CN110220602B - Switch cabinet overheating fault identification method - Google Patents
Switch cabinet overheating fault identification method Download PDFInfo
- Publication number
- CN110220602B CN110220602B CN201910548684.1A CN201910548684A CN110220602B CN 110220602 B CN110220602 B CN 110220602B CN 201910548684 A CN201910548684 A CN 201910548684A CN 110220602 B CN110220602 B CN 110220602B
- Authority
- CN
- China
- Prior art keywords
- switch cabinet
- decision tree
- fault
- temperature
- cabinet
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K1/00—Details of thermometers not specially adapted for particular types of thermometer
- G01K1/02—Means for indicating or recording specially adapted for thermometers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Abstract
The invention discloses a method for identifying overheating faults of a switch cabinet, which relates to the field of monitoring of the overheating faults of the switch cabinet, and is characterized in that temperature sensors are arranged in the switch cabinet and on the surface of the cabinet body to acquire temperature data of the switch cabinet in fault and normal operation; establishing a switch cabinet electro-magnetic-fluid coupling simulation model, and correcting the switch cabinet temperature field simulation model according to the acquired temperature measurement data; respectively simulating multiple switch cabinet overheating fault types through the corrected switch cabinet electro-magnetic-fluid coupling simulation model to obtain training samples; and a decision tree is constructed according to the training samples, and the fault of the switch cabinet is identified according to the decision tree, so that the internal temperature of the switch cabinet and the possible fault type can be indirectly and accurately diagnosed through the cabinet body temperature. The invention can be conveniently used and mastered by field workers, and can be widely used for laboratory project research and field online test identification.
Description
Technical Field
The invention relates to the field of monitoring of overheating faults of switch cabinets, in particular to a switch cabinet overheating fault identification method.
Background
In the safety production of electric power, the heating problem of the switch cabinet is always one of the hot problems concerned by researchers. When the switch cabinet is in an operating state, large current is passed through the inner conductive loop for a long time, and the conductor generates heat to enable the temperature in the cabinet to rise. Under normal operating conditions, the heat generated by the current can raise the temperature of internal components and elements, but cannot make the temperature of the internal components and elements exceed the maximum allowable operating temperature. However, if the inside of the switch cabinet breaks down, the temperature in the cabinet is too high, which can cause great potential safety hazards. Therefore, how to accurately monitor the temperature of the switch cabinet is also a focus of attention of researchers.
At present, the accurate monitoring of the internal temperature of a switch cabinet is mainly realized by internally installing a temperature sensor, and the accurate monitoring can be specifically divided into a non-direct contact type and a direct contact type. The non-direct contact type is to measure the temperature by the optical fiber sensor and transmit the signal to the background by the optical fiber, but the optical fiber is easy to break and has high cost, so the method is still in the experimental research stage. The direct contact type is that a signal is generated by a thermal sensitive element such as a thermal resistor and the like, and the signal is transmitted to a background in a wire transmission or wireless transmission mode, but a power supply office finds that the method easily causes the problem of discharge of a suspended electrode, and the safety of a power grid can be damaged.
Therefore, the temperature of the switch cabinet is accurately monitored by directly installing the sensor in the switch cabinet, but the method has potential threats to the safe and stable operation of the switch cabinet. Therefore, on the premise of not influencing the stable operation of the switch cabinet, how to quantitatively and accurately diagnose whether the interior of the switch cabinet is overheated or not by detecting the temperature of the cabinet body is an important problem to be solved in the future.
Based on the temperature field, the invention aims to systematically research the physical law between the temperature in the cabinet and the temperature of the cabinet body through numerical simulation and experimental measurement of the temperature field. On the basis, the rule is deeply learned through an artificial intelligence algorithm, and finally the internal temperature of the switch cabinet and the possible fault types are indirectly and accurately diagnosed through the cabinet body temperature.
Disclosure of Invention
The invention aims to provide a method for identifying an overheating fault of a switch cabinet, so that the defect that the temperature of the switch cabinet is more accurately monitored by directly installing a sensor in the switch cabinet in the conventional diagnosis and online monitoring of the overheating fault in the switch cabinet, but the method has potential threat to the safe and stable operation of the switch cabinet is overcome.
