A transformer working state recognition system and method based on voiceprint recognition model
Through the transformer working state recognition system based on the voiceprint recognition model, combined with ultrasonic sensors and transformer models, the local discharge fault location of the transformer is accurately identified and positioned, and the problem of inaccurate positioning in the existing technology is solved and the accuracy of identification and positioning is improved .
Patent Information
- Application Number
- CN202210905458.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-29
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-07-29
AI Technical Summary
The prior art is difficult to accurately identify and locate the local discharge fault location of the transformer, resulting in the problem of inaccurate positioning.
Using a transformer working state recognition system based on the voiceprint recognition model, real-time signals are collected through ultrasonic sensors, combined with the voiceprint recognition model, pulse module, adjustment module and partial discharge recognition module, to judge and locate the local discharge fault location, and use the transformer model and simulation The simulation was corrected, and finally the center of the circle in the smallest radius was used as the final position point for the partial discharge.
It realizes accurate identification and positioning of local discharge faults of the transformer, overcomes the problem of inaccurate positioning of a single detection point, and improves positioning accuracy.
Smart Images

Figure CN115291056B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of numerically controlled machine tools, and in particular to a transformer working state recognition system and method based on a voiceprint recognition model. Background Art
[0002] During the operation of the transformer, mechanical deformation is generated under the combined action of various factors such as internal current and magnetic field. The vibration signal is transmitted through its own structure, and this signal is transmitted through the surrounding air medium to generate the sound signal of the transformer operation. These signals can reflect the operating status of the transformer to a large extent. Therefore, by analyzing the sound signals generated during the operation of the transformer, the working status of the mechanical parts of the transformer can be identified.
[0003] After extensive searches, the applicant discovered some typical prior arts. For example, the patent with application number 202111343860.1 discloses a transformer acoustic fingerprint non-stop power detection and status warning method, which can timely warn in the non-stop power state and promptly notify the operation and maintenance personnel to quickly deal with it, thereby improving the safety performance of the equipment. Another example is the patent with application number 202210002728.2, which discloses a transformer diagnosis method and device based on acoustic wave interference. It does not require contact with the equipment to obtain the acoustic signal, which is convenient to obtain the signal. It does not generate electromagnetic signals when collecting signals, and will not interfere with the normal operation of the equipment. It is simple and convenient. Another example is the patent with application number 202210163813.7, which discloses a transformer fault detection method based on acoustic wave signals and attention networks. It uses a stacked multi-layer deep convolutional attention network, relying on the powerful feature expression ability of the deep model, and obtains deep discriminative and robust features from the acoustic wave signals generated when the transformer is working through model training and learning, thereby achieving a more accurate transformer fault detection effect.
[0004] It can be seen that there are still many unproposed technical solutions for many problems that need to be dealt with urgently in practical applications of transformers (such as how to identify the working status of the transformer). Summary of the invention
[0005] Based on this, in order to realize the recognition of the working state of the transformer, the present invention provides a transformer working state recognition system and method based on the voiceprint recognition model, and its specific technical scheme is as follows:
[0006] A transformer working state identification system based on a voiceprint recognition model comprises an ultrasonic sensor for collecting real-time ultrasonic signals generated by a target transformer; the transformer working state identification system based on a voiceprint recognition model also comprises a voiceprint recognition model, a pulse module, a regulation module and a partial discharge identification module.
[0007] The voiceprint recognition model is used to identify the working status of the target transformer according to the real-time ultrasonic signal generated by the target transformer, and to determine whether the target transformer has a partial discharge fault. If so, a pulse detection instruction is generated.
[0008] The pulse module is used to respond to the pulse inspection instruction and input a preset amplitude pulse detection signal to the target transformer at a preset frequency; the adjustment module is used to adjust and change the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to multiple different detection points.
[0009] The partial discharge identification module is used to locate the actual position point of partial discharge of the target transformer according to the detection ultrasonic signals corresponding to multiple different detection points.
