Cable protection detection method based on optical fiber vibration measurement technology

Abstract

The present invention relates to a cable protection detection method based on optical fiber vibration measurement technology, comprising the following steps: Step 1, detecting the vibration signal of the optical fiber arranged along the cable in real time by means of an optical fiber interference sensor; Step 2, bringing the vibration signal into a DCN neural network trained by a spectrum sliding detection algorithm; Step 3, the DCN neural network identifies the input vibration signal and determines the event causing the vibration. The beneficial effects of the present invention are: using the optical cable as a sensing unit, continuously collecting external vibration signals and generating alarm information. When an external damage event occurs, it can be detected in time and the alarm information can be sent to the person in charge. If necessary, the external damage event can be promptly remedied, which greatly reduces the economic losses and personnel monitoring needs caused by the external damage of the facility, and is of great significance to perimeter security. It can effectively process the aliased signal containing the single-tone vibration signal, obtain the single-tone vibration signal and determine the event causing the vibration.

Description

Technical Field

[0001] The present invention relates to the field of cable protection detection, and in particular to a cable protection detection method based on optical fiber vibration measurement technology. Background Art

[0002] With the acceleration of urban development, underground cables are becoming an increasingly common feature of urban areas. Due to their high copper content, dispersed distribution, and unattended operation, these cables have become a target for thieves, posing a significant social risk. Furthermore, due to the inherent nature of cable installation, their widespread distribution makes monitoring difficult, exposing them to increasing risk of damage. In particular, the widespread practice of unauthorized construction without prior approval complicates oversight by power authorities.

[0003] Traditionally, underground cable protection relies solely on dedicated patrols. This method, however, lacks automation and intelligence, resulting in suboptimal protection. Therefore, distributed fiber optic sensing technology is currently the primary method for monitoring underground cables and preventing unauthorized intrusion. However, signal recognition in existing distributed fiber optic sensors is susceptible to environmental factors such as weather, animals, and industrial activity, resulting in a high false alarm rate. Furthermore, the prevalence of aliased signals in real-world situations hinders the ability of distributed fiber optic sensors to detect intrusion events. Summary of the Invention

[0004] The purpose of the present invention is to provide a cable protection detection method based on optical fiber vibration measurement technology to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A cable protection detection method based on optical fiber vibration measurement technology includes the following steps:

[0007] Step 1: Detecting the vibration signal of the optical fiber arranged along the cable in real time through the optical fiber interferometer sensor;

[0008] Step 2: The vibration signal is fed into the DCN neural network trained with the spectrum sliding detection algorithm.

[0009] Step 3: The DCN neural network identifies the input vibration signal and determines the event that causes the vibration.

[0010] As a further solution of the present invention: the DCN neural network recognizes the input vibration signal, including the following steps:

[0011] Determine whether the input vibration signal is an aliased signal type or a single-tone vibration signal type; if it is a single-tone vibration signal type, extract a portion of the input vibration signal that matches the single-tone vibration signal, and determine the event causing the vibration based on the single-tone vibration signal in the portion;

[0012] If the input vibration signal is of the aliasing signal type, the DCN neural network is used to determine the single-tone vibration signal in the input vibration signal, and the frequency and probability of the single-tone vibration signal are obtained; if the probability of the single-tone vibration signal is higher than the preset threshold, it is determined that the vibration signal contains a single-tone vibration signal, and the event causing the vibration is determined based on the obtained frequency of the single-tone vibration signal.

[0013] As a further solution of the present invention: if the input vibration signal presents the characteristics of a single-tone vibration signal in the time domain exceeding a preset threshold range, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliasing signal type.

[0014] As a further solution of the present invention: if the input vibration signal presents the characteristics of a single-tone vibration signal in the time domain for more than 70%, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliasing signal type.

[0015] As a further solution of the present invention: for the type of aliased signal, a method for identifying the type of aliased signal through DCN neural network recognition includes the following steps:

[0016] Step 1, set the initial step size and initial window length to input the detector;

[0017] Step 2: Separate individual spectra of equal bandwidth from the entire spectrum at regular intervals. Spectra close to the single-tone vibration signal in the dataset are taken as positive samples, and spectra far from the signal in the dataset are taken as negative samples.

[0018] Step 3: Separate the equal-bandwidth spectrum from the entire spectrum at intervals and input it into the DCN neural network. The DCN neural network outputs the recognition result, and calculates the frequency of the single-tone vibration signal based on the recognition result.

