Apparatus for determining leakage using pipeline characteristics and noise pattern, and learning method thereof

Abstract

An apparatus for determining leakage using pipeline characteristics and a noise pattern according to the present disclosure comprises: a memory in which at least one process for determining leakage using pipeline characteristics and a noise pattern is stored; and a processor for performing an operation according to the process, wherein the processor may be configured to: preset a pattern of a leakage signal generated in a pipeline; collect a vibration signal from a sensor unit installed in the pipeline, process the vibration signal by removing a signal according to the pipeline characteristics from the collected vibration signal; compare the processed vibration signal with a preset noise pattern so as to process the vibration signal by removing a corresponding noise pattern; and make the vibration signal from which the noise pattern has been removed into learning data so as to compare the processed vibration signal with the set pattern of the leakage signal and determine whether a leakage has occurred.

Description

A device for determining water leakage using pipe characteristics and noise patterns and its learning method

[0001] The present disclosure relates to a water leak determination device and a learning method thereof, and more specifically, to a water leak determination device using pipe characteristics and noise patterns that classifies collected signals according to pipe characteristics and removes corresponding noise pattern signals from the classified signals to perform water leak determination.

[0002] The inside of pipes such as water pipes may accumulate foreign substances through repeated use, sedimentation, accumulation, and floating, which may lead to corrosion. In addition, cracks may occur due to influences from the external environment, which may cause leakage in the pipes.

[0003] In particular, water pipes used in various buildings such as apartments, single-family homes, and buildings are made of pipes made of various materials such as polybutylene pipes, metal corrugated pipes, copper pipes, stainless steel pipes, polyethylene pipes, and cross-linked polyethylene pipes.

[0004] When a leak occurs in a water pipe, a leak detection device must be used to locate the leak. However, because the characteristics of each pipe material vary, the accuracy of a single leak detection device may vary depending on the material.

[0005] In the past, a method of detecting whether there was nitrogen leakage using a nitrogen detector was used to detect the location of a leak regardless of the material of the pipeline. However, there was a problem in that it was physically difficult to detect the exact location of a leak in the pipeline by detecting the location of the leaking nitrogen.

[0006] In addition, even if an artificial intelligence model is used to learn how to find leak detection locations, it is difficult to maintain high prediction accuracy because the learning data differs depending on the water pipe material.

[0007] In addition, when collecting vibration signals to detect the location of a water leak, the noise from the surroundings or the water pipes used by consumers is also collected along with the vibration signals, which reduces the accuracy of detecting the location of a water leak.

[0008] The embodiment of the present disclosure is proposed to solve the above-described problem, and can provide a leak determination device and a learning method thereof using pipe characteristics and noise patterns that classify vibration signals according to the material characteristics and noise types of the pipe and detect an accurate leak location.

[0009] The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0010] In order to solve the above-described problem, a device for determining a water leak using a pipe characteristic and a noise pattern according to the present disclosure includes a memory storing at least one process for determining a water leak using a pipe characteristic and a noise pattern, and a processor performing an operation according to the process, wherein the processor may be configured to preset a pattern of a water leak signal generated in the pipe, collect a vibration signal from a sensor unit installed in the pipe, remove a signal according to the pipe characteristic from the collected vibration signal, compare the processed vibration signal with a preset noise pattern, remove a corresponding noise pattern, and generate a vibration signal from which the noise pattern has been removed as learning data to determine whether a water leak has occurred by comparing the pattern with the preset leak signal.

[0011] A method for generating learning data for determining a water leak using pipe characteristics and noise patterns, performed by a processor of a device according to the present disclosure for solving the above-described problem, may include the steps of: pre-setting a pattern of a water leak signal generated in a pipe; collecting a vibration signal from a sensor unit installed in the pipe; removing and processing a signal according to the pipe characteristics from the collected vibration signal; comparing the processed vibration signal with a preset noise pattern and removing and processing a corresponding noise pattern; and generating a vibration signal from which the noise pattern has been removed as learning data so as to determine whether a water leak has occurred by comparing the pattern of the preset water leak signal.

