CN102156799A - Cascadable complex event processing engine and train overhauling automatic recording method - Google Patents
Cascadable complex event processing engine and train overhauling automatic recording method Download PDFInfo
- Publication number
- CN102156799A CN102156799A CN2011100096195A CN201110009619A CN102156799A CN 102156799 A CN102156799 A CN 102156799A CN 2011100096195 A CN2011100096195 A CN 2011100096195A CN 201110009619 A CN201110009619 A CN 201110009619A CN 102156799 A CN102156799 A CN 102156799A
- Authority
- CN
- China
- Prior art keywords
- event
- incident
- complicated
- train
- processing engine
- 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.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a cascadable complex event processing engine and a train overhauling automatic recording method, wherein the complex event processing engine can receive a primitive event of a bottom layer, generate a corresponding query plan for a query rule submitted by a user, detect the primitive event according to the query plan, and report the generated complex event to upper application; the engine comprises an event pretreatment module, an event compression module, an event dispatching module and an event detection module. The train overhauling automatic recording method based on the cascadable complex event processing engine provided by the invention overcomes the laborious and time-consuming shortage and the easy error shortage caused by the conventional artificial recording method, the various statuses of the train can be automatically detected in the case of nobody on duty, the recording method is favorable for timely repairing the train, the time and labor are saved, the mistake is difficult to be made, and the application and configuration process are simple and easy.
Description
Technical field
The invention belongs to complicated event and handle (CEP) technical field, but especially a kind of complicated event processing engine of cascade and train overhaul automatic record method, but the core of this train overhaul automatic record method is the complicated event processing engine of cascade, for railway system's train robotization maintenance process is offered help, can be under unattended situation, automatically record train overhaul state is raised the efficiency.
Background technology
It is a kind of based on flow of event that complicated event is handled (CEP) technology, the emerging technology that is used for management information and structure infosystem, extensive, the real time data that different system is produced regarded as dissimilar and from the incident in different event source, therefrom find complicated event schema, help enterprise in real time environment, to discern threat, find chance, enterprise can be made a strategic decision immediately, realize the automatic intelligent operation.The inner mass incident that produces of different system often forms flow of event, certain situation has taken place in representations of events sometime, independent treat each incident in the stream, can not obtain professional value, yet under the specific context situation, pay close attention to a plurality of incidents simultaneously, can obtain significant complicated event.Such as, sequential relationship between the tense associative expression incident, the track of space correlation expression incident development, the change procedure of the incident that the space-time co expression is relevant with object, the complicated event treatment technology is exactly the association that utilizes between the event attribute, constantly filter the flow of event that arrives continuously by matched rule or algebraic manipulation, find complicated event.Flow of event is handled the real-time characteristic of concern incident and the sequential relationship between incident, traditional relational database is stored the disposal route of afterwards inquiring about earlier can not handle flow data effectively, form the bottleneck of application system deal with data, therefore the complicated event treatment technology based on stream receives much concern.The complicated event treatment technology is applicable to the business scenario that many real time datas need be flocked together and respond, and the most important aspect that relates to is: the 1. definition of the description of compound event pattern and observed incident, identification during for detection.2. the recognition mechanism of incidence relation between incident, and the generation of definite complicated event on this basis.
It is rapid, of many uses that complicated event is handled (CEP) technical development, can be used for multiple fields such as traffic monitoring, stock market, logistics net.But current complicated event treatment technology is not applied in train overhaul to be gone out in the warehouse-in system, does not also form unified ripe technical standard.At present, the record train overhaul go out the mode that the warehouse-in state mainly adopts manual record, by the staff to train go out the warehouse-in state, inspecting state carries out record, time and effort consuming is made mistakes easily.
Summary of the invention
The objective of the invention is to overcome the time and effort consuming that traditional manual record mode brought, the shortcoming of makeing mistakes easily, but a kind of complicated event processing engine and train overhaul automatic record method of cascade are provided, this method can be formed a compound complicated event processing engine with the form of cascade with a plurality of basic complicated event processing engine, receive different event sources, bring into play the processing power of each engine to greatest extent at certain specific detection rule, can in distributed environment, bring into play simultaneously the effect that complicated event is handled, improve efficiency of service.