[0010] After being trained, the voiceprint recognition model is used to identify the real-time ultrasonic signal generated by the target transformer. When a partial discharge fault occurs in the target transformer, the actual location of the partial discharge of the target transformer can be located by obtaining the detection ultrasonic signals corresponding to multiple different detection points and obtaining the positioning information corresponding to the multiple different detection points. The positioning information corresponding to multiple different detection points is comprehensively considered to overcome the problem of inaccurate positioning that may occur when using the ultrasonic signal of a single detection point to locate the target transformer.
[0011] In summary, the transformer working state recognition system based on the voiceprint recognition model can not only determine whether the target transformer has a partial discharge fault, but also can well locate the actual position point of the partial discharge of the target transformer.
[0012] Furthermore, the transformer working state identification system based on the voiceprint recognition model also includes a transformer model, a simulation module and a correction module.
[0013] The transformer model is constructed according to the target transformer; the simulation module performs partial discharge simulation on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0014] The correction module corrects the actual position point of the local discharge of the located target transformer according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0015] Furthermore, the transformer working state identification system based on the voiceprint recognition model also includes a clustering module and a positioning module.
[0016] The clustering module is used to perform clustering processing on the actual location points of the partial discharge of the target transformer located by the partial discharge identification module and the actual location points of the partial discharge of the target transformer corrected by the correction module.
[0017] The positioning module obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers, and uses the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
[0018] Furthermore, the positioning module also includes a signal source area acquisition unit, a superposition area acquisition unit and a positioning unit.
[0019] The signal source area acquisition unit is used to obtain the local discharge ultrasonic signal source areas corresponding to multiple different detection points according to the accuracy error of the ultrasonic sensor; the superposition area acquisition unit is used to obtain the superposition area of the local discharge ultrasonic signal source areas corresponding to multiple different detection points on the target transformer.
[0020] The positioning unit obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers located in the superposition area, and uses the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
[0021] A method for identifying the working state of a transformer based on a voiceprint recognition model comprises the following steps:
[0022] S1, collects the real-time ultrasonic signal generated by the target transformer.
[0023] S2, identifying the working state of the target transformer according to the real-time ultrasonic signal generated by the target transformer, and judging whether the target transformer has a partial discharge fault, and if so, generating a pulse detection instruction.
[0024] S3, responding to the pulse check instruction, inputting a pulse detection signal of a preset amplitude to the target transformer at a preset frequency.
[0025] S4, adjusting and changing the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to a plurality of different detection points.
[0026] S5, locating the actual position point of the partial discharge of the target transformer according to the detection ultrasonic signals corresponding to the multiple different detection points.
[0027] Furthermore, the transformer working state identification method based on the voiceprint recognition model also includes the following steps:
[0028] Building a transformer model according to the target transformer;
[0029] Performing partial discharge simulation on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located, so as to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points;
[0030] The actual position point of the local discharge of the located target transformer is corrected according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0031] Furthermore, the transformer working state identification method based on the voiceprint recognition model also includes the following steps:
[0032] Clustering the actual location points of partial discharge of the target transformer located by the partial discharge identification module and the actual location points of partial discharge of the target transformer corrected by the correction module;
[0033] Obtain a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers;
[0034] The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
[0035] Furthermore, the transformer working state identification method based on the voiceprint recognition model also includes the following steps:
[0036] Obtaining partial discharge ultrasonic signal source areas corresponding to a plurality of different detection points according to the accuracy error of the ultrasonic sensor;
[0037] Obtaining the superposition area of the partial discharge ultrasonic signal source areas corresponding to multiple different detection points on the target transformer;
[0038] Obtain a circular area with a minimum radius that can include the actual location points of partial discharges of all target transformers located in the superposition area after clustering;
[0039] The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
[0040] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the transformer working state identification method based on a voiceprint recognition model. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.