[0012] According to the aforementioned problem solving means of the present disclosure, when detecting a leak location using a plurality of sensor units in a pipe of different materials, the signal is classified according to the material characteristics and noise type of the pipe, and only the signal generated by the leak is used, thereby providing an effect of detecting an accurate leak location.

[0013] The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned can be clearly understood by those skilled in the art from the description below.

[0014] FIG. 1 is a block diagram schematically illustrating a configuration for detecting a leak location in a leak determination device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0015] FIG. 2 is a block diagram briefly illustrating a configuration for determining whether there is a leak in a leak determination device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0016] FIG. 3 is a conceptual diagram illustrating the positions of the first sensor unit and the second sensor unit of a water leak detection device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0017] FIG. 4 illustrates a leak signal and a leak signal including noise collected by a leak determination device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0018] FIG. 5 is a flowchart illustrating a learning method of a leak detection device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0019] FIGS. 6 to 8 are graphs illustrating noise signal patterns of a leak detection device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0020] FIG. 9 illustrates a computing device of the present disclosure.

[0021] Throughout this disclosure, the same reference numerals denote the same components. This disclosure does not describe all elements of the embodiments, and any content that is common in the technical field to which this disclosure pertains or that overlaps between embodiments is omitted. The terms "part, module, element, block" used in the specification may be implemented in software or hardware, and depending on the embodiments, multiple "parts, modules, elements, blocks" may be implemented as a single component, or a single "part, module, element, block" may include multiple components.

[0022] Throughout the specification, when a part is said to be "connected" to another part, this includes not only direct connection but also indirect connection, and indirect connection includes connection via a wireless communication network.

[0023] Additionally, when a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise specifically stated.

[0024] Throughout the specification, when we say that an element is "on" another element, this includes not only cases where the element is in contact with the other element, but also cases where another element exists between the two elements.

[0025] The terms first, second, etc. are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

[0026] Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0027] The identification codes for each step are used for convenience of explanation and do not describe the order of each step. Each step may be performed in a different order than specified unless the context clearly indicates a specific order.

[0028] The operating principle and embodiments of the present disclosure are described below with reference to the attached drawings.

[0029] As used herein, the term "device according to the present disclosure" encompasses a variety of devices capable of performing computational processing and providing results to a user. For example, the device according to the present disclosure may include a computer, a server device, and a portable terminal, or may be any one of them.

[0030] Here, the computer may include, for example, a notebook, desktop, laptop, tablet PC, slate PC, etc. equipped with a web browser.

[0031] The above server device is a server that processes information by communicating with an external device, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0032] The above portable terminal may include, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as a PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminal, a smart phone, and a wearable device such as a watch, a ring, a bracelet, an anklet, a necklace, glasses, contact lenses, or a head-mounted device (HMD).

[0033] The artificial intelligence-related functions according to the present disclosure are operated via a processor and memory. The processor may be comprised of one or more processors. In this case, one or more processors may be a general-purpose processor such as a CPU, an AP, a DSP (Digital Signal Processor), a graphics-only processor such as a GPU or a VPU (Vision Processing Unit), or an artificial intelligence-only processor such as an NPU. One or more processors control the processing of input data according to predefined operating rules or artificial intelligence models stored in memory. Alternatively, if one or more processors are artificial intelligence-only processors, the artificial intelligence-only processor may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0034] The predefined operation rules or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that the basic artificial intelligence model is learned by a learning algorithm using a plurality of learning data, thereby creating a predefined operation rules or artificial intelligence model set to perform a desired characteristic (or purpose). This learning may be performed on the device itself on which the artificial intelligence according to the present disclosure is performed, or may be performed through a separate server and / or system. Examples of the learning algorithm include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0035] An artificial intelligence model may be composed of multiple neural network layers. Each of the multiple neural network layers has multiple weight values, and performs neural network operations through operations between the operation results of the previous layer and the multiple weights. The multiple weights of the multiple neural network layers may be optimized based on the learning results of the artificial intelligence model. For example, the multiple weights may be updated so that the loss value or cost value obtained from the artificial intelligence model is reduced or minimized during the learning process. The artificial neural network may include a deep neural network (DNN), and examples thereof include, but are not limited to, a convolutional neural network (CNN), a deep neural network (DNN), a recurrent neural network (RNN), a restricted boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), or deep Q-networks.

