Health state detection method, device and equipment suitable for multiple types of iron towers

By collecting and enhancing voiceprint data on the tower and building a health status prediction model, the problem that the existing technology cannot detect the health status of the tower in a timely manner is solved, and the timely effectiveness of predicting the health status of the tower is achieved, and the tower body collapse is avoided. damage.

CN119049502APending Publication Date: 2024-11-29ECONOMIC TECH RES INST OF STATE GRID ANHUI ELECTRIC POWER +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411175484.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing personnel inspection methods cannot detect the health status of the tower in a timely and effective manner, which can easily lead to the collapse of the tower body and cause serious damage to the surrounding environment and residents.

Method used

By collecting voiceprint data on the first and second models of towers, data enhancement is carried out, and a tower health status prediction model is built to predict the health status of the tower to be detected based on the trained model.

Benefits of technology

The health status of the tower is predicted in a timely and effective manner to avoid damage to the surrounding environment and residents due to the collapse of the tower body.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049502A_ABST
    Figure CN119049502A_ABST
Patent Text Reader

Abstract

The invention provides a health state detection method, device and equipment suitable for multiple types of iron towers, and the method comprises the steps: respectively collecting voiceprint data on a first type of iron tower and a second type of iron tower, and obtaining a source domain data set and a target domain data set; performing data enhancement on the source domain data set and the target domain data set to obtain the source domain data set and the target domain data set after data enhancement; constructing an iron tower health state prediction model, and training the iron tower health state prediction model based on the source domain data set, the target domain data set and the source domain data set and the target domain data set after data enhancement to obtain a trained iron tower health state prediction model; and predicting the health state of the to-be-detected iron tower based on the trained iron tower health state prediction model. In this way, the health state of the iron tower can be effectively predicted in time, and damage to the surrounding environment and residents due to tower body collapse is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a health status detection method, device and equipment applicable to multiple types of iron towers. Background Art

[0002] In recent years, the coverage area of ​​communication networks has become increasingly wider, and a large number of iron towers have been used in communication lines. Due to some natural phenomena (such as rain, snow, strong winds, etc.) as well as coal mining, engineering construction, and human damage, the iron towers themselves are often prone to loose screws, resonance, and aging, causing the iron towers to be in a state of failure. In severe cases, the towers will collapse, causing serious damage to the surrounding environment and residents.

[0003] Since most communication towers are located in the wild, monitoring by personnel patrol will consume a lot of manpower and material resources, and will be affected and restricted by terrain and climatic conditions. It is impossible to detect the health status of the towers in a timely manner, and the reliability of implementation is difficult to guarantee. Summary of the invention

[0004] The purpose of the present invention is to provide a method, device and equipment for detecting the health status of multiple types of iron towers, aiming to solve the problem that the existing personnel patrol method cannot timely and effectively detect the health status of the iron tower, which may easily cause the tower to collapse and cause serious damage to the surrounding environment and residents.

[0005] According to one aspect of the present disclosure, a health status detection method applicable to multiple types of towers is provided, comprising: The voiceprint data are collected on the first type of iron tower and the second type of iron tower respectively to obtain a source domain data set and a target domain data set; Performing data enhancement on the source domain dataset and the target domain dataset to obtain a source domain dataset and a target domain dataset after data enhancement; Constructing a tower health status prediction model, training the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained tower health status prediction model; Based on the trained tower health status prediction model, the health status of the tower to be tested is predicted.

[0006] In some implementations of the first aspect, the collecting voiceprint data on the first model tower and the second model tower respectively to obtain a source domain dataset and a target domain dataset includes: Select faulty towers and non-faulty towers on the first model of tower to collect voiceprint data to obtain a source domain data set; A faulty tower and a non-faulty tower are selected on the second model tower to collect voiceprint data to obtain a target domain data set, wherein the target domain data set includes a target domain labeled data set and a target domain unlabeled data set.

[0007] In some implementations of the first aspect, the method of performing data enhancement on the source domain dataset and the target domain dataset includes: weak data enhancement and strong data enhancement.

[0008] In some implementations of the first aspect, performing data weak enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset includes: Gaussian white noise is superimposed on the voiceprint segments in the source domain dataset and the target domain unlabeled dataset respectively to obtain a weakly enhanced source domain dataset and a target domain unlabeled dataset.

[0009] In some implementations of the first aspect, performing data strong enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset includes: A random scaling method or a cropping and integration method is used to perform data strong enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset.

