMART ENVIRONMENTAL TECHNOLOGIES (BEIJING) Co.,Ltd.

Country or region after: China

Address before: 100176 No. 8 Boxing Road, Daxing District, Beijing Economic and Technological Development Zone

Patentee before: BEIJING ETECHWIN ELECTRIC Co.,Ltd.

Country or region before: China