[0036] According to an exemplary embodiment of the present disclosure, a processor can implement artificial intelligence. Artificial intelligence refers to a machine learning method based on an artificial neural network that imitates human neurons (biological neurons) to enable machines to learn. Artificial intelligence methodologies can be categorized into supervised learning, in which input data and output data are provided together as training data depending on the learning method, so that the solution (output data) to the problem (input data) is determined; unsupervised learning, in which only input data is provided without output data, so that the solution (output data) to the problem (input data) is not determined; and reinforcement learning, in which a reward (Reward) is provided from an external environment whenever an action (Action) is taken in the current state (State), and learning is performed in a direction to maximize this reward. In addition, artificial intelligence methodologies can be categorized according to the architecture of the learning model. The architectures of widely used deep learning technologies can be categorized into convolutional neural networks (CNNs), recurrent neural networks (RNNs), transformers, and generative adversarial networks (GANs).

[0037] The present device and system may include an artificial intelligence model. The artificial intelligence model may be a single artificial intelligence model or may be implemented as multiple artificial intelligence models. The artificial intelligence model may be composed of a neural network (or artificial neural network) and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to a model in general that has problem-solving capabilities by changing the binding strength of synapses through learning, formed by artificial neurons (nodes) that form a network by combining synapses. The neurons of the neural network may include a combination of weights or biases. The neural network may include one or more layers composed of one or more neurons or nodes. For example, the device may include an input layer, a hidden layer, and an output layer. The neural network constituting the device can infer a desired result (output) from an arbitrary input (input) by changing the weights of neurons through learning.

[0038] The processor can create a neural network, train (or learn) a neural network, perform a calculation based on received input data, generate an information signal based on the calculation result, or retrain the neural network. The models of the neural network can include various types of models such as CNN (Convolution Neural Network) such as GoogleNet, AlexNet, VGG Network, R-CNN (Region with Convolution Neural Network), RPN (Region Proposal Network), RNN (Recurrent Neural Network), S-DNN (Stacking-based deep Neural Network), S-SDNN (State-Space Dynamic Neural Network), Deconvolution Network, DBN (Deep Belief Network), RBM (Restrcted Boltzman Machine), Fully Convolutional Network, LSTM (Long Short-Term Memory) Network, Classification Network, etc., but are not limited thereto. The processor can include one or more processors for performing calculations according to the models of the neural network. For example, the neural network can be a deep neural network. It may include a deep neural network.

[0039] Neural networks include CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), perceptron, multilayer perceptron, FF (Feed Forward), RBF (Radial Basis Network), DFF (Deep Feed Forward), LSTM (Long Short Term Memory), GRU (Gated Recurrent Unit), AE (Auto Encoder), VAE (Variational Auto) Encoder), DAE (Denoising Auto Encoder), SAE (Sparse Auto Encoder), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Depp Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), Generative Adversarial Network (GAN), Liquid State Machine (LSM), Extreme Learning Machine (ELM), It will be understood by those skilled in the art that any neural network may be included, including but not limited to ESN (Echo State Network), DRN (Deep Residual Network), DNC (Differentiable Neural Computer), NTM (Neural Turning Machine), CN (Capsule Network), KN (Kohonen Network), and AN (Attention Network).