[0042] Figure 1It is a schematic diagram of the overall structure of a transformer working state recognition system based on a voiceprint recognition model in one embodiment of the present invention;
[0043] Figure 2 It is a schematic diagram of the relationship between the ultrasonic signal source positioning area and the overlapping area in a transformer partial discharge detection system in one embodiment of the present invention;
[0044] Figure 3 is an enlarged schematic diagram of an overlapping area in a transformer partial discharge detection system in one embodiment of the present invention;
[0045] Figure 4 It is a schematic diagram of the overall process of a transformer working state identification method based on a voiceprint recognition model in one embodiment of the present invention.
[0046] Description of reference numerals:
[0047] 1. Ultrasonic sensor; 2. Voiceprint recognition model; 3. Pulse module; 4. Regulation module; 5. Partial discharge recognition module; 6. Target transformer; 7. Partial discharge ultrasonic signal source area; 8. Overlay area. DETAILED DESCRIPTION
[0048] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with its embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the protection scope of the present invention.
[0049] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.
[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.
[0051] The “first” and “second” mentioned in the present invention do not represent specific quantities and orders, but are merely used to distinguish names.
[0052] Embodiment 1:
[0053] like Figure 1 As shown, a transformer working state identification system based on a voiceprint recognition model includes an ultrasonic sensor 1 for collecting real-time ultrasonic signals generated by a target transformer; the transformer working state identification system based on a voiceprint recognition model also includes a voiceprint recognition model 2, a pulse module 3, a regulation module 4 and a partial discharge identification module 5.
[0054] The voiceprint recognition model is used to identify the working state of the target transformer 6 according to the real-time ultrasonic signal generated by the target transformer, and to determine whether the target transformer has a partial discharge fault. If so, a pulse detection instruction is generated.
[0055] The voiceprint recognition model stores multiple ultrasonic fault signals corresponding to partial discharge faults, which can be a trained convolutional neural network. The voiceprint recognition model determines whether the target transformer has a partial discharge fault by comparing the similarity between the real-time ultrasonic signal generated by the target transformer and the ultrasonic fault signal.
[0056] The pulse module is used to respond to the pulse inspection instruction and input a preset amplitude pulse detection signal to the target transformer at a preset frequency; the adjustment module is used to adjust and change the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to multiple different detection points.
[0057] The purpose of the preset amplitude pulse detection signal is to provide a driving current to the target transformer to generate a corresponding ultrasonic signal.
[0058] The adjustment module includes a circular track 40 arranged around the target transformer and an XYZ three-axis sliding module 41 slidably mounted on the circular track, and the ultrasonic sensor is mounted on the XYZ three-axis sliding module. The XYZ three-axis sliding module is used to drive the ultrasonic sensor to move in the XYZ three-axis direction, and the XYZ three-axis sliding module is configured to move along the circular track to achieve movement around the target transformer.
[0059] In this way, the relative position between the ultrasonic sensor and the target transformer can be adjusted to achieve ultrasonic signal detection at multiple different detection points of the target transformer.
[0060] The partial discharge identification module is used to locate the actual position point of partial discharge of the target transformer according to the detection ultrasonic signals corresponding to multiple different detection points.
[0061] After being trained, the voiceprint recognition model is used to identify the real-time ultrasonic signal generated by the target transformer. When a partial discharge fault occurs in the target transformer, the actual location of the partial discharge of the target transformer can be located by obtaining the detection ultrasonic signals corresponding to multiple different detection points and obtaining the positioning information corresponding to the multiple different detection points. The positioning information corresponding to multiple different detection points is comprehensively considered to overcome the problem of inaccurate positioning that may occur when using the ultrasonic signal of a single detection point to locate the target transformer.
[0062] In summary, the transformer working state recognition system based on the voiceprint recognition model can not only determine whether the target transformer has a partial discharge fault, but also can well locate the actual position point of the partial discharge of the target transformer.
[0063] Embodiment 2:
[0064] like Figure 1 As shown, a transformer working state identification system based on a voiceprint recognition model includes an ultrasonic sensor for collecting real-time ultrasonic signals generated by a target transformer; the transformer working state identification system based on a voiceprint recognition model also includes a voiceprint recognition model, a pulse module, a regulation module and a partial discharge identification module.