[0010] In some implementations of the first aspect, the training of the tower health status prediction model based on the source domain dataset, the target domain dataset, and the data-enhanced source domain dataset and the target domain dataset to obtain the trained tower health status prediction model includes: Determine the loss function of the tower health status prediction model based on the source domain data set, the target domain data set, and the data-enhanced source domain data set and the target domain data set; The iron tower health status prediction model is trained according to the loss function of the iron tower health status prediction model and the stochastic gradient descent method to obtain a trained iron tower health status prediction model.

[0011] In some implementations of the first aspect, determining the loss function of the tower health status prediction model based on the source domain dataset, the target domain dataset, and the data-enhanced source domain dataset and target domain dataset includes: Determine the supervision loss of the source domain dataset according to the source domain dataset and the strongly enhanced source domain dataset ; Determine the supervision loss of the target domain labeled dataset according to the target domain labeled dataset and the strongly enhanced target domain labeled dataset ; The conditional MMD-based domain adaptation method determines the distribution difference loss between the source domain dataset and the target domain dataset ; According to the weakly enhanced target domain unlabeled dataset and the strongly enhanced target domain unlabeled dataset, the consistency loss of the target domain unlabeled dataset is determined ; According to the supervision loss , monitor losses , distribution difference loss and consistency loss Determine the loss function of the tower health status prediction model.

[0012] In some implementations of the first aspect, the loss function is:

[0013] in, for The corresponding weights, for The corresponding weight.

[0014] According to a second aspect of the present disclosure, a health status detection device applicable to multiple types of iron towers is provided, comprising: A data collection module is used to collect voiceprint data on the first type of iron tower and the second type of iron tower respectively to obtain a source domain data set and a target domain data set; A data enhancement module is used to perform data enhancement on the source domain dataset and the target domain dataset to obtain the source domain dataset and the target domain dataset after data enhancement; A model training module is used to construct a tower health status prediction model, and train the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained tower health status prediction model; The state prediction module is used to predict the health state of the tower to be detected based on the trained tower health state prediction model.

[0015] According to a third aspect of the present disclosure, there is provided an electronic device, comprising a processor, and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in a server; when the processor is running, it retrieves a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the health status detection method applicable to multiple types of towers as described above.

[0016] According to a fourth aspect of the present disclosure, there is provided a storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, implement the health status detection method applicable to multiple types of towers as described above.

[0017] According to the health status detection method applicable to multiple types of iron towers provided by the present disclosure, the health status of the iron tower can be predicted timely and effectively, avoiding damage to the surrounding environment and residents caused by tower collapse. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A flow chart of a health status detection method applicable to multiple types of iron towers according to an embodiment of the present invention is shown; Figure 2 A block diagram showing a health status detection device applicable to multiple types of iron towers according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of a computing device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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.

[0020] 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. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0021] The present invention provides a health status detection method applicable to multiple types of iron towers, specifically comprising: collecting voiceprint data on a first type of iron tower and a second type of iron tower respectively to obtain a source domain data set and a target domain data set; performing data enhancement on the source domain data set and the target domain data set to obtain a source domain data set and a target domain data set after data enhancement; constructing an iron tower health status prediction model, training the iron tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained iron tower health status prediction model; predicting the health status of the iron tower to be detected based on the trained iron tower health status prediction model. In this way, the health status of the iron tower can be predicted in a timely and effective manner to avoid damage to the surrounding environment and residents caused by the collapse of the tower body.

[0022] The following is a more detailed description of a health status detection method applicable to multiple types of towers provided by the present disclosure in conjunction with the accompanying drawings and specific embodiments.

[0023] Embodiment 1 Figure 1 A flow chart of a health status detection method applicable to multiple types of towers provided by an embodiment of the present disclosure is shown. Figure 1 As shown, the method 100 may include the following steps: S110, collecting voiceprint data on the first type of tower and the second type of tower respectively to obtain a source domain data set and a target domain data set.

[0024] In some embodiments, voiceprint data can be collected by attaching a vibration source of fixed frequency to a certain part of the tower and attaching a bone voiceprint collector to another part of the tower. The vibration source is turned on and the voiceprint collector records the voiceprint signal x for a period of time. The time length of all voiceprint signals is fixed, that is, there is a fixed number of sampling points.

[0025] Specifically, data can be collected on the first model tower first, and then the faulty tower and the non-faulty tower in the first model tower are selected for experimental data collection to obtain the source domain data set. , For a certain voiceprint data, is the corresponding tower state, The source domain dataset The number of samples in the tower; assuming there are three fault types, the tower state category space contains four state categories, namely, three fault types plus a no-fault state.

[0026] Then, data is collected on the second model of tower to obtain a labeled dataset for the target domain. , and the target domain unlabeled dataset , for A voiceprint data in is the corresponding tower state, for A voiceprint data in and Represents the number of samples in the corresponding data set, and all the data in the target domain are In this patent, the symbol x represents the voiceprint data, and y represents the tower status.

[0027] S120, performing data enhancement on the source domain dataset and the target domain dataset to obtain a source domain dataset and a target domain dataset after data enhancement.

[0028] In some embodiments, the methods of performing data enhancement on voiceprint data are divided into strong enhancement and weak enhancement.

[0029] Among them, the weak enhancement method is: Each voiceprint segment in is superimposed with Gaussian white noise to obtain .

[0030] Strong enhancement can be achieved by random scaling or cropping integration. Among them, random scaling: Each sampling point in each voiceprint segment in is multiplied by a random number sampled from a uniform distribution of [0.8, 1] to obtain a strongly enhanced dataset. .

[0031] Cutting and integration: Cut several random length segments and stretch them to the same length as of equal length, and then these randomly cut and stretched segments are Find the average together and get the enhanced , and then get .

[0032] Can and Perform similar operations to obtain a strongly enhanced dataset and .

[0033] S130, constructing a tower health status prediction model, training the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement, to obtain a trained tower health status prediction model.

[0034] In some embodiments, based on The dataset uses a semi-supervised learning method to train a neural network model, such as a convolutional neural network. The model is defined as , where f is the feature extraction layer and g is the network output layer.

[0035] based on , we can get Pseudo-labels, so Each sample in has a category label. This makes it easy to do domain adaptation based on conditional probability to train the feature extraction layer to obtain aligned features. Here, we use a domain adaptation method based on conditional MMD (Maximum mean discrepancy).

[0036] We are and First, the supervision loss is introduced as follows: for ,have

[0037] in, for The sample x in is the predicted confidence vector output by the model, is the label corresponding to the sample; similarly, , with supervised loss, as follows:

[0038] in, for The sample x in is the predicted confidence vector output by the model, is the label corresponding to the sample, represents the cross entropy loss.

[0039] Then introduce and The distribution difference loss between them, here we use conditional MMD, denoted as .

[0040] At the same time, for , we introduce consistency loss,

[0041] in, for The x in is weakly enhanced to obtain the predicted confidence vector of the sample, for The x in is strongly enhanced to obtain the prediction confidence vector of the sample, represents the cross entropy loss, is a dynamic threshold, which is in the form of

[0042] Where C is the total number of tower states, represents the training step, That is, the threshold value of the rth training iteration, is a real number between (0,1).

[0043] The final loss function is:

[0044] in, and is the corresponding weight.

[0045] S140: Predicting the health status of the tower to be inspected based on the trained tower health status prediction model.

[0046] According to the embodiments of the present invention, the following technical effects are achieved: 1. The present disclosure proposes a health status detection method applicable to multiple types of iron towers, which can timely and effectively predict the health status of iron towers and avoid damage to the surrounding environment and residents caused by tower collapse.

[0047] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0048] The above is an introduction to a method embodiment. The following is a further explanation of the solution of the present invention through an apparatus embodiment.

[0049] Embodiment 2 Figure 2 A block diagram of a health status detection device 200 applicable to multiple types of towers according to an embodiment of the present disclosure is shown. Figure 2 As shown, the device 200 includes: The data collection module 210 is used to collect voiceprint data on the first type of tower and the second type of tower respectively to obtain a source domain data set and a target domain data set; A data enhancement module 220 is used to perform data enhancement on the source domain dataset and the target domain dataset to obtain a source domain dataset and a target domain dataset after data enhancement; A model training module 230 is used to construct a tower health status prediction model, and train the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained tower health status prediction model; The state prediction module 240 is used to predict the health state of the tower to be detected based on the trained tower health state prediction model.

[0050] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0051] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0052] Embodiment 3 Figure 3 A schematic structural diagram of an electronic device 300 that can be used to implement an embodiment of the present invention is shown.

[0053] like Figure 3 As shown, the electronic device 300 includes at least one processor 311, and a memory connected to the at least one processor 311 in communication, such as a read-only memory (ROM) 312, a random access memory (RAM) 313, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 311 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 312 or the computer program loaded from the storage unit 718 to the random access memory (RAM) 313. In the RAM 313, various programs and data required for the operation of the electronic device 300 can also be stored. The processor 311, the ROM 312, and the RAM 313 are connected to each other via a bus 314. An input / output (I / O) interface 315 is also connected to the bus 314.

[0054] A number of components in the electronic device 300 are connected to the I / O interface 315, including: an input unit 316, such as a keyboard, a mouse, etc.; an output unit 317, such as various types of displays, speakers, etc.; a storage unit 318, such as a disk, an optical disk, etc.; and a communication unit 319, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 319 allows the electronic device 300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0055] The processor 311 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 311 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 311 executes the various methods and processes described above, such as a health status detection method applicable to multiple types of towers.

[0056] In some embodiments, a health status detection method applicable to multiple types of iron towers may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 318. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 300 via the ROM 312 and / or the communication unit 319. When the computer program is loaded into the RAM 313 and executed by the processor 311, one or more steps of the angle adjustment method for the leveling cylinder and the luffing cylinder described above may be executed. Alternatively, in other embodiments, the processor 311 may be configured to execute a health status detection method applicable to multiple types of iron towers in any other appropriate manner (for example, by means of firmware).

[0057] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0058] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0059] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0060] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0061] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0062] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0063] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document is not limited here. The above is only a specific implementation mode of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.

Claims

1. A health status detection method applicable to multiple types of towers, characterized in that: include: The voiceprint data are collected on the first type of iron tower and the second type of iron tower respectively to obtain a source domain data set and a target domain data set; Performing data enhancement on the source domain dataset and the target domain dataset to obtain a source domain dataset and a target domain dataset after data enhancement; Constructing a tower health status prediction model, training the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained tower health status prediction model; Based on the trained tower health status prediction model, the health status of the tower to be tested is predicted.

2. The method according to claim 1, characterized in that The voiceprint data are collected on the first type of iron tower and the second type of iron tower respectively to obtain a source domain data set and a target domain data set, including: Select faulty towers and non-faulty towers on the first model of tower to collect voiceprint data to obtain a source domain data set; A faulty tower and a non-faulty tower are selected on the second model tower to collect voiceprint data to obtain a target domain data set, wherein the target domain data set includes a target domain labeled data set and a target domain unlabeled data set.

3. The method according to claim 2, characterized in that The method of performing data enhancement on the source domain dataset and the target domain dataset includes: weak data enhancement and strong data enhancement.

4. The method according to claim 1, characterized in that: Performing data weak enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset, including: Gaussian white noise is superimposed on the voiceprint segments in the source domain dataset and the target domain unlabeled dataset respectively to obtain a weakly enhanced source domain dataset and a target domain unlabeled dataset.

5. The method according to claim 1, characterized in that Performing data strong enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset, including: A random scaling method or a cropping and integration method is used to perform data strong enhancement on the source domain dataset, the target domain labeled dataset, and the target domain unlabeled dataset.

6. The method according to claim 5, characterized in that The method of training the tower health status prediction model based on the source domain data set, the target domain data set, and the data-enhanced source domain data set and the target domain data set to obtain a trained tower health status prediction model includes: Determine the loss function of the tower health status prediction model based on the source domain data set, the target domain data set, and the data-enhanced source domain data set and the target domain data set; The iron tower health status prediction model is trained according to the loss function of the iron tower health status prediction model and the stochastic gradient descent method to obtain a trained iron tower health status prediction model.

7. The method according to claim 6, characterized in that The loss function of the tower health status prediction model is determined based on the source domain data set, the target domain data set, and the data-enhanced source domain data set and the target domain data set, including: Determine the supervision loss of the source domain dataset according to the source domain dataset and the strongly enhanced source domain dataset ; Determine the supervision loss of the target domain labeled dataset according to the target domain labeled dataset and the strongly enhanced target domain labeled dataset ; The conditional MMD-based domain adaptation method determines the distribution difference loss between the source domain dataset and the target domain dataset ; According to the weakly enhanced target domain unlabeled dataset and the strongly enhanced target domain unlabeled dataset, the consistency loss of the target domain unlabeled dataset is determined ; According to the supervision loss , monitor losses , distribution difference loss and consistency loss Determine the loss function of the tower health status prediction model.

8. The method according to claim 7, characterized in that The loss function is: , in, for The corresponding weights, for The corresponding weight.

9. A health status detection device applicable to multiple types of towers, characterized in that: include: A data collection module is used to collect voiceprint data on the first type of iron tower and the second type of iron tower respectively to obtain a source domain data set and a target domain data set; A data enhancement module is used to perform data enhancement on the source domain dataset and the target domain dataset to obtain the source domain dataset and the target domain dataset after data enhancement; A model training module is used to construct a tower health status prediction model, and train the tower health status prediction model based on the source domain data set, the target domain data set, and the source domain data set and the target domain data set after data enhancement to obtain a trained tower health status prediction model; The state prediction module is used to predict the health state of the tower to be detected based on the trained tower health state prediction model.

10. An electronic device, characterized in that: It includes a processor, and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in a server; when the processor is running, it calls a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the method described in any one of claims 1 to 7.