In order to achieve the purpose, the invention provides a method for identifying the overheating fault of a switch cabinet, which comprises the following steps:
s1, arranging temperature sensors in the switch cabinet and on the surface of the cabinet body, and acquiring temperature data of the switch cabinet in fault and normal operation and on the surface of the cabinet body through the temperature sensors;
s2, establishing a switch cabinet electro-magnetic-fluid coupling simulation model through computational fluid dynamics software according to the specific parameters and the running physical process of the switch cabinet, simulating temperature measurement data acquired in S1 with the electro-magnetic-fluid coupling simulation model brought into the switch cabinet, comparing the simulation result with the temperature measurement data acquired in S1, and further correcting the switch cabinet temperature field simulation model;
s3, simulating various switch cabinet overheating fault types respectively through the corrected switch cabinet electro-magnetic-fluid coupling simulation model to obtain training samples;
and S4, training, testing and verifying the training samples by adopting a training sample decision tree recognition algorithm to construct a decision tree, and recognizing the switch cabinet fault according to the decision tree.
Further, the method comprises S5, acquiring real-time temperature data through S1, acquiring a real-time training sample through the switch cabinet electro-magnetic-fluid coupling simulation model corrected through S2, and performing fault recognition on the real-time training sample through the decision tree to determine the fault type of the training sample; and when the decision tree is adopted for fault identification, the identification results are compared and stored all the time.
Further, the switch cabinet overheating fault types include: excessive load, poor contact and fault arcing.
Further, the training samples include: the temperature inside the switch cabinet body, the temperature and humidity of the environment where the switch cabinet is located, the operation mode of the switch cabinet, the load of the switch cabinet, the sunshine time and the wind speed.
Further, the step of constructing the decision tree comprises:
s41, determining the input and the output of the decision tree;
s42, generating a decision tree by adopting a C4.5 algorithm, and acquiring a training sample according to the input of the decision tree determined in S41;
s43, judging the training samples, and if the number of the training samples is too small, directly entering S46; otherwise, judging whether all training samples belong to the same class, if so, directly entering S46; if not, continuously judging whether the attribute of the training sample belongs to the discrete value, if so, directly entering S46, otherwise, entering S44;
s44, calculating a segmentation threshold value for the training sample obtained in the S43;
s45, calculating an information gain rate for the training sample obtained in the S44;
s46, comparing and finding out the attribute with the maximum information gain rate as the current test attribute, and dividing the sample set into a plurality of sub-sample sets according to the current test attribute;
s47, repeating S46 on each subsample set to continue segmenting until the subsample set is not segmented or a termination condition is reached, and generating an initial decision tree;
and S48, pruning the initial decision tree by using a later pruning algorithm to obtain the decision tree.
Further, the input characteristic parameters of the decision tree include: the temperature inside and outside the switch cabinet body, the environment temperature and humidity of the switch cabinet and the sunshine condition; the output of the decision tree is the corresponding code of the fault type.
Further, the formula for calculating the information gain rate is shown in formula (1), where formula (1) is a ratio of the information gain (a) to the split information entropy spliti (a), and the information gain rate obtained by dividing S by a is gainratio (a):
in the formula (1), pjIs that any sample belongs to CiM is the number of sample classes, and a is an attribute of the training sample.
Further, the pruning principle of the late pruning algorithm is that if the error rate of the root node estimation of the sub-tree after branching is larger than the error rate of the leaf node estimation before branching, the pruning is executed, otherwise, the pruning is not executed.
Further, the estimated error rate e of the leaf node is calculated by:
in the formula (2), f is the error rate in the general sense, and f ═ E/N, where E is the number of samples in the error rate in the leaf node, N is the total number of current leaf samples, and z is the confidence limit;
the estimated error rate of the root node is a weighted average sum of the estimated error rates of the leaf nodes, namely:
in the formula (3), k is the number of branches (type of failure), NiThe number of samples taken in the ith branch (the number of samples belonging to the fault type).