The objective of the invention is to solve by the following technical programs:
But the complicated event processing engine of this cascade can receive the bottom primitive event, and the rule searching that the user submitted to is generated corresponding inquiry plan, according to inquiry plan primitive event is detected, and the complicated event that generates is reported to upper layer application; This engine comprises incident pretreatment module, incident compression module, event scheduling module and event checking module;
The effect of described incident pretreatment module is to remove invalid event, and level and smooth and filter event is to eliminate the false transitions of sequence of events.By using for reference tradition based on the ESP data pre-service mechanism of SQL query model, native system adopt the pipe design pattern successively to incident gather, smoothly, filter, merge, arbitration and format.
Described incident compression module adopts lossless compression algorithm by the analysis to the characteristics of flow of event own, by setting up a string table, represents that with short code long character string realizes compression.By incident is compressed the data volume that has reduced in Network Transmission, bandwidth and propagation delay have been saved.
Described event scheduling module adopts the dynamic event priority policy and based on the preemptive type event scheduling strategy of priority, is that primitive event distributes priority according to the Event Priority model, and dynamically adjusts the priority of incident;
Described event checking module is integrated OPRAH, three kinds of different detection models of leaf and Esper; Described OPRAH is based on the method for finite-state automata and comes the detection of complex incident; The algorithm that described leaf is based on tree comes the detection of complex incident; Three kinds of different detection models of integrated this can use separately or carry out cascade according to mode arbitrarily.
The present invention proposes a kind of train overhaul automatic record method based on above complicated event processing engine, may further comprise the steps:
1) but start the complicated event processing engine of cascade;
2) select secondary cascade mode by the user, to event checking module three integrated detection models be configured, comprise that configuration receives the port of flow of event, send the address of complicated event and detect rule;
Described secondary cascade mode is: detected the train warehouse-in and overhauled two complicated events by Esper, by OPRAH overhauling train outbound incident, leaf is as the second level in the cascade engine, receiving Esper and the detected complicated event of OPRAH flows as incoming event, detect final complicated event-train overhaul and finish, give the upper strata with event report then;
3) after configuration is finished, when but the flow of event of contemporary tabular car current state arrives the complicated event processing engine of cascade, but the complicated event processing engine of cascade is according to step 2) in configuration, handle the flow of event that is received, complicated event so that detection train warehouse-in, outbound, maintenance neutralization maintenance are finished if engine successfully detects desired complicated event, then continues step 4), if do not detect, then continue incoming event stream is detected;
4) be that train overhaul is finished incident output with detected complicated event, thereby finish the record of train overhaul.
The collection of the flow of event of above-mentioned representative train current state is to realize by wireless radio-frequency.
Above-mentioned wireless radio-frequency is to be provided with the RFID label ON TRAINS, is provided with RFID label read-write equipment at maintenance station, and described RFID label read-write equipment sends the data that collect to the complicated event processing engine.
The present invention has following beneficial effect:
But since the complicated event processing engine of cascade of the present invention integrated OPRAH, three kinds of different detection models of leaf and Esper, and the characteristics of these three kinds of different detection models and performance are different, the user can detect according to detecting the different detection model of regular type selecting or the combination of detection model, has effectively improved detection efficiency and dirigibility.
But the train overhaul automatic record method of the complicated event processing engine based on cascade of the present invention has overcome the shortcoming that wastes time and energy, makes mistakes easily that traditional manual record method is brought, can under unattended situation, detect the various states of train automatically, help in time train being carried out breakdown maintenance, time saving and energy saving, be difficult for makeing mistakes, use and layoutprocedure are simple and easy to use.