[0040] According to an exemplary embodiment of the present disclosure, the processor may be configured to perform a process for generating a CNN (Convolution Neural Network) such as GoogleNet, AlexNet, VGG Network, Region with Convolution Neural Network (R-CNN), Region Proposal Network (RPN), Recurrent Neural Network (RNN), Stacking-based deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restrcted Boltzman Machine (RBM), Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT, SP-BERT, MRC / QA for natural language processing, Text Analysis, Dialog System, GPT-3, GPT-4, Visual Analytics for vision processing, Visual Understanding, Video Synthesis, ResNet for data intelligence, Anomaly Detection, Prediction, Time-Series Forecasting, Various artificial intelligence structures and algorithms, including optimization, recommendation, and data creation, can be utilized, but are not limited thereto. Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0041] FIG. 1 is a block diagram briefly illustrating a configuration for detecting a leak location in a leak determination device using pipe characteristics and noise patterns according to various embodiments of the present disclosure, and FIG. 2 is a block diagram briefly illustrating a configuration for determining whether or not there is a leak in a leak determination device using pipe characteristics and noise patterns according to various embodiments of the present disclosure.

[0042] Hereinafter, a leak determination device and a learning method thereof using pipe characteristics and noise patterns according to the present disclosure will be described with reference to FIGS. 1 to 8.

[0043] At least one component may be added or deleted to correspond to the performance of the components illustrated in Figure 1. Furthermore, it will be readily apparent to those skilled in the art that the relative positions of the components may be altered to correspond to the performance or structure of the system.

[0044] Each component illustrated in Figure 1 represents software and / or hardware components such as a Field Programmable Gate Array (FPGA) and an Application Specific Integrated Circuit (ASIC).

[0045] As illustrated in FIG. 1, a leak determination device using pipe characteristics and noise patterns according to the present disclosure may include a first sensor unit (111) positioned at one end of a pipe made of a first material, a second sensor unit (112) positioned at the other end of a pipe made of a second material, a first detection unit (121) for detecting a leak location in the pipe using the first sensor unit (111) based on the pipe made of the first material, a second detection unit (122) for detecting a leak location in the pipe using the second sensor unit (112) based on the pipe made of the second material, a first leak location correction unit (131) for reflecting a first correction value considering the characteristics of the first material to the leak location detected by the first detection unit (121), and a second leak location correction unit (132) for reflecting a second correction value considering the characteristics of the second material to the leak location detected by the second detection unit (122).

[0046] In addition, the leak location detection device according to the present disclosure may further include a leak location determination unit (140) that determines an average value of a first leak location reflecting the first correction value and a second leak location reflecting the second correction value as a final leak location.

[0047] In addition, the leak determination device using the pipe characteristics and noise pattern according to the present disclosure may include a leak determination unit (200) that determines whether a leak is actually occurring at the determined final leak location.

[0048] Specifically, as illustrated in FIG. 2, the leak determination unit (200) may include a signal collection unit (210) that collects vibration signals from a first sensor unit (111) and a second sensor unit (112) arranged in a pipe.

[0049] The vibration signal collected from the signal collection unit (210) may be a signal including various signals such as a signal generated by the characteristics of the pipeline, an internal noise signal generated inside the pipeline, an external noise signal generated outside the pipeline, and a water leakage signal generated at a location where a water leakage has occurred.

[0050] Accordingly, the pipe characteristic classification unit (220) of the leak determination unit (200) can classify vibration signals according to pre-stored pipe characteristics. The pipe characteristic may be at least one of the material, diameter, and thickness of the pipe. In particular, when the pipe is composed of multiple different materials, the characteristic according to each material may be included.

[0051] In addition, before classifying the vibration signal according to the pipe characteristics in the pipe characteristic classification unit (220), the collected vibration signal can be converted into a frequency signal, and the converted frequency signal can be divided into multiple sections.

[0052] For example, a vibration signal in the time domain can be converted into a frequency signal by frequency conversion, and the converted frequency signal can be divided into 100 Hz units by frequency band, and then classified by material, diameter, and thickness, which are the characteristics of the pipe.

[0053] Rather than comparing the entire signal, the differences can be detected in greater detail by comparing each divided unit. The smaller the number of unit units used to divide the frequency signal, the more frequency signals are generated, increasing the amount of data. Conversely, the larger the number of unit units used, the fewer frequency signals are generated, reducing the amount of data.

[0054] By segmenting the frequency signal and increasing the amount of data, more accurate and detailed values ​​can be obtained while obtaining more training data.

[0055] Therefore, the converted frequency signal can be divided by setting the size of the unit according to the converted frequency characteristics or the amount of data of the acquired signal.