Compared with the prior art, the invention has the following beneficial effects:
according to the method for identifying the overheating fault of the switch cabinet, the temperature sensors are arranged in the switch cabinet and on the surface of the cabinet body to acquire the temperature data of the switch cabinet in fault and normal operation and on the surface of the cabinet body; establishing a switch cabinet electro-magnetic-fluid coupling simulation model, and correcting the switch cabinet temperature field simulation model according to the acquired temperature measurement data; respectively simulating multiple switch cabinet overheating fault types through the corrected switch cabinet electro-magnetic-fluid coupling simulation model to obtain training samples; and a decision tree is constructed according to the training samples, and the fault of the switch cabinet is identified according to the decision tree, so that the internal temperature of the switch cabinet and the possible fault type can be indirectly and accurately diagnosed through the cabinet body temperature. The method can be conveniently used and mastered by field workers, can be widely applied to laboratory project research and field online test identification, and lays a theoretical foundation for diagnosis of the internal insulation state of the switch cabinet.
Drawings
In order to more clearly illustrate the technical solution of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only one embodiment of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.
FIG. 1 is a flow chart of a method of identifying a switchgear overheating fault of the present invention;
FIG. 2 is a flow chart of decision tree generation in accordance with the present invention.
Detailed Description
The technical solutions in the present invention are clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments 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.
As shown in fig. 1, the method for identifying an overheating fault of a switch cabinet provided by the present invention includes the following steps:
s1, arranging temperature sensors in the switch cabinet and on the surface of the cabinet body, so that temperature data in the switch cabinet and on the surface of the cabinet body are collected when the switch cabinet is in fault and normal operation, and a reliable experimental data source is provided for the accuracy of the subsequent switch cabinet electro-magnetic-fluid coupling simulation model establishment;
s2, according to the actual physical processes of heat convection, heat conduction, heat radiation and the like in the switch cabinet, establishing a switch cabinet electro-magnetic-fluid coupling simulation model through computational fluid dynamics software Fluent, bringing temperature measurement data acquired in S1 into the switch cabinet electro-magnetic-fluid coupling simulation model for simulation, comparing the temperature simulation result of the switch cabinet electro-magnetic-fluid coupling simulation model with the temperature measurement data acquired in S1, and further correcting the switch cabinet temperature field simulation model;
s3, respectively carrying out temperature, switch cabinet operation mode and other parameter simulation on three switch cabinet overheating fault types of overlarge load, poor contact and fault arc through the corrected switch cabinet electro-magnetic-fluid coupling simulation model to obtain a large number of training samples; the training samples include: the switch cabinet comprises a switch cabinet body, a switch cabinet, a control system and a control system, wherein the switch cabinet body comprises a switch cabinet body internal temperature, a switch cabinet environment temperature and humidity, a switch cabinet operation mode, a switch cabinet load, sunshine time, wind speed and other large data;
s4, training, testing and verifying the training samples by adopting a decision tree recognition algorithm to construct a decision tree, and recognizing the fault of the switch cabinet according to the decision tree; the decision tree is trained, tested and verified by adopting a decision tree recognition algorithm to construct a decision tree, so that the performance of the decision tree can be optimal, the accuracy of the final fault type judgment is improved, and the power grid loss caused by the fault of the switch cabinet is greatly reduced;
s5, acquiring real-time temperature data through S1, acquiring real-time training samples from the real-time temperature data through the switch cabinet electro-magnetic-fluid coupling simulation model corrected through S2, and performing fault recognition on the real-time training samples through a decision tree to determine the fault types of the training samples; and when the decision tree is adopted for fault identification, the identification results are compared and stored all the time.
As shown in fig. 2, the specific steps of training, testing and verifying the decision tree recognition algorithm by using the training samples are as follows:
s41, determining the input and output of the decision tree; the characteristic parameters of the input of the decision tree include: the temperature inside and outside the switch cabinet body, the environment temperature and humidity of the switch cabinet, the sunshine condition and the like are output as corresponding codes of fault types, and characteristic parameters influencing the identification result are pruned, so that the accuracy of the fault identification method is improved.
S42, generating a decision tree by adopting a C4.5 algorithm, and acquiring a training sample according to the input of the decision tree determined in S41; the C4.5 algorithm is one of the most widely used decision tree algorithms at present, is based on the ID3 algorithm proposed in the last 90 years, retains all the advantages of the ID3 algorithm, and greatly improves the performance of the ID3 algorithm by carrying out a series of improvements.