Description of drawings
But Fig. 1 is the framework of the complicated event processing engine of cascade;
But Fig. 2 is the structural drawing of the complicated event processing engine of cascade;
But Fig. 3 be cascade the complicated event processing engine integrated three kinds of detection models;
Fig. 4 is a workflow diagram of the present invention;
Fig. 5 is the topology diagram that detection model carries out cascade among the present invention.
Embodiment
The invention will be further described below in conjunction with accompanying drawing.
But extensive, real time data that the complicated event processing engine of cascade produces different system are regarded as dissimilar and from the incident in different event source, therefrom find complicated event schema, help enterprise in real time environment, to discern threat, find chance, enterprise can be made a strategic decision immediately, realize the automatic intelligent operation.It can receive the bottom primitive event of magnanimity, and the rule searching that the user submitted to is generated corresponding inquiry plan, according to inquiry plan primitive event is detected, and the complicated event that generates is reported to upper layer application.This engine comprises incident pretreatment module, incident compression module, event scheduling module and event checking module, as Fig. 1.
Wherein, the various dissimilar primitive event circulation that engine is received of incident pretreatment module is changed to the treatable unified type of engine. and its effect is to remove invalid event, and level and smooth and filter event is to eliminate the false transitions of sequence of events.By using for reference tradition based on the data pre-service mechanism of SQL query model, native system adopt the pipe design pattern successively to incident gather, smoothly, filter, merge, arbitration and format.
The incident compression module adopts lossless compression algorithm by the analysis to the characteristics of flow of event own, by setting up a string table, represents that with short code long character string realizes compression.By incident is compressed the data volume that has reduced in Network Transmission, bandwidth and propagation delay have been saved.The incident compression module is responsible for the incident that receives is carried out decompress(ion), and the complicated event of engine output is compressed.
The event scheduling module adopts the dynamic event priority policy and based on the preemptive type event scheduling strategy of priority, is that primitive event distributes priority according to the Event Priority model, and dynamically adjusts the priority of incident.Two factors of power that this engine utilizes the stand-by period of primitive event and gathers the rfid interrogator signal of primitive event are incident dynamic assignment priority.The time that primitive event is waited for is long more, and priority is just high more, and the time that incident is waited for is to deduct the time that incident produces the current time.The rfid interrogator signal of gathering primitive event is strong more, and correspondingly, the priority of primitive event is just high more.
Mutual relationship between incident pretreatment module, incident compression module, event checking module, the event scheduling module as shown in Figure 2.The incoming event stream that system received can be the incident of JavaBean type, XML type or Map type.By the incident compression module primitive event stream is carried out decompress(ion), send the incident pretreatment module then to.By the incident pretreatment module dissimilar primitive events is carried out pre-service, and send the event scheduling module to after converting the incident of unified type to.The event scheduling module passes to event checking module after distributing priority according to models of priority to incident., according to user-defined rule searching incoming event stream is inquired about by event checking module, with the detection of complex incident.After by the incident compression module complicated event that produces being compressed then, output on the address of user's appointment.
Event checking module is integrated OPRAH (Aurora), three kinds of different detection models of leaf (Leafage) and Esper are as Fig. 3.The algorithm difference that these three kinds of detection models are realized, performance and using method and detection language are different.
Wherein the Aurora method that is based on finite-state automata is come the detection of complex incident, and it becomes finite-state automata with user-defined detection rule parsing, utilizes the migration of state to come the detection of complex incident, is fit to detect the sequence incident.Simultaneously, it has defined oneself a detection language, uses simple.Aurora uses the time of origin of incident to detect, so it can handle out of order incident.
The algorithm that Leafage is based on tree comes the detection of complex incident, it is with the form of user-defined detection rule parsing Cheng Shu, primitive event is the leaf of tree, complicated event is a tree root, begin primitive event is mated from leaf node, successively report, just explanation has detected complicated event when matching root node.It uses XML to define the detection rule of oneself, for the user, as long as be familiar with XML, just can very easily write out the detection rule of oneself.Leafage uses the time of origin of incident to detect equally, so it also can handle out of order incident.
