CN115358355B - Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity - Google Patents

Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity Download PDF

Info

Publication number
CN115358355B
CN115358355B CN202211299080.6A CN202211299080A CN115358355B CN 115358355 B CN115358355 B CN 115358355B CN 202211299080 A CN202211299080 A CN 202211299080A CN 115358355 B CN115358355 B CN 115358355B
Authority
CN
China
Prior art keywords
oil temperature
curve
index
main transformer
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202211299080.6A
Other languages
Chinese (zh)
Other versions
CN115358355A (en
Inventor
张殷
王俊波
李国伟
唐琪
熊仕斌
蒋维
罗容波
王智娇
曾烨
范心明
李新
董镝
宋安琪
刘崧
黄静
陈贤熙
曾庆辉
刘少辉
刘昊
章涛
马榕嵘
张思寒
赖艳珊
陈绮琪
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.)
Foshan Power Supply Bureau of Guangdong Power Grid Corp
Original Assignee
Foshan Power Supply Bureau of Guangdong Power Grid Corp
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 Foshan Power Supply Bureau of Guangdong Power Grid Corp filed Critical Foshan Power Supply Bureau of Guangdong Power Grid Corp
Priority to CN202211299080.6A priority Critical patent/CN115358355B/en
Publication of CN115358355A publication Critical patent/CN115358355A/en
Application granted granted Critical
Publication of CN115358355B publication Critical patent/CN115358355B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
    • G01K13/00—Thermometers specially adapted for specific purposes
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Housings And Mounting Of Transformers (AREA)
  • Protection Of Transformers (AREA)

Abstract

The invention discloses a method and a device for judging the abnormality of oil temperature of a main transformer and top oil temperature, which are characterized in that the abnormal condition of the numerical value of each main transformer oil temperature meter is judged by combining a plurality of logic criteria, and a top oil temperature abnormality meter is locked; quantitatively calculating the target correlation indexes of all main transformers, and constructing target correlation index vectors of all the main transformers; calculating a historical association index vector of the main transformer to be analyzed by combining historical data, and establishing a typical association index vector of the main transformer by using a k-means algorithm; calculating the distance between a target correlation index vector of a main transformer to be analyzed and a typical correlation index vector of the main transformer, and determining a top-layer abnormal oil temperature main transformer; and aiming at the continuously updated data stream, a sliding window method is adopted to roll and develop the top oil temperature abnormity monitoring and top oil temperature meter abnormity checking process of the main transformer.