S43, judging the training samples, and if the number of the training samples is too small, directly entering S46; otherwise, judging whether all training samples belong to the same class, if so, directly entering S46; if not, continuously judging whether the attribute of the training sample belongs to the discrete value, if so, directly entering S46, otherwise, entering S44;
and S44, calculating the information gain rate of each attribute in the current sample set when each level of node selects the attribute by using the information entropy theory. The information gain ratio calculation steps are as follows: if A is an attribute of the training sample, the A is a continuous attribute; firstly, training samples in a training set S (training samples obtained from S43) are sorted from small to large according to the value of an attribute A, and if the training sample set A has v different values, the value sequence of the attribute A after sorting is { a }1,a2,...,avAnd taking the average value of adjacent values one by one as a dividing point in sequence, wherein v-1 dividing points are total, calculating the information gain rate of each dividing point respectively, selecting the dividing point with the maximum information gain rate as a local threshold value, and then, in a sequence { a }1,a2,...,avFind the value v which is not over but closest to the local thresholdmaxAs a segmentation threshold for attribute a.
S45, the formula for calculating the information gain rate is shown in formula (1), where formula (1) is the ratio of the information gain (a) to the split information entropy spliti (a), and the information gain rate obtained by dividing S by a is:
in the formula (1), pjIs that any sample belongs to CiM is the number of sample classes.
S46, comparing and finding out the attribute with the maximum information gain rate as the current test attribute, and dividing the sample set into a plurality of sub-sample sets by using the attribute;
and S47, repeating S46 on each subsample set to continue the segmentation until the subsample set is not segmented or a termination condition is reached, and generating an initial decision tree.
And S48, pruning the initial decision tree by using a later pruning algorithm to obtain a final decision tree. The pruning principle of the late pruning algorithm is that if the error rate of the root node estimation of the sub-tree after branching is larger than the error rate of the leaf node estimation before branching, the pruning is executed, otherwise, the pruning is not executed. The estimated error rate of a leaf node is calculated as:
in the formula (2), f is the error rate in the general sense, and f ═ E/N, where E is the number of samples that are error in the leaf node, N is the total number of current leaf samples, z is the confidence limit, and generally when the confidence is 0.25, z is 0.69;
the estimated error rate of the root node is the weighted average sum of the estimated error rates of the leaf nodes, namely:
in the formula (3), k is the number of branches (type of failure), NiThe number of samples taken in the ith branch (the number of samples belonging to the fault type).
The embodiment of the method for identifying the overheating fault of the switch cabinet is explained in detail so that the person skilled in the art can understand the invention more:
taking six groups of data of the internal and external temperatures of the switch cabinet body, the environment temperature and humidity of the switch cabinet, the operation mode of the switch cabinet, the load of the switch cabinet, the sunshine time and the wind speed as characteristic quantities, according to a switch cabinet overheating fault recognition method, taking 50 groups of obtained simulated data under three switch cabinet overheating faults as training samples, building and trimming a decision tree by using the switch cabinet overheating fault recognition method, setting the minimum number of nodes to be 2, setting a confidence factor to be 0.25, measuring the classification accuracy of the decision tree by adopting cross validation, and generating a program flow of the decision tree as shown in fig. 2. The results of cross-validation show that the decision tree can correctly classify 45 groups of samples, and the correct classification rate of self-cross-validation reaches 45/50-90%. According to the decision tree generation result, the finally formed decision tree only uses three characteristic quantities, namely the internal and external temperatures of the switch cabinet body, the operation mode of the switch cabinet and the load of the switch cabinet, of the three input characteristic quantities, which shows that experimental data have better discrimination, three fault types can be identified only by the three characteristic quantities, and the C4.5 algorithm selects the three characteristic quantities, namely the internal and external temperatures of the switch cabinet body, the operation mode of the switch cabinet and the load of the switch cabinet, according to the principle that the information gain rate is maximum, and omits the characteristic quantities of the environment temperature, the sunlight time and the wind speed where the switch cabinet is located, namely pruning is carried out on the decision tree.