Esper is a third-party complicated event detection model of increasing income, and is encapsulated in equally in our the complicated event processing engine.Esper has defined the detection language of oneself, is called event handling language (EPL), and EPL is a kind of type of structured query language (SQL) language, and is powerful.Esper is to use time of receiving primitive event to detect, so it can not handle out of order incident.
This complicated event processing engine integrated these three kinds of different detection models, not only can use separately, and can carry out cascade in mode arbitrarily, a complicated event that detection model detected is submitted to next detection model detects further as its primitive event.By the application programming interfaces that this engine provided, the user can realize the cascade of detection model easily.
A kind of train overhaul automatic record method based on above complicated event processing engine that the present invention proposes may further comprise the steps, as Fig. 4:
1) but start the complicated event processing engine of cascade;
2) select secondary cascade mode by the user, to event checking module three integrated detection models be configured, comprise that configuration receives the port of flow of event, send the address of complicated event and detect rule;
Described secondary cascade mode is: detected the train warehouse-in and overhauled two complicated events by Esper, by OPRAH overhauling train outbound incident, leaf is as the second level in the cascade engine, receiving Esper and the detected complicated event of OPRAH flows as incoming event, detect final complicated event-train overhaul and finish, give the upper strata with event report then;
3) enter maintenance station with the train of RFID label after, by the RFID label reader that is provided with in the maintenance station position of label is gathered, but the data that collect is sent to the complicated event processing engine of cascade.This engine is according to step 2) in configuration, handle the flow of event that is received, complicated event such as finish to detect train warehouse-in, outbound, maintenance neutralization maintenance, if engine successfully detects desired complicated event, then continue step 4), if do not detect, then continue incoming event stream is detected;
4) be that train overhaul is finished incident output with detected complicated event, thereby finish the record of train overhaul.
Cascade system in the workflow of the present invention between the basic detection model of step 2 configuration as shown in Figure 5.In order to realize this cascade system, wherein need having in this complicated event processing engine configuration:
1) at first, for this engine integrated detection model Esper, need configuration:
A) port of reception event source is used for receiving the primitive event stream that train warehouse-in process is caught by rfid interrogator;
B) port 2 of reception event source is used for receiving the primitive event stream that the train overhaul process is caught by rfid interrogator;
C) address of transmission complicated event when Esper detects desired complicated event, sends to the complicated event that produces the address of appointment.According to the cascade system described in Fig. 5, this address should be the address that leaf receives event source;
D) rule of detection of complex incident comprises rule that detects the warehouse-in incident and the rule that detects incident in the maintenance.
2) but for the complicated event processing engine of cascade integrated detection model OPRAH, need configuration:
A) port of reception event source is used for receiving the primitive event stream that train outbound process is caught by rfid interrogator;
B) address of transmission complicated event when OPRAH detects desired complicated event, sends to the complicated event that produces the address of appointment.According to the cascade system described in Fig. 5, this address should be the address that leaf receives event source;
C) detect rule, refer to the rule that detects train outbound incident here.
3) but for the complicated event processing engine of cascade integrated detection model leaf, need configuration:
A) port of reception event source is used to receive the detected complicated event of Esper and OPRAH as event source, and wherein, in order to realize cascade, the address that Esper and OPRAH send complicated event should be exactly the address at leaf place and the port of monitoring;
B) address of transmission complicated event detects final complicated event at leaf--and when incident is finished in maintenance, this complicated event is sent to this address, thereby report to upper layer application.
C) detect rule, refer to here and detect the rule that incident is finished in maintenance.