Description

Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity
Technical Field
The invention relates to the technical field of power equipment monitoring, in particular to a method and a device for judging abnormity of oil temperature of a main transformer oil temperature gauge and top oil temperature.
Background
The main transformer is used as an important primary device of a power grid, and has important influence on the safe, economic and reliable operation of the whole power system. The thermal state of the transformer is an important factor for determining the service life and the load capacity of the main transformer and is an important parameter for reflecting the running state of the main transformer.
In actual production operation, an operator usually determines the thermal state of the transformer through the internal temperature of the main transformer, and the internal temperature of the main transformer, such as the top oil temperature, is an important parameter for monitoring the state of the transformer, and the current top oil temperature of the main transformer can be collected by a top oil temperature representation field and is uploaded to a master station background for display through an electric SCADA system. And comparing the real-time top oil temperature monitoring value with a preset main transformer top oil temperature threshold value to realize abnormal monitoring of the main transformer top oil temperature.
However, different transformers have different operating conditions and operating environments, the unified threshold is too rough and single and is difficult to set scientifically and reasonably, and it is more difficult to comprehensively summarize diversified abnormal scenes of the top oil temperature of the main transformer. Based on a fixed threshold value, the differential monitoring and refined abnormal alarm of the top layer oil temperature of the main transformer are difficult to flexibly and meticulously carry out, the method is not suitable for judging the abnormal state of the top layer oil temperature which does not reach the limit value, the abnormal state of the top layer oil temperature of the main transformer cannot be timely found, and potential safety hazards are left for safe and reliable operation of equipment.
Meanwhile, the state of the top oil temperature meter needs to be determined by combining with a power failure preventive test which is periodically carried out, the obtained offline test result reflects the discrete characteristic of the state of equipment, and the real-time monitoring and abnormal rolling check of the state of the top oil temperature meter of the main transformer cannot be realized. Due to the existence of the oil temperature gauge state monitoring vacuum period, the abnormal state of the top oil temperature gauge is difficult to find in time, and the potential safety hazard of unreliable oil temperature monitoring, unreal monitoring result and unknown equipment state exists, so that the safe and reliable operation of the main transformer is seriously influenced. In addition, manual field tests consume manpower and material resources, the process and the time consumption are long, the advantages of top-layer oil temperature online monitoring and the potential of robot replacement cannot be fully released, and negative influences are brought to safe operation and reliable power supply of a power grid due to equipment power failure and adjustment of the power grid operation mode.
Disclosure of Invention
The invention provides a method and a device for judging the abnormity of a main transformer oil temperature gauge and a top layer oil temperature, which solve the problems that the existing fixed threshold value is difficult to flexibly and meticulously carry out differential monitoring and refined abnormity alarm of the main transformer top layer oil temperature, the existing fixed threshold value is not suitable for judging the abnormity state that the top layer oil temperature does not reach the limit value, the abnormity state of the main transformer top layer oil temperature cannot be found in time, potential safety hazards are left for safe and reliable operation of equipment, meanwhile, the potential safety hazards of 'unreliable oil temperature monitoring, unreal monitoring result and unknown equipment state' exist in the oil temperature gauge state monitoring vacuum period, the safe and reliable operation of a main transformer is seriously influenced, in addition, manual field tests consume manpower and material resources, the flow and the time are long, the advantages of online monitoring of the top layer oil temperature and the potential of a robot are not fully released, and the negative influence is brought to the safe operation and the reliable power supply of a power grid due to the equipment power failure along with the adjustment of a power grid operation mode.
The invention provides a method for judging the abnormity of oil temperature of a main transformer oil temperature gauge and a top layer oil temperature, which comprises the following steps:
acquiring basic state data of a plurality of main transformers to be judged in a preset period, preprocessing the basic state data, and generating a main transformer state curve for calculation at the current moment;
extracting a plurality of curve state data from the main transformer state curve, and calculating a top oil temperature table state index based on the curve state data;
comparing the state index of the top oil temperature meter with a preset standard state index, and screening the main transformers to be analyzed except the abnormal top oil temperature meter from the plurality of main transformers to be judged according to the comparison result;
intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector;
determining typical associated index vectors according to clustering results of historical associated index vectors of the main transformer to be analyzed, and calculating target associated index vectors and target category distance values of the typical associated index vectors;
comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the main transformer to be analyzed has top layer oil temperature abnormity;
and counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results.
Optionally, the step of obtaining and preprocessing the basic state data of a plurality of main transformers to be judged in a preset period to generate a main transformer state curve for calculation at the current time includes:
acquiring the target data and the historical data of a plurality of main transformers to be judged in a preset period;
sequencing the target data and the historical data respectively based on data types to generate a main transformer state curve for calculation at the current moment; the data types include top oil temperature, load, and ambient temperature.
Optionally, the step of extracting a plurality of curve status data from the main transformer status curve and calculating the top oil temperature table status index based on the curve status data includes the steps of:
selecting a plurality of curve state data of a preset quantity from the main transformer state curve according to a time reverse sequence from the current moment;
determining a corresponding state index of the first top oil temperature meter by using the curve state data and a preset oil temperature limit value;
determining a corresponding state index of the second top oil temperature meter by using the curve state data;
determining the state index of the third top-layer oil temperature meter by adopting the curve state data corresponding to two adjacent top-layer oil temperature meters in the main transformer to be judged;
selecting data of two adjacent moments in the curve state data, and determining corresponding state indexes of the fourth top-layer oil temperature meter;
selecting data of three adjacent moments in the curve state data, and determining corresponding state indexes of the fifth top-layer oil temperature meter;
and determining the corresponding state index of the sixth top oil temperature meter by adopting the curve state data.
Optionally, the step of comparing the state index of the top oil temperature gauge with a preset standard state index, and screening the main transformers to be analyzed from the plurality of main transformers to be judged according to the comparison result, except for the abnormality of the top oil temperature gauge, includes:
comparing the state index of the top oil temperature meter with a corresponding preset standard state index;
if any one of the top-layer oil temperature meter state indexes is larger than the corresponding preset standard state index, judging that the top-layer oil temperature meter of the main transformer to be judged is abnormal;
and determining all the main transformers to be judged, of which the state indexes of the top oil temperature meter are less than or equal to the corresponding preset standard state indexes, as main transformers to be analyzed for abnormal oil temperature.
Optionally, the step of intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating target associated indexes of each main transformer to be analyzed according to the curve to be processed, and constructing a target associated index vector includes:
intercepting the target top layer oil temperature curve and the target load curve in a preset time period as a curve to be processed, and calculating the corresponding first correlation index according to the curve to be processed;
intercepting the target top layer oil temperature curve and the target environment temperature curve in a preset time period as a curve to be processed, and calculating corresponding second correlation indexes according to the curve to be processed;
intercepting the target top layer oil temperature curve and the historical top layer oil temperature curve in a preset time period as a curve to be processed, and calculating a corresponding third correlation index according to the curve to be processed;
intercepting the target load curve and the historical load curve in a preset time period as curves to be processed, and calculating corresponding fourth correlation indexes according to the curves to be processed;
intercepting the target top layer oil temperature curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding fifth correlation indexes according to the curves to be processed;
intercepting the target load curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding sixth correlation indexes according to the curves to be processed;
and constructing corresponding target associated index vectors by adopting the first associated index, the second associated index, the third associated index, the fourth associated index, the fifth associated index and the sixth associated index.
Optionally, the step of determining a typical associated indicator vector according to a clustering result of the historical associated indicator vectors of the main transformers to be analyzed, and calculating a target associated indicator vector and a target category distance value of each typical associated indicator vector includes:
selecting initial clustering centers according with the number of centers from a plurality of historical associated index vectors of each main transformer to be analyzed;
calculating an initial distance between each historical relevance index vector and each initial clustering center, and classifying each historical relevance index vector into a clustering category with the minimum initial distance;
calculating a middle clustering center of each clustering category;
if the intermediate clustering centers are not changed, calculating effective clustering indexes according to the initial distance and the intermediate clustering centers;
and if the number of the cluster types is larger than a preset type threshold value, and the cluster effective index is larger than or equal to the cluster effective index corresponding to the previous cluster type number, taking a middle cluster center corresponding to the previous cluster type number as a typical association index vector, and calculating a target association index vector and a target type distance value of each typical association index vector.
Optionally, the method further comprises:
and if the number of the cluster categories is less than or equal to a preset category threshold value, or the cluster effective index is less than the cluster effective index corresponding to the previous cluster category number, increasing the number of the centers, and skipping to execute the step of selecting the initial cluster centers meeting the number of the centers from the plurality of historical associated index vectors of each main transformer to be analyzed.
Optionally, the step of comparing the target category distance value with a plurality of preset category distance thresholds and determining whether the main transformer to be analyzed has top layer oil temperature abnormality includes:
comparing the target class distance value with a plurality of preset class distance thresholds;
if the target category distance value is smaller than or equal to any category distance threshold value, judging that the main transformer to be analyzed does not have top layer oil temperature abnormity;
and if the target category distance values are all larger than all the category distance threshold values, judging that the top layer oil temperature abnormity exists in the main transformer to be analyzed.
Optionally, the method further comprises:
calculating an initial distance value between a plurality of historical relevance index vectors and each typical relevance index vector;
respectively selecting the maximum initial distance value in the clustering category to which each typical correlation index vector belongs;
and respectively calculating the multiplication value between each maximum initial distance value and a preset threshold coefficient as the class distance threshold of the cluster class at the current moment.
The second aspect of the present invention provides a device for determining an oil temperature anomaly of a main transformer oil temperature gauge and a top oil temperature anomaly, comprising:
the data preprocessing module is used for acquiring and preprocessing basic state data of a plurality of main transformers to be judged in a preset period to generate a main transformer state curve for calculation at the current moment;
the top oil temperature table state index calculating module is used for extracting a plurality of curve state data from the main transformer state curve and calculating the top oil temperature table state index based on the curve state data;
the abnormal top oil temperature meter determining module is used for comparing the state index of the top oil temperature meter with a preset standard state index and screening the main transformers to be analyzed except the abnormal top oil temperature meter from the plurality of main transformers to be judged according to the comparison result;
the target associated index vector construction module is used for intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating a target associated index of each main transformer to be analyzed according to the curve to be processed, and constructing a target associated index vector;
the target category distance value acquisition module is used for determining typical associated index vectors according to clustering results of historical associated index vectors of the main transformer to be analyzed and calculating target associated index vectors and target category distance values of the typical associated index vectors;
the top layer oil temperature abnormity determining module is used for comparing the target category distance value with a plurality of preset category distance thresholds and judging whether the main transformer to be analyzed has top layer oil temperature abnormity;
and the abnormal result generation module is used for counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter and generating an abnormal result.
According to the technical scheme, the invention has the following advantages:
acquiring basic state data of a plurality of main transformers to be judged in a preset period, preprocessing the basic state data, and generating a main transformer state curve for calculation at the current moment; extracting a plurality of curve state data from a main transformer state curve, and calculating a top oil temperature table state index based on the curve state data; comparing the state index of the top oil temperature meter with a preset standard state index, and screening main transformers to be analyzed except the abnormality of the top oil temperature meter from the plurality of main transformers to be judged according to the comparison result; intercepting a main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector; determining typical correlation index vectors according to clustering results of historical correlation index vectors of the main transformer to be analyzed, and calculating target correlation index vectors and target category distance values of the typical correlation index vectors; comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the top layer oil temperature of the main transformer to be analyzed is abnormal or not; counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results; the method solves the problems that the existing fixed threshold is difficult to flexibly and meticulously carry out main transformer top layer oil temperature differential monitoring and refined abnormal alarm, is not suitable for judging the abnormal state that the top layer oil temperature does not reach the limit value, cannot find the abnormal state of the main transformer top layer oil temperature in time, leaves potential safety hazards for safe and reliable operation of equipment, has a vacuum period for monitoring the state of an oil temperature meter, is difficult to find the abnormal state of the top layer oil temperature meter in time, has the potential safety hazards of unreliable oil temperature monitoring, unreal monitoring results and unknown equipment states, seriously influences the safe and reliable operation of a main transformer, in addition, manual field tests consume manpower and material resources, have long process and time consumption, cannot fully release the advantages of online monitoring of the top layer oil temperature and the potential of a machine to replace people, and brings negative influences on the safe operation and reliable power supply of a power grid due to the power failure of the equipment along with the adjustment of a power grid operation mode; the top layer oil temperature abnormity monitoring method based on data analysis replaces a fixed monitoring threshold value, individual and common characteristics of the top layer oil temperature running state of each transformer are finely excavated, the top layer oil temperature abnormity refined monitoring of the main transformer is achieved, meanwhile, the top layer oil temperature table abnormity checking method based on data excavation replaces periodic power failure pre-test, the top layer oil temperature online monitoring advantage and the robot potential are fully released, the real-time monitoring and abnormal rolling checking of the state of the top layer oil temperature table of the main transformer are achieved, the state evaluation of the top layer oil temperature table is changed from power failure test to online monitoring, and the purposes of improving equipment safety, reducing equipment power failure, reducing labor cost and increasing checking efficiency are achieved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive labor.
Fig. 1 is a flowchart illustrating steps of a method for determining an abnormal oil temperature of a main transformer oil temperature gauge and a top oil temperature according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating steps of a method for determining an abnormal main transformer oil temperature gauge and an abnormal top layer oil temperature according to a second embodiment of the present invention;
fig. 3 is a flowchart illustrating steps of a method for determining an abnormal main transformer oil temperature gauge and an abnormal top layer oil temperature according to a second embodiment of the present invention;
fig. 4 is a block diagram of a structure of a device for determining an abnormal oil temperature of a main transformer oil temperature gauge and a top oil temperature provided by a third embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a main transformer oil temperature gauge and a top layer oil temperature abnormity judging method and device, which are used for solving the technical problems that the existing fixed threshold value is difficult to flexibly and meticulously carry out main transformer top layer oil temperature differentiation monitoring and refined abnormity alarm, the method is not suitable for judging the abnormal state that the top layer oil temperature does not reach the limit value, the abnormal state of the main transformer top layer oil temperature cannot be timely found, potential safety hazards are left for safe and reliable operation of equipment, meanwhile, the abnormal state of the top layer oil temperature gauge is difficult to timely find in a vacuum oil temperature gauge state monitoring period, the potential safety hazards of 'unreliable oil temperature monitoring, unreal monitoring results and unknown equipment states' exist, the safe and reliable operation of a main transformer is seriously influenced, in addition, manual field tests consume manpower and material resources, the process and the time consumption are long, the advantages of online monitoring of the top layer oil temperature and the potential of robots cannot be fully released, and the negative influence is brought to the safe operation and the reliable power supply of the equipment along with the adjustment of a power grid.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating steps of a method for determining an abnormal oil temperature of a main transformer oil temperature gauge and an abnormal oil temperature of a top layer according to an embodiment of the present invention.
The invention provides a method for judging the abnormity of oil temperature of a main transformer oil temperature gauge and a top layer oil temperature, which comprises the following steps:
step 101, obtaining basic state data of a plurality of main transformers to be judged in a preset period, preprocessing the basic state data, and generating a main transformer state curve for calculation at the current moment.
And the basic state data refers to top oil temperature, load and environment temperature data of the main transformer to be judged, which are acquired based on the electric SCADA system. Historical top oil temperature, load and environmental temperature data of a plurality of main transformers to be judged in the 1 st year after the main transformers are put into operation are obtained through an electric SCADA system.
The preset period refers to a time period set in advance for data acquisition.
And preprocessing, namely sequencing top oil temperature, load and environment temperature data of the main transformer to be judged, which are acquired based on the electric SCADA system, according to types and time, and then assembling into corresponding curve data.
In the embodiment of the invention, top layer oil temperature, load and environment temperature data of a plurality of main transformers to be judged within preset time are obtained through an electric SCADA system, and the obtained top layer oil temperature, load and environment temperature data are sequenced according to types and time to assemble a main transformer state curve used for calculation at the current moment.
And 102, extracting a plurality of curve state data from the main transformer state curve, and calculating the state index of the top oil temperature table based on the curve state data.
The top oil temperature meter state index is used for judging whether the top oil temperature meter is abnormal or not.
In the embodiment of the invention, a plurality of curve state data are extracted from the obtained main transformer state curve, and the state index of the top layer oil temperature table for judging whether the top layer oil temperature table is abnormal or not is calculated according to the plurality of curve state data.
And 103, comparing the state indexes of the top oil temperature meter with preset standard state indexes, and screening the main transformers to be analyzed except the top oil temperature meter abnormity from the plurality of main transformers to be judged according to the comparison result.
And the standard state index is used for judging whether the top oil temperature meter is abnormal or not.
It is worth mentioning that the state indexes of the top oil temperature meter are compared with the preset standard state indexes, the main transformer with the abnormal top oil temperature meter is screened out from the main transformers to be judged according to the comparison result, and the other main transformers with the abnormal oil temperature are to be analyzed.
In the embodiment of the invention, the calculated state index of the top oil temperature meter is compared with the preset standard state index, and the main transformers to be analyzed except the top oil temperature meter abnormity are screened from the plurality of main transformers to be judged according to the comparison result.
And 104, intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector.
The curve to be processed refers to curve data used for calculating the target correlation index.
In the embodiment of the invention, a main transformer state curve in a preset time interval is intercepted and used as a curve to be processed, the target correlation index of each main transformer to be analyzed is calculated according to the curve to be processed, and a corresponding target correlation index vector is constructed based on the target correlation index.
And 105, determining typical associated index vectors according to clustering results of the historical associated index vectors of the main transformer to be analyzed, and calculating target associated index vectors and target category distance values of the typical associated index vectors.
The historical association index vector is constructed based on the historical association index obtained through calculation.
It is worth mentioning that the clustering uses a K-means clustering algorithm (K-means).
In the embodiment of the invention, historical associated indexes of the main transformers to be analyzed are calculated, corresponding historical associated index vectors are constructed on the basis of the historical associated indexes, the historical associated index vectors are clustered, typical associated index vectors are determined according to clustering results, and target category distance values of the target associated index vectors and the typical associated index vectors are calculated.
And 106, comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the top layer oil temperature of the main transformer to be analyzed is abnormal or not.
And the category distance threshold refers to parameters of the main transformer for judging the oil temperature abnormity.
In the embodiment of the invention, the target category distance value is compared with a plurality of preset category distance thresholds; if the target category distance value is smaller than or equal to any category distance threshold value, judging that the top layer oil temperature abnormity does not exist in the main transformer to be analyzed; and if the target category distance values are all larger than all category distance threshold values, judging that the top layer oil temperature of the main transformer to be analyzed is abnormal.
And 107, counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results.
In the embodiment of the invention, the main transformers with abnormal top oil temperature and the main transformers with abnormal top oil temperature meter are counted and summarized, and abnormal results are output.
In the embodiment of the invention, basic state data of a plurality of main transformers to be judged in a preset period are obtained and preprocessed, and a main transformer state curve for calculation at the current moment is generated; extracting a plurality of curve state data from a main transformer state curve, and calculating a top oil temperature table state index based on the curve state data; comparing the state index of the top oil temperature meter with a preset standard state index, and screening main transformers to be analyzed except the abnormality of the top oil temperature meter from the plurality of main transformers to be judged according to the comparison result; intercepting a main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector; determining typical correlation index vectors according to clustering results of historical correlation index vectors of the main transformer to be analyzed, and calculating target correlation index vectors and target category distance values of the typical correlation index vectors; comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the top layer oil temperature of the main transformer to be analyzed is abnormal or not; counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results; the method solves the problems that the existing fixed threshold is difficult to flexibly and meticulously carry out differential monitoring and refined abnormal alarm of the top oil temperature of the main transformer, is not suitable for judging the abnormal state of the top oil temperature which does not reach the limit value, cannot find the abnormal state of the top oil temperature of the main transformer in time, leaves potential safety hazards for safe and reliable operation of equipment, has a vacuum period for monitoring the state of an oil temperature meter, is difficult to find the abnormal state of the top oil temperature meter in time, has the safety hazards of unreliable oil temperature monitoring, unreal monitoring results and unknown equipment states, seriously influences the safe and reliable operation of the main transformer, in addition, manual field tests consume manpower and material resources, have long process and time consumption, cannot fully release the advantages of online monitoring of the top oil temperature and the potential of a robot, and bring negative effects to the safe operation and reliable power supply of a power grid when the equipment is powered off along with the adjustment of a power grid operation mode; the top layer oil temperature abnormity monitoring method based on data analysis replaces a fixed monitoring threshold value, individual and common characteristics of the top layer oil temperature running state of each transformer are finely excavated, the top layer oil temperature abnormity refined monitoring of the main transformer is achieved, meanwhile, the top layer oil temperature table abnormity checking method based on data excavation replaces periodic power failure pre-test, the top layer oil temperature online monitoring advantage and the robot potential are fully released, the real-time monitoring and abnormal rolling checking of the state of the top layer oil temperature table of the main transformer are achieved, the state evaluation of the top layer oil temperature table is changed from power failure test to online monitoring, and the purposes of improving equipment safety, reducing equipment power failure, reducing labor cost and increasing checking efficiency are achieved.
Referring to fig. 2, fig. 2 is a flowchart illustrating steps of a method for determining an abnormal oil temperature of a main transformer oil temperature gauge and an abnormal oil temperature of a top layer according to a second embodiment of the present invention.
The invention provides a method for judging the abnormity of oil temperature of a main transformer oil temperature gauge and a top layer oil temperature, wherein basic state data comprises target data and historical data, and the method comprises the following steps:
step 201, acquiring target data and historical data of a plurality of main transformers to be judged in a preset period.
In the embodiment of the invention, top oil temperature, load and environment temperature data of a plurality of main transformers to be judged in a preset time are obtained through an electric SCADA system. Historical top oil temperature, load and environmental temperature data of a plurality of main transformers to be judged in the 1 st year after operation are obtained through an electric SCADA system.
Step 202, respectively sequencing target data and historical data based on data types to generate a main transformer state curve for calculation at the current moment; the data types include top oil temperature, load, and ambient temperature.
In one example of the invention, time is obtained based on a power SCADA systemT 1 Top layer oil temperature of main transformer with internal to-be-analyzed structureTOil, loadLAnd ambient temperatureTAnd (7) en data. For time of dayt 1 To extractt 1 -T 1 +1 tot 1 The data in between are assembled into calculation data.T 1 Can be selected according to actual needs, such as 30 days. Wherein, the data acquisition frequency is 15 min/point, 30 days correspond to 30 × 96 data points。
