CN101958802A - Double-network online switching method of sampling values - Google Patents

Double-network online switching method of sampling values Download PDF

Info

Publication number
CN101958802A
CN101958802A CN201010273732XA CN201010273732A CN101958802A CN 101958802 A CN101958802 A CN 101958802A CN 201010273732X A CN201010273732X A CN 201010273732XA CN 201010273732 A CN201010273732 A CN 201010273732A CN 101958802 A CN101958802 A CN 101958802A
Authority
CN
China
Prior art keywords
data
frame
sampled value
network
information
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
Application number
CN201010273732XA
Other languages
Chinese (zh)
Inventor
杨玉昆
余文辉
何锦雄
陈宏辉
胡铁斌
张学强
戴丽娜
周俊涛
李广华
文继锋
李力
冯亚东
陈松林
李九虎
高新华
陈炯聪
杨颖安
马凯
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.)
NR Electric Co Ltd
Electric Power Research Institute of Guangdong Power Grid Co Ltd
Maoming Power Supply Bureau of Guangdong Power Grid Co Ltd
Original Assignee
NR Electric Co Ltd
Electric Power Research Institute of Guangdong Power Grid Co Ltd
Maoming Power Supply Bureau of Guangdong Power Grid 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 NR Electric Co Ltd, Electric Power Research Institute of Guangdong Power Grid Co Ltd, Maoming Power Supply Bureau of Guangdong Power Grid Co Ltd filed Critical NR Electric Co Ltd
Priority to CN201010273732XA priority Critical patent/CN101958802A/en
Publication of CN101958802A publication Critical patent/CN101958802A/en
Pending legal-status Critical Current

Links

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention relates to a double-network online switching method of sampling values, comprising a double-network redundant data fast recognition algorithm and a double-network data online switching method. The double-network redundant data fast recognition algorithm achieves the purpose of fast recognition through caching some state information of the received data, and can fast and effectively recognize a double-network redundant message through utilizing the firstly arrived network transmission characteristic and the continuous characteristic of a sampling counter, analyzing the network source of the message, slightly decoding the redundant message and simply comparing. The double-network data online switching method updates a reception buffering area as soon as possible via the sampling value content which firstly arrives at a reception device, realizes the frame loss compensation via another network data which arrives more slowly when the sampling value data which arrives faster loses frame, and guarantees the correct reception of the sampling value when a single network has fault or the double networks have faults alternately. According to the invention, the reliable transmission of the sampling value data is guaranteed when any network has fault.

Description

The online changing method of the two nets of sampled value
Technical field
The present invention relates to the method for the online switching of the two net of digital transformer substation sampled value, especially with two methods of netting online switching based on the two net sampled data quick identification of sampled value network.
Background technology
In automation of transformation substations field, the development of intelligent electric equipment, particularly the maturation of the appearance of intelligent switch, photoelectric type instrument transformer electromechanical integration equipment and Fast Ethernet mechanics of communication is used, and automation of transformation substations technology has entered the new stage of digital applications.The key foundation of digital transformer substation communication is an IEC61850 communication protocol, and this agreement is divided into station level, wall, process layer with transformer station, stipulates to communicate by letter with the inner Fast Ethernet that adopts of layer between each layer.Compare with the integrated automation converting station of routine, it is the digitlization of process layer that digital transformer substation changes the best part: the cable of having cancelled between primary equipment such as traditional instrument transformer, switch and the secondary device connects, and the optical fiber that has progressively become between the equipment such as photoelectric type instrument transformer, intelligence interface, merge cells connects.
IEC61850-9-2 is the part of IEC61850 sampled value transmission standard, this part of standards has realized transmitting between process layer and the wall particular communication service mapping of sampled value fully according to the sampled value data model and the relevant ACSI service definition thereof of IEC61850 regulation.It is an abstract model based on the hybrid protocol stack: for the visit of the attribute of the sampled value controll block of transmission sampled value setting is the MMS mapping of adopting the IEC61850-8-1 definition; For directly visit ISO/IEC 8802-3 link maps is adopted in the transmission of sampled value.
Whole services in the sampled value model ACSI service that the IEC61850-9-2 stipulations are supported to define among the IEC61850-7-2.Simultaneously, its broad sense is also supported the whole relevant ACSI service of data server in communication, association, LD, LN, DATA and DataSet.The DataSet model of IEC61850-9-2 can flexible configuration, can be according to requirement of engineering and the compound mode of tissues sampled channel data flexibly, be convenient to realize stride the transmission of the sampled value of secondary function at interval.The IEC61850-9-2 stipulations have been stipulated two types sampled value controll block: MSVCB (multiplex broadcasting sampled value controll block class) and USVCB (single channel transmission sampled value class).USVCB realizes the sampled value transmission in point-to-point mode.MSVCB adopts publisher/subscriber's model, realizes the sampled value transmission with the Ethernet networking model; This transmission mode can effectively be supported the reception of multi-compartment sampled value, and maximization realization data sharing, has obtained more use in actual engineering.
Transformer substation case (GOOSE) towards general object is the mechanism that is used to satisfy electric substation automation system rapid message demand in IEC 61850 standards.Transformer station's configuration computer supervisory control system, bear that the operations staff normally controls, the function of each side such as supervision, signal, measurement and data statistic analysis, supervisory control system adopts the IEC61850 communication standard, utilize the Fast Ethernet characteristic, realize information exchange and supervision interval connection blocking function between the protection by GOOSE (object-oriented transformer station general purpose event), with protection system unified Modeling, unified networking, share unified information platform, improve fail safe, the reliability of electrical secondary system; The stipulations conversion equipment has been saved in the application of IEC61850, has cancelled the preposition intermediate communication link that waits, and reduces operation, maintenance, maintenance workload, and the secondary device of saving repetition is to reach cost-effective purpose.
Sustainable development and constantly engineering practice along with the digital transformer substation technology, sampled value networking transmission has become the technique direction of follow-up developments, this also is the objective requirement of data sharing, how improving the reliability of sampled value networking transmission, is one of most important technological difficulties during digital transformer substation is implemented.
Transmitting the particular communication service mapping of sampled value between two network process layers of digital transformer substation SV (sampled value) and the wall, is an abstract model based on the hybrid protocol stack: for the visit of the attribute of the sampled value controll block of transmission sampled value setting is the MMS mapping of adopting the IEC61850-8-1 definition; For directly visit ISO/IEC 8802-3 link maps is adopted in the transmission of sampled value.
Summary of the invention
The objective of the invention is: in the IEC61850-9-2 of double-network redundant sampled value transmission plan, provide a kind of based on the online handover scheme of two nets of two netting indexs according to quick identification, guarantee the reliable transmission of arbitrary network failure situation down-sampled values data, improve the sampled value treatment effeciency simultaneously.
The object of the invention also is to propose the sampled value transmission means based on the online switching of two net sampled data quick identification, arbitrary network instantaneity occurs or permanent fault does not all influence the correct transmission of sampled value in two nets, and improves the ability of anti-switch network frame losing to greatest extent; Simultaneously in order to improve the sampled value disposal ability of correlation reception equipment, sampled value double-network redundant data Fast Recognition Algorithm is proposed targetedly, whether Frame is carried out under the situation of complete decoding not needing, distinguishing is the redundancy sampling data, improves the sampled value treatment effeciency.
Technical scheme of the present invention is: the transmission method of the online switching of the two net of sampled value, and the online changing method of the two nets of this sampled value is made up of according to online changing method double-network redundant data Fast Recognition Algorithm and two netting index:
Double-network redundant data Fast Recognition Algorithm:
Two netting indexs reach the purpose of quick identification according to several state informations of Fast Recognition Algorithm by buffer memory reception data; Wherein each state information all comprises following content:
1.bValid:True represent other members initialization of this state information, Flase represents no initializtion;
2.NetSource: the network information of record sampled value Data Source, A net or B net;
3.smpCnt: the sample counter information of record sampled value data.
4.smpSyn: record sampled value Synchronization Status Message.
Maximum network postpones poorly between the number of buffer status information and two net, and sampling time interval is relevant; Minimal cache state information number=(maximum network postpones poor+sampling time interval-1)/sampling time interval;
As the current valid frame information of Status1 state recording; Status2 writes down the last valid frame information of current valid frame, and its pair net fast-speed redundancy recognizer computational process is as follows:
(1) device initial information: Status1.bValid=Status2.bValid=false;
(2) do not receive the sampled value message continuously: enter " the online switch logic of two nets "; Status1.bValid=Status2.bValid=false is set subsequently; Current valid frame record no initializtion is set in expression;
(3) no initializtion Status1: resolve this frame data, enter " the online switch logic of two nets " and further handle the sampled value data, upgrade Status1 subsequently
Status1.b?Valid=ture
Status1.NetSource=curNet
Expression record sampled value source, sample counter information, Synchronization Status Message;
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
(4) the sampled value source is identical with Status1.NetSourse: this frame data Non-redundant data, resolve this frame data, and enter
" the online switch logic of two nets " further handles the sampled value data; Upgrade Status1, Status2 subsequently
Status1=Status2
Status1.NetSource=curNet
Record sampled value source, sample counter information, the synchronous regime letter of representing current valid frame
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
Breath;
(5) the sampled value source is different with Status1.NetSourse: expression is different with the record sampled value of current valid frame, resolves the sample counter information of this frame;
(6) Status1.smpCnt==curSmpCnt: the record sampled value counter information of representing current valid frame is identical, and redundant data abandons and withdraws from;
(7) (Status2.bValid==ture) ﹠amp; ﹠amp; (Status2.smpCnt==curSmpCnt): the record sampled value initialization of representing current valid frame is identical, and redundant data abandons and withdraws from;
(8) resolve this frame data, enter " the online switch logic of two nets " and further handle the sampled value data;
Double-network redundant data Fast Recognition Algorithm has been utilized the first continuous characteristic of Network Transmission characteristic, sample counter that arrives earlier that sends, by analyzing the network originating of message, by a small amount of decoding to redundant message, and simple compare operation, just can effectively discern the double-network redundant message fast;
Two netting indexs, prevent because network delay, the network delay difference of the A/B net that a variety of causes such as Route Selection, network congestion cause are shaken the harmful effect that two netting indexs are caused according to identification by the number of buffer status information is set according to Fast Recognition Algorithm;
Two netting indexs are according to online changing method: upgrade the reception buffering area as early as possible with the sampled value content that arrives receiving system at first; Occur under the frame loss condition in the very fast sampled value data that arrive, realize the frame losing compensation with another network data that arrives more slowly; It is correct to guarantee that when single net fault or two net replace fault sampled value receives;
Sampled value is new criterion more: according to increasing progressively continuously under identical synchronous regime with the sampled value counter, in the characteristic of to a certain degree reversing, judge whether the sampled value counter upgrades.
Under identical synchronous regime, consider that of short duration frame loss condition may take place in the sampled value reception, its sampled value more new criterion is as follows:
1)LastSmpCnt>(MaxSmpCnt-MaxLostNum):
During CurSmpCnt>LastSmpCnt||CurSmpCnt<((LastSmpCnt+MaxLostNum) %MaxSmpCnt), think that CurSmpCnt upgrades; Otherwise, think the generation saltus step;
2) CurSmpCnt>LastSmpCnt﹠amp; ﹠amp; During CurSmpCnt<(LastSmpCnt+MaxLostNum), think that sampled value upgrades; Otherwise, think the generation saltus step;