Claims (5)
1. but the complicated event processing engine of a cascade, it is characterized in that, described complicated event processing engine can receive the bottom primitive event, the rule searching that the user submitted to is generated corresponding inquiry plan, according to inquiry plan primitive event is detected, the complicated event that generates is reported to upper layer application; This engine comprises incident pretreatment module, incident compression module, event scheduling module and event checking module;
The various dissimilar primitive event circulation that engine is received of described incident pretreatment module is changed to the treatable unified type of engine;
Described incident compression module adopts lossless compression algorithm, by setting up a string table, represents that with short code long character string realizes the compression to flow of event;
Described event scheduling module adopts the dynamic event priority policy and based on the preemptive type event scheduling strategy of priority, is that primitive event distributes priority according to the Event Priority model, and dynamically adjusts the priority of incident;
Described event checking module is integrated OPRAH, three kinds of different detection models of leaf and Esper; Described OPRAH is based on the method for finite-state automata and comes the detection of complex incident; The algorithm that described leaf is based on tree comes the detection of complex incident; Three kinds of different detection models of integrated this can use separately or carry out cascade according to mode arbitrarily.
2. but the complicated event processing engine of cascade according to claim 1 is characterized in that, described incident pretreatment module also adopt the pipe design pattern successively to incident gather, smoothly, filter, merge, arbitration and format.
3. but the train overhaul automatic record method based on the complicated event processing engine of claim 1 cascade is characterized in that, may further comprise the steps:
1) but start the complicated event processing engine of cascade;
2) select secondary cascade mode by the user, to event checking module three integrated detection models be configured, comprise that configuration receives the port of flow of event, send the address of complicated event and detect rule;
Described secondary cascade mode is: detected the train warehouse-in and overhauled two complicated events by Esper, by OPRAH overhauling train outbound incident, leaf is as the second level in the cascade engine, receiving Esper and the detected complicated event of OPRAH flows as incoming event, detect final complicated event-train overhaul and finish, give the upper strata with event report then;
3) after configuration is finished, when but the flow of event of contemporary tabular car current state arrives the complicated event processing engine of cascade, but the complicated event processing engine of cascade is according to step 2) in configuration, handle the flow of event that is received, complicated event so that detection train warehouse-in, outbound, maintenance neutralization maintenance are finished if engine successfully detects desired complicated event, then continues step 4), if do not detect, then continue incoming event stream is detected;
4) be that train overhaul is finished incident output with detected complicated event, thereby finish the record of train overhaul.
4. train overhaul automatic record method according to claim 3 is characterized in that, the collection of the flow of event of described representative train current state is to realize by wireless radio-frequency.
5. train overhaul automatic record method according to claim 4, it is characterized in that, described wireless radio-frequency is to be provided with the RFID label ON TRAINS, is provided with RFID label read-write equipment at maintenance station, and described RFID label read-write equipment sends the data that collect to the complicated event processing engine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100096195A CN102156799A (en) | 2011-01-17 | 2011-01-17 | Cascadable complex event processing engine and train overhauling automatic recording method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100096195A CN102156799A (en) | 2011-01-17 | 2011-01-17 | Cascadable complex event processing engine and train overhauling automatic recording method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102156799A true CN102156799A (en) | 2011-08-17 |
Family
ID=44438293
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100096195A Pending CN102156799A (en) | 2011-01-17 | 2011-01-17 | Cascadable complex event processing engine and train overhauling automatic recording method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102156799A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567803A (en) * | 2011-12-27 | 2012-07-11 | 中国科学院自动化研究所 | Complex event scheduling system and method based on priority-assigned event graph |