Obtaining the same date of the 1 st year after the operation of the main transformer to be analyzed based on the electric SCADA systemT 1 Historical top oil temperature, load, and ambient temperature data over a period of time. If the current primary profile date is 20220716,T 1 30 days later, the current analysis data are 20220616-20220716, and if the operation date of the main transformer is 20150101, the same date is used in the 1 st year after the operation of the main transformerT 1 The time period is 20150616-20150716.
And sequencing each type of data of each main transformer to be judged according to the time sequence, and assembling a main transformer state curve for calculation consisting of top oil temperature, load and ambient temperature curves.
And 203, extracting a plurality of curve state data from the main transformer state curve, and calculating the state index of the top oil temperature table based on the curve state data.
Further, the top oil temperature table status indexes include a first top oil temperature table status index, a second top oil temperature table status index, a third top oil temperature table status index, a fourth top oil temperature table status index, a fifth top oil temperature table status index, and a sixth top oil temperature table status index, and step 203 may include the following steps:
and S11, selecting a plurality of curve state data of a preset quantity from the main transformer state curve according to the time reverse sequence from the current time.
And S12, determining a corresponding first top oil temperature table state index by using the curve state data and a preset oil temperature limit value.
First top oil thermometer state index, referring to timeT 1 And the times of the numerical value of the oil temperature of the inner top layer exceeding the limit value are used for judging the times of the reading of a top layer oil temperature meter in the main transformer to be judged exceeding the preset oil temperature limit value.
The preset oil temperature limit value refers to a top layer oil temperature normal lower limit value and a top layer oil temperature normal upper limit value.
Calculating timeT 1 The times of the oil temperature value of the inner top layer exceeding the limit value are as follows:
Figure 302975DEST_PATH_IMAGE001
，
Figure 531700DEST_PATH_IMAGE002
in the formula,
Figure 373754DEST_PATH_IMAGE003
is time of dayT 1 The times that the oil temperature value of the inner top layer exceeds the limit value,
Figure 768963DEST_PATH_IMAGE004
to judge the value of the top oil temperature table of the main transformer at the moment t,
Figure 396385DEST_PATH_IMAGE005
is the normal lower limit value of the top layer oil temperature,
Figure 281164DEST_PATH_IMAGE006
is the upper limit value of the top layer oil temperature,
Figure 548197DEST_PATH_IMAGE007
is a counting identifier.
And S13, determining the corresponding second top oil temperature table state index by adopting the curve state data.
Second top-level oil temperature gauge status indicator, referring to timeT 1 The number of times the internal load changes but the top layer oil temperature value does not change.
Calculating timeT 1 Number of times that internal load changes but top layer oil temperature value does not change:
Figure 49893DEST_PATH_IMAGE008
，
Figure 781089DEST_PATH_IMAGE002
in the formula,
Figure 774452DEST_PATH_IMAGE009
is time of dayT 1 The number of times that the internal load changes but the top layer oil temperature value does not change,
Figure 341831DEST_PATH_IMAGE010
for the numerical value of the top oil temperature table of the main transformer at the moment t-1 to be judged,
Figure 78843DEST_PATH_IMAGE011
to be able to determine the value of the main transformer load at time t,
Figure 664545DEST_PATH_IMAGE012
the value of the main transformer load at the time t-1.
And S14, determining the corresponding state index of a third top-layer oil temperature meter by adopting curve state data corresponding to two adjacent top-layer oil temperature meters in the main transformer to be judged.
The third top oil temperature meter state index refers to timeT 1 The number of times that the numerical difference value of 2 top-layer oil temperature meters of the main transformer to be judged internally exceeds 5 ℃ is determined.
Calculating timeT 1 The times that the numerical difference value of 2 top-layer oil temperature meters of the main transformer to be judged internally exceeds 5 ℃ are as follows:
Figure 405974DEST_PATH_IMAGE013
，
Figure 647599DEST_PATH_IMAGE002
in the formula,
Figure 985040DEST_PATH_IMAGE014
is time of dayT 1 The times that the numerical difference value of 2 top oil temperature meters of the main transformer to be judged exceeds 5 ℃ are counted,
Figure 441560DEST_PATH_IMAGE015
to be judged the value of the top oil temperature table 1 of the main transformer at the time t,
Figure 776726DEST_PATH_IMAGE016
the value of the top-layer oil temperature table 2 of the main transformer at the moment t is to be judged.
And S15, selecting data of two adjacent moments in the curve state data, and determining the corresponding state index of the fourth top-layer oil temperature meter.
Fourth top-level oil temperature gauge status indicator, which refers to timeT 1 And the numerical value of the oil temperature of the inner top layer changes the times.
Calculating timeT 1 Inner top layer oil temperature numerical value mutation times:
Figure 567965DEST_PATH_IMAGE017
，
Figure 646779DEST_PATH_IMAGE002
in the formula,
Figure 459271DEST_PATH_IMAGE018
is time of dayT 1 And the numerical value of the oil temperature of the inner top layer changes the times.
And S16, selecting data of three adjacent moments in the curve state data, and determining the corresponding state index of the fifth top oil temperature table.
The fifth top oil temperature table state index refers to timeT 1 The number of inflection points of the oil temperature numerical curve of the inner top layer.
Calculating timeT 1 The number of inflection points of the oil temperature numerical curve of the inner top layer is as follows:
Figure 293235DEST_PATH_IMAGE019
，
Figure 243873DEST_PATH_IMAGE002
in the formula,
Figure 939428DEST_PATH_IMAGE020
is time of dayT 1 The number of inflection points of the oil temperature numerical curve of the inner top layer,
Figure 291912DEST_PATH_IMAGE021
the numerical value of the top oil temperature table of the main transformer at the moment t-2 is to be judged.
And S17, determining a corresponding sixth top layer oil temperature table state index by using curve state data.
The sixth top oil temperature table state index refers to timeT 1 And the number of times that the oil temperature value of the inner top layer is lower than the ambient temperature.
Calculating timeT 1 The times that the oil temperature value of the inner top layer is lower than the ambient temperature are as follows:
Figure 296777DEST_PATH_IMAGE022
，
Figure 734711DEST_PATH_IMAGE002
in the formula,
Figure 466913DEST_PATH_IMAGE023
is time of dayT 1 The number of times that the oil temperature value of the inner top layer is lower than the ambient temperature,
Figure 1800DEST_PATH_IMAGE024
is the value of the ambient temperature at time t.
And 204, comparing the state indexes of the top oil temperature meter with preset standard state indexes, and screening the main transformers to be analyzed except the top oil temperature meter abnormity from the main transformers to be judged according to the comparison result.
Further, step 204 may include the following sub-steps:
and S21, comparing the state index of the top oil temperature table with the corresponding preset standard state index.
And S22, if the state index of any top oil temperature meter is larger than the corresponding standard state index, judging that the top oil temperature meter of the main transformer to be judged is abnormal.
And S23, determining all the main transformers to be judged, of which the state indexes of the top oil temperature meter are less than or equal to the corresponding standard state indexes, as the main transformers to be analyzed, of which the oil temperatures are abnormal.
In the embodiment of the invention, the preset standard state indexes comprise an oil temperature value overrun frequency threshold value, a load change but top layer oil temperature invariable frequency threshold value, a difference overrun frequency threshold value, a mutation occurrence frequency threshold value, an inflection point occurrence frequency threshold value and an oil temperature value lower than an annular temperature value frequency threshold value.
In one example of the invention:
judging whether the reading of the top oil temperature meter exceeds a normal limit value or not, specifically comprising the following steps: if it is
Figure 115249DEST_PATH_IMAGE025
Represents timeT 1 And if the number of times of the oil temperature value of the inner top layer exceeds the threshold value of the number of times of the oil temperature value exceeding, judging that the top-layer thermometer is abnormal. Wherein,
Figure 853529DEST_PATH_IMAGE026
the threshold value of the number of times of the oil temperature value overrun can be selected according to actual needs, such as 0.05%T 1 。
The linkage condition of the top oil temperature of the main transformer to be judged and the load of the main transformer to be judged is judged, and the linkage condition comprises the following steps: if it is
Figure 77837DEST_PATH_IMAGE027
Represents timeT 1 And if the load change occurs in the top layer oil temperature table but the top layer oil temperature constant frequency is larger than the threshold value of the load change but the top layer oil temperature constant frequency, judging that the top layer thermometer is abnormal. Wherein,
Figure 201651DEST_PATH_IMAGE028
the threshold value of the number of times of the load change but the top oil temperature is not changed can be selected according to actual needs, such as 0.05 ×T 1 。
The difference conditions among the 2 top-layer oil temperature meters are judged as follows: if it is
Figure 486002DEST_PATH_IMAGE029
Indicates that the main transformer to be judged has2 top oil temperature gauges and timeT 1 And if the times that the internal difference value exceeds 5 ℃ is greater than the difference value overrun time threshold value, judging that the top oil temperature meter is abnormal. Wherein,
Figure 213043DEST_PATH_IMAGE030
the threshold value for the number of times of difference overrun can be selected according to actual needs, such as 0.05%T 1 。
Judging the top oil temperature mutation condition specifically as follows: if it is
Figure 303358DEST_PATH_IMAGE031
Represents timeT 1 And if the mutation frequency of the oil temperature numerical value of the inner top layer is larger than the mutation frequency threshold value, judging that the oil temperature table of the top layer is abnormal. Wherein,
Figure 484941DEST_PATH_IMAGE032
the threshold value for the number of mutations can be selected according to practical requirements, such as 0.05%T 1 。
Judging the fluctuation condition of the top oil temperature as follows: if it is
Figure 487663DEST_PATH_IMAGE033
Represents timeT 1 And if the inflection point occurrence frequency of the inner top layer oil temperature curve is greater than the inflection point occurrence frequency threshold value, judging that the top layer oil temperature gauge is abnormal. Wherein,
Figure 387486DEST_PATH_IMAGE034
the threshold value for the number of times of inflection point occurrence can be selected according to practical requirements, such as 0.1%T 1 。
Whether the oil temperature of the top layer of the main transformer is lower than the ambient temperature is judged, and the method specifically comprises the following steps: if it is
Figure 15913DEST_PATH_IMAGE035
Represents timeT 1 And if the times that the oil temperature value of the inner top layer is lower than the environmental temperature are greater than the threshold value of the times that the oil temperature value is lower than the ring temperature value, judging that the top layer oil temperature meter is abnormal. Wherein,
Figure 629166DEST_PATH_IMAGE036
the threshold value of the times of the oil temperature value being lower than the ring temperature value can be selected according to actual needs, such as 0.05%T 1 。
And sequentially selecting all main transformers to be judged for judgment. Here, for example, a main transformer to be determined is selected for determination, which includes the following steps:
comparing the state index of the top oil temperature meter with the corresponding preset standard state index, such as sequentially comparing
Figure 989740DEST_PATH_IMAGE003
And
Figure 439176DEST_PATH_IMAGE026
、
Figure 808978DEST_PATH_IMAGE009
and
Figure 512623DEST_PATH_IMAGE028
、
Figure 309677DEST_PATH_IMAGE014
and
Figure 980830DEST_PATH_IMAGE030
、
Figure 468837DEST_PATH_IMAGE018
and
Figure 479518DEST_PATH_IMAGE032
、
Figure 509791DEST_PATH_IMAGE020
Figure 605923DEST_PATH_IMAGE034
、
Figure 130576DEST_PATH_IMAGE023
and
Figure 58081DEST_PATH_IMAGE036
。
if it occurs
Figure 196938DEST_PATH_IMAGE025
、
Figure 91951DEST_PATH_IMAGE027
、
Figure 341667DEST_PATH_IMAGE029
、
Figure 389257DEST_PATH_IMAGE031
、
Figure 699016DEST_PATH_IMAGE033
、
Figure 582789DEST_PATH_IMAGE035
In one of the six conditions, judging that the top oil temperature meter of the main transformer to be judged is abnormal;
and determining all main transformers to be judged, of which the state indexes of the top oil temperature meter are less than or equal to the corresponding standard state indexes, as main transformers to be analyzed for oil temperature abnormity. In the present example, the preset standard status index refers to
Figure 636196DEST_PATH_IMAGE026
、
Figure 538293DEST_PATH_IMAGE028
、
Figure 345186DEST_PATH_IMAGE030
、
Figure 637627DEST_PATH_IMAGE032
、
Figure 557041DEST_PATH_IMAGE034
And
Figure 251328DEST_PATH_IMAGE036
the top oil temperature gauge state index refers to
Figure 715938DEST_PATH_IMAGE003
、
Figure 761255DEST_PATH_IMAGE009
、
Figure 218781DEST_PATH_IMAGE014
、
Figure 79158DEST_PATH_IMAGE018
、
Figure 167200DEST_PATH_IMAGE020
And
Figure 496550DEST_PATH_IMAGE023
。
and step 205, intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating target correlation indexes of the main transformers to be analyzed according to the curve to be processed, and constructing target correlation index vectors.
Further, the main transformer state curve includes a target top oil temperature curve, a target load curve, a target ambient temperature curve, a historical top oil temperature curve, a historical load curve and a historical ambient temperature curve, the target associated indexes include a first associated index, a second associated index, a third associated index, a fourth associated index, a fifth associated index and a sixth associated index, and step 205 may include the following steps:
and S31, intercepting a target top layer oil temperature curve and a target load curve in a preset time period as a curve to be processed, and calculating a corresponding first correlation index according to the curve to be processed.
For time of dayt 1 Intercept oft 1 -T 2 +1 tot 1 And assembling top layer oil temperature, load and environment temperature data of the main transformer into a calculated data curve.T 2 In order to calculate the length of the data,T 2 can be selected according to actual needs, such as 1 day. Wherein the data acquisition is performed at a frequencyThe rate was 15 min/point, and 1 day corresponded to 96 data points.
Calculating a first correlation index of the top oil temperature of the main transformer and the load of the main transformer:
Figure 695450DEST_PATH_IMAGE037
in the formula,
Figure 911799DEST_PATH_IMAGE038
the correlation index of the top layer oil temperature curve of the main transformer and the load curve of the main transformer is shown,
Figure 233059DEST_PATH_IMAGE039
for the main transformer i in length ofT 2 The a-th value in the top oil temperature curve of (1),
Figure 987388DEST_PATH_IMAGE040
the main transformer i has a length ofT 2 The a-th value in the load curve of (1).
And S32, intercepting a target top layer oil temperature curve and a target environment temperature curve in a preset time period as a curve to be processed, and calculating a corresponding second correlation index according to the curve to be processed.
Calculating a second correlation index of the top oil temperature and the ambient temperature of the main transformer:
Figure 38914DEST_PATH_IMAGE042
in the formula,
Figure 562300DEST_PATH_IMAGE043
is the correlation index of the top layer oil temperature curve and the environment temperature curve of the main transformer i,
Figure 54461DEST_PATH_IMAGE044
is in a length ofT 2 The a-th value in the ring temperature curve of (1).