[0065] The voiceprint recognition model is used to identify the working status of the target transformer according to the real-time ultrasonic signal generated by the target transformer, and to determine whether the target transformer has a partial discharge fault. If so, a pulse detection instruction is generated.
[0066] The voiceprint recognition model stores multiple ultrasonic fault signals corresponding to partial discharge faults, which can be a trained convolutional neural network. The voiceprint recognition model determines whether the target transformer has a partial discharge fault by comparing the similarity between the real-time ultrasonic signal generated by the target transformer and the ultrasonic fault signal.
[0067] The pulse module is used to respond to the pulse inspection instruction and input a preset amplitude pulse detection signal to the target transformer at a preset frequency; the adjustment module is used to adjust and change the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to multiple different detection points.
[0068] The purpose of the preset amplitude pulse detection signal is to provide a driving current to the target transformer to generate a corresponding ultrasonic signal.
[0069] The adjustment module includes a circular track arranged around the target transformer and an XYZ three-axis sliding module slidably mounted on the circular track, and the ultrasonic sensor is mounted on the XYZ three-axis sliding module. The XYZ three-axis sliding module is used to drive the ultrasonic sensor to move in the XYZ three-axis direction, and the XYZ three-axis sliding module is configured to move along the circular track to achieve movement around the target transformer.
[0070] In this way, the relative position between the ultrasonic sensor and the target transformer can be adjusted to achieve ultrasonic signal detection at multiple different detection points of the target transformer.
[0071] The partial discharge identification module is used to locate the actual position point of partial discharge of the target transformer according to the detection ultrasonic signals corresponding to multiple different detection points.
[0072] After being trained, the voiceprint recognition model is used to identify the real-time ultrasonic signal generated by the target transformer. When a partial discharge fault occurs in the target transformer, the actual location of the partial discharge of the target transformer can be located by obtaining the detection ultrasonic signals corresponding to multiple different detection points and obtaining the positioning information corresponding to the multiple different detection points. The positioning information corresponding to multiple different detection points is comprehensively considered to overcome the problem of inaccurate positioning that may occur when using the ultrasonic signal of a single detection point to locate the target transformer.
[0073] In summary, the transformer working state recognition system based on the voiceprint recognition model can not only determine whether the target transformer has a partial discharge fault, but also can well locate the actual position point of the partial discharge of the target transformer.
[0074] Generally speaking, due to the measurement accuracy error of ultrasonic sensors, the specific position coordinates of multiple actual local discharge locations of the target transformer obtained based on the ultrasonic signals obtained by the ultrasonic sensor on the target transformer may not all be consistent or overlap at one point, but will generally fall within a certain area.
[0075] In order to better accurately locate the actual location point of partial discharge of the target transformer, a maximum radius value can be preset, and a minimum radius circle smaller than the maximum radius value can be used to include as many actual location points of partial discharge as possible, and finally the center of the minimum radius circle is used as the final location point of partial discharge.
[0076] In this case, when multiple actual position points of partial discharge of the target transformer obtained based on the ultrasonic signal obtained by the ultrasonic sensor do not completely overlap, the above method can still be used to comprehensively consider multiple actual position points of partial discharge of the target transformer to accurately locate the actual position point of partial discharge of the target transformer.
[0077] The general method for locating the partial discharge position of a transformer through ultrasonic signals is to obtain the difference between the time point when the ultrasonic signal is generated and the time point when the ultrasonic sensor receives the ultrasonic signal, and then multiply the difference by the ultrasonic signal transmission speed to obtain the ultrasonic signal transmission distance. Finally, the actual location of the partial discharge is obtained based on the ultrasonic transmission distance and the position coordinates of the ultrasonic sensor.
[0078] However, the above method does not take into account the subtle changes in ultrasonic transmission speed caused by changes in the transformer structure medium, so the actual location of the partial discharge obtained will have a certain deviation.