| CN102571424A (en) * | 2011-12-29 | 2012-07-11 | 华为技术服务有限公司 | Processing method, device and system for engineering event |
| CN102594871A (en) * | 2011-12-31 | 2012-07-18 | 山东中创软件工程股份有限公司 | Data packaging method and system, adapter and Internet of things middleware |
| CN103455611A (en) * | 2013-09-06 | 2013-12-18 | 北京邮电大学 | Data processing method for RFID (Radio Frequency Identification) middleware |
| CN105653734A (en) * | 2016-02-26 | 2016-06-08 | 国网江苏省电力公司电力科学研究院 | Uncertain complex event processing system and method |
| CN106250471A (en) * | 2016-07-29 | 2016-12-21 | 东北大学 | A kind of data for train ATP automatically extract and store system and method |
| CN106600446A (en) * | 2016-11-07 | 2017-04-26 | 中国科学院自动化研究所 | Dynamic-priority-based real-time processing system and method of product field event |
-
2011
- 2011-01-17 CN CN2011100096195A patent/CN102156799A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567803A (en) * | 2011-12-27 | 2012-07-11 | 中国科学院自动化研究所 | Complex event scheduling system and method based on priority-assigned event graph |
| CN102567803B (en) * | 2011-12-27 | 2015-04-15 | 中国科学院自动化研究所 | Complex event scheduling system and method based on priority-assigned event graph |
| CN102571424A (en) * | 2011-12-29 | 2012-07-11 | 华为技术服务有限公司 | Processing method, device and system for engineering event |
| CN102594871A (en) * | 2011-12-31 | 2012-07-18 | 山东中创软件工程股份有限公司 | Data packaging method and system, adapter and Internet of things middleware |
| CN103455611A (en) * | 2013-09-06 | 2013-12-18 | 北京邮电大学 | Data processing method for RFID (Radio Frequency Identification) middleware |
| CN105653734A (en) * | 2016-02-26 | 2016-06-08 | 国网江苏省电力公司电力科学研究院 | Uncertain complex event processing system and method |
| CN105653734B (en) * | 2016-02-26 | 2019-03-19 | 国网江苏省电力公司电力科学研究院 | A kind of processing system and method for uncertainty complicated event |
| CN106250471A (en) * | 2016-07-29 | 2016-12-21 | 东北大学 | A kind of data for train ATP automatically extract and store system and method |
| CN106600446A (en) * | 2016-11-07 | 2017-04-26 | 中国科学院自动化研究所 | Dynamic-priority-based real-time processing system and method of product field event |
| CN106600446B (en) * | 2016-11-07 | 2021-04-02 | 中国科学院自动化研究所 | Production field event real-time processing system and method based on dynamic priority |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102156799A (en) | Cascadable complex event processing engine and train overhauling automatic recording method | |
| CN107886238B (en) | Business process management system and method based on mass data analysis | |
| US20210192389A1 (en) | Method for ai optimization data governance | |
| CN107038162A (en) | Real time data querying method and system based on database journal | |
| CN110008201A (en) | A kind of quality of data towards big data checks monitoring method | |
| CN110390596A (en) | A kind of financial futures trading system | |
| CN106777703A (en) | A kind of bus passenger real-time analyzer and its construction method | |
| CN105913656B (en) | Based on the frequent method and system for crossing vehicle of distributed statistics | |
| CN102509001B (en) | Method for automatically removing time sequence data outlier point | |
| CN104899314A (en) | Pedigree analysis method and device of data warehouse | |
| CN107247799A (en) | Data processing method, system and its modeling method of compatible a variety of big data storages | |
| CN112181955B (en) | Data standard management method for information sharing of heavy haul railway comprehensive big data platform | |
| CN102902752A (en) | Method and system for monitoring log | |
| CN107103064A (en) | Data statistical approach and device | |
| CN112182077B (en) | Intelligent operation and maintenance system based on data middling platform technology | |
| CN113111140A (en) | Method for rapidly analyzing multi-source marine business observation data | |
| CN112905685A (en) | Framework management and control system and equipment for informatization construction | |
| CN110427298A (en) | A kind of Automatic Feature Extraction method of distributed information log | |
| CN109815254A (en) | Cross-region method for scheduling task and system based on big data | |
| CN111680027A (en) | Method and system for realizing intelligent cloud management based on knowledge drive | |
| CN109308290A (en) | A kind of efficient data cleaning conversion method based on CIM | |
| CN114297290A (en) | Electric power marketing data sharing application platform | |
| CN109033330A (en) | Big data cleaning method, device and server | |
| CN114757448B (en) | Manufacturing inter-link optimal value chain construction method based on data space model | |
| CN107729410A (en) | A kind of database engine suitable for power equipment big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110817 |