And S33, intercepting a target top layer oil temperature curve and a historical top layer oil temperature curve in a preset time period as a curve to be processed, and calculating a corresponding third correlation index according to the curve to be processed.
S33a, extracting the same date and time length of the current main transformer to be analyzed in the 1 st year after operationT 1 The historical data of the main transformer to be analyzed is divided into n time lengthsT 2 Respectively constructing a main transformer top layer oil temperature curve under each historical scene. Wherein,n=T 1 /T 2 。
s33b, respectively calculating correlation indexes of the top oil temperature of the main transformer to be analyzed and the top oil temperature curve of the main transformer in the n historical scenes, and calculating an index average value.
Calculating a third correlation index of the top oil temperature of the main transformer and the historical top oil temperature of the main transformer:
Figure 30507DEST_PATH_IMAGE046
in the formula,
Figure 384259DEST_PATH_IMAGE047
the correlation index of the top layer oil temperature curve of the main transformer i and the historical top layer oil temperature curve of the main transformer i is obtained,
Figure 90047DEST_PATH_IMAGE048
the correlation index of the top oil temperature curve of the main transformer i and the top oil temperature curve of the main transformer i in the historical scene j,
Figure 425213DEST_PATH_IMAGE049
for a historical scene j, the length of a main transformer i isT 2 The a-th value in the top oil temperature curve of (1).
And S34, intercepting the target load curve and the historical load curve in the preset time period as to-be-processed curves, and calculating corresponding fourth correlation indexes according to the to-be-processed curves.
S34a, acquiring n historical calculation scenes according to the method of S33a, and respectively establishing main transformer load curves under the historical scenes;
and S34b, respectively calculating the associated indexes of the main transformer load and the main transformer load curves in the n historical scenes, and calculating the average value of the indexes.
Calculating a fourth correlation index of the main transformer load and the main transformer historical load:
Figure 465720DEST_PATH_IMAGE050
Figure 544534DEST_PATH_IMAGE051
in the formula,
Figure 104828DEST_PATH_IMAGE052
the correlation index of the main transformer i load curve and the main transformer i historical load curve is obtained,
Figure 423945DEST_PATH_IMAGE053
the correlation index of the main transformer i load curve and the main transformer i load curve in the historical scene j,
Figure 640163DEST_PATH_IMAGE054
the length of a main transformer i under a historical scene j isT 2 The a-th value in the load curve of (1).
And S35, intercepting target top layer oil temperature curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding fifth correlation indexes according to the curves to be processed.
Calculating a fifth correlation index of the top layer oil temperature of the main transformer and the top layer oil temperature of the adjacent main transformer of the same transformer substation:
Figure 584985DEST_PATH_IMAGE056
in the formula,
Figure 937469DEST_PATH_IMAGE057
for main transformer i topThe correlation indexes of the layer oil temperature curve and the top layer oil temperature curve of the adjacent main transformer of the same transformer substation are obtained;
Figure 194532DEST_PATH_IMAGE058
the length of adjacent main transformers i' of the same transformer station isT 2 The a-th value in the top layer oil temperature curve of (1).
In the present embodiment, 2 main substations of the same substation are exemplified, and only 1 substation is calculated
Figure 632466DEST_PATH_IMAGE057
And (4) indexes. If a transformer substation has a plurality of adjacent main transformers, a plurality of main transformers are calculated
Figure 115400DEST_PATH_IMAGE057
And (4) indexes.
And S36, intercepting target load curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding sixth correlation indexes according to the curves to be processed.
Calculating a sixth correlation index of the main transformer load and the adjacent main transformer load of the same transformer substation:
Figure 401019DEST_PATH_IMAGE059
in the formula,
Figure 248889DEST_PATH_IMAGE060
the correlation index of a main transformer i load curve and an adjacent main transformer load curve of the same transformer substation is obtained;
Figure 236437DEST_PATH_IMAGE061
the length of adjacent main transformers i' of the same transformer station isT 2 The a-th value in the load curve of (1).
Similarly, in the present embodiment, 2 main substations in the same substation are exemplified, and only 1 substation is calculated
Figure 460745DEST_PATH_IMAGE060
And (4) indexes. If a transformer substation has a plurality of adjacent main transformers, a plurality of main transformers are calculated
Figure 99405DEST_PATH_IMAGE060
And (4) indexes.
And S37, constructing a corresponding target associated index vector by adopting the first associated index, the second associated index, the third associated index, the fourth associated index, the fifth associated index and the sixth associated index.
In the embodiment of the invention, for the main transformer i, a target associated index vector is constructed by combining the main transformer top layer oil temperature monitoring characteristic index of the steps S31 to S36
Figure 383756DEST_PATH_IMAGE038
、
Figure 858600DEST_PATH_IMAGE043
、
Figure 434069DEST_PATH_IMAGE047
、
Figure 615652DEST_PATH_IMAGE052
、
Figure 133221DEST_PATH_IMAGE057
、
Figure 33043DEST_PATH_IMAGE060
]. And all main transformers to be analyzed sequentially construct target correlation index vectors.
And step 206, determining typical correlation index vectors according to the clustering result of the historical correlation index vectors of the main transformers to be analyzed, and calculating target correlation index vectors and target category distance values of the typical correlation index vectors.
Further, step 206 comprises the following sub-steps:
and S41, selecting initial clustering centers according with the number of centers from a plurality of historical associated index vectors of the main transformer to be analyzed.
And S42, calculating the initial distance between each historical associated index vector and each initial clustering center, and classifying each historical associated index vector to a clustering class with the minimum initial distance.
And S43, calculating the middle clustering center of each clustering class.
And S44, if the middle clustering centers are not changed, calculating effective clustering indexes according to the initial distance and the middle clustering centers.
And step S45, if the number of the cluster types is larger than a preset type threshold value and the effective clustering index is larger than or equal to the effective clustering index corresponding to the number of the previous cluster types, taking a middle cluster center corresponding to the number of the previous cluster types as a typical correlation index vector, and calculating a target correlation index vector and a target type distance value of each typical correlation index vector.
Calculating a target class distance value:
Figure 913668DEST_PATH_IMAGE062
in the formula,
Figure 12074DEST_PATH_IMAGE063
is mainly changed into it 1 The distance between the target association index vector at the moment and the f-th typical association index vector of the main transformer;
Figure 638228DEST_PATH_IMAGE064
is mainly changed into it 1 The xth characteristic index in the target association index vector at the moment;
Figure 838396DEST_PATH_IMAGE065
and the characteristic index is the x-th characteristic index in the f-th typical correlation index vector of the main transformer i.
Further, step 206 further comprises the following sub-steps:
and S46, if the number of the cluster types is less than or equal to a preset type threshold value, or the cluster effective index is less than the cluster effective index corresponding to the previous cluster type number, increasing the number of centers, and skipping to execute the step of selecting the initial cluster centers which accord with the number of centers from a plurality of historical associated index vectors of the main transformer to be analyzed.
In one example of the present invention, please refer to fig. 3 in detail as follows:
A. extracting the 1 st annual history data of the main transformer to be analyzed after operation one by one, and assembling the top oil temperature, load and environment temperature data of the main transformer into a calculation data curve.
B. According to step 205, a historical association index vector of each historical time of the main transformer to be analyzed is calculated.
C. And determining a plurality of corresponding clustering centers by using a k-means algorithm in combination with the historical associated index vector of the main transformer to be analyzed.
In this example, the primary transformers are all referred to as primary transformers, and the eigenvectors are all referred to as historical association index vectors.
C1, enabling the serial number of a main transformer i =1 and the serial number of a feature vector e =1, counting the total number of main transformers to be analyzed to be N, the total number of feature vectors to be analyzed of the main transformer i to be M, and setting the feature vector set of the main transformer i to be a ready pocketI i1 ,…,I ie ,…,I iM }. Wherein, the ith characteristic vector of the main transformer is as follows:
Figure 942618DEST_PATH_IMAGE066
in the formula,
Figure 161110DEST_PATH_IMAGE067
、
Figure 269749DEST_PATH_IMAGE068
、
Figure 878585DEST_PATH_IMAGE069
、
Figure 114394DEST_PATH_IMAGE070
、
Figure 125076DEST_PATH_IMAGE071
and
Figure 906081DEST_PATH_IMAGE072
and step 205, calculating characteristic indexes of the e-th moment of the main transformer i respectively.
And C2, if i is less than or equal to N, enabling the number of the clustering centers to be k =2, and otherwise, executing the step C11.
C3, randomly extracting k eigenvectors from the eigenvector set of the main transformer i as k initial clustering centersu 1 ,…,u f ,…,u k And initializing k cluster categories as great faceC 1 ,…,C f ,…,C k }。
C4, if e is less than or equal to M, calculating the distance between the characteristic vector e and the clustering center fd ef （f=1,…,k) Otherwise, executing step C6;
wherein, the distance between the characteristic vector e and the clustering center f
Figure 2213DEST_PATH_IMAGE073
The calculation expression is:
Figure 510555DEST_PATH_IMAGE074
in the formula,
Figure 944117DEST_PATH_IMAGE073
is the distance of the feature vector e from the cluster center f,
Figure 82974DEST_PATH_IMAGE075
the x-th characteristic index in the characteristic vector e of the main transformer i is obtained;
Figure 728719DEST_PATH_IMAGE076
is the x-th characteristic index of the clustering center f.
C5, orderd em =min{d e1 ,…,d ef ,…,d ek M is the cluster center number closest to the characteristic vector e, the characteristic vector e is classified into a cluster type m, and the cluster type m is updatedC m Let e = e +1, execute step C4.
C6, recalculating the clustering centers of all clustering categories:
Figure 978435DEST_PATH_IMAGE077
in the formula,
Figure 776758DEST_PATH_IMAGE078
for the f-th cluster center, the cluster center,
Figure 148833DEST_PATH_IMAGE079
for the f-th cluster category,
Figure 219558DEST_PATH_IMAGE080
and the e-th feature vector of the main transformer i.
And C7, if the k clustering centers are not changed, executing a step C8, otherwise, enabling e =1, and executing a step C4.
C8, calculating the clustering effectiveness index according to the clustering result
Figure 584549DEST_PATH_IMAGE081
；
Figure 486646DEST_PATH_IMAGE082
Figure 967306DEST_PATH_IMAGE083
Figure 72796DEST_PATH_IMAGE084
Figure 929893DEST_PATH_IMAGE085
In the formula,
Figure 686497DEST_PATH_IMAGE086
、
Figure 652572DEST_PATH_IMAGE087
for the effectiveness index 1 and the index 2,
Figure 697889DEST_PATH_IMAGE088
the distance between the cluster center f and the cluster center y,
Figure 155415DEST_PATH_IMAGE089
is the x-th characteristic index of the clustering center y.
C9 if k>2, comparison
Figure 704208DEST_PATH_IMAGE081
And
Figure 605299DEST_PATH_IMAGE090
numerical value and step C10 is performed, otherwise, k = k +1, step C3 is performed.
C10, if
Figure 872332DEST_PATH_IMAGE081
＜
Figure 867970DEST_PATH_IMAGE090
If so, let k = k +1, execute step C3, otherwise, output the clustering Result when k-1 i And the number Num of cluster centers i And taking the clustering center at the moment as a typical correlation index vector Sample of the main transformer i i Let i = i +1, step C2 is executed.
Wherein,
Figure 848433DEST_PATH_IMAGE090
is a cluster effective index corresponding to the number of the previous cluster types,
Figure 107376DEST_PATH_IMAGE081
is a clustering effective index.
And C11, obtaining typical correlation index vectors of all main transformers to be analyzed, namely obtaining a plurality of corresponding clustering centers.
Further, still include:
s1, calculating initial distance values between a plurality of historical associated index vectors and each typical associated index vector.
And S2, respectively selecting the maximum initial distance value in the cluster category to which each typical correlation index vector belongs.
And S3, respectively calculating the multiplication value between each maximum initial distance value and a preset threshold coefficient to serve as the category distance threshold of the cluster category at the current moment.
In an example of the invention, a main transformer i target association index vector clustering Result is obtained by combining with the calculation of the step C10 i Calculating a category distance threshold for each cluster category
Figure 658443DEST_PATH_IMAGE091
。
Calculating a category distance threshold:
Figure 661034DEST_PATH_IMAGE092
Figure 997469DEST_PATH_IMAGE093
in the formula,
Figure 224051DEST_PATH_IMAGE094
clustering categories for primary transformers i
Figure 465676DEST_PATH_IMAGE079
A category distance threshold of (a); a is a threshold coefficient, which can be selected according to actual needs, such as a value of 1.2.
And step 207, comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the top layer oil temperature of the main transformer to be analyzed is abnormal.
Further, step 207 may comprise the following sub-steps:
and S51, comparing the target class distance value with a plurality of preset class distance thresholds.
And S52, if the target category distance value is smaller than or equal to any category distance threshold value, judging that the top layer oil temperature abnormity does not exist in the main transformer to be analyzed.
And S53, if the target category distance values are all larger than all category distance threshold values, judging that top layer oil temperature abnormity exists in the main transformer to be analyzed.
In one example of the present invention, a target class distance value is determined
Figure 320893DEST_PATH_IMAGE095
Distance from class threshold
Figure 761102DEST_PATH_IMAGE094
Determining the abnormal main transformer of the top oil temperature:
calculating a target class distance value
Figure 361848DEST_PATH_IMAGE063
Distance from class threshold
Figure 638239DEST_PATH_IMAGE094
Target determination value therebetween:
Figure 982633DEST_PATH_IMAGE096
in the formula,
Figure 542927DEST_PATH_IMAGE097
is a target judgment value, and is,
Figure 360579DEST_PATH_IMAGE063
is a distance value for the object class,
Figure 576797DEST_PATH_IMAGE094
is a preset threshold value of the category distance,
Figure 521619DEST_PATH_IMAGE098
the number of clustering centers.
When the target category distance value is smaller than or equal to the category distance threshold, the value is 1, and when the target category distance value is larger than the category distance threshold, the value is 0.
If the target class distance value is less than or equal to any preset class distance threshold value, the target judgment value
Figure 874103DEST_PATH_IMAGE097
If not, judging that the top layer oil temperature abnormality does not exist in the main transformer to be analyzed;
if the target category distance values are all larger than all category distance threshold values, the target judgment value
Figure 364122DEST_PATH_IMAGE097
And if the oil temperature is equal to 0, judging that the top oil temperature abnormity exists in the main transformer to be analyzed.
And step 208, counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results.
In the embodiment of the invention, the main transformers with abnormal top oil temperature and the main transformers with abnormal top oil temperature meter are counted and summarized, and abnormal results are output.