The above disclosure is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of changes or modifications within the technical scope of the present invention, and shall be covered by the scope of the present invention.
Claims (9)
1. A method for identifying an overheating fault of a switch cabinet is characterized by comprising the following steps: the method comprises the following steps:
s1, arranging temperature sensors in the switch cabinet and on the surface of the cabinet body, and acquiring temperature data of the switch cabinet in fault and normal operation and on the surface of the cabinet body through the temperature sensors;
s2, establishing a switch cabinet electro-magnetic-fluid coupling simulation model through computational fluid dynamics software according to the specific parameters and the running physical process of the switch cabinet, simulating temperature measurement data acquired in S1 with the electro-magnetic-fluid coupling simulation model brought into the switch cabinet, comparing the simulation result with the temperature measurement data acquired in S1, and further correcting the switch cabinet temperature field simulation model;
s3, simulating various switch cabinet overheating fault types respectively through the corrected switch cabinet electro-magnetic-fluid coupling simulation model to obtain training samples;
and S4, training, testing and verifying the training samples by adopting a training sample decision tree recognition algorithm to construct a decision tree, and recognizing the switch cabinet fault according to the decision tree.
2. The method for identifying an overheating fault of a switchgear cabinet according to claim 1, wherein: the method further comprises S5, acquiring real-time temperature data through S1, acquiring real-time training samples through the switch cabinet electro-magnetic-fluid coupling simulation model corrected through S2, and performing fault recognition on the real-time training samples through the decision tree to determine the fault types of the training samples; and when the decision tree is adopted for fault identification, the identification results are compared and stored in real time.
3. The method for identifying an overheating fault of a switchgear cabinet according to claim 1, wherein: the switch cabinet overheating fault types comprise: excessive load, poor contact and fault arcing.
4. The method for identifying an overheating fault of a switchgear cabinet according to claim 1, wherein: the training sample includes: the temperature inside the switch cabinet body, the temperature and humidity of the environment where the switch cabinet is located, the operation mode of the switch cabinet, the load of the switch cabinet, the sunshine time and the wind speed.
5. The method for identifying an overheating fault of a switchgear cabinet according to claim 1, wherein: the step of constructing the decision tree comprises:
s41, determining the input and the output of the decision tree;
s42, generating a decision tree by adopting a C4.5 algorithm, and acquiring a training sample according to the input of the decision tree determined in S41;
s43, judging the training samples, and if the number of the training samples is too small, directly entering S46; otherwise, judging whether all training samples belong to the same class, if so, directly entering S46; if not, continuously judging whether the attribute of the training sample belongs to the discrete value, if so, directly entering S46, otherwise, entering S44;
s44, calculating a segmentation threshold value for the training sample obtained in the S43;
s45, calculating an information gain rate for the training sample obtained in the S44;
s46, comparing and finding out the attribute with the maximum information gain rate as the current test attribute, and dividing the sample set into a plurality of sub-sample sets according to the current test attribute;
s47, repeating S46 on each subsample set to continue segmenting until the subsample set is not segmented or a termination condition is reached, and generating an initial decision tree;
and S48, pruning the initial decision tree by using a later pruning algorithm to obtain the decision tree.
6. The method for identifying an overheating fault of a switchgear cabinet according to claim 5, wherein: the input characteristic parameters of the decision tree comprise: the temperature inside and outside the switch cabinet body, the environment temperature and humidity of the switch cabinet and the sunshine condition; the output of the decision tree is the corresponding code of the fault type.
7. The method for identifying an overheating fault of a switchgear cabinet according to claim 5, wherein: the formula for calculating the information gain rate is shown in formula (1), where formula (1) is the ratio of the information gain (a) to the split information entropy split i (a), and the information gain rate obtained by dividing S by a is gainratio (a):
in the formula (1), pjIs that any sample belongs to CiM is the number of sample classes, and a is an attribute of the training sample.
8. The method for identifying an overheating fault of a switchgear cabinet according to claim 5, wherein: the pruning principle of the late pruning algorithm is that if the error rate of the root node estimation of the sub-tree after branching is larger than the error rate of the leaf node estimation before branching, the pruning is executed, otherwise, the pruning is not executed.
9. The method for identifying an overheating fault of a switchgear cabinet according to claim 8, wherein: the estimated error rate e of the leaf node is calculated as:
in the formula (2), f is the error rate in the general sense, and f ═ E/N, where E is the number of samples in the error rate in the leaf node, N is the total number of current leaf samples, and z is the confidence limit;
the estimated error rate of the root node is a weighted average sum of the estimated error rates of the leaf nodes, namely:
in the formula (3), k is the number of branches, NiThe number of samples in the ith branch.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910548684.1A CN110220602B (en) | 2019-06-24 | 2019-06-24 | Switch cabinet overheating fault identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910548684.1A CN110220602B (en) | 2019-06-24 | 2019-06-24 | Switch cabinet overheating fault identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110220602A CN110220602A (en) | 2019-09-10 |
| CN110220602B true CN110220602B (en) | 2020-08-25 |
Family
ID=67814426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910548684.1A Active CN110220602B (en) | 2019-06-24 | 2019-06-24 | Switch cabinet overheating fault identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110220602B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111121971B (en) * | 2019-12-18 | 2021-09-03 | 广西电网有限责任公司电力科学研究院 | Method for judging fault type through surface temperature change of high-voltage switch cabinet |
| CN112231949A (en) * | 2020-09-27 | 2021-01-15 | 大唐水电科学技术研究院有限公司 | Intelligent recognition method for defects of generator outlet circuit breaker contacts |
| CN112417669B (en) * | 2020-11-17 | 2022-06-24 | 广西电网有限责任公司电力科学研究院 | Method for realizing simulation correction of thermal fault of switch cabinet based on equivalent heat source |
| CN113074831A (en) * | 2020-11-17 | 2021-07-06 | 中国南方电网有限责任公司超高压输电公司广州局 | Converter valve internal temperature monitoring method, overheating fault positioning method and device |
| CN112417674B (en) * | 2020-11-19 | 2024-03-19 | 广西电网有限责任公司电力科学研究院 | Method for simulating thermal faults of dynamic correction switch cabinet |
| CN112417675B (en) * | 2020-11-19 | 2024-03-19 | 广西电网有限责任公司电力科学研究院 | Thermal fault simulation precision correction method |
| CN112417686A (en) * | 2020-11-20 | 2021-02-26 | 广西电网有限责任公司电力科学研究院 | Method for correcting thermal fault simulation through distribution of temperature fields of switch cabinet |
| CN112417690B (en) * | 2020-11-23 | 2024-03-19 | 广西电网有限责任公司电力科学研究院 | Thermal fault simulation correcting method through environmental temperature field distribution |
| CN113091949B (en) * | 2021-02-18 | 2023-04-07 | 深圳供电局有限公司 | Cable state detection method, device and equipment |
| CN113219330B (en) * | 2021-05-26 | 2022-08-16 | 广西电网有限责任公司电力科学研究院 | Method and system for detecting state of isolating switch |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000009976A1 (en) * | 1998-08-14 | 2000-02-24 | Applied Materials, Inc. | Tuning a substrate temperature measurement system |
| GB2409529A (en) * | 2003-12-17 | 2005-06-29 | Hewlett Packard Development Co | Electronic device environmental effect prediction |
| CN103440803A (en) * | 2013-08-30 | 2013-12-11 | 国家电网公司 | True-voltage simulation-current closed-loop all-true training system for 10kV power distribution device |
| CN105547508A (en) * | 2016-01-20 | 2016-05-04 | 国网浙江武义县供电公司 | Switch cabinet temperature detection device and method based on cabinet door external measurement |
| CN107817427A (en) * | 2017-10-31 | 2018-03-20 | 国网重庆市电力公司电力科学研究院 | Decision tree recognition methods based on sulfur hexafluoride gas shelf depreciation |
| CN108805295A (en) * | 2018-03-26 | 2018-11-13 | 海南电网有限责任公司电力科学研究院 | A kind of method for diagnosing faults based on decision Tree algorithms |
| CN109459639A (en) * | 2018-12-12 | 2019-03-12 | 云南电网有限责任公司电力科学研究院 | A kind of appraisal procedure for preventing GIS device temperature overheating |
-
2019
- 2019-06-24 CN CN201910548684.1A patent/CN110220602B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000009976A1 (en) * | 1998-08-14 | 2000-02-24 | Applied Materials, Inc. | Tuning a substrate temperature measurement system |