[0056] In other words, the leak determination unit (200) can adjust the unit for dividing the converted frequency signal according to the number of collected vibration signals.

[0057] The noise filtering unit (230) of the water leak determination unit (200) can filter the noise frequency signal within the frequency signal by comparing the frequency signal classified according to the pipe characteristics with the corresponding section of the preset water leak signal.

[0058] The noise filtering unit (230) can calculate a noise signal by comparing a frequency signal containing noise with a reference leakage signal pattern set as a reference leakage signal.

[0059] If the calculated noise frequency signal is within the noise band, it can be filtered using a bandpass filter in the frequency domain. Alternatively, if the difference between the reference leakage signal and the noise frequency signal is outside the noise band, the noise can be minimized by calculating the difference in signal intensity.

[0060] The learning unit (240) of the leak determination unit (200) performs learning using the processed signal, and can determine whether there is a leak based on whether the processed signal includes a leak signal corresponding to the reference leak signal.

[0061] As an example, the noise frequency signal may vary depending on the noise type.

[0062] The above noise type may include internal noise occurring inside the pipe and external noise occurring outside the pipe.

[0063] For example, the internal noise may include the sound of water being used or the cracking sound of water hitting the pipe, and the external noise may include at least one of the driving sound of a control device controlling the pipe, the driving sound of a transformer, the driving sound of a fan, and the sound of a train passing by. In addition, noises generated inside and outside the pipe may be included.

[0064] Accordingly, the water leak determination unit (200) can perform water leak determination using an artificial intelligence model, and input a time-converted signal of a frequency signal with the noise frequency signal removed into the learned artificial intelligence model to determine whether a water leak has occurred in the pipe.

[0065] In relation to the pipe characteristic classification section (220), as shown in FIG. 3, it is assumed that the first material is material A (11), the second material is material B (12), the pipe length of material A (11) is La, the pipe length of material B (12) is Lb, and the length of the entire pipe section where a leak occurs is L.

[0066] A first sensor unit (111) may be installed at one end of material A (11), and a second sensor unit (112) may be installed at one end of material B (12). At this time, one end of material B (12) may be an end spaced apart from material A (11).

[0067] As an example, the first sensor unit (111) or the second sensor unit (112) may be a vibration sensor. The location of the water leak can be detected using information obtained by transmitting a vibration waveform to the location of the water leak.

[0068] For example, when a leak location occurs in material A (11), the vibration waveform generated from the first sensor unit (111) passes only through material A (11) to reach the leak location, and the vibration waveform generated from the second sensor unit (112) passes through material B (12) and material A (11) to reach the leak location.

[0069] As a result, a signal passing only through material A (11) is only affected by the pipe signal characteristic value of material A (11), and a signal passing through material A (11) and material B (12) can be affected by both the pipe signal characteristic value of material A (11) and the pipe signal characteristic value of material B (12).

[0070] Therefore, in a pipeline composed of different materials, a correction value that takes into account the characteristics of each material can be reflected in the measured value of the sensor section to detect the exact location of the leak.

[0071] In addition, the leak determination unit (200) can calculate a leak characteristic value for each pipe material by multiplying the vibration transmission speed in the first material or the second material by the length of the pipe that is the first material or the length of the pipe that is the second material among the total pipe length.

[0072] The pipe characteristic classification unit (220) of the above-mentioned leak determination unit (200) can classify the leak characteristic values ​​by pipe material and remove the pipe material-specific characteristics from the leak signal, and the learning unit (240) can train an artificial intelligence model to output a leak determination by learning a leak signal that does not include the pipe material-specific characteristics.

[0073] The reference leak value can be set by extracting it from a database containing existing leak measurement results.

[0074] With respect to the noise filtering unit (230), as illustrated in FIG. 4, noise filtering can be performed by comparing a leakage signal with a leakage signal containing noise. At this time, the leakage signal may be a reference leakage signal, and the leakage signal containing noise may be a frequency signal converted from a vibration signal collected by the signal collection unit (210).

[0075] Referring to FIG. 4, the noise filtering unit (230) can divide the leakage signal and the leakage signal containing noise into three sections in 100 Hz units and compare the corresponding frequency bands. Since the maximum signal frequency of the leakage signal is 1192 Hz and the maximum signal frequency of the leakage signal containing noise is 1417 Hz, it is known that the leakage signal containing noise contains noise and noise filtering can be performed.

[0076] If the remaining signal components of the leak signal containing noise are matched with the leak occurrence signal stored in the DB after noise filtering, the leak determination unit (200) can determine that a leak is occurring and provide an alarm to the user.

[0077] Through the above-described process, it is possible to accurately determine a water leak by removing all external factors other than the signal component generated by the water leak.

[0078] Accordingly, a device for determining a water leak using a pipe characteristic and a noise pattern according to the present disclosure may include a memory storing at least one process for determining a water leak using the pipe characteristic and a noise pattern, and a processor performing an operation according to the process. The processor of the device for determining a water leak using a pipe characteristic and a noise pattern according to the present disclosure may be configured to preset a pattern of a water leak signal generated in the pipe, collect a vibration signal from a sensor unit installed in the pipe, remove a signal according to the pipe characteristic from the collected vibration signal, compare the processed vibration signal with a preset noise pattern to remove a corresponding noise pattern, and generate a vibration signal from which the noise pattern has been removed as learning data to determine whether a water leak has occurred by comparing the pattern with the preset leak signal.

[0079] As illustrated in Fig. 5, signals can be collected from the first sensor unit (111) or the second sensor unit (112) installed in the pipeline (S510), and the collected signals can be classified by pipeline material (S520). At this time, signal characteristics by pipeline material can be obtained from the pipeline material DB.

[0080] Additionally, signals classified by pipe material can be classified by pipe thickness (S530). At this time, pipe thickness signal characteristics can be obtained from the pipe thickness DB.

[0081] In addition, noise removal can be performed on signals classified according to pipe material and thickness, and no difference occurs depending on the pipe material and thickness (S540). Noise can be removed by considering noise patterns stored in the noise DB.

[0082] For example, as shown in FIGS. 6 to 8, frequency signals of water usage noise generated when water hits the pipe as residents use the tap water, transformer noise resonating in the pipe as the transformer is operated, curved pipe noise generated when water hits the pipe, fan noise generated when the fan is operated, electric train noise transmitted to the pipe as an electric train passes nearby, and control equipment noise generated when the control equipment is operated can be stored in a noise DB by converting them into a DB.

[0083] Noise can be removed based on the corresponding noise pattern in signals classified according to conduit material and thickness, taking into account the frequency band where the maximum signal frequency occurs or the signal strength. By classifying and removing noise by type, removal efficiency can be improved.

[0084] In addition to detecting and removing noise in response to the above-described noise pattern, it is also possible to remove signals in a frequency band of 400 Hz or less and signals in a frequency band of 1400 Hz or more.

[0085] Then, the characteristics of each material and thickness can be removed from the collected signal, and a signal with noise removed can be generated as learning data (S550).

[0086] As an example, the leak determination unit (200) may band-filter a frequency band including the noise frequency signal to remove the noise frequency signal from the frequency signal, or may amplify the signal according to the difference in signal intensity between a frequency band including the noise frequency signal and a frequency band not including the noise frequency signal.

[0087] Meanwhile, a learning method of a water leak determination device using pipe characteristics and noise patterns according to the present disclosure may include a step of presetting a pattern of a water leak signal generated in a pipe, a step of collecting a vibration signal from a sensor unit installed in the pipe, a step of removing and processing a signal according to the pipe characteristics from the collected vibration signal, a step of comparing the processed vibration signal with a preset noise pattern to remove and process a corresponding noise pattern, and a step of generating a vibration signal from which the noise pattern has been removed as learning data so as to determine whether a water leak has occurred by comparing the processed vibration signal with the preset pattern of the water leak signal. For the sake of brevity of the specification, any content overlapping with the above will be omitted.

