CN117612345A - Power equipment state monitoring and alarming system and method - Google Patents
Power equipment state monitoring and alarming system and method Download PDFInfo
- Publication number
- CN117612345A CN117612345A CN202311710845.5A CN202311710845A CN117612345A CN 117612345 A CN117612345 A CN 117612345A CN 202311710845 A CN202311710845 A CN 202311710845A CN 117612345 A CN117612345 A CN 117612345A
- Authority
- CN
- China
- Prior art keywords
- power
- time sequence
- feature
- temperature
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 74
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000002159 abnormal effect Effects 0.000 claims abstract description 39
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 239000013598 vector Substances 0.000 claims description 203
- 238000000605 extraction Methods 0.000 claims description 31
- 238000001514 detection method Methods 0.000 claims description 23
- 230000004927 fusion Effects 0.000 claims description 22
- 230000005856 abnormality Effects 0.000 claims description 11
- 238000003062 neural network model Methods 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 11
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 238000005457 optimization Methods 0.000 claims description 10
- 238000009826 distribution Methods 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 13
- 238000004422 calculation algorithm Methods 0.000 abstract description 11
- 230000008859 change Effects 0.000 description 15
- 230000005540 biological transmission Effects 0.000 description 10
- 238000012423 maintenance Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 239000000284 extract Substances 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000003449 preventive effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003345 natural gas Substances 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/185—Electrical failure alarms
Landscapes
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention discloses a power equipment state monitoring alarm system and a method, which are used for monitoring and collecting operation parameters of power equipment, such as a temperature value and a power value of the power equipment in real time, introducing a data processing and analyzing algorithm at the rear end to carry out time sequence collaborative analysis of the temperature and the power of the power equipment so as to judge whether the equipment works normally or not, and if the abnormal condition of the working state is found, the system can timely send an alarm signal to inform related personnel to carry out processing. Therefore, the power equipment state can be monitored in real time and timely alarmed, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
Description
Technical Field
The invention relates to the technical field of intelligent monitoring and alarming, in particular to a power equipment state monitoring and alarming system and method.
Background
Electrical equipment plays a vital role in electrical systems, however, various faults and anomalies such as overloads, short circuits, excessive temperatures, etc. may occur during operation of the electrical equipment. Timely monitoring of the status of the power equipment can help to discover these problems and take timely action to repair them to avoid equipment damage, accidents, or greater impact on the power system.
However, the conventional power equipment monitoring alarm system mainly relies on manual inspection and periodic maintenance, and the mode is low in efficiency and high in cost. Moreover, manual inspection consumes a great deal of time and manpower resources, and cannot monitor the state of equipment in real time, so that potential problems are easily ignored.
Accordingly, an optimized power device condition monitoring alarm system is desired.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a power equipment state monitoring alarm system and a power equipment state monitoring alarm method, which are used for monitoring and collecting operation parameters of power equipment, such as a temperature value and a power value of the power equipment in real time, and introducing a data processing and analyzing algorithm at the rear end to perform time sequence collaborative analysis of the temperature and the power of the power equipment so as to judge whether the equipment works normally or not, and if the abnormal condition of the working state is found, the system can timely send an alarm signal to inform related personnel to process. Therefore, the power equipment state can be monitored in real time and timely alarmed, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
The invention adopts the following technical scheme.
The invention provides a power equipment state monitoring alarm system, which comprises:
the data acquisition module is used for acquiring temperature values and power values of the monitored power equipment at a plurality of set time points in a set time period;
the data time sequence arrangement module is used for arranging the temperature values and the power values of the plurality of set time points into a temperature time sequence input vector and a power time sequence input vector according to a time dimension;
the power time sequence feature analysis module is used for carrying out local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors;
the temperature time sequence feature extraction module is used for carrying out feature extraction on the temperature time sequence input vector to obtain a temperature time sequence feature vector;
the data time sequence semantic feature fusion updating module is used for updating the temperature time sequence feature vector by using the sequence of the power local time sequence feature vector to obtain updated temperature time sequence features;
and the power equipment working state detection module is used for determining whether the working state of the monitored power equipment is abnormal or not based on the updated temperature time sequence characteristic.
Preferably, the power timing characteristic analysis module includes:
the power time sequence vector segmentation unit is used for carrying out vector segmentation on the power time sequence input vector to obtain a sequence of the power local time sequence input vector;
and the power local time sequence feature extraction unit is used for carrying out feature extraction on the sequence of the power local time sequence input vector to obtain the sequence of the power local time sequence feature vector.
Preferably, the power local time sequence feature extraction unit is further configured to perform feature extraction on the sequence of power local time sequence input vectors by using a power time sequence feature extractor based on a one-dimensional convolution layer to obtain a sequence of power local time sequence feature vectors.
Preferably, the temperature time sequence feature extraction module is further configured to perform feature extraction on the temperature time sequence input vector by using a temperature time sequence feature extractor based on a deep neural network model.
Preferably, the deep neural network model is a one-dimensional convolutional neural network model.
Preferably, the fusion formula used in the semantic feature fusion type update is:
wherein v is the temperature time sequence feature vector, h i Is each power local time sequence characteristic vector in the sequence of the power local time sequence characteristic vectors, A is 1 XN w B is 1 XN h Matrix of (N) w And N h Is the temperature timing feature vector and the scale of each power local timing feature vector, N is the total number of vectors of the plurality of power local timing feature vectors, sigma (·) is a Sigmoid function, M w (. Cndot.) and M h (. Cndot.) represents the point convolution function, v' is the updated temperature timing feature vector.
Preferably, the power equipment working state detection module further includes:
the feature optimization unit is used for carrying out feature distribution optimization on the updated temperature time sequence feature to obtain an optimized updated temperature time sequence feature vector;
the working state abnormality detection unit is used for enabling the optimized updating temperature time sequence feature vector to pass through the classifier to obtain a classification result, and the classification result is used for indicating whether the working state of the monitored power equipment is abnormal or not.
Preferably, the operation state abnormality detection unit includes:
the full-connection coding subunit is used for carrying out full-connection coding on the optimized updating temperature time sequence feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector;
and the classification subunit is used for inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
The invention also provides a power equipment state monitoring and alarming method, which is a use method of the power equipment state monitoring system and comprises the following steps:
step 210, obtaining temperature values and power values of the monitored power equipment at a plurality of set time points in a set time period;
step 220, arranging the temperature values and the power values of the plurality of set time points into a temperature time sequence input vector and a power time sequence input vector according to a time dimension;
step 230, performing local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors;
step 240, extracting the characteristics of the temperature time sequence input vector through a temperature time sequence characteristic extractor based on a deep neural network model to obtain a temperature time sequence characteristic vector;
step 250, based on the sequence of the power local time sequence feature vector, carrying out semantic feature fusion update on the temperature time sequence feature vector to obtain updated temperature time sequence features;
step 260, determining whether the operating state of the monitored power equipment is abnormal based on the updated temperature time sequence characteristic.
Preferably, in step 230, the local timing characteristic analysis is performed on the power timing input vector to obtain a sequence of power local timing characteristic vectors, including:
step 231: vector segmentation is carried out on the power time sequence input vector to obtain a sequence of power local time sequence input vectors;
step 232: and extracting the characteristics of the sequence of the power local time sequence input vectors to obtain the sequence of the power local time sequence characteristic vectors.
Compared with the prior art, the invention provides the power equipment state monitoring alarm system and the power equipment state monitoring alarm method, which are used for monitoring and collecting the operation parameters of the power equipment, such as the temperature value and the power value of the power equipment in real time, and introducing a data processing and analyzing algorithm at the rear end to perform time sequence collaborative analysis of the temperature and the power of the power equipment so as to judge whether the equipment works normally, and if the abnormal working state is found, the system can timely send an alarm signal to inform related personnel to process. Therefore, the power equipment state can be monitored in real time and timely alarmed, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
Drawings
FIG. 1 is a block diagram of a power equipment status monitoring alarm system provided in an embodiment of the present invention;
FIG. 2 is a flowchart of a method for monitoring and alarming a power device status according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a system architecture of a power equipment status monitoring and alarming method according to an embodiment of the present invention;
fig. 4 is an application scenario diagram of a power equipment status monitoring alarm system provided in an embodiment of the present invention;
100-power equipment state monitoring and alarming system, 110-data acquisition module, 120-data time sequence arrangement module, 130-power time sequence feature analysis module, 140-temperature time sequence feature extraction module, 150-time sequence semantic feature fusion type updating module and 160-power equipment working state detection module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments of the invention are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art without making any inventive effort, are within the scope of the present invention.
In describing embodiments of the present invention, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present invention is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the invention described herein may be practiced in sequences other than those illustrated or described herein.
Power devices refer to various devices used in power systems for generating, transmitting, distributing and controlling electrical energy, which play a vital role in the power system, ensuring reliable supply and safe operation of electrical energy. The following are some common electrical devices, for example:
a power generation apparatus comprising: the generator, the equipment for converting mechanical energy into electric energy, commonly comprises a gas turbine, a water turbine, a wind driven generator and the like. Power plants, centralized power generation facilities, typically composed of multiple generators, can use different energy sources to produce electrical energy, such as coal, natural gas, nuclear energy, and the like.
A power transformation device, comprising: and the transformer is used for changing the voltage level of the alternating current and realizing the transmission and distribution of electric energy. The transformer is divided into a power generation transformer of a power plant and a transformer station transformer on a power transmission line. Circuit breakers are used to open or close electrical circuits in electrical power systems to protect equipment and personnel.
A power transmission apparatus comprising: and the power transmission line is used for transmitting the electric energy generated by the power plant from the power plant to each user. Transmission lines typically use High Voltage Direct Current (HVDC) or High Voltage Alternating Current (HVAC) technology. And the power transmission tower is used for supporting the structure of the power transmission line and suspending the power transmission line in the air so as to avoid contact with the ground.
A power distribution apparatus comprising: distribution transformers transform high voltage electrical energy transmitted by transmission lines into electrical energy suitable for domestic, commercial and industrial use. Switching devices, including switches, sockets, contactors, etc., are used to control the distribution and transmission of electrical energy.
Control and protection device comprising: an automatic system is used for monitoring, controlling and protecting equipment of a power system and comprises a monitoring instrument, an automatic adjusting device, a protective relay and the like. And the monitoring and data acquisition system is used for monitoring and controlling the running state of the power system in real time.
However, various malfunctions and abnormal conditions of the power equipment during operation, such as overload, short circuit, excessive temperature, etc., may occur. The state of the power equipment is monitored in time to be an important link of operation and maintenance of the power system, and faults, anomalies and potential problems of the equipment can be found early by monitoring the operation parameters and the state of the equipment so as to take appropriate measures for repairing and preventing.
Conventional power equipment monitoring alarm systems suffer from several drawbacks, including the following: traditional power equipment monitoring mainly relies on manual inspection and periodic maintenance, and this kind of mode inefficiency and cost are higher, and manual inspection needs to consume a large amount of time and manpower resources, can't real-time supervision equipment's state moreover, can only rely on periodic maintenance to discover the problem. Traditional monitoring systems rely mainly on manual experience and judgment to detect equipment faults and anomalies, and this approach is prone to deviation in subjective judgment, which can lead to neglect or misjudgment of potential problems.
Conventional monitoring systems are usually periodically patrol or periodically maintained, and cannot monitor the status of the equipment in real time, which means that if the equipment fails or is abnormal during the patrol period, the time for discovery and processing may be delayed, and the risk of failure expansion increases. The traditional monitoring system can only find out the faults or abnormal conditions which occur, and the occurrence of equipment faults can not be predicted in advance, so that maintenance personnel can only passively wait for the occurrence of the faults and then take measures, preventive maintenance can not be carried out, and the influence of the equipment faults on a power system is increased. The traditional monitoring system has limited capability in data acquisition, storage and analysis, cannot process a large amount of equipment data, and can perform accurate fault diagnosis and prediction. This limits the overall monitoring and analysis of the status of the device and does not allow for the discovery of any hidden problems.
Therefore, in the present invention, an optimized power equipment status monitoring alarm system, such as a data analysis and intelligent algorithm, is provided, using a data acquisition system and cloud platform, to collect, store and analyze monitored equipment data, to identify abnormal status and trends of equipment using machine learning and artificial intelligent algorithms, to predict equipment failures in advance, and to provide preventive maintenance advice.
In embodiment 1 of the present invention, fig. 1 is a block diagram of a power equipment status monitoring alarm system provided in embodiment 1 of the present invention. As shown in fig. 1, a power equipment status monitoring alarm system 100 according to embodiment 1 of the present invention includes:
a data acquisition module 110, configured to acquire temperature values and power values of the monitored power device at a plurality of predetermined time points within a predetermined time period;
in the data acquisition module 110, it is ensured that the data acquisition module can accurately acquire temperature values and power values of the monitored power equipment at a plurality of preset time points within a preset time period, and proper sensors and data acquisition equipment are selected in consideration of accuracy and instantaneity of data acquisition. By accurately acquiring the temperature value and the power value of the power equipment, a necessary data basis is provided for subsequent data analysis and state monitoring.
A data timing arrangement module 120, configured to arrange the temperature values and the power values of the plurality of predetermined time points into a temperature timing input vector and a power timing input vector according to a time dimension;
in the data timing arrangement module 120, it is ensured that the temperature values and the power values at a plurality of predetermined time points are correctly arranged according to the time dimension, so as to ensure the continuity and the order of the timing data. And arranging the temperature value and the power value into time sequence input vectors according to the time dimension, and providing an ordered data structure for subsequent feature analysis and extraction.
The power timing sequence feature analysis module 130 is configured to perform local timing sequence feature analysis on the power timing sequence input vector to obtain a sequence of power local timing sequence feature vectors;
in the power timing feature analysis module 130, when performing local timing feature analysis, a suitable analysis method and algorithm, such as sliding window, timing clustering, etc., are selected to extract a sequence of power local timing feature vectors. By carrying out local time sequence feature analysis on the power time sequence input vector, the change trend and abnormal condition of the power can be captured, and useful feature information is provided for subsequent state detection.
A temperature time sequence feature extraction module 140, configured to perform feature extraction on the temperature time sequence input vector by using a temperature time sequence feature extractor based on a deep neural network model to obtain a temperature time sequence feature vector;
in the temperature timing feature extraction module 140, an appropriate neural network structure and training method is selected to extract the useful features of the temperature timing input vector using a temperature timing feature extractor based on a deep neural network model. By extracting the temperature time sequence characteristics, the characteristics related to the state of the equipment, such as the trend, periodicity and the like of temperature change, can be extracted from the temperature data, so that more comprehensive information is provided for subsequent state detection.
The data timing semantic feature fusion updating module 150 is configured to perform semantic feature fusion updating on the temperature timing feature vector based on the sequence of the power local timing feature vector to obtain an updated temperature timing feature;
in the data timing semantic feature fusion updating module 150, when the semantic feature fusion updating is performed on the temperature timing feature, how to reasonably fuse the sequence of the power local timing feature vector with the temperature timing feature vector is considered, so as to obtain the updated temperature timing feature. By means of semantic feature fusion type updating, information of power local time sequence features and temperature time sequence features can be comprehensively utilized, and detection accuracy and robustness of equipment states are improved.
The power equipment operating state detection module 160 is configured to determine whether an abnormality exists in the operating state of the monitored power equipment based on the updated temperature timing characteristic.
In the power equipment operating state detection module 160, based on the updated temperature timing characteristics, a suitable method and algorithm are selected to determine whether an abnormality exists in the operating state of the monitored power equipment, such as setting a threshold, using a machine learning model, and the like. By detecting the working state of the power equipment, the abnormal state of the equipment can be timely found, potential faults are predicted, corresponding maintenance measures are taken, and the reliability and the operation efficiency of the power equipment are improved.
Aiming at the technical problems, the technical conception of the invention is that the operation parameters of the electric power equipment, such as the temperature value and the power value of the electric power equipment, are monitored and collected in real time, and a data processing and analyzing algorithm is introduced at the rear end to carry out the time sequence collaborative analysis of the temperature and the power of the electric power equipment, so as to judge whether the equipment works normally or not, and if the abnormal working state is found, the system can send out an alarm signal in time to inform related personnel to carry out processing. Therefore, the power equipment state can be monitored in real time and timely alarmed, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
Specifically, in the technical scheme of the invention, first, temperature values and power values of the monitored power equipment at a plurality of preset time points in a preset time period are obtained. Then, taking into consideration that the temperature value and the power value of the monitored power equipment have a time sequence dynamic change rule in the time dimension, namely, the temperature value and the power value of the plurality of preset time points respectively have a time sequence association relation in the time dimension. Therefore, in order to analyze and characterize the temperature time sequence change and the power time sequence change of the power equipment, in the technical scheme of the invention, the temperature values and the power values at a plurality of preset time points are required to be arranged into a temperature time sequence input vector and a power time sequence input vector according to a time dimension so as to integrate the distribution information of the temperature values and the power values in time sequence respectively.
And then, extracting the characteristic of the temperature time sequence input vector in a temperature time sequence characteristic extractor based on a one-dimensional convolutional neural network model so as to extract time sequence dynamic characteristic information of the temperature value of the power equipment in a time dimension, thereby obtaining a temperature time sequence characteristic vector.
The deep neural network model is a one-dimensional convolutional neural network model.
The one-dimensional convolutional neural network model can effectively capture time sequence dynamic characteristics in temperature time sequence data, including trend, periodicity, fluctuation and the like of temperature, and can extract characteristic information related to equipment state change by learning a time sequence mode of the temperature data. The one-dimensional convolutional neural network model has strong feature extraction capability, can automatically learn abstract feature representation in data, and can capture features on different time scales by applying convolutional operation on temperature time sequence data, so that the change of the equipment temperature is more comprehensively described.
The one-dimensional convolutional neural network model can reduce the dimension of the input temperature time sequence data in a pooling operation mode and the like, extracts more compact feature vectors, is beneficial to reducing the dimension of the features and improves the efficiency of subsequent processing and analysis. The influence of noise and abnormal values can be reduced based on the feature extraction of the one-dimensional convolutional neural network, the robustness of the features is improved, and the model can filter or reduce the influence of the noise and the abnormal values through learning the local mode and the global correlation of data, so that more accurate and stable features are extracted.
The temperature time sequence input vector is subjected to feature extraction by the temperature time sequence feature extractor based on the one-dimensional convolutional neural network model, so that time sequence dynamic feature information of the temperature value of the power equipment in the time dimension can be obtained, the detection and analysis capability of the equipment state is improved, accurate monitoring and anomaly detection of the power equipment are facilitated, and the safety and reliability of a power system are improved.
It should be appreciated that the power parameters of the power devices will typically change over time, and that such changes may include some local patterns and laws. For example, a power device may experience transient power fluctuations or anomalies over a period of time, which fluctuations may only manifest themselves in local timing. Therefore, in order to analyze the power time sequence change condition of the monitored power equipment more fully and accurately, in the technical scheme of the invention, after vector segmentation is performed on the power time sequence input vector to obtain a sequence of power local time sequence input vectors, the sequence of the power local time sequence input vector is further subjected to feature extraction in a power time sequence feature extractor based on a one-dimensional convolution layer so as to extract power local time sequence feature information related to the power equipment under each local time sequence segment respectively, thereby obtaining the sequence of the power local time sequence feature vector.
In embodiment 2 of the present invention, the power timing characteristic analysis module includes: the power time sequence vector segmentation unit is used for carrying out vector segmentation on the power time sequence input vector so as to obtain a sequence of power local time sequence input vectors; and the power local time sequence feature extraction unit is used for enabling the sequence of the power local time sequence input vectors to pass through a power time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the power local time sequence feature vectors.
The power time sequence vector segmentation unit can segment the long-time power time sequence input vector into a plurality of sequences of power local time sequence input vectors, so that the power data in a long time range can be divided into a plurality of shorter time periods, and the power data in each time period is more centralized and manageable. After slicing the power timing data into a sequence of local timing input vectors, each local timing input vector represents power data for a particular time period. In this way, the power local timing feature extraction unit may extract features for each local timing input vector, capturing power variation features over the period of time, including trends, fluctuations, etc. of the power.
The power timing feature extractor based on the one-dimensional convolution layer can effectively capture timing dynamic features in the power local timing input vector sequence. Through convolution operation, the model can learn power change modes on different time scales, and time sequence characteristic information related to equipment state change is extracted. The one-dimensional convolution layer has stronger feature extraction capability, can automatically learn abstract feature representation in power time sequence data, and can capture a local time sequence mode by applying convolution operation on a power local time sequence input vector sequence, thereby extracting more comprehensive and rich power change features.
The power local time sequence feature extraction unit can reduce the dimension of each local time sequence input vector through operations such as pooling and the like, extracts a more compact power local time sequence feature vector sequence, is beneficial to reducing the dimension of the feature and improves the efficiency of subsequent processing and analysis. The details of the power change can be described more precisely by extracting the power local time sequence feature vector sequence, so that the capability of abnormality detection is improved, the abnormal power change is often reflected in the local time sequence features, and the abnormal condition of the working state of the equipment can be detected and identified more accurately by analyzing and comparing the features.
In electrical equipment, power and temperature are often closely related. Power changes in the device may result in an increase or decrease in temperature, and abnormal changes in temperature may also be indicative of power problems. Therefore, in order to comprehensively utilize time sequence associated characteristic information between two parameters of power and temperature to improve the judging accuracy of the state of the power equipment, in the technical scheme of the invention, the temperature time sequence characteristic vector is further subjected to semantic characteristic fusion type updating based on the sequence of the power local time sequence characteristic vector to obtain an updated temperature time sequence characteristic vector. The sequence of the power local time sequence characteristic vector and the temperature time sequence characteristic vector are fused and updated, so that the working state of the power equipment can be better reflected. In particular, the fusion mode can fully utilize the association information between the power and the temperature of the power equipment and improve the expression capability of the characteristics, so that whether the working state of the power equipment is abnormal or not can be judged more accurately.
In embodiment 3 of the present invention, the data timing semantic feature fusion updating module is configured to: based on the sequence of the power local time sequence feature vector, carrying out semantic feature fusion type update on the temperature time sequence feature vector by using the following fusion formula to obtain an updated temperature time sequence feature vector as the updated temperature time sequence feature; wherein, the fusion formula is:
wherein v is the temperature time sequence feature vector, h i Is each power local time sequence characteristic vector in the sequence of the power local time sequence characteristic vectors, A is 1 XN w B is 1 XN h Matrix of (N) w And N h Is the temperature timing feature vector and the scale of each power local timing feature vector, N is the total number of vectors of the plurality of power local timing feature vectors, sigma (·) is a Sigmoid function, M w (. Cndot.) and M h (. Cndot.) represents the point convolution function, v' is the updated temperature timing feature vector.
In embodiment 4 of the present invention, the power equipment operation state detection module includes:
the feature optimization unit is used for carrying out feature distribution optimization on the updated temperature time sequence feature vector so as to obtain an optimized updated temperature time sequence feature vector;
the working state abnormality detection unit is used for enabling the updated and optimized temperature time sequence feature vector to pass through the classifier to obtain a classification result, and the classification result is used for indicating whether the working state of the monitored power equipment is abnormal or not.
Specifically, the feature optimization unit is configured to: and optimizing the updated temperature time sequence feature vector based on the temperature time sequence feature vector to obtain the optimized updated temperature time sequence feature vector.
In particular, in the technical solution of the present invention, the sequence of the power local time sequence feature vector expresses a local time sequence correlation feature of the power value in a local time domain determined by vector slicing in a global time domain, and the temperature time sequence feature vector expresses a local time sequence correlation feature of the temperature value in the global time domain, so when the semantic feature fusion type update is performed on the temperature time sequence feature vector based on the sequence of the power local time sequence feature vector, the class regression is performed on the obtained updated temperature time sequence feature vector through a classifier, and the updated temperature time sequence feature vector is an interpolation type time sequence correlation feature mixture for a class regression target based on the temperature time sequence feature vector.
In this way the first and second light sources, in order to promote the power value local time domain time sequence correlation characteristic strengthening expression effect of the updated temperature time sequence characteristic vector on the basis of the expression consistency of the global time domain local time sequence correlation characteristic of the temperature time sequence characteristic vector, optimizing the updated temperature timing feature vector based on the temperature timing feature vector, expressed as: performing feature distribution optimization on the updated temperature time sequence feature vector based on the temperature time sequence feature vector by using the following optimization formula to obtain an optimized updated temperature time sequence feature vector; wherein, the optimization formula is:
wherein V is 1 Is the temperature time sequence characteristic vector, V 2 Is the updated temperature time sequence characteristic vector, v 1max -1 And v 2max -1 Respectively represent the temperature time sequence characteristic vector V 1 And institute(s)The updated temperature time sequence feature vector V 2 Is the reciprocal of the global maximum of (1), I is the unit vector, and V 2 ⊙-1 Representing the updated temperature timing feature vector V 2 Taking reciprocal of position-by-position characteristic value of (V)' 2 Is the optimized updated temperature timing feature vector,representing addition by position +.>Indicating subtraction by position, +..
Specifically, for interpolation type time sequence associated feature mixing of a regression target in a feature extraction process, based on the idea of interpolation regularization, a feature mapping of outlier features is unmixed, so that a high-dimensional feature manifold is restored to a manifold geometry based on weak enhancement based on induced deviation, consistent feature enhancement mapping of interpolation samples and interpolation prediction based on feature extraction is realized, and a time sequence associated feature enhancement expression effect in a power value local time domain is obtained while the expression consistency of the updated temperature time sequence feature vector in the local time sequence associated feature of the temperature time sequence feature vector is maintained, so that the time sequence feature expression effect of the updated temperature time sequence feature vector is improved, and the accuracy of a classification result obtained by optimizing the updated temperature time sequence feature vector through a classifier is improved. Therefore, the power equipment state can be monitored in real time and the alarm can be given in time, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
And then, the optimized updating temperature time sequence feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the working state of the monitored power equipment is abnormal or not. That is, the temperature time sequence characteristic information updated by the power time sequence change of the power equipment is utilized to carry out classification processing, so as to judge whether the monitored power equipment works normally, and if the abnormal working state is found, the system can timely send out an alarm signal to inform related personnel to carry out processing.
In embodiment 5 of the present invention, the operation state abnormality detection unit includes:
the full-connection coding subunit is used for carrying out full-connection coding on the updated and optimized temperature time sequence feature vector by using a plurality of full-connection layers of the classifier so as to obtain a coding classification feature vector;
and the classification subunit is used for passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
The temperature time sequence feature vector is classified by the classifier, so that the abnormal condition of the working state of the power equipment can be timely detected, and if the classification result is abnormal, a corresponding early warning signal can be sent out to remind operators or related departments to take necessary maintenance and treatment measures so as to avoid equipment faults or accidents. By introducing the classifier to automatically classify the working state, the dependence and intervention on manpower can be reduced, whether the working state of the equipment is normal or not can be automatically judged, and the subjectivity and uncertainty of the artificial judgment are reduced.
The accuracy and the efficiency of detection can be improved by classifying based on the optimized and updated temperature time sequence feature vector. By using a proper classification algorithm and model training, the classifier has higher discrimination capability and accurately distinguishes between normal states and abnormal states, thereby realizing more reliable equipment state detection. The classifier can classify the temperature time sequence feature vector in real time, and rapidly gives a judging result of the working state, so that abnormal conditions of the equipment state can be found in time, and corresponding measures can be rapidly taken to ensure the normal operation and safety of the power equipment.
By classifying the optimized and updated temperature time sequence feature vectors through the classifier, the classification result of whether the working state of the monitored power equipment is abnormal or not can be obtained, intelligent monitoring and abnormality detection of the equipment state are facilitated, and the reliability and safety of the power system are improved.
In summary, the power equipment state monitoring alarm system 100 according to the embodiment of the invention is illustrated, which can realize real-time monitoring and timely alarm of the power equipment state, so as to effectively identify the abnormal state of the power equipment and improve the safety and reliability of the power system.
As described above, the power device state monitoring alarm system 100 according to the embodiment of the present invention may be implemented in various terminal devices, such as a server for power device state monitoring alarm, and the like. In one example, the power device condition monitoring alarm system 100 according to embodiments of the present invention may be integrated into a terminal device as a software module and/or hardware module. For example, the power device status monitoring alarm system 100 may be a software module in the operating system of the terminal device or may be an application developed for the terminal device; of course, the power device status monitoring alarm system 100 could equally be one of a number of hardware modules of the terminal device.
Preferably, in another example, the power device status monitoring alarm system 100 and the terminal device may also be separate devices, and the power device status monitoring alarm system 100 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.
Fig. 2 is a flowchart of a power equipment status monitoring and alarming method provided in embodiment 6 of the present invention. Fig. 3 is a schematic diagram of a system architecture of a power equipment status monitoring and alarming method provided in embodiment 6 of the present invention. As shown in fig. 2 and 3, a power equipment state monitoring and alarming method includes:
step 210, acquiring temperature values and power values of the monitored power equipment at a plurality of preset time points in a preset time period;
step 220, arranging the temperature values and the power values of the plurality of preset time points into a temperature time sequence input vector and a power time sequence input vector according to a time dimension;
step 230, performing local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors;
step 240, extracting features of the temperature time sequence input vector through a temperature time sequence feature extractor based on a deep neural network model to obtain a temperature time sequence feature vector;
step 250, based on the sequence of the power local time sequence feature vectors, performing semantic feature fusion updating on the temperature time sequence feature vectors to obtain updated temperature time sequence features;
step 260, determining whether the operating state of the monitored power equipment is abnormal based on the updated temperature time sequence characteristic.
In the power equipment state monitoring and alarming method, performing local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors, including: vector segmentation is carried out on the power time sequence input vector so as to obtain a sequence of power local time sequence input vectors; and passing the sequence of the power local time sequence input vectors through a power time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of the power local time sequence feature vectors.
In the power equipment state monitoring and alarming method, the deep neural network model is a one-dimensional convolutional neural network model.
It will be appreciated by those skilled in the art that the specific operation of the individual steps in the above-described power device state monitoring alarm method has been described in detail in the above description of the power device state monitoring alarm system with reference to fig. 1, and thus, a repetitive description thereof will be omitted.