| GB2409529A (en) * | 2003-12-17 | 2005-06-29 | Hewlett Packard Development Co | Electronic device environmental effect prediction |
| CN103440803A (en) * | 2013-08-30 | 2013-12-11 | 国家电网公司 | True-voltage simulation-current closed-loop all-true training system for 10kV power distribution device |
| CN105547508A (en) * | 2016-01-20 | 2016-05-04 | 国网浙江武义县供电公司 | Switch cabinet temperature detection device and method based on cabinet door external measurement |
| CN107817427A (en) * | 2017-10-31 | 2018-03-20 | 国网重庆市电力公司电力科学研究院 | Decision tree recognition methods based on sulfur hexafluoride gas shelf depreciation |
| CN108805295A (en) * | 2018-03-26 | 2018-11-13 | 海南电网有限责任公司电力科学研究院 | A kind of method for diagnosing faults based on decision Tree algorithms |
| CN109459639A (en) * | 2018-12-12 | 2019-03-12 | 云南电网有限责任公司电力科学研究院 | A kind of appraisal procedure for preventing GIS device temperature overheating |
Non-Patent Citations (2)
| Title |
|---|
| 基于有限元法的固体绝缘开关柜温度场仿真研究;张凯 等;《高压电器》;20181216;第54卷(第12期);76-81 * |
| 基于自然对流作用的开关柜热仿真及改进设计;何荣涛;《电气开关》;20170630(第6期);27-30 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110220602A (en) | 2019-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110220602B (en) | Switch cabinet overheating fault identification method | |
| CN112699913B (en) | Method and device for diagnosing abnormal relationship of household transformer in transformer area | |
| CN105846780B (en) | A kind of photovoltaic module method for diagnosing faults based on decision-tree model | |
| CN110320468B (en) | Vector machine-based switch cabinet overheating fault diagnosis method | |
| CN110971677B (en) | Electric power internet of things terminal equipment side channel safety monitoring method based on countermeasure reinforcement learning | |
| CN115271000B (en) | State monitoring method and system for cable tunnel | |
| CN115455746B (en) | Nuclear power device operation monitoring data anomaly detection and correction integrated method | |
| CN113177646A (en) | Power distribution equipment online monitoring method and system based on self-adaptive edge proxy | |
| CN116308304A (en) | New energy intelligent operation and maintenance method and system based on meta learning concept drift detection | |
| CN116390137A (en) | Intelligent terminal wireless network signal abnormity monitoring method | |
| CN116418117A (en) | Equipment detection system for intelligent power grid | |
| CN116295948A (en) | Abnormality detection method, system and storage medium of industrial temperature sensor in large temperature difference environment | |
| CN112926686B (en) | BRB and LSTM model-based power consumption anomaly detection method and device for big power data | |
| CN114595762A (en) | Photovoltaic power station abnormal data sequence extraction method | |
| Dong et al. | Fault diagnosis and classification in photovoltaic systems using scada data | |
| CN116614177B (en) | Optical fiber state multidimensional parameter monitoring system | |
| CN115586402B (en) | Power distribution network fault diagnosis and processing method | |
| CN116401532A (en) | Method and system for recognizing frequency instability of power system after disturbance | |
| CN114611566A (en) | Method for identifying overheating fault of high-voltage power distribution cabinet | |
| US20180087489A1 (en) | Method for windmill farm monitoring | |
| CN112037084B (en) | Method and system for testing fault processing, testing and analyzing of power distribution main station | |
| CN114581699A (en) | Transformer state evaluation method based on deep learning model in consideration of multi-source information | |
| Chokr et al. | Feature extraction-reduction and machine learning for fault diagnosis in PV panels | |
| CN112381130A (en) | Cluster analysis-based power distribution room multivariate data anomaly detection method | |
| CN111044176A (en) | Method for monitoring temperature abnormity of generator |
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 |