[0088] FIG. 9 illustrates a computing device according to one embodiment of the present disclosure. As illustrated in FIG. 9, the water leak detection device (200) according to the present disclosure may be implemented in the form of a recording medium storing computer-executable instructions. The instructions may be stored in the form of program code, and when executed by a processor, may generate program modules to perform the operations of the disclosed embodiments. The recording medium may be implemented as a computer-readable recording medium.

[0089] The leak determination device according to the present disclosure may correspond to a computing device (12), and the computing device (12) may include at least one processor (14), a computer-readable storage medium (16) including a program (20), and a communication bus (18).

[0090] Additionally, the computing device (12) may include one or more input / output interfaces (22) and one or more network communication interfaces (26) that provide interfaces for input / output devices (24).

[0091] Computer-readable storage media include all types of storage media that store instructions that can be deciphered by a computer. Examples include read-only memory (ROM), random access memory (RAM), magnetic tape, magnetic disks, flash memory, and optical data storage devices.

[0092] The disclosed embodiments have been described with reference to the attached drawings as described above. Those skilled in the art will understand that the present disclosure can be implemented in forms other than the disclosed embodiments without altering the technical spirit or essential features of the present disclosure. The disclosed embodiments are illustrative and should not be construed as limiting.

Claims

1. A memory storing at least one process for determining a leak using pipe characteristics and noise patterns; and Includes a processor that performs operations according to the above process, The above processor, Pre-set the pattern of the leakage signal generated in the above pipe, Collect vibration signals from the sensor unit installed in the above pipe, The signal according to the above pipe characteristics is removed from the collected vibration signal, and The processed vibration signal is compared with a preset noise pattern and the corresponding noise pattern is removed. A leak determination device using pipeline characteristics and noise patterns configured to generate a vibration signal with the noise pattern removed as learning data to determine whether a leak has occurred by comparing it with the pattern of the set leak signal.

2. In paragraph 1, The above processor A leak detection device using pipe characteristics and noise patterns, which is further configured to convert a collected vibration signal into a frequency and divide the converted frequency signal into preset units.

3. In paragraph 2, The above processor A leak determination device using a pipe characteristic and a noise pattern, configured to classify the divided frequency signal according to the pipe characteristic, wherein the pipe characteristic includes at least one of a material, a diameter, and a thickness of the pipe.

4. In paragraph 3, The above processor A leak detection device using pipe characteristics and noise patterns configured to compare a frequency signal classified according to the pipe characteristics with a corresponding section of a preset leak signal and output a noise frequency signal within the frequency signal.

5. In paragraph 4, The above processor, A leak detection device using pipe characteristics and noise patterns configured to determine whether a leak has occurred in the pipe by inputting a time-converted frequency signal from which the noise frequency signal has been removed into an artificial intelligence model.

6. In paragraph 5, The above noise frequency signal is A leak detection device using pipe characteristics and noise patterns that are characterized differently depending on the type of noise.

7. In paragraph 6, The above noise type is A leak detection device using pipe characteristics and noise patterns including internal noise generated inside the pipe and external noise generated outside the pipe.

8. In paragraph 5, The above internal noise includes the sound of water being used or the sound of water hitting the pipe, A leak detection device using the pipe characteristics and noise pattern, wherein the external noise includes at least one of a control device driving sound controlling the pipe, a transformer driving sound, a fan driving sound, and a train passing sound.

9. In paragraph 2, The above processor A leak detection device using pipe characteristics and noise patterns that adjusts the unit for dividing the converted frequency signal according to the number of collected vibration signals.

10. A method for generating learning data for determining a water leak using pipe characteristics and noise patterns, performed by a processor of a device, A step for presetting the pattern of a leakage signal generated in a pipeline; A step of collecting a vibration signal from a sensor unit installed in the above pipe; A step of removing and processing a signal according to the pipe characteristics from the collected vibration signal; A step of comparing the processed vibration signal with a preset noise pattern and removing the corresponding noise pattern; and A method comprising the step of generating a vibration signal from which the noise pattern has been removed as learning data to determine whether a leak has occurred by comparing it with the pattern of the set leak signal.

Images