CN116317151A - Signal tag method and system based on multi-source heterogeneous power grid information feature extraction - Google Patents

Abstract

The invention discloses a signal tag method and a system based on multi-source heterogeneous power grid information feature extraction, wherein the method comprises the following steps: acquiring multi-source heterogeneous power grid information; identifying the multi-source heterogeneous power grid information according to an abnormal information feature extraction module, wherein the abnormal information feature extraction module comprises a module for identifying whether the multi-source heterogeneous power grid information is invalid power grid information; and if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, marking the multi-source heterogeneous power grid information with an invalid power grid information label. The method utilizes key technologies such as abnormal data feature extraction, intelligent power grid data identification and the like to construct an abnormal data intelligent identification rule base, carries out labeling treatment on accessed multi-source heterogeneous information, intelligently identifies data abnormal problems, marks abnormal data and forms an effective data set of power grid operation information.

Description

Signal tag method and system based on multi-source heterogeneous power grid information feature extraction

Technical Field

The application relates to the field of power grid system monitoring, in particular to a signal tag method and a system based on multi-source heterogeneous power grid information feature extraction.

Background

The traditional power monitoring system has unfriendly man-machine interaction interface and poor openness, and the operation monitoring only depends on two measures: firstly, displaying the running state of the power equipment and the abnormal condition of the display of the optical character plate through a graphical interface; secondly, the alarm information is listed in time sequence through alarm windows, and one alarm window is generally divided into two alarm information fields, wherein one alarm window is an unacknowledged information field, and the other alarm window is all alarm fields. The traditional mode can also meet the monitoring requirement under the condition of a single transformer substation, but in the monitoring center mode, the number of transformer substations and the number of equipment are greatly increased, the traditional mode can not realize comprehensive monitoring, and the problems that alarm signals are not classified clearly, the alarm window format is single, the signal display is not comprehensive, the alarm information is generated in the test process of overhaul equipment to cause interference and the like exist.

The regular test of the power equipment is a basic requirement for ensuring the safe operation of the power equipment, corresponding alarm information is necessarily accompanied in the test process, the traditional alarm information is easy to be interfered by various test signals, and when a real equipment abnormal signal is mixed in the test signal, the real equipment abnormal signal is easy to be missed, so that the equipment is caused to continuously run abnormally, and accidents are caused. The traditional monitoring system has insufficient statistical analysis capability on the current abnormal signal, can not effectively judge whether the abnormal signal is effective for the abnormal condition of a certain device, can only check the abnormal condition of the total station device one by one, has low efficiency, can not provide a data analysis basis for comprehensive perception, dynamic analysis and timely early warning of the running states of the device and the power grid, realizes the further development of the running state prediction of the device, realizes the feasibility of early warning and active treatment of the fault of the device, reduces the running risk of the device and the power grid to the minimum, and can not ensure the safe running of the power grid. The method can not process the multi-source heterogeneous power grid information in time and mark the information in real time.

Disclosure of Invention

In order to solve the technical problems, one of the purposes of the invention is to provide a signal tag method based on multi-source heterogeneous power grid information feature extraction, which can solve the problems that the traditional monitoring system has insufficient statistical analysis capability on current abnormal signals, can not effectively judge whether the abnormal signals are effective for the abnormal condition of a certain device, can only check the abnormal condition of the total station device one by one, has low efficiency, can not provide a data analysis basis for comprehensive perception, dynamic analysis and timely early warning of the running states of the device and the power grid, realizes further development of the running state prediction of the device, realizes the feasibility of early warning and active treatment of the device, reduces the running risk of the device and the power grid to the minimum, and can not ensure the safe running of the power grid.

The second purpose of the invention is to provide a signal tag system based on multi-source heterogeneous power grid information feature extraction, which can solve the problems that the traditional monitoring system has insufficient statistical analysis capability on the current abnormal signal, can not effectively judge whether the abnormal signal is effective for the abnormal condition of a certain device, can only check the abnormal condition of the total station device one by one, has low efficiency, can not provide a data analysis basis for comprehensive perception and dynamic analysis and timely early warning of the running states of the device and the power grid, realizes further prediction of the running states of the device, realizes the feasibility of early warning and active treatment of the fault of the device, minimizes the running risks of the device and the power grid, and can not ensure the safe running of the power grid.

One of the purposes of the invention is realized by adopting the following technical scheme:

a signal tag method based on multi-source heterogeneous power grid information feature extraction comprises the following steps:

s1, acquiring multi-source heterogeneous power grid information;

s2, identifying the multi-source heterogeneous power grid information according to an abnormal information feature extraction module, wherein the abnormal information feature extraction module comprises a module for identifying whether the multi-source heterogeneous power grid information is invalid power grid information or not;

s3, if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, marking the multi-source heterogeneous power grid information with an invalid power grid information label.