It should be mentioned that, in this embodiment, the basic state data of the main transformers to be determined in the preset period is obtained as the to-be-processed data at the current time. And aiming at the continuously updated data stream, rolling to carry out the main transformer top layer oil temperature abnormity monitoring and top layer oil temperature meter abnormity checking process by adopting a sliding window method.
a. The fixed window width is set asT 1 For time of day, sliding window oft 1 By sliding window extraction of fixed window widtht 1 -T 1 +1 tot 1 Assembling the data into calculation data; for time of dayt 1 +1, by fixingWindow width sliding window extractiont 1 -T 1 +2 tot 1 +And assembling the data between 1 into calculation data, and so on.
b. And (3) rolling and assembling a main transformer top oil temperature, load and ambient temperature curve by combining a sliding window, and screening an abnormal top oil temperature table and an abnormal oil temperature main transformer to be analyzed in step 204.
c. The fixed window width is set asT 2 The sliding window is used for rolling and assembling a top oil temperature, load and ambient temperature curve of the main transformer, and the main transformer with abnormal oil temperature is screened out by adopting the steps 205 to 207.
In the embodiment of the invention, basic state data of a plurality of main transformers to be judged in a preset period are obtained and preprocessed, and a main transformer state curve for calculation at the current moment is generated; extracting a plurality of curve state data from a main transformer state curve, and calculating a top oil temperature table state index based on the curve state data; comparing the state index of the top oil temperature meter with a preset standard state index, and screening main transformers to be analyzed except the top oil temperature meter abnormality from the plurality of main transformers to be judged according to the comparison result; intercepting a main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector; determining typical correlation index vectors according to clustering results of historical correlation index vectors of main transformers to be analyzed, and calculating target correlation index vectors and target category distance values of the typical correlation index vectors; comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the top layer oil temperature of the main transformer to be analyzed is abnormal or not; counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results; the method solves the problems that the existing fixed threshold is difficult to flexibly and meticulously carry out differential monitoring and refined abnormal alarm of the top oil temperature of the main transformer, is not suitable for judging the abnormal state of the top oil temperature which does not reach the limit value, cannot find the abnormal state of the top oil temperature of the main transformer in time, leaves potential safety hazards for safe and reliable operation of equipment, has a vacuum period for monitoring the state of an oil temperature meter, is difficult to find the abnormal state of the top oil temperature meter in time, has the safety hazards of unreliable oil temperature monitoring, unreal monitoring results and unknown equipment states, seriously influences the safe and reliable operation of the main transformer, in addition, manual field tests consume manpower and material resources, have long process and time consumption, cannot fully release the advantages of online monitoring of the top oil temperature and the potential of a robot, and bring negative effects to the safe operation and reliable power supply of a power grid when the equipment is powered off along with the adjustment of a power grid operation mode; the top oil temperature abnormity monitoring method based on data analysis replaces a fixed monitoring threshold value, individuation and common characteristics of top oil temperature running states of all transformers are carefully excavated, refined monitoring of top oil temperature abnormity of a main transformer is achieved, meanwhile, a top oil temperature table abnormity checking method based on data excavation replaces periodic power outage prerun, online monitoring advantages of top oil temperature and robot potential are fully released, real-time monitoring and abnormal rolling checking of states of the top oil temperature table of the main transformer are achieved, state evaluation of the top oil temperature table is changed from ' power outage test ' to ' online monitoring ', and the purposes of improving equipment safety, reducing equipment power outage, reducing labor cost and increasing checking efficiency ' are achieved.
Referring to fig. 4, fig. 4 is a block diagram illustrating a main transformer oil temperature gauge and a device for determining an abnormal top oil temperature according to a third embodiment of the present invention.
The invention provides a device for judging the abnormity of oil temperature of a main transformer oil temperature gauge and a top oil temperature, which comprises:
the data preprocessing module 301 is configured to acquire and preprocess basic state data of a plurality of main transformers to be determined in a preset period, and generate a main transformer state curve for calculation at a current time. The top oil temperature table state index calculating module 302 is configured to extract a plurality of curve state data from the main transformer state curve, and calculate a top oil temperature table state index based on the curve state data. And the abnormal top oil temperature meter determining module 303 is used for comparing the state indexes of the top oil temperature meter with preset standard state indexes and screening the main transformers to be analyzed except the abnormal top oil temperature meter from the plurality of main transformers to be judged according to the comparison result. And a target correlation index vector construction module 304, configured to intercept the main transformer state curve in a preset time period as a to-be-processed curve, calculate a target correlation index of each to-be-analyzed main transformer according to the to-be-processed curve, and construct a target correlation index vector.
The target category distance value obtaining module 305 is configured to determine a typical correlation index vector according to a clustering result of the historical correlation index vectors of the main transformers to be analyzed, and calculate a target correlation index vector and a target category distance value of each typical correlation index vector. And a top layer oil temperature anomaly determination module 306, configured to compare the target category distance value with a plurality of preset category distance thresholds, and determine whether the top layer oil temperature anomaly exists in the main transformer to be analyzed. An abnormal result generating module 307, configured to count the main transformer with the abnormal top oil temperature and the main transformer with the abnormal top oil temperature table, and generate an abnormal result.
Further, the basic state data includes target data and historical data, and the data preprocessing module 301 includes: and the data acquisition submodule is used for acquiring target data and historical data of the main transformers to be judged in a preset period. The data processing submodule is used for respectively sequencing the target data and the historical data based on the data types and generating a main transformer state curve for calculation at the current moment; the data types include top oil temperature, load, and ambient temperature.
Further, the top-layer oil temperature table status indexes include a first top-layer oil temperature table status index, a second top-layer oil temperature table status index, a third top-layer oil temperature table status index, a fourth top-layer oil temperature table status index, a fifth top-layer oil temperature table status index, and a sixth top-layer oil temperature table status index, and the top-layer oil temperature table status index calculation module 302 includes: and the curve state data acquisition submodule is used for selecting a plurality of curve state data of a preset quantity from the main transformer state curve according to a time reverse sequence from the current moment. And the first top layer oil temperature meter state index acquisition submodule is used for determining a corresponding first top layer oil temperature meter state index by adopting curve state data and a preset oil temperature limit value. And the second top-layer oil temperature meter state index acquisition submodule is used for determining the corresponding second top-layer oil temperature meter state index by adopting curve state data. And the third top layer oil temperature meter state index acquisition submodule is used for determining the corresponding third top layer oil temperature meter state index by adopting curve state data corresponding to two adjacent top layer oil temperature meters in the main transformer to be judged. And the fourth top oil temperature meter state index acquisition submodule is used for selecting data of two adjacent moments in the curve state data and determining the corresponding fourth top oil temperature meter state index. And the fifth top-layer oil temperature meter state index acquisition submodule is used for selecting data of three adjacent moments in the curve state data and determining the corresponding fifth top-layer oil temperature meter state index. And the sixth top-layer oil temperature meter state index acquisition submodule is used for determining the corresponding sixth top-layer oil temperature meter state index by adopting curve state data.
Further, the abnormal top oil temperature table determining module 303 includes: and the first comparison sub-module is used for comparing the state index of the top oil temperature table with the corresponding preset standard state index. And the first judgment submodule is used for judging that the top oil temperature meter of the main transformer to be judged is abnormal if the state index of any top oil temperature meter is greater than the corresponding standard state index. And the second stator judging module is used for determining all the main transformers to be judged, of which the state indexes of the top oil temperature meter are less than or equal to the corresponding standard state indexes, as the main transformers to be analyzed, of which the oil temperatures are abnormal.
Further, the main transformer state curve includes a target top oil temperature curve, a target load curve, a target ambient temperature curve, a historical top oil temperature curve, a historical load curve and a historical ambient temperature curve, the target associated indexes include a first associated index, a second associated index, a third associated index, a fourth associated index, a fifth associated index and a sixth associated index, and the target associated index vector construction module 304 includes: and the first correlation index calculation submodule is used for intercepting a target top layer oil temperature curve and a target load curve in a preset time period as a curve to be processed and calculating a corresponding first correlation index according to the curve to be processed. And the second correlation index calculation submodule is used for intercepting a target top layer oil temperature curve and a target environment temperature curve in a preset time period as a curve to be processed and calculating a corresponding second correlation index according to the curve to be processed. And the third correlation index calculation submodule is used for intercepting a target top layer oil temperature curve and a historical top layer oil temperature curve in a preset time period as a curve to be processed and calculating a corresponding third correlation index according to the curve to be processed. And the fourth correlation index calculation submodule is used for intercepting the target load curve and the historical load curve in a preset time period as the curves to be processed and calculating a corresponding fourth correlation index according to the curves to be processed. And the fifth correlation index calculation submodule is used for intercepting target top layer oil temperature curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding fifth correlation indexes according to the curves to be processed. And the sixth correlation index calculation submodule is used for intercepting target load curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed and calculating corresponding sixth correlation indexes according to the curves to be processed. And the target correlation index vector construction submodule is used for constructing a corresponding target correlation index vector by adopting the first correlation index, the second correlation index, the third correlation index, the fourth correlation index, the fifth correlation index and the sixth correlation index.
Further, the object class distance value obtaining module 305 includes: and the selection submodule is used for selecting initial clustering centers according with the number of the centers from a plurality of historical association index vectors of the main transformer to be analyzed. And the classification submodule is used for calculating the initial distance between each historical association index vector and each initial clustering center and classifying each historical association index vector to the clustering category with the minimum initial distance. And the middle clustering center calculating submodule is used for calculating the middle clustering centers of all clustering categories. And the effective clustering index calculating submodule is used for calculating effective clustering indexes according to the initial distance and the middle clustering center if the middle clustering center is unchanged. And the target category distance value submodule is used for taking a middle clustering center corresponding to the previous clustering category quantity as a typical correlation index vector and calculating a target correlation index vector and a target category distance value of each typical correlation index vector if the quantity of the clustering categories is greater than a preset category threshold value and the clustering effective index is greater than or equal to the clustering effective index corresponding to the previous clustering category quantity.
Further, the object class distance value obtaining module 305 further includes: and the skipping submodule is used for increasing the number of centers if the number of the clustering categories is less than or equal to a preset category threshold value or the effective clustering index is less than the effective clustering index corresponding to the number of the previous clustering categories, and skipping to execute the step of selecting the initial clustering centers which accord with the number of centers from a plurality of historical associated index vectors of the main transformer to be analyzed.
Further, the top layer oil temperature anomaly determination module 306 includes: and the comparison submodule is used for comparing the target class distance value with a plurality of preset class distance thresholds. And the first top layer oil temperature abnormity determining submodule is used for determining that the main transformer to be analyzed has no top layer oil temperature abnormity if the target category distance value is smaller than or equal to any category distance threshold value. And the second top layer oil temperature abnormity determination submodule is used for determining that the top layer oil temperature abnormity exists in the main transformer to be analyzed if the target category distance values are all larger than all the category distance threshold values.
Further, still include: and the initial distance value calculation module is used for calculating initial distance values between the historical associated index vectors and the typical associated index vectors. The maximum initial distance value determining module is used for respectively selecting the maximum initial distance values in the clustering categories to which the typical associated index vectors belong; and the category distance threshold calculation module is used for calculating the multiplication value between each maximum initial distance value and a preset threshold coefficient as a category distance threshold of the cluster category at the current moment.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (9)