MaxSmpCnt is illustrated in the numerical value number of the sample counter under the fixed sampling frequency, is MaxLostNum receiving the sampled value counter number of transitions that its maximum possible takes place in the data break continuously;
Compare by data and current data to last frame, two nets of sampled value switch the judgement that realizes the sampled values renewal; Sampled value is ageing data simultaneously, will be used to after the buffering that these data only can be of short duration calculate use; So several frame data that the frame losing compensation only needs compensation to lose recently get final product.For guaranteeing the ageing of frame losing information, each frame losing information all has a timer, can compensate the frame losing data in timer expiry;
Last frame information, frame losing information all need to write down several specific state informations; Last frame receives data needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, Frame source netSource; Frame losing informational needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, overtime timer Timer.
When compensating two frame data, LastRx record last frame information, Lost1 sign is the frame losing of close current data, and Lost2 represents time the frame losing near current data, and the online handoff procedure of its pair net is as follows;
(1) do not receive message continuously: the data buffer zone is provided with the network interrupt identification; Simultaneously
LastRx.b Valid=Lost1.bValid=Lost2.b Valid=false, and withdraw from;
(2) do not upgrade lastRx information for a long time: the data buffer zone is provided with interrupt identification, while lastRx.bValid=false, and withdraw from;
(3) LastRx.bValid==false: jump to 8;
(4) LastRx.NetSourse==CurNet: jump to 7;
(5) CurSmpSyn ≠ LastRx.smpSyn: abandon this information;
(6) CurSmpCnt upgrades than LastRx.smpCnt: enter 7; Otherwise these frame data are used for compensating frame losing information:
1) Lost1.bValid==True: if CurSmpSyn==Lost1.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost1.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost1.Valid=Lost2.Valid=False;
2) Lost2.bValid==True: if CurSmpSyn==Lost2.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost2.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost2.Valid=False;
(7) CurSmpCnt ≠ ((LastRx.smpCnt+1) %MaxSmpCnt): there is frame losing that (if the frame losing of multiframe information takes place, directly writing down two up-to-date frame frame losing information) takes place,
1)Lost1.bValid==True:Lost2=Lost1;
2) upgrade up-to-date frame losing information;
Lost1.bValid=true
Lost1.smpCnt=lostSmpCnt
lost1.smpSyn=curSyn
SET?Lost1.Timer
3) the frame losing content is write the data buffer zone with invalid flag;
4) if more frame losing data are arranged, Enter:1);
(8) these frame data are write the data buffer zone; And upgrade LastRx information with following step:
LastRx.b?Valid=True
LastRx.smpCnt=CurSmpCnt
LastRx.NetSource=CurNet
LastRx.smpSyn=CurSyn
(9) timer of inspection Lost1 and Lost2:
1)Lost1.Timer==TimeOut:Lost1.Valid=Lost2.Valid=False;
2)Lost2.Timer==TimeOut:Lost2.Valid=False;
(10)Return。
Change moment in sampled value message synchronous regime, saltus step takes place in sampled value counter automatically, the sampled value continuity criterion of this moment lost efficacy, and the A/B net sampled value data that device receives may continuous a few frame generation synchronous regimes, the situation of sample counter while saltus step.For preventing that algorithm is judged by accident in this case, the online handoff algorithms of two nets only according to consolidated network Data Update data buffer zone, promptly changes the Data Receiving network of preceding last frame synchronously under the situation that sampled value message synchronous regime changes; When having identical synchronous regime simultaneously, recover normal two net switching modes up to two netting indexs.
Sampled data flow in the digital transformer substation is very big, and the data traffic that typical sample frequency is the merge cells of 4k approximates 5Mb/s.So the data traffic under the sample frequency is very high to the real-time processing requirements of web-transporting device, data receiver.Concerning needs received the equipment of striding at interval of many merge cellses data, the sampled value data of handling two nets simultaneously were unusual acid tests to system loading, sometimes even cause system's overload.
Digital transformer substation, based on the two net sampled datas of SV network, adopt double-network redundant transmission sampled value data, the sampled data of arbitrary transmission over networks all with the another one network on data content basic identical: might be different except that the vlan domain of sign heterogeneous networks route, the data segment content be identical.Adopt the receiving system of two net sampled value communications can receive the bulk redundancy message, two netting indexs provide the recognition methods of the redundant data of two nets fast according to Fast Recognition Algorithm; The computational complexity of relative two single net complete decoding, the decoding complex degree of this double-network redundant data Fast Recognition Algorithm reduces greatly---and only being equivalent to has increased an a small amount of complexity of decoding on the basis of single net complete decoding, has improved two net decoding efficiencies.
The invention has the beneficial effects as follows: arbitrary network instantaneity occurs or permanent fault does not all influence the correct transmission of sampled value in two nets, and improves the ability of anti-switch network frame losing to greatest extent; Simultaneously in order to improve the sampled value disposal ability of correlation reception equipment, sampled value double-network redundant data Fast Recognition Algorithm is proposed targetedly, whether Frame is carried out under the situation of complete decoding not needing, distinguishing is the redundancy sampling data, improves the sampled value treatment effeciency.
On the basis of double-network redundant data Fast Recognition Algorithm, two netting indexs only use a data buffering area according to online switching Fast Recognition Algorithm, have reduced the use of system resource at double.Two netting indexs are according to the sampled value Data Update data buffer zone of online handoff algorithms to receive at first, and the sampled value data can access the fastest renewal; Simultaneously in case a netting index according to frame losing occurring, can reduce the influence that frame losing receives sampled value to greatest extent with other net data frame compensation.Two netting indexs only are on the basis of single netting index according to processing according to the work complexity of online handoff algorithms, have increased the complexity of a dynamic frame losing compensation, far below the complexity of two single netting indexs according to processing.
Embodiment
One, double-network redundant Fast Recognition Algorithm and the precondition of two netting indexs according to online handoff algorithms realization:
1. merge cells (MU) sends the continuity of sampled data: its sample counter (smpCnt) all adds an operation after each sampling; And [0, move in circles between MaxSmpCnt) (the MaxSmpCnt=sample frequency is an example with the sample frequency of 4k, and its MaxSmpCnt is 4000).
Should send identical two netting indexs simultaneously according to (the VALN content that only is used for the network route can be inconsistent) 2.MU send the identical of two netting index certificates: MU.
3. Network Transmission guarantees: the sampled value data that assurance MU sends earlier under normal condition are received equipment earlier and receive, and are received after the data that the back sends.
4. the network delay between two nets is poor: the same sampled value message of two net transmission (A, B net) is owing to the not equal reason of network route may successively arrive receiving equipment.Two net network delay differences are that A, B net the time difference that same sampled value message arrives receiving equipment.The difference of the network delay time difference of A, B net should be not excessive, and its maximum should be controlled within the specific limits.
5. the identification of sampled value message network originating: from A net or B net? can distinguish by dual mode:
If a. the A/B net is from the heterogeneous networks interface of device, and can discern the physical interface in sampling message source, then can be by identification physical interface identification sampled value message source.
If b. A/B net sampled value message has different VLAN signs, and this sign content is received the device reception simultaneously, then can indicate the message source of discerning sampled value by discerning different VLAN.
Two, fast-speed redundancy message recognizer
Two netting indexs reach the purpose of quick identification according to several state informations of Fast Recognition Algorithm by buffer memory reception data.Wherein each state information all comprises following content:
1.bValid:True represent other members initialization of this state information, Flase represents no initializtion;
2.NetSource: the network information of record sampled value Data Source, A net or B net;
3.smpCnt: the sample counter information of record sampled value data.
4.smpSyn: record sampled value Synchronization Status Message.
Maximum network postpones poorly between the number of buffer status information and two net, and sampling time interval (1s/ sample frequency) is relevant.Minimal cache state information number=(maximum network postpones poor+sampling time interval-1)/sampling time interval.
To cushion 2 frame state information is example, and wherein Status1 writes down current valid frame information; Status2 writes down the last valid frame information of current valid frame, and its pair net fast-speed redundancy recognizer computational process is as follows:
1. device initialization: Status1.bValid=Status2.bValid=false;
2. do not receive the sampled value message continuously: enter " the online switch logic of two nets "; Status1.bValid=Status2.bValid=false is set subsequently;
3. no initializtion Status1: resolve this frame data, enter " the online switch logic of two nets " and further handle the sampled value data, upgrade Status1 subsequently
Status1.bValid=ture
Status1.NetSource=curNet
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
Wherein curNet is the network number of current use, and curSmplCnt is the Sample Counter of current use, and curSmplSyn is the data sync quality position of current use.
4. the sampled value source is identical with Status1.NetSourse: this frame data Non-redundant data, resolve this frame data, and enter " the online switch logic of two nets " and further handle the sampled value data; Upgrade Status1, Status2 subsequently
Status1=Status2
Status1.NetSource=curNet
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
5. the sampled value source is different with Status1.NetSourse: the sample counter information of resolving this frame.
6.Status1.smpCnt==curSmpCnt (what meaning of this statement?): redundant data abandons and withdraws from;
7. (Status2.bValid==ture) ﹠amp; ﹠amp; (Status2.smpCnt==curSmpCnt): (what meaning of this statement?) redundant data, abandon and withdraw from;
8. resolve these frame data, enter " the online switch logic of two nets " (what is two net switch logics, there there explanation?) further handle the sampled value data.
Double-network redundant data Fast Recognition Algorithm has been utilized the first continuous characteristic of Network Transmission characteristic, sample counter that arrives earlier that sends, by analyzing the network originating of message, by a small amount of decoding to redundant message, and simple compare operation, just can effectively discern the double-network redundant message fast, improve the decoding efficiency of two net sampled value messages greatly.Simultaneously two netting indexs, can effectively prevent because network delay, the network delay difference of the A/B net that a variety of causes such as Route Selection, network congestion cause are shaken the impression that two netting indexs are caused according to identification by the number of buffer status information rationally is set according to Fast Recognition Algorithm.
Three, two netting indexs are according to online switching
Two netting indexs comprise according to online switch logic main contents: upgrade the reception buffering area as early as possible with the sampled value content that arrives receiving system at first; Occur under the frame loss condition in the very fast sampled value data that arrive, realize the frame losing compensation with another network data that arrives more slowly; It is correct to guarantee that when single net fault or two net replace fault sampled value receives.
Sampled value is new criterion more: according to increasing progressively continuously under identical synchronous regime with the sampled value counter, to the characteristic of to a certain degree reversing, judge whether the sampled value counter upgrades.Under identical synchronous regime, consider that sampled value receives possible cause of short duration frame loss condition takes place, its sampled value more new criterion is as follows:
1.LastSmpCnt>(MaxSmpCnt-MaxLostNum):
During CurSmpCnt>LastSmpCnt||CurSmpCnt<((LastSmpCnt+MaxLostNum) %MaxSmpCnt), think that CurSmpCnt upgrades; Otherwise, think the generation saltus step.
2.CurSmpCnt>LastSmpCnt﹠amp; ﹠amp; During CurSmpCnt<(LastSmpCnt+MaxLostNum), think that sampled value upgrades; Otherwise, think the generation saltus step.
MaxSmpCnt is illustrated in the numerical value number of the sample counter under the fixed sampling frequency, is example with the 4k sample rate,
MaxSmpCnt=4000;smpCnt∈[0,3999]。According to the continuous reception criterion of " two netting indexs are according to quick identification ", be MaxLostNum receiving the sampled value counter number of transitions that its maximum possible takes place in the data break continuously.
Compare by data and current data to last frame, two nets of sampled value switch the judgement that realizes the sampled values renewal.Sampled value is ageing data simultaneously, will be used to after the buffering that these data only can be of short duration calculate use; So several frame data that the frame losing compensation only needs compensation to lose recently get final product.For guaranteeing the ageing of frame losing information, each frame losing information all has a timer, can compensate the frame losing data in timer expiry.
Last frame information, frame losing information all need to write down several specific state informations.Last frame receives data needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, Frame source netSource; Frame losing informational needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, overtime timer Timer.
With maximum compensation 2 frame data is example, and LastRx record last frame information, Lost1 sign is the frame losing of close current data, and Lost2 represents time the frame losing near current data, and the online handoff procedure of its pair net is as follows;
1. do not receive message continuously: the data buffer zone is provided with the network interrupt identification; While lastRx.bValid=Lost1.bValid=Lost2.bValid=false, and withdraw from;
2. do not upgrade lastRx information for a long time: the data buffer zone is provided with interrupt identification, while lastRx.bValid=false, and withdraw from;
3.LastRx.bValid==false: jump to 8;
4.LastRx.NetSourse==CurNet: jump to 7;
5.CurSmpSyn ≠ LastRx./mpSyn: abandon this information;
6.CurSmpCnt upgrade than LastRx.smpCnt: enter 7; Otherwise these frame data are used for compensating frame losing information:
1) Lost1.bValid==True: if CurSmpSyn==Lost1.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost1.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost1.Valid=Lost2.Valid=False;
2) Lost2.bValid==True: if CurSmpSyn==Lost2.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost2.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost2.Valid=False;
7.CurSmpCnt ≠ ((LastRx.smpCnt+1) %MaxSmpCnt): there is frame losing that (if the frame losing of multiframe information takes place, directly writing down two up-to-date frame frame losing information) takes place,
1)Lost1.bValid==True:Lost2=Lost1;
2 upgrade up-to-date frame losing information;
Lost1.bValid=true
Lost1.smpCnt=lostSmpCnt
lost?1.smpSyn=curSyn
SET?Lost1.Timer
3) the frame losing content is write the data buffer zone with invalid flag;
4) if more frame losing data are arranged, Enter:1);
8. these frame data are write the data buffer zone; And renewal LastRx information:
LastRx.bValid=True
LastRx.smpCnt=CurSmpCnt
LastRx.NetSource=CurNet
LastRx.smpSyn=CurSyn
9. check the timer of Lost1 and Lost2:
1)Lost1.Timer==TimeOut:Lost1.Valid=Lost2.Valid=False;
2)Lost2.Timer==TimeOut:Lost2.Valid=False;
10.Return;
Change moment in sampled value message synchronous regime, saltus step takes place in sampled value counter automatically, the sampled value continuity criterion of this moment lost efficacy, and the A/B net sampled value data that device receives may continuous a few frame generation synchronous regimes, the situation of sample counter while saltus step.For preventing that algorithm is judged by accident in this case, the online handoff algorithms of two nets only according to consolidated network Data Update data buffer zone, promptly changes the Data Receiving network of preceding last frame synchronously under the situation that sampled value message synchronous regime changes; When having identical synchronous regime simultaneously, recover normal two net switching modes up to two netting indexs.

Claims (4)

1. the online changing method of the two nets of sampled value is characterized in that being made up of according to online changing method double-network redundant data Fast Recognition Algorithm and two netting indexs:
Double-network redundant data Fast Recognition Algorithm:
Two netting indexs reach the purpose of quick identification according to several state informations of Fast Recognition Algorithm by buffer memory reception data; Wherein each state information all comprises following content:
(1) bValid:True represents other members initialization of this state information, and Flase represents no initializtion;
(2) NetSource: the network information of record sampled value Data Source, A net or B net;
(3) smpCnt: the sample counter information of record sampled value data;
(4) smpSyn: record sampled value Synchronization Status Message;
Maximum network postpones poorly between the number of buffer status information and two net, and sampling time interval is relevant; Minimal cache state information number=(maximum network postpones poor+sampling time interval-1)/sampling time interval;
As the current valid frame information of Status1 state recording; Status2 writes down the last valid frame information of current valid frame, and its pair net fast-speed redundancy recognizer computational process is as follows:
(1) device initial information: Status1.bValid=Status2.bValid=false;
(2) do not receive the sampled value message continuously: enter " the online switch logic of two nets "; Status1.bValid=Status2.bValid=false is set subsequently; Current valid frame record no initializtion is set in expression;
(3) no initializtion Status1: resolve this frame data, enter " the online switch logic of two nets " and further handle the sampled value data, upgrade Status1 subsequently
Status1.b?Valid=ture
Status1.NetSource=curNet
Expression record sampled value source, sample counter information, Synchronization Status Message;
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
(4) the sampled value source is identical with Status1.NetSourse: this frame data Non-redundant data, resolve this frame data, and enter " the online switch logic of two nets " and further handle the sampled value data; Upgrade Status1, Status2 subsequently
Status1=Status2
Status1.NetSource=curNet
Record sampled value source, sample counter information, the synchronous regime letter of representing current valid frame
Status1.smpCnt=curSmpCnt
Status1.smpSyn=curSmpSyn
Breath;
(5) the sampled value source is different with Status1.NetSourse: expression is different with the record sampled value of current valid frame, resolves the sample counter information of this frame;
(6) Status1.smpCnt==curSmpCnt: the record sampled value counter information of representing current valid frame is identical, and redundant data abandons and withdraws from;
(7) (Status2.bValid==ture) ﹠amp; ﹠amp; (Status2.smpCnt==curSmpCnt): the record sampled value initialization of representing current valid frame is identical, and redundant data abandons and withdraws from;
(8) resolve this frame data, enter " the online switch logic of two nets " and further handle the sampled value data;
1. double-network redundant data Fast Recognition Algorithm has been utilized the first continuous characteristic of Network Transmission characteristic, sample counter that arrives earlier that sends, by analyzing the network originating of message, by a small amount of decoding to redundant message, and simple compare operation, just can effectively discern the double-network redundant message fast;
Two netting indexs, prevent because network delay, the network delay difference of the A/B net that a variety of causes such as Route Selection, network congestion cause are shaken the harmful effect that two netting indexs are caused according to identification by the number of buffer status information is set according to Fast Recognition Algorithm;
Two netting indexs are according to online changing method: upgrade the reception buffering area as early as possible with the sampled value content that arrives receiving system at first; Occur under the frame loss condition in the very fast sampled value data that arrive, realize the frame losing compensation with another network data that arrives more slowly; It is correct to guarantee that when single net fault or two net replace fault sampled value receives;
Sampled value is new criterion more: according to increasing progressively continuously under identical synchronous regime with the sampled value counter, in the characteristic of to a certain degree reversing, judge whether the sampled value counter upgrades.
2. the online changing method of the two nets of sampled value according to claim 1 is characterized in that under identical synchronous regime, considers that of short duration frame loss condition may take place in the sampled value reception, and its sampled value more new criterion is as follows:
1)LastSmpCnt>(MaxSmpCnt-MaxLostNum):
During CurSmpCnt>LastSmpCnt||CurSmpCnt<((LastSmpCnt+MaxLostNum) %MaxSmpCnt), think that CurSmpCnt upgrades; Otherwise, think the generation saltus step;
2) CurSmpCnt>LastSmpCnt﹠amp; ﹠amp; During CurSmpCnt<(LastSmpCnt+MaxLostNum), think that sampled value upgrades; Otherwise, think the generation saltus step;
MaxSmpCnt is illustrated in the numerical value number of the sample counter under the fixed sampling frequency, is MaxLostNum receiving the sampled value counter number of transitions that its maximum possible takes place in the data break continuously;
Compare by data and current data to last frame, two nets of sampled value switch the judgement that realizes the sampled values renewal; Sampled value is ageing data simultaneously, will be used to after the buffering that these data only can be of short duration calculate use; So several frame data that the frame losing compensation only needs compensation to lose recently get final product.For guaranteeing the ageing of frame losing information, each frame losing information all has a timer, can compensate the frame losing data in timer expiry;
Last frame information, frame losing information all need to write down several specific state informations; Last frame receives data needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, Frame source netSource; Frame losing informational needs record: effective marker bValid, frame losing sample counter SmpCnt, frame losing synchronous regime smpSyn, overtime timer Timer.
3. the online changing method of the two nets of sampled value according to claim 1 and 2, when it is characterized in that compensating two frame data, LastRx record last frame information, Lost1 sign is the frame losing of close current data, Lost2 represents time frame losing of close current data, and the online handoff procedure of its pair net is as follows;
(1) do not receive message continuously: the data buffer zone is provided with the network interrupt identification; Simultaneously
LastRx.bValid=Lost1.bValid=Lost2.b Valid=false, and withdraw from;
(2) do not upgrade lastRx information for a long time: the data buffer zone is provided with interrupt identification, while lastRx.bValid=false, and withdraw from;
(3) LastRx.bValid==false: jump to 8;
(4) LastRx.NetSourse==CurNet: jump to 7;
(5) CurSmpSyn ≠ LastRx.smpSyn: abandon this information;
(6) CurSmpCnt upgrades than LastRx.smpCnt: enter 7; Otherwise these frame data are used for compensating frame losing information:
1) Lost1.bValid==True: if CurSmpSyn==Lost1.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost1.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost1.Valid=Lost2.Valid=False;
2) Lost2.bValid==True: if CurSmpSyn==Lost2.smpSyn﹠amp; ﹠amp; CurSmpCnt==Lost2.smpCnt inserts the relevant position, data buffer zone with this frame information; While Lost2.Valid=False;
(7) CurSmpCnt ((LastRx.smpCnt+1) %MaxSmpCnt): there is frame losing that (if the frame losing of multiframe information takes place, directly writing down two up-to-date frame frame losing information) takes place,
1)Lost1.bValid==True:Lost2=Lost1;
2) upgrade up-to-date frame losing information;
Lost1.bValid=true
Lost1.smpCnt=lostSmpCnt
lost1.smpSyn=curSyn
SET?Lost1.Timer
3) the frame losing content is write the data buffer zone with invalid flag;
4) if more frame losing data are arranged, Enter:1);
(8) these frame data are write the data buffer zone; And upgrade LastRx information with following step:
LastRx.bValid=True
LastRx.smpCnt=CurSmpCnt
LastRx.NetSource=CurNet
LastRx.smpSyn=CurSyn
(9) timer of inspection Lost1 and Lost2:
1)Lost1.Timer==TimeOut:Lost1.Valid=Lost2.Valid=False;
2)Lost2.Timer==TimeOut:Lost2.Valid=False;
(10)Return。
4. the online changing method of the two nets of sampled value according to claim 1 and 2, it is characterized in that changing moment in sampled value message synchronous regime, saltus step takes place in sampled value counter automatically, the sampled value continuity criterion of this moment lost efficacy, and the A/B net sampled value data that device receives may continuous a few frame generation synchronous regimes, the situation of sample counter while saltus step.For preventing that algorithm is judged by accident in this case, the online handoff algorithms of two nets only according to consolidated network Data Update data buffer zone, promptly changes the Data Receiving network of preceding last frame synchronously under the situation that sampled value message synchronous regime changes; When having identical synchronous regime simultaneously, recover normal two net switching modes up to two netting indexs.
CN201010273732XA 2010-09-06 2010-09-06 Double-network online switching method of sampling values Pending CN101958802A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201010273732XA CN101958802A (en) 2010-09-06 2010-09-06 Double-network online switching method of sampling values

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201010273732XA CN101958802A (en) 2010-09-06 2010-09-06 Double-network online switching method of sampling values

Publications (1)

Publication Number Publication Date
CN101958802A true CN101958802A (en) 2011-01-26

Family

ID=43485924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201010273732XA Pending CN101958802A (en) 2010-09-06 2010-09-06 Double-network online switching method of sampling values

Country Status (1)

Country Link
CN (1) CN101958802A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102437923A (en) * 2011-10-18 2012-05-02 南京南瑞继保电气有限公司 Method for switching sampling value data during mixed application of peer-to-peer mode and networking mode
CN103117599A (en) * 2013-02-04 2013-05-22 南京南瑞继保电气有限公司 Intelligent transformer substation link redundancy sampling value online seamless switching method
CN106789510A (en) * 2016-12-21 2017-05-31 中国船舶重工集团公司第七研究所 The data communications method of redundant CAN bus communication system
CN111447253A (en) * 2020-02-28 2020-07-24 清华大学 Buffer management method, system, device and storage medium
CN114124672A (en) * 2021-10-12 2022-03-01 南方电网数字电网研究院有限公司 Process layer double-network sampling data processing method of intelligent substation and protection device thereof
CN115065444A (en) * 2022-05-30 2022-09-16 中国电子科技集团公司第十研究所 On-orbit dual-redundancy data processing method of satellite-borne management and control computer

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102437923A (en) * 2011-10-18 2012-05-02 南京南瑞继保电气有限公司 Method for switching sampling value data during mixed application of peer-to-peer mode and networking mode
CN102437923B (en) * 2011-10-18 2014-08-20 南京南瑞继保电气有限公司 Method for switching sampling value data during mixed application of peer-to-peer mode and networking mode
CN103117599A (en) * 2013-02-04 2013-05-22 南京南瑞继保电气有限公司 Intelligent transformer substation link redundancy sampling value online seamless switching method
CN103117599B (en) * 2013-02-04 2015-03-25 南京南瑞继保电气有限公司 Intelligent transformer substation link redundancy sampling value online seamless switching method
CN106789510A (en) * 2016-12-21 2017-05-31 中国船舶重工集团公司第七研究所 The data communications method of redundant CAN bus communication system
CN106789510B (en) * 2016-12-21 2020-03-24 中国船舶重工集团公司第七一一研究所 Data communication method of redundant CAN bus communication system
CN111447253A (en) * 2020-02-28 2020-07-24 清华大学 Buffer management method, system, device and storage medium
CN111447253B (en) * 2020-02-28 2021-06-25 清华大学 Buffer management method, system, device and storage medium
CN114124672A (en) * 2021-10-12 2022-03-01 南方电网数字电网研究院有限公司 Process layer double-network sampling data processing method of intelligent substation and protection device thereof
CN114124672B (en) * 2021-10-12 2024-01-23 南方电网数字电网研究院有限公司 Process layer double-network sampling data processing method and protection device for intelligent substation
CN115065444A (en) * 2022-05-30 2022-09-16 中国电子科技集团公司第十研究所 On-orbit dual-redundancy data processing method of satellite-borne management and control computer

Similar Documents

Publication Publication Date Title
CN101958802A (en) Double-network online switching method of sampling values
CN103729806B (en) Transformer substation alarming graph gateway minimizing system and achieving method thereof
CN101057483B (en) Reception of redundant and non-redundant frames
CN102333028B (en) Method and communication equipment for sending messages by using layered bi-layer virtual private network
CN102025560B (en) Method for testing process-level network test method
CN103779966B (en) Based on intelligent terminal monitoring system and the method for the empty terminal mapping mechanism of input
CN104601375B (en) A kind of intelligent substation communication network monitoring system, setting method and monitoring method
CN106300678A (en) A kind of digital protective relay system supporting unification of three nets network
CN107069681B (en) A kind of multi-point circuit differential protecting method and system
CN105099777A (en) Method suitable for communication-gateway double-machine redundancy in rail transit power monitoring system
CN105356432B (en) More contact distribution network failure dead electricity region topology identifying systems and method based on Multi Agent
CN101609332B (en) Online three-layer synchronous anti-error locking system
CN101521406B (en) GOOSE dual-net redundant communication signal transmission method
KR20210044467A (en) Highly Reliable, Low Cost Next Generation Network System For Digital Substation
CN101621550A (en) Method and device for recycling MAC address in VPLS
CN111682508A (en) Differential protection and rapid self-healing method suitable for regional power distribution network
CN104065160A (en) Method for processing abnormal message in electric power system
CN103117599B (en) Intelligent transformer substation link redundancy sampling value online seamless switching method
CN101741145B (en) Traditional switch intelligent interface device for digitized transformer substation
CN111244919A (en) Distribution network topology self-adaptive identification method suitable for intelligent distributed FA
CN105305602B (en) The standby power system of railway digital traction substation and its haul oneself willingly into method
CN202978439U (en) Process level networking structure applicable to centralized station domain protection communication of intelligent substations
CN113036731B (en) Point-to-point communication power distribution network feeder fault isolation method and self-healing system
CN101304181B (en) Double-network control method for transformer substation case mechanism based on facing all-purpose object
CN102263820A (en) Intelligent transformer substation network communication structure based on protection independence and information sharing

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: 20110126