Further, the step S2 further includes the following step S20: identifying the alarm information according to an invalid alarm information feature extraction module, wherein the invalid alarm information feature extraction module comprises a module for identifying whether the alarm information is invalid alarm information;

the step S3 further includes the following step S30: if the alarm information accords with the invalid alarm information feature in the invalid alarm information feature extraction module, marking the alarm information with an invalid alarm information label;

the invalid warning information tag comprises one or more of the following tags: companion tags, commissioning tags, manual tags, ignore tags, unproductive tags, overhaul tags, telemetry-induced action tags or normally lit tags.

Further, the step S30 further includes the following steps:

and if the alarm information comprises switch deflection signals with the same interval in the front and back set time or bus side knife deflection signals with the same voltage in the front and back set time, marking the alarm information with associated labels.

Further, the step S30 further includes the following steps:

and if the alarm information belongs to the debugging responsibility area or the alarm information comprises a debugging identifier, marking the debugging operation label with the alarm information.

Further, the step S30 further includes the following steps:

and if the alarm information comprises deflection alarms which exist at the same interval in a specific time, wherein the deflection alarms are alarms of manual remote control, and the deflection states are brake-off, and the difference between the deflection time and the alarm occurrence time is smaller than the specific time, the alarm information is marked with a manual operation label.

Further, the step S30 further includes the following steps:

if the alarm information comprises one or more of standby, standby interval signals, main gear shift operation signals, voltage telemetry out-of-limit signals in a power failure state, line TV voltage loss abnormal alarm signals, bus connection and bus-split control loop disconnection abnormal alarm signals and reclosing charging completion signals, marking the alarm information with an neglected tag;

if the alarm information comprises a non-production-start mark and the time for recording the production-start mark is earlier than the time for sending the alarm information, marking the alarm information as a non-production-start mark.

If the alarm information comprises one or more of a cold standby identifier, an overhaul identifier, an associated operation identifier and an associated overhaul identifier, marking the alarm information with an overhaul label;

further, if the alarm information comprises a configuration of telemetry verification associated with a telemetry ID and the telemetry alarm requirement is verified to be met through a telemetry value verification formula, marking a telemetry-induced action tag, wherein the telemetry value verification formula comprises a digital comparison formula; and if the alarm information comprises a normally-bright mark and the non-reset alarm information is sent out, marking the non-reset alarm information with a normally-bright label.

Further, the step S2 includes the following steps: the multi-source heterogeneous power grid information is obtained in parallel, and different abnormal information feature extraction sub-modules in the abnormal information feature extraction module are matched according to the source of the multi-source heterogeneous power grid information; the abnormal information feature extraction submodules are respectively used for processing power grid information of different sources, the abnormal information feature extraction submodules are in a parallel processing mode, and S3 is executed after the multi-source heterogeneous power grid information is identified as invalid power grid information;

further, the step S3 further includes the following steps: and if a plurality of invalid grid information is identified at the same time, marking the plurality of invalid grid information as a plurality of invalid grid information labels correspondingly, and ignoring alarm information corresponding to the grid information labels when grid event diagnosis is carried out.

The second purpose of the invention is realized by adopting the following technical scheme:

signal tag system based on heterogeneous electric wire netting information characteristic extraction of multisource includes:

the receiving module is used for acquiring multi-source heterogeneous power grid information;

the abnormal information feature extraction module is used for identifying whether the multi-source heterogeneous power grid information is invalid power grid information or not;

and the marking module is used for marking the multi-source heterogeneous power grid information with an invalid power grid information label if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module.

Further, the signal tag system based on multi-source heterogeneous power grid information feature extraction further comprises:

the abnormal information feature extraction module further comprises an invalid alarm information feature extraction module for identifying whether the alarm information is invalid alarm information;

the marking module is used for marking the alarm information with an invalid alarm information label if the alarm information accords with the invalid alarm information characteristics in the invalid alarm information characteristic extracting module;

the invalid warning information tag comprises one or more of the following tags: companion tags, commissioning tags, manual tags, ignore tags, unproductive tags, overhaul tags, telemetry-induced action tags or normally lit tags.

Compared with the prior art, the invention has the beneficial effects that:

(1) And constructing an abnormal data intelligent identification rule base by utilizing key technologies such as abnormal data feature extraction and intelligent power grid data identification, carrying out labeling treatment on the accessed multi-source heterogeneous information, intelligently identifying abnormal data, and marking the abnormal data to form an effective data set of power grid operation information.

(2) The data mining analysis capability is improved, a data analysis basis can be provided for comprehensive perception, dynamic analysis and timely early warning of the running states of equipment and a power grid, the running state prediction of the equipment is further developed, the feasibility of early warning and active treatment of the equipment fault is realized, the running risk of the equipment and the power grid is reduced to the minimum, and the safe running of the large power grid is ensured.