1. A main transformer oil temperature gauge and a top layer oil temperature abnormity judging method are characterized by comprising the following steps:
acquiring basic state data of a plurality of main transformers to be judged in a preset period, preprocessing the basic state data, and generating a main transformer state curve for calculation at the current moment;
extracting a plurality of curve state data from the main transformer state curve, and calculating the state index of the top oil temperature table based on the curve state data;
comparing the state index of the top oil temperature meter with a preset standard state index, and screening main transformers to be analyzed except the top oil temperature meter abnormality from the plurality of main transformers to be judged according to the comparison result;
intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating a target correlation index of each main transformer to be analyzed according to the curve to be processed, and constructing a target correlation index vector;
determining typical associated index vectors according to clustering results of historical associated index vectors of the main transformer to be analyzed, and calculating target associated index vectors and target category distance values of the typical associated index vectors;
comparing the target category distance value with a plurality of preset category distance thresholds, and judging whether the main transformer to be analyzed has top layer oil temperature abnormity;
counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter to generate abnormal results;
the top oil temperature table state indexes comprise a first top oil temperature table state index, a second top oil temperature table state index, a third top oil temperature table state index, a fourth top oil temperature table state index, a fifth top oil temperature table state index and a sixth top oil temperature table state index, and the step of extracting a plurality of curve state data from the main transformer state curve and calculating the top oil temperature table state indexes based on the curve state data comprises the following steps:
selecting a plurality of curve state data of a preset quantity from the main transformer state curve according to a time reverse sequence from the current moment;
determining a corresponding state index of the first top-layer oil temperature meter by adopting target curve data and a preset oil temperature limit value;
determining a corresponding second top-layer oil temperature meter state index by adopting the target curve data;
determining the state index of the third top-layer oil temperature meter by adopting the target curve data corresponding to two adjacent top-layer oil temperature meters in the main transformer to be judged;
selecting data of two adjacent moments in the target curve data, and determining corresponding state indexes of the fourth top-layer oil temperature meter;
selecting data of three adjacent moments in the target curve data, and determining corresponding state indexes of the fifth top-layer oil temperature meter;
and determining the corresponding state index of the sixth top-layer oil temperature meter by using the target curve data.
2. The method for determining the main transformer oil temperature gauge and the top layer oil temperature abnormality according to claim 1, wherein the basic state data includes target data and historical data, and the step of obtaining and preprocessing the basic state data of a plurality of main transformers to be determined in a preset period to generate a main transformer state curve for calculating at the current time includes:
acquiring the target data and the historical data of a plurality of main transformers to be judged in a preset period;
respectively sequencing the target data and the historical data based on data types to generate a main transformer state curve for calculation at the current moment; the data types include top oil temperature, load, and ambient temperature.
3. The method according to claim 1, wherein the step of comparing the status indicator of the top oil temperature gauge with a predetermined standard status indicator and selecting the main transformers to be analyzed from the plurality of main transformers to be determined according to the comparison result, comprises:
comparing the state index of the top oil temperature meter with corresponding preset standard state data;
if any one of the top oil temperature gauge state indexes is larger than the corresponding standard state data, judging that the top oil temperature gauge of the main transformer to be judged is abnormal;
and determining all the main transformers to be judged, of which the state indexes of the top oil temperature meter are less than or equal to the corresponding standard state data, as the main transformers to be analyzed for abnormal oil temperature.
4. The method for determining the main transformer oil temperature gauge and the top layer oil temperature abnormality according to claim 1, wherein the main transformer state curve includes a target top layer oil temperature curve, a target load curve, a target environment temperature curve, a historical top layer oil temperature curve, a historical load curve and a historical environment temperature curve, the target associated indicators include a first associated indicator, a second associated indicator, a third associated indicator, a fourth associated indicator, a fifth associated indicator and a sixth associated indicator, the step of intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating the target associated indicator of each main transformer to be analyzed according to the curve to be processed, and constructing a target associated indicator vector includes:
intercepting the target top layer oil temperature curve and the target load curve in a preset time period as a curve to be processed, and calculating the corresponding first correlation index according to the curve to be processed;
intercepting the target top layer oil temperature curve and the target environment temperature curve in a preset time period as a curve to be processed, and calculating corresponding second correlation indexes according to the curve to be processed;
intercepting the target top layer oil temperature curve and the historical top layer oil temperature curve in a preset time period as a curve to be processed, and calculating a corresponding third correlation index according to the curve to be processed;
intercepting the target load curve and the historical load curve in a preset time period as curves to be processed, and calculating corresponding fourth correlation indexes according to the curves to be processed;
intercepting the target top layer oil temperature curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding fifth correlation indexes according to the curves to be processed;
intercepting the target load curves corresponding to two adjacent main transformers to be analyzed of the same transformer substation in a preset time period as curves to be processed, and calculating corresponding sixth correlation indexes according to the curves to be processed;
and constructing corresponding target relevance index vectors by adopting the first relevance index, the second relevance index, the third relevance index, the fourth relevance index, the fifth relevance index and the sixth relevance index.
5. The method according to claim 1, wherein the step of determining a typical associated indicator vector according to a clustering result of historical associated indicator vectors of the main transformers to be analyzed and calculating a target associated indicator vector and a target category distance value of each typical associated indicator vector comprises:
selecting initial clustering centers according with the number of centers from a plurality of historical associated index vectors of each main transformer to be analyzed;
calculating an initial distance between each historical relevance index vector and each initial clustering center, and classifying each historical relevance index vector into a clustering category with the minimum initial distance;
calculating a middle clustering center of each clustering category;
if the intermediate clustering centers are not changed, calculating effective clustering indexes according to the initial distance and the intermediate clustering centers;
and if the number of the cluster types is larger than a preset type threshold value and the effective clustering index is larger than or equal to the effective clustering index corresponding to the number of the previous cluster types, taking a middle cluster center corresponding to the number of the previous cluster types as a typical correlation index vector, and calculating a target correlation index vector and a target type distance value of each typical correlation index vector.
6. The method for determining the main transformer oil temperature gauge and the top oil temperature abnormality according to claim 5, further comprising:
and if the number of the clustering categories is less than or equal to a preset category threshold value, or the effective clustering index is less than the effective clustering index corresponding to the previous clustering category number, increasing the number of the centers, and skipping to execute the step of selecting the initial clustering centers which accord with the number of the centers from the plurality of historical associated index vectors of each main transformer to be analyzed.
7. The method for determining the main transformer oil temperature gauge and the top layer oil temperature anomaly according to claim 1, wherein the step of comparing the target category distance value with a plurality of predetermined category distance thresholds to determine whether the main transformer to be analyzed has the top layer oil temperature anomaly comprises:
comparing the target class distance value with a plurality of preset class distance thresholds;
if the target category distance value is smaller than or equal to any one category distance threshold value, judging that the main transformer to be analyzed does not have top layer oil temperature abnormity;
and if the target category distance values are all larger than all the category distance threshold values, judging that the top layer oil temperature abnormity exists in the main transformer to be analyzed.
8. The method for determining an abnormal main transformer oil temperature gauge and top oil temperature as claimed in claim 1, further comprising:
calculating an initial distance value between a plurality of historical relevance index vectors and each typical relevance index vector;
respectively selecting the maximum initial distance value in the clustering category to which each typical correlation index vector belongs;
and respectively calculating the multiplication value between each maximum initial distance value and a preset threshold coefficient as the class distance threshold of the cluster class at the current moment.
9. The utility model provides a main transformer oil temperature table and top layer oil temperature anomaly's judgement device which characterized in that includes:
the data preprocessing module is used for acquiring and preprocessing basic state data of a plurality of main transformers to be judged in a preset period to generate a main transformer state curve for calculation at the current moment;
the top oil temperature table state index calculating module is used for extracting a plurality of curve state data from the main transformer state curve and calculating the top oil temperature table state index based on the curve state data;
the abnormal top oil temperature meter determining module is used for comparing the state index of the top oil temperature meter with a preset standard state index and screening main transformers to be analyzed except the top oil temperature meter abnormality from the plurality of main transformers to be judged according to the comparison result;
the target associated index vector construction module is used for intercepting the main transformer state curve in a preset time period as a curve to be processed, calculating a target associated index of each main transformer to be analyzed according to the curve to be processed, and constructing a target associated index vector;
the target category distance value acquisition module is used for determining typical associated index vectors according to clustering results of historical associated index vectors of the main transformer to be analyzed and calculating target associated index vectors and target category distance values of the typical associated index vectors;
the top layer oil temperature abnormity determining module is used for comparing the target category distance value with a plurality of preset category distance thresholds and judging whether the main transformer to be analyzed has top layer oil temperature abnormity;
the abnormal result generation module is used for counting the main transformer with abnormal top oil temperature and the main transformer with abnormal top oil temperature meter and generating an abnormal result;
the top oil temperature table state indexes comprise a first top oil temperature table state index, a second top oil temperature table state index, a third top oil temperature table state index, a fourth top oil temperature table state index, a fifth top oil temperature table state index and a sixth top oil temperature table state index, and the top oil temperature table state index calculation module comprises:
the curve state data acquisition submodule is used for selecting a plurality of curve state data of a preset quantity from the main transformer state curve according to a time reverse sequence from the current moment;
the first top oil temperature meter state index acquisition submodule is used for determining a corresponding first top oil temperature meter state index by adopting target curve data and a preset oil temperature limit value;
the second top-layer oil temperature meter state index acquisition submodule is used for determining a corresponding second top-layer oil temperature meter state index by adopting the target curve data;
the third top-layer oil temperature meter state index acquisition submodule is used for determining the corresponding third top-layer oil temperature meter state index by adopting the target curve data corresponding to two adjacent top-layer oil temperature meters in the main transformer to be judged;
the fourth top oil temperature meter state index acquisition sub-module is used for selecting data of two adjacent moments in the target curve data and determining corresponding fourth top oil temperature meter state indexes;
the fifth top-layer oil temperature meter state index acquisition submodule is used for selecting data of three adjacent moments in the target curve data and determining corresponding fifth top-layer oil temperature meter state indexes;
and the sixth top layer oil temperature meter state index acquisition submodule is used for determining the corresponding sixth top layer oil temperature meter state index by adopting the target curve data.
CN202211299080.6A 2022-10-24 2022-10-24 Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity Active CN115358355B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211299080.6A CN115358355B (en) 2022-10-24 2022-10-24 Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211299080.6A CN115358355B (en) 2022-10-24 2022-10-24 Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity

Publications (2)

Publication Number Publication Date
CN115358355A CN115358355A (en) 2022-11-18
CN115358355B true CN115358355B (en) 2023-01-24

Family

ID=84008766

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211299080.6A Active CN115358355B (en) 2022-10-24 2022-10-24 Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity

Country Status (1)

Country Link
CN (1) CN115358355B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116108760B (en) * 2023-04-12 2023-06-27 广东电网有限责任公司佛山供电局 Main heating characteristic monitoring method and system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111562036A (en) * 2020-05-14 2020-08-21 广东电网有限责任公司 Online calibration method for transformer oil temperature gauge
CN112082670A (en) * 2020-08-06 2020-12-15 中国电力科学研究院有限公司 Distributed optical fiber sensing-based method and system for judging temperature rise state of transformer winding
CN112541634A (en) * 2020-12-16 2021-03-23 国网江苏省电力有限公司检修分公司 Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111679949B (en) * 2020-04-23 2024-10-18 平安科技（深圳）有限公司 Abnormality detection method based on equipment index data and related equipment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111562036A (en) * 2020-05-14 2020-08-21 广东电网有限责任公司 Online calibration method for transformer oil temperature gauge
CN112082670A (en) * 2020-08-06 2020-12-15 中国电力科学研究院有限公司 Distributed optical fiber sensing-based method and system for judging temperature rise state of transformer winding
CN112541634A (en) * 2020-12-16 2021-03-23 国网江苏省电力有限公司检修分公司 Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium

Also Published As

Publication number Publication date
CN115358355A (en) 2022-11-18

Similar Documents

Publication Publication Date Title
CN110763929A (en) Intelligent monitoring and early warning system and method for convertor station equipment
Seem Using intelligent data analysis to detect abnormal energy consumption in buildings
CN110097297A (en) A kind of various dimensions stealing situation Intellisense method, system, equipment and medium
CN110222991B (en) Metering device fault diagnosis method based on RF-GBDT
US20190369570A1 (en) System and method for automatically detecting anomalies in a power-usage data set
JP2003242212A (en) Apparatus and method for determining day of the week with similar utility consumption profile
CN110309981A (en) A kind of power station Decision-making of Condition-based Maintenance system based on industrial big data
CN113032454A (en) Interactive user power consumption abnormity monitoring and early warning management cloud platform based on cloud computing
CN115358355B (en) Method and device for judging main transformer oil temperature gauge and top layer oil temperature abnormity
CN113036913B (en) Method and device for monitoring state of comprehensive energy equipment
CN117078017A (en) Intelligent decision analysis system for monitoring power grid equipment
CN109919432A (en) A kind of substation equipment failure analysis of Influential Factors method based on big data
CN104834305B (en) Distribution automation terminal remote measurement exception analysis system and method based on DMS systems
CN112434942A (en) Intelligent early warning for preventing electricity stealing and user electricity utilization behavior analysis method
CN118152784B (en) Modularized substation equipment data feature extraction method
CN111612019A (en) Method for identifying and analyzing fault abnormality of intelligent electric meter based on big data model
CN113128707A (en) Situation risk assessment method for distribution automation terminal
CN112417627A (en) Power distribution network operation reliability analysis method based on four-dimensional index system
CN115809805A (en) Power grid multi-source data processing method based on edge calculation
CN117560300B (en) Intelligent internet of things flow prediction and optimization system
CN117495106A (en) Real-time risk screening and predicting method and system for intelligent electric meter
CN112785456A (en) High-loss line electricity stealing detection method based on vector autoregressive model
CN112783939A (en) Low-voltage distribution network running state evaluation method based on data mining
CN115689071B (en) Equipment fault fusion prediction method and system based on associated parameter mining
CN117132225A (en) Digital intelligent management platform for laboratory

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant