Intelligent measurement method for mechanical characteristics of pole-mounted circuit breaker

Abstract

The invention discloses an intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker, and belongs to the technical field of power equipment fault prediction. The method solves the problem that time and labor are wasted due to manual measurement in an existing method, predicts possible fault types and positions of the pole-mounted circuit breaker by constructing a prediction model, and collects historical mechanical characteristic data of the pole-mounted circuit breaker to form a historical database, so as to achieve the prediction of the fault types and positions of the pole-mounted circuit breaker. The comprehensive and deep analysis on the mechanical characteristics of the pole-mounted circuit breaker is facilitated; acquiring real-time mechanical characteristic data of the pole-mounted circuit breaker by adopting a non-contact detection technology, and comparing the real-time mechanical characteristic data with historical data to judge whether the pole-mounted circuit breaker is abnormal or not; determining the processing result of the abnormal early warning by analyzing the abnormal factors and combining the predicted fault type and position; and finally, the intelligent control system is used for carrying out remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing result, and the target of intelligent management and maintenance is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of power equipment fault prediction, and in particular to an intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker. Background Art

[0002] Pole mounted circuit breakers are switchgear installed on power poles in power transmission and distribution networks. They are important power equipment, mainly used to automatically cut off power supply in the event of power failure or poor conditions to protect equipment and personnel from electric shock. Their working principle is to automatically cut off the circuit when a circuit fault is detected to prevent the fault from spreading and protect the stable operation of the power grid.

[0003] During operation, regular maintenance and inspection are required. Among them, the inspection of mechanical properties is a very important task. However, traditional mechanical property inspection methods usually rely on manual inspection, which is time-consuming and laborious.

[0004] Therefore, the existing needs are not met, and we propose an intelligent measurement method for the mechanical characteristics of pole-mounted circuit breakers. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent measurement method for the mechanical characteristics of a pole-mounted circuit breaker, which predicts the possible fault types and locations of the pole-mounted circuit breaker by constructing a prediction model, thereby providing a possibility for timely discovery of potential faults; collecting historical mechanical characteristic data of the pole-mounted circuit breaker to form a historical database, which is helpful for a comprehensive and in-depth analysis of the mechanical characteristics of the pole-mounted circuit breaker; using non-contact detection technology to obtain real-time mechanical characteristic data of the pole-mounted circuit breaker, and comparing it with historical data to determine whether there is an abnormality; by analyzing abnormal factors, combined with the predicted fault type and location, determining the processing result of this abnormal warning; finally, using an intelligent control system to perform remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the warning processing result, achieving the goal of intelligent management and maintenance, and solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The intelligent measurement method of mechanical characteristics of a pole-mounted circuit breaker comprises the following steps:

[0008] Step 1: Collect historical status information of the pole-mounted circuit breaker, and build a convolutional neural network model to train and test the historical status information to obtain a prediction model for fault diagnosis of the pole-mounted circuit breaker;

[0009] Step 2: Monitor the real-time status information of the pole-mounted circuit breaker online through the sensor equipment, perform fault prediction on the real-time status information according to the prediction model, and obtain the possible fault type and fault location of the pole-mounted circuit breaker;

[0010] Step 3: Collect historical mechanical characteristic data of the pole-mounted circuit breaker, clean and save the historical mechanical characteristic data, and form a historical database;

[0011] Step 4: Use non-contact detection technology to perform high-precision, non-destructive measurement of key parts of the pole-mounted circuit breaker to obtain real-time mechanical characteristic data of the pole-mounted circuit breaker;

[0012] Step 5: Compare the real-time mechanical characteristic data of the pole-mounted circuit breaker with the historical mechanical characteristic data in the historical database to determine whether there is an abnormality in the real-time mechanical characteristic data of the pole-mounted circuit breaker; and analyze the abnormal factors to determine the processing result of this abnormal warning;

[0013] Step 6: Feedback the predicted fault type, fault location and abnormal information of the real-time mechanical characteristic data of the pole-mounted circuit breaker to the intelligent control system. The intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing results to achieve intelligent management and maintenance.

[0014] Furthermore, in the step 1, a prediction model for fault diagnosis of a pole-mounted circuit breaker is obtained, which specifically includes the following steps:

[0015] By collecting historical status information of the pole-mounted circuit breaker in different states, the historical status information includes: the number of times the pole-mounted circuit breaker is opened, the load condition and the ambient temperature; preprocessing the collected historical status information, including: data cleaning, normalization and extraction of feature information, to ensure effective removal of noise, elimination of interference, and extraction of useful feature information; analyzing the fault type corresponding to the historical status information, and setting up labels for each fault type; then dividing the historical status information and fault labels into a training set and a test set, the training set is used to train the convolutional neural network model, and the test set is used to evaluate the generalization ability of the convolutional neural network model; constructing a convolutional neural network model, The training set is used for training to establish the feature layer of the convolutional neural network model and output the training results. The test set is then imported into the trained convolutional neural network model for testing and the test results are output. The test results are used to verify whether the training results of the convolutional neural network model are accurate. If the training results are consistent with the test results, it indicates that the prediction ability of the convolutional neural network model is normal. If the training results are inconsistent with the test results, it indicates that the prediction ability of the convolutional neural network model is abnormal. The training set and test set need to be re-divided and the convolutional neural network model needs to be retrained and tested until the prediction ability of the convolutional neural network model is normal.

[0016] Furthermore, in the step 2, the possible fault type and fault location of the pole-mounted circuit breaker are obtained, which specifically includes the following steps:

[0017] The real-time status information of the pole-mounted circuit breaker is collected through sensor equipment. The real-time status information includes: temperature parameters, pressure parameters, current parameters and vibration parameters. The sensor equipment includes: temperature sensor, pressure sensor and vibration sensor, which can capture various state changes of the pole-mounted circuit breaker; when the real-time status information of the pole-mounted circuit breaker arrives, it is directly input into the trained convolutional neural network model to predict the fault type, and obtain the possible fault type and fault location of the pole-mounted circuit breaker.

[0018] Furthermore, in step three, historical mechanical characteristic data of the pole-mounted circuit breaker is collected to form a historical database, which specifically includes the following steps:

[0019] The historical mechanical characteristic data of the pole-mounted circuit breaker is obtained through a third-party monitoring agency. The historical mechanical characteristic data includes: thermal images, opening and closing, speed and stroke, load conditions, current waveforms, vibration frequencies and operating temperatures of the pole-mounted circuit breaker; the obtained raw data is cleaned to remove invalid and duplicate data; data cleaning includes: data type checking, missing value processing, abnormal value processing and duplicate value processing; the cleaned data is saved in the database in time series, and the historical mechanical characteristic data is then associated with the device information and geographic location information to form a historical database, so that the data of specific equipment can be easily called and viewed in subsequent analysis; the historical database is encrypted with a symmetric key, and the access objects of the historical database are limited; the historical database is cleaned and updated regularly to ensure the accuracy and completeness of the data.

[0020] Furthermore, the historical mechanical property data is associated with the equipment information and the geographical location information to form a historical database, including:

[0021] Dividing the historical mechanical characteristic data based on data types to obtain a plurality of first data, and marking the first data based on the data types to obtain first marked data;

[0022] Dividing the historical mechanical characteristic data based on the data time feature to obtain a plurality of second data, and marking the second data based on the data time feature to obtain second marked data;

[0023] Dividing the historical mechanical characteristic data based on the equipment information to obtain a plurality of third data, and marking the third data based on the equipment information to obtain third marked data;

[0024] Dividing the historical mechanical characteristic data based on the geographic location information to obtain a plurality of fourth data, and marking the fourth data based on the geographic location information to obtain fourth marked data;

[0025] Based on the data content, the first marked data, the second marked data, the third marked data and the fourth marked data are associated with each other to obtain content association information;

[0026] Matching a storage address for each historical mechanical characteristic data in data order, and establishing tag-address association information of first tag data, second tag data, third tag data, and fourth tag data, respectively, based on content association information;

[0027] Based on each historical mechanical characteristic data and its matching storage address, a historical database is established. Based on the tag-address association information, a data directory display result based on data type, data time characteristics, equipment information and geographic location information is established for the historical database.

[0028] Furthermore, in step 4, non-contact detection technology is used to perform high-precision, non-destructive measurement on key parts of the pole-mounted circuit breaker, which specifically includes the following steps:

[0029] Prepare non-contact detection equipment, which includes: infrared thermal imager, acoustic sensor and electromagnetic sensor; determine the key parts of the pole-mounted circuit breaker, and place infrared thermal imager, acoustic sensor and electromagnetic sensor at the key parts respectively, capture the temperature distribution information on the surface of the pole-mounted circuit breaker by infrared thermal imager, and convert the information into a visual image for display; collect the opening and closing time, speed and stroke in real time by the acoustic sensor installed on the pole-mounted circuit breaker, and analyze the parameters to obtain possible problems with the pole-mounted circuit breaker; capture the changes in the magnetic field around the pole-mounted circuit breaker by electromagnetic sensor, and calculate the conductivity of the surface of the pole-mounted circuit breaker according to the changes in the magnetic field; fuse the data collected by the infrared thermal imager, acoustic sensor and electromagnetic sensor to obtain the mechanical property data of the key parts of the pole-mounted circuit breaker.

[0030] Furthermore, in step 5, determining whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal specifically includes the following steps:

[0031] The collected historical mechanical characteristic data are classified by attributes, and the attribute name is marked for each type of data; each type of data in the real-time mechanical characteristic data is compared with the historical mechanical characteristic data with the same attribute to determine whether the various types of data in the real-time mechanical characteristic data are similar to the historical data; data with low similarity are screened out and analyzed to obtain whether there is any abnormality in the real-time mechanical characteristic data of the pole-mounted circuit breaker.

[0032] Furthermore, in step 5, the processing result of the abnormal warning is determined, which specifically includes the following steps:

[0033] Based on the evaluation results of the mechanical characteristic data of the pole-mounted circuit breaker, the factors that cause abnormalities in the real-time mechanical characteristic data are analyzed, and the processing results of this abnormal warning are obtained based on the possible fault types and fault locations of the pole-mounted circuit breaker.

[0034] Furthermore, the filtering out of data with low similarity and analyzing the data to obtain a result of whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal includes:

[0035] Obtaining data types of data with low similarity, and determining a similarity value under each data, and determining a type weight for each data type based on the degree of influence of the data type on pole-mounted circuit breaker fault detection;

[0036] Based on the similarity value under the data type and the type weight of each data type, the abnormal judgment value of the current mechanical characteristic data is obtained according to the following formula;

[0037]

[0038] Wherein, K represents the abnormal judgment value of the current mechanical characteristic data; ε represents the type weight of the data type of the current mechanical characteristic data, and its value is (0, 1); F represents the similarity value between the current mechanical characteristic data and the historical mechanical characteristic data, and its value is (0, 1); Indicates the average similarity value between all historical characteristic data and historical data under the data type of the current mechanical characteristic data;

[0039] When the abnormality judgment value of the current mechanical characteristic data is greater than the preset abnormality judgment value, it is determined that the current mechanical characteristic data is abnormal;

[0040] Based on the abnormal judgment value of each mechanical characteristic data, the comprehensive judgment value of the data with low similarity is obtained according to the following formula;

[0041]

[0042] Where Y represents the comprehensive judgment value of the data with low similarity; n represents the amount of data with low similarity; n0 represents the amount of abnormal data in the real-time mechanical property data of the data with low similarity; K i represents the abnormal judgment value of the i-th real-time mechanical characteristic data; H i Indicates whether the i-th real-time mechanical characteristic data is abnormal. If so, the value is 1, otherwise the value is 0; C represents a constant, and the value is 5;

[0043] When the comprehensive judgment value of the data with lower similarity is greater than the preset comprehensive judgment value, it is determined that the data with lower similarity is abnormal as a whole.

[0044] Furthermore, in step six, the intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing result, which specifically includes the following steps:

[0045] The processing result of this abnormal warning is obtained through wireless communication technology, and the intelligent control system provides feedback through different channels according to the content of the processing result. If the processing result shows that there is a fault in the pole-mounted circuit breaker and the pole-mounted circuit breaker itself needs to be handled, the intelligent control system will feedback the instruction to the controller to which the pole-mounted circuit breaker belongs, and the pole-mounted circuit breaker itself will be inspected and maintained through the controller; if the processing result shows that the fault factor of the pole-mounted circuit breaker is the working environment, the intelligent control system will feedback the instruction to the on-site human-computer interaction terminal, and the manager of the human-computer interaction terminal will perform operation and maintenance on the working environment of the pole-mounted circuit breaker.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention constructs a convolutional neural network model to conduct deep learning training on historical status information, thereby obtaining a fault prediction model that can accurately predict the real-time status information of a pole-mounted circuit breaker; the prediction model can be used to predict the possible fault types and locations of the pole-mounted circuit breaker, thereby providing a possibility for timely discovery of potential faults; at the same time, historical mechanical characteristic data of the pole-mounted circuit breaker is collected to form a rich historical database containing a large amount of data, which is helpful for a comprehensive and in-depth analysis of the mechanical characteristics of the pole-mounted circuit breaker; in addition, non-contact detection technology is used to obtain the real-time mechanical characteristic data of the pole-mounted circuit breaker, and the data is compared with the historical data to determine whether there is an abnormality; by analyzing the abnormal factors and combining the predicted fault type and location, the processing result of the abnormal warning is determined; finally, the intelligent control system is used to perform remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the warning processing result, thereby achieving the goal of intelligent management and maintenance, thereby improving the operation efficiency of the power system, reducing the failure rate of the pole-mounted circuit breaker, and ensuring the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The figure is a flow chart of the intelligent measurement method of mechanical characteristics of a pole-mounted circuit breaker according to the present invention. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In order to solve the technical problem that the mechanical characteristics detection method of the pole-mounted circuit breaker in the current technology usually needs to rely on manual inspection, which is time-consuming and laborious, please refer to Figure 1 , this embodiment provides the following technical solutions:

[0051] The intelligent measurement method of mechanical characteristics of a pole-mounted circuit breaker comprises the following steps:

[0052] Step 1: Collect historical status information of the pole-mounted circuit breaker, and build a convolutional neural network model to train and test the historical status information to obtain a prediction model for fault diagnosis of the pole-mounted circuit breaker; specifically, the following steps are included:

[0053] The historical status information of the pole-mounted circuit breaker in different states is collected through web crawler technology, including: the number of times the pole-mounted circuit breaker is opened, the load conditions and the ambient temperature; the above historical status information includes multiple different time periods to ensure the generalization ability of the convolutional neural network model; the collected historical status information is preprocessed, including: data cleaning, normalization, and feature information extraction to ensure effective removal of noise and interference, so as to convert the original data into a form suitable for the convolutional neural network model and extract useful feature information; the fault type corresponding to the historical status information is analyzed, and a label is set for each fault type; the historical status information and fault labels are then divided into a training set and a test set. The training set is used to train the convolutional neural network model. The purpose of training is to allow the model to learn the corresponding pattern from the input data; the test set is used to evaluate the generalization ability of the convolutional neural network model, and the test is a test of the model performance; a convolutional neural network model is constructed, and the training set is used for training to establish the convolutional neural network. The feature layer of the neural network model is imported, and the training results are output; the test set is then imported into the trained convolutional neural network model for testing, and the test results are output; the test results are used to verify whether the training results of the convolutional neural network model are accurate; if the training results are consistent with the test results, it indicates that the prediction ability of the convolutional neural network model is in a normal state; if the training results are inconsistent with the test results, it indicates that the prediction ability of the convolutional neural network model is in an abnormal state; the training set and the test set need to be re-divided, and the convolutional neural network model is re-trained and tested until the prediction ability of the convolutional neural network model is in a normal state; when the verification of the convolutional neural network model is completed, the convolutional neural network model is deployed in the actual application environment to perform fault diagnosis on the new pole-mounted circuit breaker state; through the above steps, a prediction model that can predict whether the pole-mounted circuit breaker has a fault can be obtained; when the model is trained, we can obtain a model that can predict the fault condition of the pole-mounted circuit breaker. This model can be used to detect whether the new pole-mounted circuit breaker status information has a fault, thereby achieving the purpose of fault diagnosis; through the above operation, the potential faults of the pole-mounted circuit breaker can be identified quickly and accurately, providing maintenance personnel with targeted maintenance plans, improving maintenance efficiency and quality, so that timely measures can be taken to avoid accidents.

[0054] Step 2: monitor the real-time status information of the pole-mounted circuit breaker online through the sensor equipment, predict the fault of the real-time status information according to the prediction model, and obtain the possible fault type and fault location of the pole-mounted circuit breaker; specifically, the following steps are included:

[0055] The temperature parameters, pressure parameters, current parameters and vibration parameters of the pole-mounted circuit breaker are collected through sensor equipment. The sensor equipment includes: temperature sensor, pressure sensor and vibration sensor, which can capture various state changes of the pole-mounted circuit breaker; when the real-time status information of the pole-mounted circuit breaker arrives, it is directly input into the trained convolutional neural network model to predict the fault type, and the possible fault type and fault location of the pole-mounted circuit breaker are obtained, and the existence and location of the fault are further confirmed; according to the possible fault type and location, corresponding fault handling guidance is given; such as: replacing a certain component or performing repairs, etc., so as to provide operation and maintenance personnel with a fast and effective fault handling solution; by predicting faults based on the real-time status information of the pole-mounted circuit breaker, potential fault risks can be discovered early; and timely repairs or replacements can be carried out to avoid adverse consequences such as power outages caused by faults, thereby helping maintenance personnel to quickly locate the cause of the fault and improve maintenance efficiency.

[0056] The beneficial effects achieved by the above content: Through technologies such as machine learning and artificial intelligence, intelligent management and maintenance of pole-mounted circuit breakers can be achieved, thereby helping users predict possible failures before the pole-mounted circuit breakers fail, assisting users to repair or replace them in advance, avoiding accidents caused by equipment failures, greatly improving the effectiveness of maintenance and management, reducing the risks of human errors and improper operations, and providing the possibility for timely detection of potential failures.

[0057] Step 3: Collect historical mechanical characteristic data of the pole-mounted circuit breaker, clean and save the historical mechanical characteristic data, and form a historical database; specifically, the following steps are included:

[0058] The historical mechanical characteristic data of the pole-mounted circuit breaker is obtained through a third-party monitoring agency. The historical mechanical characteristic data includes: thermal images, opening and closing, speed and stroke, load conditions, current waveforms, vibration frequencies and operating temperatures of the pole-mounted circuit breaker; the obtained raw data is cleaned to remove invalid and duplicate data to eliminate the differences between different data sets; data cleaning includes: data type checking, missing value processing, outlier processing and duplicate value processing; the cleaned data is saved in the database in time series, and then the historical mechanical characteristic data is associated with the device information and geographic location information to form a historical database, so as to facilitate the call and viewing of data of specific devices in subsequent analysis; and the historical database is encrypted with a symmetric key and the access objects of the historical database are limited, thereby improving the security and privacy of data storage in the database and preventing malicious access or tampering; the historical database is cleaned and updated regularly to ensure the accuracy and integrity of the data; through the above steps, we can collect a large amount of historical mechanical characteristic data of the pole-mounted circuit breaker and save it in the historical database, providing a basis for future data analysis.

[0059] The beneficial effects achieved by the above content are: by collecting the historical mechanical characteristic data of the pole-mounted circuit breaker, a rich historical database containing a large amount of data is formed, which helps to conduct a comprehensive and in-depth analysis of the mechanical characteristics of the pole-mounted circuit breaker, and can promptly discover abnormal conditions in the real-time mechanical characteristic data of the pole-mounted circuit breaker, and take timely measures for early warning processing, thereby avoiding adverse consequences such as power outages caused by equipment failures.

[0060] In one embodiment, associating the historical mechanical property data with the equipment information and the geographic location information to form a historical database includes:

[0061] Dividing the historical mechanical characteristic data based on data types to obtain a plurality of first data, and marking the first data based on the data types to obtain first marked data;

[0062] Dividing the historical mechanical characteristic data based on the data time feature to obtain a plurality of second data, and marking the second data based on the data time feature to obtain second marked data;

[0063] Dividing the historical mechanical characteristic data based on the equipment information to obtain a plurality of third data, and marking the third data based on the equipment information to obtain third marked data;

[0064] Dividing the historical mechanical characteristic data based on the geographic location information to obtain a plurality of fourth data, and marking the fourth data based on the geographic location information to obtain fourth marked data;

[0065] Based on the data content, the first marked data, the second marked data, the third marked data and the fourth marked data are associated with each other to obtain content association information;

[0066] Matching a storage address for each historical mechanical characteristic data in data order, and establishing tag-address association information of first tag data, second tag data, third tag data, and fourth tag data, respectively, based on content association information;

[0067] Based on each historical mechanical characteristic data and its matching storage address, a historical database is established. Based on the tag-address association information, a data directory display result based on data type, data time characteristics, equipment information and geographic location information is established.

[0068] In this embodiment, the content-related information is used to determine the marking features of the data type, data time feature, device information, and geographic location information corresponding to each historical mechanical characteristic data.

[0069] In this embodiment, the data directory display results are different directory display results generated based on data type, data time characteristics, device information and geographic location information, so that users can use different directories to search according to different needs.

[0070] The beneficial effects of the above design scheme are: historical mechanical characteristic data are analyzed and processed according to data type, data time characteristics, equipment information and geographic location information to establish the relationship between storage address and data content, and at the same time, a historical database is established based on each historical mechanical characteristic data and its matching storage address, and based on the tag-address association information, a data directory display result based on the data type, data time characteristics, equipment information and geographic location information of the historical database is established respectively, so as to realize the accurate and orderly establishment of the historical database of historical mechanical characteristic data, provide convenience for searching the historical mechanical characteristic data in the historical database, realize the efficient and accurate use of the historical database, and provide a database foundation for the intelligent measurement of mechanical characteristics of pole-mounted circuit breakers.

[0071] Step 4: Use non-contact detection technology to perform high-precision, non-destructive measurement of key parts of the pole-mounted circuit breaker to obtain real-time mechanical characteristic data of the pole-mounted circuit breaker; specifically, the following steps are included:

[0072] Prepare non-contact detection equipment, which includes: infrared thermal imager, acoustic sensor and electromagnetic sensor; specifically, the non-contact detection equipment is operated by remote control to ensure that it can be measured from different angles and distances without contacting the object being measured; determine the key parts of the pole-mounted circuit breaker, and place infrared thermal imager, acoustic sensor and electromagnetic sensor at the parts respectively, capture the temperature distribution information on the surface of the pole-mounted circuit breaker by infrared thermal imager, and convert the information into a visual image for display; determine the position and size of the hot spot on the surface of the pole-mounted circuit breaker, as well as the temperature distribution of the surface by observing the image generated by the infrared thermal imager; collect the opening and closing time, speed and stroke in real time through the acoustic sensor installed on the pole-mounted circuit breaker, and analyze the parameters to obtain possible problems of the pole-mounted circuit breaker, such as: stuck hysteresis, wear, etc., so as to ensure the stable operation of the power system; the magnetic field changes around the pole-mounted circuit breaker are captured by the electromagnetic sensor, and the conductivity of the surface of the pole-mounted circuit breaker is calculated according to the magnetic field changes; the conductivity characteristics of the surface of the pole-mounted circuit breaker, as well as the resistance and inductance parameters inside the object are determined by analyzing the data generated by the electromagnetic sensor; the various data collected by the infrared thermal imager, acoustic sensor and electromagnetic sensor are integrated and processed to obtain the real-time mechanical characteristic data of the key parts of the pole-mounted circuit breaker; through comparative analysis, the temperature, vibration and electromagnetic field characteristics and deformation of the key parts of the pole-mounted circuit breaker are obtained, so as to complete the high-precision and non-destructive measurement process of the mechanical properties of the pole-mounted circuit breaker; specifically, through the above method, high-precision and non-destructive measurement of the key parts of the pole-mounted circuit breaker can be achieved, and real-time mechanical characteristic data can be obtained. These data can provide a reliable reference for the maintenance and management of the equipment, help to timely discover and solve potential problems, and improve the safety and reliability of the equipment.

[0073] Step 5: Compare the real-time mechanical characteristic data of the pole-mounted circuit breaker with the historical mechanical characteristic data in the historical database to determine whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal; and analyze the abnormal factors to determine the processing result of this abnormal warning; specifically, the following steps are included:

[0074] The collected historical mechanical characteristic data are classified by attributes, and the attribute name is marked for each type of data; each type of data in the real-time mechanical characteristic data is compared with the historical mechanical characteristic data of the same attribute, and it is determined whether each type of data in the real-time mechanical characteristic data is similar to the historical mechanical characteristic data; the data with low similarity is screened out, and the data is analyzed to obtain the result of whether there is an abnormality in the real-time mechanical characteristic data of the pole-mounted circuit breaker; specifically, in this embodiment, for example: a characteristic curve of the pole-mounted circuit breaker to be evaluated during the opening and closing action is obtained by an acoustic sensor installed on the pole-mounted circuit breaker, and the characteristic curve includes: a travel curve and a current curve collected during the opening, closing, closing and reclosing actions of the pole-mounted circuit breaker; the obtained characteristic curve is divided into several sections according to the physical process of opening and closing, such as: opening section, Closing section; then compare the similarity of the characteristic curve of the pole-mounted circuit breaker to be evaluated with the characteristic curve of the standard pole-mounted circuit breaker in the historical mechanical characteristic data. Specifically, the similarity of the characteristic curve of the pole-mounted circuit breaker to be evaluated and the characteristic curves of all standard pole-mounted circuit breakers in the historical mechanical characteristic data can be calculated, and various similarity measurement methods can be used for calculation, such as: cosine similarity, Euclidean distance, etc., and the standard pole-mounted circuit breaker refers to a circuit breaker whose mechanical characteristic status meets the standard; based on the result of the similarity comparison, if the similarity is higher, the mechanical characteristic status of the pole-mounted circuit breaker to be evaluated is better; if the similarity is lower, the mechanical characteristic status is worse, and further inspection and maintenance are required; through the above steps, the real-time mechanical characteristic data of the pole-mounted circuit breaker can be effectively compared and analyzed to ensure the stability and reliability of its mechanical characteristics.

[0075] Secondly, based on the evaluation result of the real-time mechanical characteristic data of the pole-mounted circuit breaker, the factors causing the abnormality in the real-time mechanical characteristic data are analyzed, and the processing result of the abnormal warning is obtained in combination with the possible fault types and fault locations of the pole-mounted circuit breaker; in connection with the above embodiment, different measures may be taken according to the mechanical characteristic state of the pole-mounted circuit breaker to be evaluated; if the mechanical characteristic state of the pole-mounted circuit breaker to be evaluated is good, the operation may continue, and the changes in the real-time mechanical characteristic data may be monitored regularly; if the mechanical characteristic state of the pole-mounted circuit breaker to be evaluated is poor, the factors are analyzed to obtain the processing result, such as taking measures such as replacing parts for repair, and the obtained processing result is fed back to the intelligent control system.

[0076] In one embodiment, the filtering out of data with low similarity and analyzing the data to obtain a result of whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal includes:

[0077] Obtaining data types of data with low similarity, and determining a similarity value under each data, and determining a type weight for each data type based on the degree of influence of the data type on pole-mounted circuit breaker fault detection;

[0078] Based on the similarity value under the data type and the type weight of each data type, the abnormal judgment value of the current mechanical characteristic data is obtained according to the following formula;

[0079]

[0080] Among them, K represents the abnormal judgment value of the current mechanical characteristic data; ε represents the type weight of the data type of the current mechanical characteristic data, and its value is (0, 1); F represents the similarity value between the current mechanical characteristic data and the historical data, and its value is (0, 1); Indicates the average similarity value between all historical characteristic data and historical data under the data type of the current mechanical characteristic data;

[0081] When the abnormality judgment value of the current mechanical characteristic data is greater than the preset abnormality judgment value, it is determined that the current mechanical characteristic data is abnormal;

[0082] Based on the abnormal judgment value of each mechanical characteristic data, the comprehensive judgment value of the data with low similarity is obtained according to the following formula;

[0083]

[0084] Among them, Y represents the comprehensive judgment value of the data with low similarity; n represents the amount of data with low similarity; n0 represents the amount of abnormal mechanical property data in the data with low similarity; K i H represents the abnormality judgment value of the i-th mechanical characteristic data; i Indicates whether the i-th mechanical characteristic data is abnormal. If so, the value is 1, otherwise the value is 0; C represents a constant, and the value is 5;

[0085] When the comprehensive judgment value of the data with lower similarity is greater than the preset comprehensive judgment value, it is determined that the data with lower similarity is abnormal as a whole.

[0086] In this embodiment, all historical characteristic data are data extracted from historical data that are consistent with the data type of current mechanical characteristic data. The larger the ratio of the average similarity value of all historical characteristic data under the data type of current mechanical characteristic data and historical data to the similarity value between the current mechanical characteristic data and historical data, the greater the possibility that the corresponding current mechanical characteristic data is abnormal.

[0087] The beneficial effect of the above design scheme is: by obtaining the data type of data with low similarity, and determining the similarity value under each data, and based on the degree of influence of the data type on the pole-mounted circuit breaker fault detection, determining the type weight for each data type, and obtaining the abnormal judgment value of the current mechanical characteristic data, when the abnormal judgment value of the current mechanical characteristic data is greater than the preset abnormal judgment value, it is determined that the current mechanical characteristic data has an abnormality, and the comprehensive judgment value of the data with low similarity, when the comprehensive judgment value of the data with low similarity is greater than the preset comprehensive judgment value, it is determined that the data with low similarity as a whole has an abnormality, thereby realizing the abnormal judgment result of the real-time mechanical characteristic data of the pole-mounted circuit breaker, and providing a basis for subsequent pole-mounted circuit breaker fault detection.

[0088] Step 6: Feedback the fault type, fault location and abnormal information of mechanical characteristic data predicted by the pole-mounted circuit breaker to the intelligent control system. The intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing results to realize intelligent management and maintenance; specifically, the following steps are included:

[0089] The processing result of this abnormal warning is obtained through wireless communication technology, and the intelligent control system provides feedback through different channels according to the content of the processing result. If the processing result shows that: there is a fault in the pole-mounted circuit breaker and the pole-mounted circuit breaker itself needs to be processed for the fault, the intelligent control system will feedback the instruction to the controller to which the pole-mounted circuit breaker belongs, and the pole-mounted circuit breaker itself is inspected and maintained by the controller; if the processing result shows that: the fault factor of the pole-mounted circuit breaker is the working environment, the intelligent control system will feedback the instruction to the on-site human-computer interaction terminal, and the manager of the human-computer interaction terminal will perform operation and maintenance on the working environment of the pole-mounted circuit breaker; in connection with the above embodiment, if: the processing result shows that: measures such as replacing parts are taken for repair, and the processing result is fed back to the intelligent control system, the intelligent control system analyzes the result and obtains that the processing method is: maintaining the pole-mounted circuit breaker itself, then the instruction is fed back to the controller to which the pole-mounted circuit breaker belongs, and the controller replaces the specified parts of the pole-mounted circuit breaker, thereby realizing the remote operation and maintenance and fault handling process.