The foregoing description is only an overview of the present invention, and is intended to be more clearly understood as the present invention, which can be practiced according to the present description, by referring to the following detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:

fig. 1 is a schematic flow chart of a signal tag method based on multi-source heterogeneous power grid information feature extraction according to the present invention.

Fig. 2 is a flow chart of a signal tag method based on alarm information feature extraction according to the present invention.

Fig. 3 is a flowchart of a signal tag method based on associated alarm signal feature extraction in an embodiment of the present invention.

Fig. 4 is a flowchart of a signal tag method based on debugging operation alarm signal feature extraction in an embodiment of the present invention.

Fig. 5 is a flowchart of a signal tag method based on manually operated alert signal feature extraction in an embodiment of the present invention.

Fig. 6 is a flow chart of a signal tag method based on standby, stand-by interval signal feature extraction in an embodiment of the invention.

Fig. 7 is a flowchart of a signal tag method for extracting an alarm signal feature based on a main shift operation according to an embodiment of the present invention.

Fig. 8 is a flowchart of a signal tag method based on voltage telemetry out-of-limit alarm signal feature extraction in an embodiment of the invention.

Fig. 9 is a flowchart of a signal tag method based on line TV voltage loss anomaly alert signal feature extraction in an embodiment of the present invention.

Fig. 10 is a flow chart of a signal tag method for extracting abnormal alarm signal characteristics of a control loop based on a busbar and a busbar component in an embodiment of the invention.

Fig. 11 is a flowchart of a signal tag method for completing alarm signal feature extraction based on reclosing charging in an embodiment of the present invention.

Fig. 12 is a flowchart of a signal tag method based on non-commissioned warning signal feature extraction in an embodiment of the invention.

Fig. 13 is a flow chart of a signal tag method based on cold standby and service alarm signal feature extraction in an embodiment of the invention.

FIG. 14 is a flow chart of a method for extracting signal tags based on associated ticket alert signal features in an embodiment of the present invention.

Fig. 15 is a flowchart of a signal tag method based on the associated service ticket alert signal feature extraction in an embodiment of the present invention.

FIG. 16 is a flow chart of a method for signal tagging based on telemetry-induced alert signal feature extraction in an embodiment of the present invention.

FIG. 17 is a flowchart of a signal tag method based on non-return alert signal feature extraction in accordance with an embodiment of the present invention.

Fig. 18 is a block diagram of a signal tag system based on multi-source heterogeneous power grid information feature extraction according to the present invention.

Detailed Description

For the purposes, technical solutions and advantages of the present application will be clearly and completely described below through embodiments, and it is apparent that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

Example 1

The embodiment provides a signal tag method based on multi-source heterogeneous power grid information feature extraction, which is applied to the field of power grid system monitoring, and as shown in fig. 1, the method can comprise the following steps:

s1, acquiring multi-source heterogeneous power grid information.

In this embodiment, the multi-source heterogeneous power grid information may be a switch position signal, an alarm signal, a protection action signal, etc., which reflects actual operation states of various electrical devices of the substation; the multi-source heterogeneous power grid information is from a plurality of data sources in a power grid system, including data sets acquired by different database systems and different devices in operation. The operating system and the management system where different data sources are located (for example, OA, CRM, HR, MES, SCM) are different, the storage mode and the logic structure of the data are different, and the generation time, the use place, the code protocol and the like of the data are also different, so that the characteristic of multiple sources of the data is caused.

Preferably, as shown in fig. 2, S1 may further include the following step S10: and acquiring alarm information.

In this embodiment, the alarm information may be a fatal error of all devices, for example: communication interruption, error checking of a fixed value, error protection of a pressing plate area and the like; the control loops of all devices are disconnected; PT wire breakage and CT wire breakage of all devices; all devices failed to trip; a main light gas alarm action; main transformer overload operation, etc.

S2, identifying the multi-source heterogeneous power grid information according to an abnormal information feature extraction module, wherein the abnormal information feature extraction module comprises a module for identifying whether the multi-source heterogeneous power grid information is invalid power grid information.

In this embodiment, the abnormal information feature extraction module may be used to identify whether the multi-source heterogeneous power grid information is invalid power grid information, and the abnormal information feature extraction module may be an invalid alarm information feature extraction module or the like; identifying the multi-source heterogeneous power grid information according to the abnormal information feature extraction module, and identifying the multi-source heterogeneous power grid information as effective power grid information when the multi-source heterogeneous power grid information does not accord with the ineffective power grid information feature in the abnormal information feature extraction module; and when the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, identifying the multi-source heterogeneous power grid information as invalid power grid information.

Preferably, as shown in fig. 2, S2 may further include the following step S20: and identifying the alarm information according to an invalid alarm information feature extraction module, wherein the invalid alarm information feature extraction module comprises a module for identifying whether the alarm information is invalid alarm information.

In this embodiment, the invalid alarm information feature extraction module may be configured to identify whether the alarm information is invalid alarm information, identify the alarm information according to the invalid alarm information feature extraction module, and identify the alarm information as valid alarm information when the alarm information does not conform to the invalid alarm information feature in the invalid alarm information feature extraction module; and when the alarm information accords with the invalid alarm information characteristic in the invalid alarm information characteristic extraction module, identifying the alarm information as invalid alarm information.

S3, if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, marking the multi-source heterogeneous power grid information with an invalid power grid information label.

Preferably, as shown in fig. 2, S3 may further include the following step S30: and if the alarm information accords with the invalid alarm information feature in the invalid alarm information feature extraction module, marking the alarm information with an invalid alarm information label.

In this embodiment, the invalid alarm information tag may be an associated tag, a debug operation tag, a manual operation tag, an ignore tag, a non-commission tag, an overhaul tag, a telemetry-induced action tag, a normally-on tag, and the like. Identifying the alarm information according to the invalid alarm information feature extraction module, and identifying the alarm information as valid alarm information when the alarm information does not accord with the invalid alarm information feature in the invalid alarm information feature extraction module; when the alarm information accords with the invalid alarm information feature in the invalid alarm information feature extraction module, the alarm information is identified as invalid alarm information, the invalid alarm information label is marked, and the alarm information is ignored in the diagnosis of the power grid alarm event.

Preferably, S30 may further include the following step S301: and if the alarm information comprises switch deflection signals with the same interval in the front and back set time or bus side knife deflection signals with the same voltage in the front and back set time, marking the alarm information with associated labels.

In this embodiment, as shown in fig. 3, the set time may be 10 seconds, firstly, judging whether the alarm information is suspected of being associated with a signal, if the alarm information is a remote signaling displacement signal within 10 seconds before and after, then the alarm information is suspected of being associated with a signal, further judging the type of the associated signal, if the alarm information includes a switch displacement signal with the same interval within 10 seconds before and after, etc., judging that the alarm information is a switch displacement associated signal, belonging to invalid alarm information, marking the alarm information with an associated tag, and omitting the alarm information in the diagnosis of the grid alarm event; if the alarm information comprises bus side knife deflection signals with the same voltage within 10 seconds before and after the alarm information, the bus side knife deflection signals are judged to be bus associated signals, the bus side knife deflection signals belong to invalid alarm information, the alarm information is marked with associated labels, and the alarm information is ignored in the diagnosis of the power grid alarm event. The accompanying signals refer to signals which are immediately restored after the actions in a short time generated by contact shake and the like in the switching operation of the switch, the disconnecting link and the like, and the signals have several outstanding characteristics: always accompanying operation generation; the time from the action to the return is very short, and most of the time is in milliseconds; the amount of generated signals is relatively large; the associated signal for each occurrence is relatively fixed.

Preferably, S30 may further include the following step S302: and if the alarm information belongs to the debugging responsibility area or the alarm information comprises a debugging identifier, marking the debugging operation label with the alarm information.

In this embodiment, as shown in fig. 4, if the alarm information belongs to the debugging responsibility area, the alarm information belongs to invalid alarm information, the alarm information is marked with a debugging operation label, and the alarm information is ignored in the diagnosis of the power grid alarm event; the debug flag may be a debug flag, which is an invalid alert if the alert contains a debug flag, and the alert is marked with a debug operation tag, which ignores such alert in the grid alert event diagnosis.

Preferably, S30 may further include the following step S303: and if the alarm information comprises deflection alarms which exist at the same interval in a specific time, wherein the deflection alarms are alarms of manual remote control, and the deflection states are brake-off, and the difference between the deflection time and the alarm occurrence time is smaller than the specific time, the alarm information is marked with a manual operation label.

In this embodiment, as shown in fig. 5, the specific time may be 2 minutes, firstly, judging whether the alarm information needs to be verified by manual remote control operation through a regular expression, if the alarm information needs to be verified by manual remote control operation, then judging the deflection alarm signals at the same interval, if the alarm information meets the following three conditions: a deflection alarm signal exists at the same interval within 2 minutes; the displacement alarm signal is an alarm signal of manual remote control, the displacement state is a brake-off state, the displacement time and the alarm signal occurrence time difference is less than 2 minutes, the alarm signal belongs to invalid alarm information, the alarm information is marked with a manual operation label, and the alarm information is ignored in the diagnosis of the power grid alarm event.

Preferably, S30 may further include the following step S304: and if the alarm information comprises one or more of standby, standby interval signals, main gear shift operation signals, voltage telemetry out-of-limit signals in a power failure state, line TV voltage loss abnormal alarm signals, bus connection and bus-split control loop disconnection abnormal alarm signals and reclosing charging completion signals, marking the alarm information with an neglected tag.

In this embodiment, as shown in fig. 6, the feature extraction of the standby interval signal may be to firstly determine whether the alarm signal name includes the busbar knife & & air chamber, if the alarm signal name does not include the busbar knife & air chamber, then continuously determine whether the alarm signal interval name includes "standby" or "standby", if the alarm signal interval name includes "standby" or "standby", then the alarm signal is indicated to be a standby interval signal, and belongs to invalid alarm information, the alarm information is marked with an ignore tag, and such alarm information is ignored in the diagnosis of the alarm event of the power grid.

As shown in fig. 7, the feature extraction of the main shift operating signal may be performed by determining whether the alarm information content includes a "main shift", and if the alarm information content includes a "main shift", it indicates that the main shift operating signal belongs to invalid alarm information, and the alarm information is marked with an ignore tag, so that the alarm information is ignored in the diagnosis of the grid alarm event.

As shown in fig. 8, the feature extraction of the voltage telemetry out-of-limit signal in the power outage state may be to firstly determine whether the alarm information is a telemetry out-of-limit alarm signal, then determine whether the alarm information is a voltage telemetry out-of-limit alarm signal in the power outage state if the alarm information is a voltage telemetry out-of-limit alarm signal, and if the alarm information content includes keywords such as "phase voltage" & can "be separated" or "line voltage" & can "be separated", it is indicated that the alarm information is a voltage telemetry out-of-limit alarm signal in the power outage state, and belongs to invalid alarm information, and then the alarm information is marked to ignore the tag, and the alarm information is ignored in the diagnosis of the power grid alarm event. The feature extraction of the voltage telemetry out-of-limit signal in the power failure state can also judge whether the alarm information is the telemetry out-of-limit alarm signal or not, if the alarm information is the voltage telemetry out-of-limit alarm signal, the bus is judged whether to hang up or not through the power grid topology, if the bus is judged to not hang up through the power grid topology, the alarm information is marked with an neglect tag, and the alarm information is ignored in the power grid alarm event diagnosis.

As shown in fig. 9, the feature extraction of the abnormal voltage loss alarm signal of the circuit TV may be to firstly determine whether the alarm information content includes keywords such as "TV voltage loss" & gt, TV disconnection "or" TV secondary "& gt, and" open and trip ", if the keywords include keywords such as" TV voltage loss "& gt, TV disconnection" or "TV secondary" & gt, and "open and trip", the feature extraction may be to indicate the abnormal voltage loss alarm signal of the circuit TV, and then determine whether the switches on each side of the circuit are on or off, if the switches on each side of the circuit are on, the abnormal voltage loss alarm signal of the circuit TV is indicated to be invalid alarm information, and mark the alarm information with an neglect tag, so that the alarm information is ignored in the diagnosis of the alarm event of the power grid.

As shown in fig. 10, the feature extraction of the control loop disconnection abnormality alarm signal of the bus-tie and the bus-tie can firstly determine whether the alarm information belongs to the control loop disconnection abnormality alarm signal, if the alarm information is the control loop disconnection abnormality alarm signal, then determine whether the alarm information content contains keywords such as the bus-tie, the section and the like, if the alarm information content contains the keywords such as the bus-tie, the section and the like, then determine whether the alarm information is the bus interconnection operation signal of the same voltage level of the current station, if the alarm information is the bus interconnection operation signal of the same voltage level of the current station and the occurrence time of the bus interconnection operation signal and the control loop disconnection signal is within a certain threshold value, the alarm information is the control loop disconnection abnormality alarm signal of the bus-tie and the bus-tie, belongs to the invalid alarm information, the alarm information is marked with a neglect tag, and the alarm information is ignored in the diagnosis of the power grid alarm event.

As shown in fig. 11, the feature extraction of the reclosing charging completion signal may first determine whether the alarm information is a reclosing charging completion alarm signal, if yes, then determine whether the interval to which the alarm information belongs is a normal running state, if the interval to which the alarm information belongs is a normal running state, then indicate that the alarm information is a reclosing charging completion signal, and belongs to invalid alarm information, mark the alarm information with an neglect tag, and ignore the alarm information in the diagnosis of the grid alarm event.

Preferably, if the alarm information includes a non-commission identifier and the time of recording the commission identifier is earlier than the time of sending the alarm information, the alarm information is marked as a non-commission tag.

In this embodiment, as shown in fig. 12, the non-productive label may be a non-productive label, if the device hanging the non-productive label is the same as the device of the remote signaling alarm and the time of hanging the non-productive label is before the alarm occurs, the non-productive label is marked by the alarm information, and the alarm information is ignored in the diagnosis of the power grid alarm event.

Preferably, S30 may further include the following step S305: and if the alarm information comprises one or more of a cold standby identifier, an overhaul identifier, an associated operation identifier and an associated overhaul identifier, marking the alarm information with an overhaul label.

In this embodiment, as shown in fig. 13, the feature extraction of the cold standby identifier may be that all the knife switches in the alarm interval state are in the opening state or that the switch hand knife in the alarm interval state is in the opening state, if one of the conditions is met, the cold standby identifier is the cold standby identifier, the alarm information containing the cold standby identifier belongs to invalid alarm information, the alarm information is marked to repair the tag, and the alarm information is ignored in the diagnosis of the power grid alarm event; if the characteristics are not met, judging whether the alarm equipment is hung with the overhaul plate, if so, marking the alarm information with the overhaul plate as an overhaul label, wherein the alarm information containing the hung overhaul plate belongs to invalid alarm information. The feature extraction of the overhaul identifier can be that all the disconnecting link in the alarm interval state are in the disconnection state, the switch hand knife in the alarm interval state is in the disconnection state, the switch in the alarm interval state is in the closing state, and meanwhile, the three conditions are met, the overhaul identifier is the overhaul identifier, the alarm information containing the overhaul identifier belongs to invalid alarm information, the alarm information is marked with an overhaul label, and the alarm information is ignored in the diagnosis of the power grid alarm event; if the characteristics are not met, judging whether the alarm equipment is hung with a maintenance card, if the alarm equipment is hung with the maintenance card, the alarm information containing the hung maintenance card belongs to invalid alarm information, marking the alarm information with the maintenance label, and ignoring the alarm information in the diagnosis of the power grid alarm event.

As shown in fig. 14, the association operation identifier may be an association operation ticket, the feature extraction of the association operation ticket may be that the station to which the alarm information belongs, the interval name is included in the operation order, the alarm time is after the down time of the operation order, before the dispatch of the return order, and the operation order work is stopped or turned into cold standby or inverted, etc., and the above three conditions are satisfied, if the alarm information including the association operation ticket belongs to invalid alarm information, the alarm information is marked with an overhaul label, and such alarm information is ignored in the diagnosis of the grid alarm event.

As shown in fig. 15, the associated maintenance identifier may also be an associated maintenance ticket, the feature extraction of the associated maintenance ticket may be a factory station to which the alarm information belongs, the interval name is contained in the work unit, the work content, the safety measure and other work content in the maintenance ticket, and the alarm information is marked on the maintenance label when the completion time is empty and the time is determined to be standard when the completion time is empty before the completion time is determined to be full, and the two conditions are satisfied at the same time.

If the alarm information comprises a cold standby identifier, an overhaul identifier, an associated operation identifier, an associated overhaul identifier and the like, the alarm information is marked with an overhaul label, and the alarm information is ignored in the diagnosis of the power grid alarm event.

Preferably, S30 may further include the following step S306: if the alarm information comprises a configuration of telemetry verification associated with a telemetry ID and the telemetry alarm requirement is met through telemetry value verification formula verification, marking a telemetry induced action tag, wherein the telemetry value verification formula comprises a digital comparison formula.

In this embodiment, as shown in fig. 16, the telemetry value checking formula may be a digital comparison formula, firstly, it is determined whether the configuration of the telemetry verification is related to the telemetry signaling ID, that is, the feature extraction of telemetry alarming and telemetry value checking is maintained in the power grid system, if the corresponding feature can be found through the telemetry signaling ID, it is indicated that the checking needs to be performed through the telemetry value, then, the return result of the digital comparison formula is determined, if the return result is True, it is an alarm caused by telemetry, and the alarm information is marked with a telemetry-caused action tag, and such alarm information is ignored in the diagnosis of the power grid alarming event.

Preferably, S30 may further include the following step S307: and if the alarm information comprises a normally-bright mark and the non-reset alarm information is sent out, marking the non-reset alarm information with a normally-bright label.

In this embodiment, as shown in fig. 17, the normally-bright label may be a normally-bright label, and the device that sets the normally-bright label according to the association of the device that sends the non-reset alarm information, marks the alarm information of the device that can be associated with the normally-bright label, and ignores the alarm information in the diagnosis of the grid alarm event.

Preferably, the step S2 includes the steps of: the multi-source heterogeneous power grid information is obtained in parallel, and different abnormal information feature extraction sub-modules in the abnormal information feature extraction module are matched according to the source of the multi-source heterogeneous power grid information; the abnormal information feature extraction submodules are respectively used for processing power grid information of different sources, the abnormal information feature extraction submodules are in a parallel processing mode, and S3 is executed after the multi-source heterogeneous power grid information is identified as invalid power grid information; because the alarm information transmitted by the multi-source equipment is randomly sent, in order to alarm timely and identify false alarm information timely, the embodiment adopts parallel receiving of multi-source heterogeneous power grid information sent by the multi-source equipment, and the implementation mode can be as follows: the FPGA module is adopted to realize the access and logic judgment of the information of the multipath equipment; likewise, a plurality of controllers can also be adopted to respectively operate different abnormal information feature extraction sub-modules so as to realize the identification of invalid power grid information; and only if invalid grid information is identified, the marking is selected. If the multi-source heterogeneous power grid information is found to be effective and real alarm information after judgment, the alarm information is not intercepted any more and is directly transmitted to a monitoring system so as to realize timely alarm on the running states of equipment and a power grid.

Further, the step S3 further includes the following steps: and if a plurality of invalid grid information is identified at the same time, marking the plurality of invalid grid information as a plurality of invalid grid information labels correspondingly, and ignoring alarm information corresponding to the grid information labels when grid event diagnosis is carried out. The method has the advantages that when invalid power grid information is marked, the problem that alarm information is eliminated slowly can be reduced by adopting a parallel marking mode, so that a fault monitoring system cannot timely acquire whether equipment or a power grid truly has faults, the attention time of staff is long, and the workload of the staff is further increased.

The embodiment has the following advantages:

(1) And constructing an abnormal data intelligent identification rule base by utilizing key technologies such as abnormal data feature extraction and intelligent power grid data identification, carrying out labeling treatment on the accessed multi-source heterogeneous information, intelligently identifying abnormal data, and marking the abnormal data to form an effective data set of power grid operation information.

(2) The data mining analysis capability is improved, a data analysis basis can be provided for comprehensive perception, dynamic analysis and timely early warning of the running states of equipment and a power grid, the running state prediction of the equipment is further developed, the feasibility of early warning and active treatment of the equipment fault is realized, the running risk of the equipment and the power grid is reduced to the minimum, and the safe running of the large power grid is ensured.

Example two

The present embodiment provides a signal tag system based on multi-source heterogeneous power grid information feature extraction, as shown in fig. 18, the signal tag system 100 based on multi-source heterogeneous power grid information feature extraction includes:

and the receiving module 10 is used for acquiring the multi-source heterogeneous power grid information.

In this embodiment, the multi-source heterogeneous power grid information may be a switch position signal, an alarm signal, a protection action signal, etc., which reflects actual operation states of various electrical devices of the substation; the multi-source heterogeneous power grid information is from a plurality of data sources in a power grid system, including data sets acquired by different database systems and different devices in operation. The operating system and the management system where different data sources are located (for example, OA, CRM, HR, MES, SCM) are different, the storage mode and the logic structure of the data are different, and the generation time, the use place, the code protocol and the like of the data are also different, so that the characteristic of multiple sources of the data is caused.

The abnormal information feature extraction module 20 is configured to identify whether the multi-source heterogeneous power grid information is invalid power grid information.

In this embodiment, the abnormal information feature extraction module may be used to identify whether the multi-source heterogeneous power grid information is invalid power grid information, and the abnormal information feature extraction module may be an invalid alarm information feature extraction module or the like; identifying the multi-source heterogeneous power grid information according to the abnormal information feature extraction module, and identifying the multi-source heterogeneous power grid information as effective power grid information when the multi-source heterogeneous power grid information does not accord with the ineffective power grid information feature in the abnormal information feature extraction module; and when the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, identifying the multi-source heterogeneous power grid information as invalid power grid information.

And the marking module 30 is configured to mark the multi-source heterogeneous power grid information with an invalid power grid information tag if the multi-source heterogeneous power grid information accords with the invalid power grid information feature in the abnormal information feature extraction module.

In this embodiment, the invalid alarm information tag may be an associated tag, a debug operation tag, a manual operation tag, an ignore tag, a non-commission tag, an overhaul tag, a telemetry-induced action tag, a normally-on tag, and the like. Identifying the alarm information according to the invalid alarm information feature extraction module, and identifying the alarm information as valid alarm information when the alarm information does not accord with the invalid alarm information feature in the invalid alarm information feature extraction module; when the alarm information accords with the invalid alarm information feature in the invalid alarm information feature extraction module, the alarm information is identified as invalid alarm information, the invalid alarm information label is marked, and the alarm information is ignored in the diagnosis of the power grid alarm event.

The product can execute the signal tag method based on the multi-source heterogeneous power grid information feature extraction in the embodiment 1, and has the corresponding functional modules and beneficial effects of the execution method.

The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way; those skilled in the art can smoothly practice the invention as shown in the drawings and described above; however, those skilled in the art will appreciate that many modifications, adaptations, and variations of the present invention are possible in light of the above teachings without departing from the scope of the invention; meanwhile, any equivalent changes, modifications and evolution of the above embodiments according to the essential technology of the present invention still fall within the scope of the present invention.