[0079] In order to take into account the subtle changes in the ultrasonic transmission speed caused by changes in the transformer structural medium, so as to better accurately locate the local discharge point of the target transformer, in this embodiment, the transformer working state identification system based on the voiceprint recognition model also includes a transformer model, a simulation module and a correction module.
[0080] The transformer model is constructed according to the target transformer; the simulation module performs partial discharge simulation on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0081] The correction module corrects the actual position point of the local discharge of the located target transformer according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0082] Here, the transformer model can first obtain the specific structure and specific material of the target transformer, and then obtain the specific structural material medium of the target transformer to construct the transformer model.
[0083] After the target transformer has been working for a period of time, the performance of its structural medium will change, and the propagation speed of the ultrasonic signal on the structural medium will also change slightly. Therefore, for the target transformer and the same partial discharge point of the transformer model, the ultrasonic signals received by the ultrasonic sensor at the same position will also be different, which is specifically reflected in the different arrival times on the same propagation path.
[0084] That is to say, there will be a certain difference between the time when the ultrasonic sensor receives the simulated ultrasonic signal at the same detection point and the time when the real-time ultrasonic signal is received. The transformer model obtains the specific structural medium through which the ultrasonic transmission path between the partial discharge location and the ultrasonic sensor passes according to the working time of the target transformer, and considers the changing relationship between the elastic density of the specific structural medium and the working time to obtain the actual propagation speed of the ultrasonic signal in different specific structural media.
[0085] Specifically, the actual ultrasonic propagation velocities of different specific structural media of the transformer at different working times can be obtained, and a transformer model can be constructed according to the actual ultrasonic propagation velocities of different specific structural media of the transformer at different working times and the specific structural media of the target transformer. Then, the local discharge simulation position points of the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points on the transformer model can be calculated.
[0086] In this embodiment, the transformer working state identification system based on the voiceprint recognition model further includes a clustering module and a positioning module.
[0087] The clustering module is used to perform clustering processing on the actual location points of the partial discharge of the target transformer located by the partial discharge identification module and the actual location points of the partial discharge of the target transformer corrected by the correction module.
[0088] The positioning module obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers, and uses the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
[0089] Specifically, the positioning module also includes a signal source area acquisition unit, a superposition area acquisition unit and a positioning unit.
[0090] like Figure 2 as well as Figure 3 As shown, the signal source area acquisition unit is used to obtain the local discharge ultrasonic signal source areas 7 corresponding to multiple different detection points according to the accuracy error of the ultrasonic sensor; the superposition area acquisition unit is used to obtain the superposition area 8 of the local discharge ultrasonic signal source areas corresponding to multiple different detection points on the target transformer.
[0091] The positioning unit obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers located in the superposition area, and uses the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
[0092] The present application constructs a transformer model, calculates the local discharge simulation position points of the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points on the transformer model, and then corrects the local discharge actual position points of the located target transformer according to the local discharge simulation position points. The application can take into account the subtle changes in the ultrasonic transmission speed caused by changes in the transformer structure medium, thereby achieving accurate positioning of the local discharge position points of the target transformer.
[0093] Embodiment three:
[0094] like Figure 3 As shown, this embodiment provides a method for identifying the working state of a transformer based on a voiceprint recognition model, which includes the following steps:
[0095] S1, collects the real-time ultrasonic signal generated by the target transformer.
[0096] S2, identifying the working state of the target transformer according to the real-time ultrasonic signal generated by the target transformer, and judging whether the target transformer has a partial discharge fault, and if so, generating a pulse detection instruction.
[0097] S3, responding to the pulse check instruction, inputting a pulse detection signal of a preset amplitude to the target transformer at a preset frequency.
[0098] S4, adjusting and changing the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to a plurality of different detection points.
[0099] S5, locating the actual position point of the partial discharge of the target transformer according to the detection ultrasonic signals corresponding to the multiple different detection points.