Fig. 4 is an application scenario diagram of a power equipment status monitoring alarm system provided in embodiment 7 of the present invention. As shown in fig. 4, in this application scenario, first, temperature values (e.g., C1 as illustrated in fig. 4) and power values (e.g., C2 as illustrated in fig. 4) of the monitored power device at a plurality of predetermined time points within a predetermined period of time are acquired; the obtained temperature value and power value are then input into a server (e.g., S as illustrated in fig. 4) where a power device status monitoring alarm algorithm is deployed, wherein the server is capable of processing the temperature value and the power value based on the power device status monitoring alarm algorithm to determine whether an abnormality exists in the operating status of the monitored power device.
Compared with the prior art, the invention provides the power equipment state monitoring alarm system and the power equipment state monitoring alarm method, which are used for monitoring and collecting the operation parameters of the power equipment, such as the temperature value and the power value of the power equipment in real time, and introducing a data processing and analyzing algorithm at the rear end to perform time sequence collaborative analysis of the temperature and the power of the power equipment so as to judge whether the equipment works normally, and if the abnormal working state is found, the system can timely send an alarm signal to inform related personnel to process. Therefore, the power equipment state can be monitored in real time and timely alarmed, so that the abnormal state of the power equipment can be effectively identified, and the safety and reliability of a power system are improved.
Finally, it should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the claims.
Claims (10)
1. A power equipment condition monitoring alarm system comprising:
the data acquisition module is used for acquiring temperature values and power values of the monitored power equipment at a plurality of set time points in a set time period;
the data time sequence arrangement module is used for arranging the temperature values and the power values of the plurality of set time points into a temperature time sequence input vector and a power time sequence input vector according to a time dimension;
the power time sequence feature analysis module is used for carrying out local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors;
the temperature time sequence feature extraction module is used for carrying out feature extraction on the temperature time sequence input vector to obtain a temperature time sequence feature vector;
the data time sequence semantic feature fusion updating module is used for updating the temperature time sequence feature vector by using the sequence of the power local time sequence feature vector to obtain updated temperature time sequence features;
and the power equipment working state detection module is used for determining whether the working state of the monitored power equipment is abnormal or not based on the updated temperature time sequence characteristic.
2. A power equipment condition monitoring alarm system according to claim 1 wherein:
the power timing sequence feature analysis module comprises:
the power time sequence vector segmentation unit is used for carrying out vector segmentation on the power time sequence input vector to obtain a sequence of the power local time sequence input vector;
and the power local time sequence feature extraction unit is used for carrying out feature extraction on the sequence of the power local time sequence input vector to obtain the sequence of the power local time sequence feature vector.
3. A power equipment condition monitoring alarm system according to claim 2 wherein:
the power local time sequence feature extraction unit is further used for performing feature extraction on the sequence of the power local time sequence input vectors by using a power time sequence feature extractor based on a one-dimensional convolution layer so as to obtain the sequence of the power local time sequence feature vectors.
4. A power equipment condition monitoring alarm system according to claim 3 wherein: the temperature time sequence feature extraction module is further used for performing feature extraction on the temperature time sequence input vector by using a temperature time sequence feature extractor based on a deep neural network model.
5. A power equipment condition monitoring alarm system as defined in claim 4 wherein:
the deep neural network model is a one-dimensional convolutional neural network model.
6. A power equipment condition monitoring alarm system according to claim 5 wherein:
the fusion formula used in the semantic feature fusion type update is as follows:
wherein v is the temperature time sequence feature vector, h i Is each power local time sequence characteristic vector in the sequence of the power local time sequence characteristic vectors, A is 1 XN w B is 1 XN h Matrix of (N) w And N h Is the temperature timing feature vector and the scale of each power local timing feature vector, N is the total number of vectors of the plurality of power local timing feature vectors, sigma (·) is a Sigmoid function, M w (. Cndot.) and M h (. Cndot.) represents the point convolution function, v' is the updated temperature timing feature vector.
7. A power equipment condition monitoring alarm system in accordance with claim 6 wherein:
the power equipment working state detection module further comprises:
the feature optimization unit is used for carrying out feature distribution optimization on the updated temperature time sequence feature to obtain an optimized updated temperature time sequence feature vector;
the working state abnormality detection unit is used for enabling the optimized updating temperature time sequence feature vector to pass through the classifier to obtain a classification result, and the classification result is used for indicating whether the working state of the monitored power equipment is abnormal or not.
8. A power equipment condition monitoring alarm system in accordance with claim 7 wherein:
the operating state abnormality detection unit includes:
the full-connection coding subunit is used for carrying out full-connection coding on the optimized updating temperature time sequence feature vector by using a plurality of full-connection layers of the classifier to obtain a coding classification feature vector;
and the classification subunit is used for inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.
9. A power equipment state monitoring and alarming method, which is a method for using the power equipment state monitoring system according to the previous claims 1-8, comprising:
step 210, obtaining temperature values and power values of the monitored power equipment at a plurality of set time points in a set time period;
step 220, arranging the temperature values and the power values of the plurality of set time points into a temperature time sequence input vector and a power time sequence input vector according to a time dimension;
step 230, performing local time sequence feature analysis on the power time sequence input vector to obtain a sequence of power local time sequence feature vectors;
step 240, extracting the characteristics of the temperature time sequence input vector through a temperature time sequence characteristic extractor based on a deep neural network model to obtain a temperature time sequence characteristic vector;
step 250, based on the sequence of the power local time sequence feature vector, carrying out semantic feature fusion update on the temperature time sequence feature vector to obtain updated temperature time sequence features;
step 260, determining whether the operating state of the monitored power equipment is abnormal based on the updated temperature time sequence characteristic.
10. The power equipment state monitoring and alarming method according to claim 9, wherein:
step 230 performs local timing characteristic analysis on the power timing input vector to obtain a sequence of power local timing characteristic vectors, including:
step 231: vector segmentation is carried out on the power time sequence input vector to obtain a sequence of power local time sequence input vectors;
step 232: and extracting the characteristics of the sequence of the power local time sequence input vectors to obtain the sequence of the power local time sequence characteristic vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311710845.5A CN117612345A (en) | 2023-12-13 | 2023-12-13 | Power equipment state monitoring and alarming system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311710845.5A CN117612345A (en) | 2023-12-13 | 2023-12-13 | Power equipment state monitoring and alarming system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117612345A true CN117612345A (en) | 2024-02-27 |
Family
ID=89959770
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311710845.5A Pending CN117612345A (en) | 2023-12-13 | 2023-12-13 | Power equipment state monitoring and alarming system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117612345A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970856A (en) * | 2024-03-29 | 2024-05-03 | 杭州七所科技有限公司 | Automatic control system and method for unattended washing module |
| CN118117761A (en) * | 2024-04-29 | 2024-05-31 | 北电科技(广东)有限公司 | Intelligent safety power consumption management equipment and method |
| CN118134269A (en) * | 2024-05-10 | 2024-06-04 | 中国电建集团山东电力建设第一工程有限公司 | Land power generation safety intelligent supervision system based on data analysis |
| CN118230519A (en) * | 2024-04-23 | 2024-06-21 | 中山市恒昱智畅智能科技有限公司 | Abnormality alarm method and system based on intelligent lock control system |
-
2023
- 2023-12-13 CN CN202311710845.5A patent/CN117612345A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970856A (en) * | 2024-03-29 | 2024-05-03 | 杭州七所科技有限公司 | Automatic control system and method for unattended washing module |
| CN117970856B (en) * | 2024-03-29 | 2024-07-05 | 杭州七所科技有限公司 | Automatic control system and method for unattended washing module |
| CN118230519A (en) * | 2024-04-23 | 2024-06-21 | 中山市恒昱智畅智能科技有限公司 | Abnormality alarm method and system based on intelligent lock control system |
| CN118117761A (en) * | 2024-04-29 | 2024-05-31 | 北电科技(广东)有限公司 | Intelligent safety power consumption management equipment and method |
| CN118117761B (en) * | 2024-04-29 | 2024-07-16 | 北电科技(广东)有限公司 | Intelligent safety power consumption management equipment and method |
| CN118134269A (en) * | 2024-05-10 | 2024-06-04 | 中国电建集团山东电力建设第一工程有限公司 | Land power generation safety intelligent supervision system based on data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117612345A (en) | Power equipment state monitoring and alarming system and method | |
| US20200292608A1 (en) | Residual-based substation condition monitoring and fault diagnosis | |
| CN109101906A (en) | A kind of converting station electric power equipment infrared image exception real-time detection method and device | |
| CN106597231A (en) | GIS fault detection system and method based on multi-source information fusion and deep learning network | |
| CN117674140B (en) | Power distribution network measurement and control system and method | |
| CN107219453B (en) | A kind of substation relay protection hidden failure diagnostic method based on Multidimensional and Hybrid amount | |
| CN117590223B (en) | Online monitoring system and method for circuit breaker | |
| CN113763667B (en) | Fire disaster early warning and state monitoring device and method based on 5G edge calculation | |
| CN117110798B (en) | Fault detection method and system for intelligent power distribution network | |
| KR102604708B1 (en) | Switchboard diagnosis system based on artificial intelligence and switchboard diagnosis method based on artificial intelligence | |
| CN111555776B (en) | Fusion sensing and joint diagnosis method, system and device for power transmission line | |
| CN115077605A (en) | Intelligent substation management system | |
| CN111999591B (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| CN116169778A (en) | Processing method and system based on power distribution network anomaly analysis | |
| Dong et al. | Fault diagnosis and classification in photovoltaic systems using scada data | |
| Sun et al. | Transmission Line Fault Diagnosis Method Based on Improved Multiple SVM Model | |
| CN114355182A (en) | Online monitoring and fault diagnosis method for high-voltage circuit breaker | |
| CN114487705A (en) | Power grid equipment fault positioning detection method | |
| KR102695753B1 (en) | System and method of fault detection and diagnosis for photovoltaic power generation | |
| CN117892252B (en) | Intelligent park operation management platform based on big data | |
| CN111915855B (en) | Alarm analysis method for alarm system of million-kilowatt ultra-supercritical unit | |
| CN118191682A (en) | Transformer running state detection system and method | |
| CN118506018A (en) | Electrical equipment state monitoring informationized management system and method | |
| Zhang et al. | Application of Chaotic Particle Swarm Optimization Algorithm in Remote Fault Detection of Electrical System | |
| Napier et al. | An automated fault analysis system for SP energy networks: Requirements, design and implementation |
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 |