Claims (10)

1. The signal tag method based on multi-source heterogeneous power grid information feature extraction is characterized by comprising the following steps of:

s1, acquiring multi-source heterogeneous power grid information;

s2, identifying the multi-source heterogeneous power grid information according to an abnormal information feature extraction module, wherein the abnormal information feature extraction module comprises a module for identifying whether the multi-source heterogeneous power grid information is invalid power grid information or not;

s3, if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module, marking the multi-source heterogeneous power grid information with an invalid power grid information label.

2. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 1, wherein said S2 further comprises the following step S20: identifying the alarm information according to an invalid alarm information feature extraction module, wherein the invalid alarm information feature extraction module comprises a module for identifying whether the alarm information is invalid alarm information;

the step S3 further includes the following step S30: if the alarm information accords with the invalid alarm information feature in the invalid alarm information feature extraction module, marking the alarm information with an invalid alarm information label;

the invalid warning information tag comprises one or more of the following tags: companion tags, commissioning tags, manual tags, ignore tags, unproductive tags, overhaul tags, telemetry-induced action tags or normally lit tags.

3. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 2, wherein the step S30 further comprises the steps of:

and if the alarm information comprises switch deflection signals with the same interval in the front and back set time or bus side knife deflection signals with the same voltage in the front and back set time, marking the alarm information with associated labels.

4. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 2, wherein the step S30 further comprises the steps of:

and if the alarm information belongs to the debugging responsibility area or the alarm information comprises a debugging identifier, marking the debugging operation label with the alarm information.

5. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 2, wherein the step S30 further comprises the steps of:

and if the alarm information comprises deflection alarms which exist at the same interval in a specific time, wherein the deflection alarms are alarms of manual remote control, and the deflection states are brake-off, and the difference between the deflection time and the alarm occurrence time is smaller than the specific time, the alarm information is marked with a manual operation label.

6. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 2, wherein the step S30 further comprises the steps of:

if the alarm information comprises one or more of standby, standby interval signals, main gear shift operation signals, voltage telemetry out-of-limit signals in a power failure state, line TV voltage loss abnormal alarm signals, bus connection and bus-split control loop disconnection abnormal alarm signals and reclosing charging completion signals, marking the alarm information with an neglected tag;

if the alarm information comprises a non-production-start mark and the time for recording the production-start mark is earlier than the time for sending the alarm information, marking the alarm information as a non-production-start mark.

7. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to claim 2, wherein the step S30 further comprises the steps of:

if the alarm information comprises one or more of a cold standby identifier, an overhaul identifier, an associated operation identifier and an associated overhaul identifier, marking the alarm information with an overhaul label;

and/or, if the alarm information comprises a configuration of telemetry verification associated with a telemetry ID and the telemetry alarm requirement is verified to be met through a telemetry value verification formula, marking a telemetry-induced action tag, wherein the telemetry value verification formula comprises a digital comparison formula; and if the alarm information comprises a normally-bright mark and the non-reset alarm information is sent out, marking the non-reset alarm information with a normally-bright label.

8. The signal tag method based on multi-source heterogeneous power grid information feature extraction according to any one of claims 1 to 7, wherein S2 comprises the steps of: the multi-source heterogeneous power grid information is obtained in parallel, and different abnormal information feature extraction sub-modules in the abnormal information feature extraction module are matched according to the source of the multi-source heterogeneous power grid information; the abnormal information feature extraction submodules are respectively used for processing power grid information of different sources, the abnormal information feature extraction submodules are in a parallel processing mode, and S3 is executed after the multi-source heterogeneous power grid information is identified as invalid power grid information;

and/or, the step S3 further comprises the following steps: and if a plurality of invalid grid information is identified at the same time, marking the plurality of invalid grid information as a plurality of invalid grid information labels correspondingly, and ignoring alarm information corresponding to the grid information labels when grid event diagnosis is carried out.

9. The signal tag system based on multi-source heterogeneous power grid information feature extraction is characterized by comprising:

the receiving module is used for acquiring multi-source heterogeneous power grid information;

the abnormal information feature extraction module is used for identifying whether the multi-source heterogeneous power grid information is invalid power grid information or not;

and the marking module is used for marking the multi-source heterogeneous power grid information with an invalid power grid information label if the multi-source heterogeneous power grid information accords with the invalid power grid information characteristics in the abnormal information characteristic extraction module.

10. The signal tag system based on multi-source heterogeneous power grid information feature extraction of claim 9, further comprising:

the abnormal information feature extraction module further comprises an invalid alarm information feature extraction module for identifying whether the alarm information is invalid alarm information;

the marking module is used for marking the alarm information with an invalid alarm information label if the alarm information accords with the invalid alarm information characteristics in the invalid alarm information characteristic extracting module;

the invalid warning information tag comprises one or more of the following tags: companion tags, commissioning tags, manual tags, ignore tags, unproductive tags, overhaul tags, telemetry-induced action tags or normally lit tags.

Images