[0100] After being trained, the voiceprint recognition model is used to identify the real-time ultrasonic signal generated by the target transformer. When a partial discharge fault occurs in the target transformer, the actual location of the partial discharge of the target transformer can be located by obtaining the detection ultrasonic signals corresponding to multiple different detection points and obtaining the positioning information corresponding to the multiple different detection points. The positioning information corresponding to multiple different detection points is comprehensively considered to overcome the problem of inaccurate positioning that may occur when using the ultrasonic signal of a single detection point to locate the target transformer.
[0101] In summary, the transformer working state identification method based on the voiceprint recognition model can not only determine whether the target transformer has a partial discharge fault, but also can well locate the actual position point of the partial discharge of the target transformer.
[0102] Embodiment 4:
[0103] It should be understood that this embodiment at least includes all the technical features of the above embodiments, and is further described in detail based on the above embodiments.
[0104] In this embodiment, the transformer working state identification method based on the voiceprint recognition model further includes the following steps:
[0105] The first step is to build a transformer model based on the target transformer.
[0106] In the second step, partial discharge simulation is performed on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0107] The third step is to correct the actual position point of the local discharge of the target transformer located according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
[0108] Specifically, the transformer working state identification method based on the voiceprint recognition model also includes the following steps:
[0109] Clustering the actual location points of partial discharge of the target transformer located by the partial discharge identification module and the actual location points of partial discharge of the target transformer corrected by the correction module;
[0110] Obtain a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers;
[0111] The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
[0112] Among them, the specific method of obtaining the minimum radius circular area that can include the actual location points of partial discharge of all target transformers after clustering, and taking the center of the minimum radius circular area as the final location point of partial discharge of the target transformer, and obtaining the partial discharge ultrasonic signal source area corresponding to multiple different detection points according to the accuracy error of the ultrasonic sensor includes the following steps:
[0113] Obtaining the superposition area of the partial discharge ultrasonic signal source areas corresponding to multiple different detection points on the target transformer;
[0114] Obtain a circular area with a minimum radius that can include the actual location points of partial discharges of all target transformers located in the superposition area after clustering;
[0115] The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
[0116] By constructing a transformer model, the local discharge simulation position points of the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points on the transformer model are calculated, and then the actual local discharge position points of the located target transformer are corrected according to the local discharge simulation position points. The subtle changes in the ultrasonic transmission speed caused by changes in the transformer structure medium can be taken into account, thereby achieving accurate positioning of the local discharge position points of the target transformer.
[0117] This embodiment also provides a computer-readable storage medium storing a computer program, which implements the transformer working state identification method based on the voiceprint recognition model when the computer program is executed by a processor.
[0118] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0119] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.
Claims
1. A transformer working state recognition system based on a voiceprint recognition model, comprising an ultrasonic sensor for collecting real-time ultrasonic signals generated by a target transformer, characterized in that: The transformer working state recognition system based on the voiceprint recognition model also includes: The voiceprint recognition model is used to identify the working state of the target transformer according to the real-time ultrasonic signal generated by the target transformer, and to determine whether the target transformer has a partial discharge fault. If so, a pulse detection instruction is generated; A pulse module, used to respond to a pulse inspection instruction and input a pulse inspection signal of a preset amplitude to a target transformer at a preset frequency; An adjustment module is used to adjust and change the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to multiple different detection points one by one; the adjustment module includes a circular track arranged around the target transformer and an XYZ three-axis sliding module slidably mounted on the circular track, and the ultrasonic sensor is mounted on the XYZ three-axis sliding module; the XYZ three-axis sliding module is used to drive the ultrasonic sensor to move in the XYZ three-axis direction, and the XYZ three-axis sliding module is configured to be able to move along the circular track to achieve movement around the target transformer; The partial discharge identification module is used to locate the actual position point of partial discharge of the target transformer according to the detection ultrasonic signals corresponding to multiple different detection points.
2. A transformer working state recognition system based on voiceprint recognition model as claimed in claim 1, characterized in that: The transformer working state recognition system based on the voiceprint recognition model also includes: Transformer model, built according to the target transformer; A simulation module performs partial discharge simulation on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located, so as to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points; The correction module corrects the actual position point of the local discharge of the located target transformer according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
3. A transformer working state recognition system based on voiceprint recognition model as claimed in claim 2, characterized in that: The transformer working state recognition system based on the voiceprint recognition model also includes: A clustering module, used for clustering the actual location points of partial discharge of the target transformer located by the partial discharge identification module and the actual location points of partial discharge of the target transformer corrected by the correction module; The positioning module obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers, and uses the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
4. A transformer working state recognition system based on voiceprint recognition model as claimed in claim 3, characterized in that: The positioning module also includes: A signal source region acquisition unit, used for acquiring partial discharge ultrasonic signal source regions corresponding to a plurality of different detection points according to the accuracy error of the ultrasonic sensor; A superposition region acquisition unit, used to acquire the superposition region of the partial discharge ultrasonic signal source regions corresponding to a plurality of different detection points on the target transformer; The positioning unit obtains a circular area with a minimum radius that can include the actual location points of partial discharge of all the target transformers located in the superposition area after clustering, and takes the center of the circular area with the minimum radius as the final location point of partial discharge of the target transformer.
5. A method for identifying the working state of a transformer based on a voiceprint recognition model, characterized in that: The transformer working state identification method based on the voiceprint recognition model comprises the following steps: S1, collects the real-time ultrasonic signal generated by the target transformer; S2, identifying the working state of the target transformer according to the real-time ultrasonic signal generated by the target transformer, judging whether the target transformer has a partial discharge fault, and if so, generating a pulse detection instruction; S3, responding to the pulse inspection instruction, inputting a pulse inspection signal of a preset amplitude to the target transformer at a preset frequency; S4, adjusting and changing the relative position between the ultrasonic sensor and the target transformer to obtain detection ultrasonic signals corresponding to a plurality of different detection points; the adjustment module includes a circular track arranged around the target transformer and an XYZ three-axis sliding module slidably mounted on the circular track, and the ultrasonic sensor is mounted on the XYZ three-axis sliding module; the XYZ three-axis sliding module is used to drive the ultrasonic sensor to move in the XYZ three-axis direction, and the XYZ three-axis sliding module is configured to be able to move along the circular track to achieve movement around the target transformer; S5, locating the actual position point of the partial discharge of the target transformer according to the detection ultrasonic signals corresponding to the multiple different detection points.
6. A transformer working state identification method based on voiceprint recognition model as claimed in claim 5, characterized in that: The transformer working state identification method based on the voiceprint recognition model also includes the following steps: Building a transformer model according to the target transformer; Performing partial discharge simulation on the transformer model according to the transformer model and the actual position of partial discharge of the target transformer located, so as to obtain simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points; The actual position point of the local discharge of the located target transformer is corrected according to the simulated ultrasonic signals obtained by the ultrasonic sensor at multiple different detection points.
7. A transformer working state identification method based on voiceprint recognition model as claimed in claim 6, characterized in that: The transformer working state identification method based on the voiceprint recognition model also includes the following steps: Clustering the actual location points of partial discharge of the target transformer located by the partial discharge identification module and the actual location points of partial discharge of the target transformer corrected by the correction module; Obtain a circular area with a minimum radius that can include the actual location points of partial discharge of all clustered target transformers; The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
8. A transformer working state identification method based on voiceprint recognition model as claimed in claim 7, characterized in that: The transformer working state recognition method based on the voiceprint recognition model also includes the following steps: Obtaining partial discharge ultrasonic signal source areas corresponding to a plurality of different detection points according to the accuracy error of the ultrasonic sensor; Obtaining the superposition area of the partial discharge ultrasonic signal source areas corresponding to multiple different detection points on the target transformer; Obtain a circular area with a minimum radius that can include the actual location points of partial discharges of all target transformers located in the superposition area after clustering; The center of the circular area with the minimum radius is taken as the final location point of partial discharge of the target transformer.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for identifying the working state of a transformer based on a voiceprint recognition model according to any one of claims 5 to 8 is implemented.