[0019] Step 4: Output the probability of a single-tone vibration signal in the spectrum;

[0020] Step 5: If the probability of the single-tone vibration signal is higher than the preset threshold, the event causing the vibration is determined based on the frequency of the single-tone vibration signal; if the probability of the single-tone vibration signal is not higher than the preset threshold, the single-tone vibration signal is exempted from being determined as an invalid signal.

[0021] As a further solution of the present invention: the DCN neural network is trained using a labeled single-tone vibration signal training data set.

[0022] As a further solution of the present invention: the vibration signal of the optical fiber interference sensor is collected into a computer through an analog-to-digital converter.

[0023] As a further solution of the present invention: the vibration signal is pre-processed and then introduced into the DCN neural network trained by the spectrum sliding detection algorithm.

[0024] As a further solution of the present invention: preprocessing of the vibration signal includes: normalization processing.

[0025] As a further solution of the present invention: the preprocessing of the vibration signal also includes: smoothing and noise reduction processing.

[0026] The present invention is beneficial in that it uses optical cables as sensing units to continuously collect external vibration signals and generate alarm information. When a breach occurs, the alarm is promptly detected and sent to responsible personnel, allowing for prompt remediation if necessary. This significantly reduces the economic losses and personnel monitoring required by facility breaches, significantly improving perimeter security.

[0027] It can effectively process aliased signals containing single-tone vibration signals and obtain single-tone vibration signals to determine the events that cause vibration.

[0028] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a flow chart of a cable protection detection method based on optical fiber vibration measurement technology of the present invention. DETAILED DESCRIPTION

[0030] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0031] like Figure 1 As shown, in an embodiment of the present invention, a cable protection detection method based on optical fiber vibration measurement technology includes the following steps:

[0032] Step 1: Detecting the vibration signal of the optical fiber arranged along the cable in real time through the optical fiber interferometer sensor;

[0033] Step 2: The vibration signal is fed into the DCN neural network trained with the spectrum sliding detection algorithm.

[0034] Step 3: The DCN neural network identifies the input vibration signal and determines the event that causes the vibration.

[0035] Different types of external damage can be determined based on different signal frequencies. Using optical cables as sensing units, the system continuously collects external vibration signals and generates alarm information. When an external damage event occurs, the system promptly detects it and sends an alarm message to the responsible personnel, enabling prompt remediation if necessary. This significantly reduces the economic losses caused by facility damage and the need for personnel monitoring, significantly enhancing perimeter security.

[0036] As a specific implementation, the DCN neural network recognizes the input vibration signal, including the following steps:

[0037] Determine whether the input vibration signal is an aliased signal type or a single-tone vibration signal type; if it is a single-tone vibration signal type, extract a portion of the input vibration signal that matches the single-tone vibration signal, and determine the event causing the vibration based on the single-tone vibration signal in the portion;

[0038] If the input vibration signal is of the aliasing signal type, the DCN neural network is used to determine the single-tone vibration signal in the input vibration signal, and the frequency and probability of the single-tone vibration signal are obtained; if the probability of the single-tone vibration signal is higher than the preset threshold, it is determined that the vibration signal contains a single-tone vibration signal, and the event causing the vibration is determined based on the obtained frequency of the single-tone vibration signal.

[0039] As a specific implementation, if the input vibration signal exhibits the characteristics of a single-tone vibration signal within a range exceeding a preset threshold in the time domain, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliased signal type. Specifically, if the input vibration signal exhibits the characteristics of a single-tone vibration signal within a range exceeding 70% in the time domain, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliased signal type.

[0040] As a specific implementation, for the aliased signal type, a method for identifying the aliased signal type through DCN neural network recognition includes the following steps:

[0041] Step 1, set the initial step size and initial window length to input the detector;

[0042] Step 2: Separate individual spectra of equal bandwidth from the entire spectrum at regular intervals. Spectra close to the single-tone vibration signal in the dataset are taken as positive samples, and spectra far from the signal in the dataset are taken as negative samples.

[0043] Step 3: Separate the equal-bandwidth spectrum from the entire spectrum at intervals and input it into the DCN neural network. The DCN neural network outputs the recognition result, and calculates the frequency of the single-tone vibration signal based on the recognition result.

[0044] Step 4: Output the probability of a single-tone vibration signal in the spectrum;

[0045] Step 5: If the probability of the single-tone vibration signal is higher than the preset threshold, the event causing the vibration is determined based on the frequency of the single-tone vibration signal; if the probability of the single-tone vibration signal is not higher than the preset threshold, the single-tone vibration signal is exempted from being determined as an invalid signal.

[0046] It can effectively process aliased signals containing single-tone vibration signals and obtain single-tone vibration signals to determine the events that cause vibration.

[0047] As a specific implementation, the DCN neural network is trained using a labeled single-tone vibration signal training data set.

[0048] Specifically, the vibration signal of the fiber optic interferometer sensor is collected into a computer through an analog-to-digital converter.

[0049] As a specific implementation method, the vibration signal is pre-processed and then fed into a DCN neural network trained with a spectrum sliding detection algorithm. Specifically, the vibration signal pre-processing includes normalization and smoothing noise reduction.

[0050] The DCN neural network is trained and tested on a test set consisting of 4,800 spectra, including three types of aliased signals and eight single-tone vibration signal frequencies. The training set consists of 2,400 spectra, including three types of named signals.

[0051] For field testing of the system, the calibrated fiber optic vibration detection equipment was transported to the test site for vibration measurement of buried optical fibers. The fiber was buried at a depth of 10 cm. An impact drill was used to continuously impact externally broken rocks at locations 65 cm and 120 cm from the fiber axis, simulating cable theft. The fiber signals collected on-site demonstrated that externally broken rocks can be detected by the buried optical fiber vibration measurement system, and that the vibration signals are significantly attenuated at relatively close distances. The system can identify the type of externally broken rocks and determine the event causing the vibration.

[0052] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

[0053] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A cable protection detection method based on optical fiber vibration measurement technology, characterized in that: The following steps are involved: Step 1: Detecting the vibration signal of the optical fiber arranged along the cable in real time through the optical fiber interferometer sensor; Step 2: The vibration signal is fed into the DCN neural network trained with the spectrum sliding detection algorithm. Step 3: The DCN neural network identifies the input vibration signal and determines the event that causes the vibration; The DCN neural network recognizes the input vibration signal, including the following steps: Determine whether the input vibration signal is an aliased signal type or a single-tone vibration signal type; if it is a single-tone vibration signal type, extract a portion of the input vibration signal that matches the single-tone vibration signal, and determine the event causing the vibration based on the single-tone vibration signal in the portion; If the input vibration signal is of the aliasing signal type, the DCN neural network is used to determine the single-tone vibration signal in the input vibration signal, and the frequency and probability of the single-tone vibration signal are obtained. If the probability of the single-tone vibration signal is higher than a preset threshold, it is determined that the vibration signal contains a single-tone vibration signal, and the event causing the vibration is determined based on the obtained frequency of the single-tone vibration signal. If the input vibration signal exhibits the characteristics of a single-tone vibration signal within a preset threshold range in the time domain, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliased signal type; For the aliased signal type, the method for identifying the aliased signal through the DCN neural network includes the following steps: Step 1, set the initial step size and initial window length to input the detector; Step 2: Separate individual spectra of equal bandwidth from the entire spectrum at regular intervals. Spectra close to the single-tone vibration signal in the dataset are taken as positive samples, and spectra far from the signal in the dataset are taken as negative samples. Step 3: Separate the equal-bandwidth spectrum from the entire spectrum at intervals and input it into the DCN neural network. The DCN neural network outputs the recognition result, and calculates the frequency of the single-tone vibration signal based on the recognition result. Step 4: Output the probability of a single-tone vibration signal in the spectrum; Step 5: If the probability of the single-tone vibration signal is higher than the preset threshold, the event causing the vibration is determined based on the frequency of the single-tone vibration signal; if the probability of the single-tone vibration signal is not higher than the preset threshold, the single-tone vibration signal is exempted from being determined as an invalid signal.

2. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: If the input vibration signal exhibits the characteristics of a single-tone vibration signal in the time domain for more than 70% of the range, the vibration signal is determined to be a single-tone vibration signal type; if the vibration signal is not determined to be a single-tone vibration signal type, the vibration signal is identified as an aliasing signal type.

3. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: The DCN neural network is trained using a labeled single-tone vibration signal training dataset.

4. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: The vibration signal of the fiber optic interferometer sensor is collected into the computer through an analog-to-digital converter.

5. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: After preprocessing, the vibration signal is introduced into the DCN neural network trained with the spectrum sliding detection algorithm.

6. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: The preprocessing of vibration signals includes normalization processing.

7. The cable protection detection method based on optical fiber vibration measurement technology according to claim 1 is characterized in that: The preprocessing of vibration signals also includes: smoothing and noise reduction processing.

Images