[0090] The beneficial effects achieved by the above content are: using non-contact detection technology to obtain real-time mechanical characteristic data of the pole-mounted circuit breaker, and comparing its data with historical data to determine whether there is any abnormality; determining the processing result of the abnormal warning through analysis of abnormal factors, combined with the predicted fault type and location; and using an intelligent control system to perform remote operation and maintenance and fault processing of the pole-mounted circuit breaker based on the warning processing result, achieving the goal of intelligent management and maintenance, thereby improving the operating efficiency of the power system, reducing the failure rate of the pole-mounted circuit breaker, and ensuring the safe and stable operation of the power grid.

[0091] Working principle: The historical status information is trained through the convolutional neural network model to obtain the prediction model; the real-time status information of the pole-mounted circuit breaker is used to predict faults according to the prediction model to obtain the possible fault types and fault locations of the pole-mounted circuit breaker; the historical mechanical characteristic data of the pole-mounted circuit breaker are collected to form a historical database; the real-time mechanical characteristic data of the pole-mounted circuit breaker is measured using non-contact detection technology, and the real-time mechanical characteristic data is compared with the historical mechanical characteristic data to determine whether there is any abnormality in the real-time mechanical characteristic data of the pole-mounted circuit breaker; the abnormal factors are analyzed, and the processing results of this abnormal warning are established in combination with the predicted fault type and location; the intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the warning processing results to achieve intelligent management and maintenance.

[0092] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0093] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. Intelligent measurement method for mechanical characteristics of pole-mounted circuit breakers, characterized in that: The following steps are involved: Step 1: Collect historical status information of the pole-mounted circuit breaker, and build a convolutional neural network model to train and test the historical status information to obtain a prediction model for fault diagnosis of the pole-mounted circuit breaker; Step 2: Monitor the real-time status information of the pole-mounted circuit breaker online through the sensor equipment, perform fault prediction on the real-time status information according to the prediction model, and obtain the possible fault type and fault location of the pole-mounted circuit breaker; Step 3: Collect historical mechanical characteristic data of the pole-mounted circuit breaker, clean and save the historical mechanical characteristic data, and form a historical database; Step 4: Use non-contact detection technology to measure key parts of the pole-mounted circuit breaker to obtain real-time mechanical characteristic data of the pole-mounted circuit breaker; Step 5: Compare the real-time mechanical characteristic data of the pole-mounted circuit breaker with the historical mechanical characteristic data in the historical database to determine whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal; And analyze the abnormal factors and determine the processing results of this abnormal warning; Step 6: Feedback the predicted fault type, fault location and abnormal information of real-time mechanical characteristic data of the pole-mounted circuit breaker to the intelligent control system, and the intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing results.

2. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In the step 1, a prediction model for fault diagnosis of a pole-mounted circuit breaker is obtained, which specifically includes the following steps: By collecting historical status information of the pole-mounted circuit breaker in different states, the historical status information includes: the number of times the pole-mounted circuit breaker is opened, the load condition and the ambient temperature; preprocessing the collected historical status information, including: data cleaning, normalization and extraction of feature information, analyzing the fault type corresponding to the historical status information, and setting up labels for each fault type; dividing the historical status information and fault labels into training sets and test sets, constructing a convolutional neural network model, and using the training set for training, establishing the feature layer of the convolutional neural network model, and outputting the training results; then importing the test set into the trained convolutional neural network model for testing, and outputting the test results; using the test results to verify whether the training results of the convolutional neural network model are accurate.

3. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In the step 2, the possible fault type and fault location of the pole-mounted circuit breaker are obtained, which specifically includes the following steps: The real-time status information of the pole-mounted circuit breaker is collected through sensor equipment. The real-time status information includes: temperature parameters, pressure parameters, current parameters and vibration parameters. When the real-time status information of the pole-mounted circuit breaker arrives, it is directly input into the trained convolutional neural network model to predict the fault type and obtain the possible fault type and fault location of the pole-mounted circuit breaker.

4. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In the step 3, the historical mechanical characteristic data of the pole-mounted circuit breaker is collected to form a historical database, which specifically includes the following steps: The historical mechanical characteristic data of the pole-mounted circuit breaker is obtained through a third-party monitoring agency. The historical mechanical characteristic data includes: thermal images, opening and closing, speed and stroke, load conditions, current waveforms, vibration frequencies and operating temperatures of the pole-mounted circuit breaker; the obtained raw data is cleaned to remove invalid and duplicate data; data cleaning includes: data type checking, missing value processing, abnormal value processing and duplicate value processing; the cleaned data is saved in the database in time series, and the historical mechanical characteristic data is then associated with the equipment information and geographic location information to form a historical database; the historical database is encrypted with a symmetric key, and the access objects of the historical database are limited; the historical database is cleaned and updated regularly to ensure the accuracy and completeness of the data.

5. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 4 is characterized in that: The historical mechanical property data is associated with the equipment information and the geographical location information to form a historical database, including: Dividing the historical mechanical characteristic data based on data types to obtain a plurality of first data, and marking the first data based on the data types to obtain first marked data; Dividing the historical mechanical characteristic data based on the data time feature to obtain a plurality of second data, and marking the second data based on the data time feature to obtain second marked data; Dividing the historical mechanical characteristic data based on the equipment information to obtain a plurality of third data, and marking the third data based on the equipment information to obtain third marked data; Dividing the historical mechanical characteristic data based on the geographic location information to obtain a plurality of fourth data, and marking the fourth data based on the geographic location information to obtain fourth marked data; Based on the data content, the first marked data, the second marked data, the third marked data and the fourth marked data are associated with each other to obtain content association information; Matching a storage address for each historical mechanical characteristic data in data order, and establishing tag-address association information of first tag data, second tag data, third tag data, and fourth tag data, respectively, based on content association information; Based on each historical mechanical characteristic data and its matching storage address, a historical database is established. Based on the tag-address association information, a data directory display result based on data type, data time characteristics, equipment information and geographic location information is established for the historical database.

6. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In the step 4, non-contact detection technology is used to measure the key parts of the pole-mounted circuit breaker, which specifically includes the following steps: Prepare non-contact detection equipment, which includes: infrared thermal imager, acoustic sensor and electromagnetic sensor; determine the key parts of the pole-mounted circuit breaker, and place infrared thermal imager, acoustic sensor and electromagnetic sensor at the key parts respectively, capture the temperature distribution information on the surface of the pole-mounted circuit breaker by infrared thermal imager, and convert the information into a visual image for display; collect the opening and closing time, speed and stroke in real time by the acoustic sensor installed on the pole-mounted circuit breaker, and analyze the parameters to obtain possible problems with the pole-mounted circuit breaker; capture the changes in the magnetic field around the pole-mounted circuit breaker by electromagnetic sensor, and calculate the conductivity of the surface of the pole-mounted circuit breaker according to the changes in the magnetic field; fuse the data collected by the infrared thermal imager, acoustic sensor and electromagnetic sensor to obtain the mechanical property data of the key parts of the pole-mounted circuit breaker.

7. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In the step 5, judging whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal specifically includes the following steps: The collected historical mechanical characteristic data are classified by attributes, and the attribute name is marked for each type of data; each type of data in the real-time mechanical characteristic data is compared with the historical mechanical characteristic data with the same attribute to determine whether the various types of data in the real-time mechanical characteristic data are similar to the historical data; data with low similarity are screened out and analyzed to obtain whether there is any abnormality in the real-time mechanical characteristic data of the pole-mounted circuit breaker.

8. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 7, characterized in that: The method of screening out data with low similarity and analyzing the data to obtain a result of whether the real-time mechanical characteristic data of the pole-mounted circuit breaker is abnormal includes: Obtaining data types of data with low similarity, and determining a similarity value under each data, and determining a type weight for each data type based on the degree of influence of the data type on pole-mounted circuit breaker fault detection; Based on the similarity value under the data type and the type weight of each data type, the abnormal judgment value of the current mechanical characteristic data is obtained according to the following formula; Wherein, K represents the abnormal judgment value of the current mechanical characteristic data; ε represents the type weight of the data type of the current mechanical characteristic data, and its value is (0, 1); F represents the similarity value between the current mechanical characteristic data and the historical mechanical characteristic data, and its value is (0, 1); Indicates the average similarity value between all historical characteristic data and historical data under the data type of the current mechanical characteristic data; When the abnormality judgment value of the current mechanical characteristic data is greater than the preset abnormality judgment value, it is determined that the current mechanical characteristic data is abnormal; Based on the abnormal judgment value of each mechanical characteristic data, the comprehensive judgment value of the data with low similarity is obtained according to the following formula; Where Y represents the comprehensive judgment value of the data with low similarity; n represents the amount of data with low similarity; n0 represents the amount of abnormal data in the real-time mechanical property data among the data with low similarity; K i represents the abnormal judgment value of the i-th real-time mechanical characteristic data; H i Indicates whether the i-th real-time mechanical characteristic data is abnormal. If so, the value is 1, otherwise the value is 0; C represents a constant, and the value is 5; When the comprehensive judgment value of the data with lower similarity is greater than the preset comprehensive judgment value, it is determined that the data with lower similarity is abnormal as a whole.

9. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In step 5, the processing result of the abnormal warning is determined, which specifically includes the following steps: Based on the evaluation results of the mechanical characteristic data of the pole-mounted circuit breaker, the factors that cause abnormalities in the real-time mechanical characteristic data are analyzed, and the processing results of this abnormal warning are obtained based on the possible fault types and fault locations of the pole-mounted circuit breaker.

10. The intelligent measurement method for mechanical characteristics of a pole-mounted circuit breaker according to claim 1, characterized in that: In step 6, the intelligent control system performs remote operation and maintenance and fault processing on the pole-mounted circuit breaker according to the early warning processing result, which specifically includes the following steps: The processing result of this abnormal warning is obtained through wireless communication technology, and the intelligent control system provides feedback through different channels according to the content of the processing result. If the processing result shows that there is a fault in the pole-mounted circuit breaker and the pole-mounted circuit breaker itself needs to be handled, the intelligent control system will feedback the instruction to the controller to which the pole-mounted circuit breaker belongs, and the pole-mounted circuit breaker itself will be inspected and maintained through the controller; if the processing result shows that the fault factor of the pole-mounted circuit breaker is the working environment, the intelligent control system will feedback the instruction to the on-site human-computer interaction terminal, and the manager of the human-computer interaction terminal will perform operation and maintenance on the working environment of the pole-mounted circuit breaker.

Images