Leakage detection system and method based on urban water supply pipeline leak detection robot

Abstract

The invention discloses a leak detection system and method based on an urban water supply pipeline leak detection robot, the system comprises a leak detection robot transceiver, a leak detection robot and a data processing computer, the leak detection robot is used for detecting a pipeline leak source and realizing positioning; the leak detection robot receiving and transmitting devices are arranged at the two ends of a to-be-detected pipeline and used for putting and withdrawing the leak detection robot, the leak detection robot is arranged in the to-be-detected pipeline and used for detecting the leakage position of the to-be-detected pipeline, and the data processing computer is arranged outside the to-be-detected pipeline. The receiving module is used for receiving offline signals recorded in the advancing process of the leak detection robot in the to-be-detected pipeline. The method is realized based on the system. The method has the advantages that efficient and accurate leakage detection and positioning of the urban water supply pipeline can be realized, the leakage rate of the urban water supply pipeline is reduced, and intelligent development of leakage detection of the urban water supply pipeline is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of acoustic detection of water supply pipelines, and in particular to a leakage detection system and method based on a leakage detection robot for urban water supply pipelines. Background Art

[0002] As my country's urbanization speed accelerates and the scale of cities continues to expand, the urban water supply network is also gradually expanding, which makes it more challenging to maintain the safe operation of the water supply system. However, due to the gradual aging of the water supply network, material corrosion, and changing geological conditions, the pipeline system often leaks. These leaks not only threaten the drinking water safety and quality of life of residents, but also cause waste of water resources and bring serious economic burdens. In particular, for urban water supply networks buried underground, although current technology can easily detect large-scale leaks, micro-leakage caused by small cracks is difficult to detect. Compared with large-scale leaks, these tiny leaks are difficult to detect in time and have a long existence time, which may eventually lead to more significant economic losses and continuous loss of water resources. In addition, for pipelines buried underground, the traditional methods of detecting their leaks are mostly sound detection or water pressure measurement, but these methods are usually inefficient and have limited accuracy. Summary of the invention

[0003] In view of the above defects or improvement needs of the prior art, the present invention aims to provide a leakage detection system and method based on a city water supply pipeline leakage detection robot, so as to realize efficient and accurate leakage detection and positioning of the city water supply pipeline, reduce the leakage rate of the city water supply pipeline, and promote the intelligent development of leakage detection of the city water supply pipeline; wherein, the pipeline leakage detection robot is an intelligent robot that can move inside the water supply pipeline and use multiple sensor fusion and other detection equipment to identify and locate the leakage point. The use of the pipeline leakage detection robot helps to improve the maintenance level of the urban pipe network system and greatly reduce the waste of water resources.

[0004] To achieve the above-mentioned purpose, one aspect of the present invention provides a leakage detection system based on a leakage detection robot for urban water supply pipelines, the system comprising a leakage detection robot transceiver, a leakage detection robot and a data processing computer, the leakage detection robot transceiver is arranged at both ends of the pipeline to be detected when in use, for launching and retrieving the leakage detection robot, the leakage detection robot is arranged inside the pipeline to be detected when in use, for detecting the location where the leakage occurs in the pipeline to be detected, and the data processing computer is arranged outside the pipeline to be detected when in use, for receiving an offline signal recorded by the leakage detection robot during its movement inside the pipeline to be detected;

[0005] The missed detection robot transceiver comprises a robot delivery claw, a robot recovery claw and a robot capture net, wherein the robot delivery claw is used to deliver the missed detection robot into the pipeline to be inspected, the robot capture net is used to capture the missed detection robot, and the robot recovery claw is used to recover the missed detection robot;

[0006] Among them, the missed detection robot includes a waterproof shell, anti-collision wheels and a laser scanner arranged on the top of the waterproof shell, and a hydrophone, a sonar speed meter, a nine-axis IMU sensor, a data acquisition main control board and a storage USB disk arranged inside the waterproof shell. The laser scanner, hydrophone, sonar speed meter and nine-axis IMU sensor are all electrically connected to the data acquisition main control board, and the data acquisition main control board is electrically connected to the storage USB disk.

[0007] To achieve the above object, another aspect of the present invention provides a leakage detection method based on a city water supply pipeline leakage detection robot, the method comprising the following steps:

[0008] S1. Put the missed detection robot into the target detection section of the pipeline to be detected, and drive the missed detection robot to move in the target detection section of the pipeline to be detected. During the movement of the missed detection robot, use the detection device to collect acoustic signals and position information of the missed detection robot. When the missed detection robot moves to the detection end point of the target detection section of the pipeline to be detected, take out the missed detection robot and use the data processing computer to receive the offline signal recorded during its movement;

[0009] S2, processing and identifying the acoustic signal collected in step S1, and determining the specific moving time of the missed detection robot when the leakage occurs;

[0010] S3, using the position information of the missed detection robot collected in step S1, combined with the position layout information of the target detection section of the pipeline to be detected and the size information of the pipeline to be detected, constructing the corresponding relationship between the position and time of the missed detection robot in the target detection section of the pipeline to be detected;

[0011] S4. Using the correspondence between the position and time of the missed detection robot in the target detection section of the pipeline to be detected constructed in step S2, combined with the specific moving time of the missed detection robot when the leak occurs obtained in step S2, the position where the leak occurs in the pipeline to be detected is determined, thereby realizing the detection of the leak.

[0012] Further, in step S1, the detection device is composed of a missed detection robot transceiver, a laser scanner, a hydrophone, a sonar velocimeter, a nine-axis IMU sensor, a data acquisition main control board and a storage USB flash drive; wherein, the missed detection robot transceiver is used for launching and recovering the missed detection robot, and the hydrophone is used for recording the ambient sound signal of the missed detection robot during its self-drifting process in the pipeline to be detected with the help of the signal amplifier; the laser scanner, the sonar velocimeter and the nine-axis IMU sensor are used for recording the position information of the missed detection robot.

[0013] Furthermore, in step S1, the position information of the missed detection robot includes parameter information of the nine-axis IMU sensor, the number of welds crossed by the missed detection robot recorded by the laser scanner, and the speed curve of the missed detection robot collected by the sonar speed meter.

[0014] Furthermore, in step S2, the collected acoustic signal is processed and identified by using an adaptive variational mode decomposition method, which specifically includes the following steps:

[0015] S21, setting the number of decomposition layers and penalty factors of the variational mode decomposition method, and decomposing the acoustic signal into a plurality of decomposition signals according to different decomposition parameters;

[0016] S22, calculating the mutual information entropy of each of the decomposed signals, and using multi-objective optimization iterative calculation to obtain the optimal decomposition parameter corresponding to the minimum mutual information entropy and the optimal decomposition signal corresponding to the optimal decomposition parameter;

[0017] S23, performing Fourier transform and screening denoising on the optimal decomposition signal to obtain a reconstructed signal, and calculating the bubble entropy value of the reconstructed signal to construct a feature vector of pipeline leakage;

[0018] S24. Using an extreme learning machine algorithm to identify the reconstructed signal, obtain a signal corresponding to the pipeline leakage and a time corresponding to the occurrence of the leakage.

[0019] Furthermore, in step S3, the method for establishing the corresponding relationship between the position and time of the missed detection robot in the target detection section of the pipeline to be detected is as follows:

[0020] S31. Use the nine-axis IMU sensor to collect the acceleration, angular velocity, and relative angle of the magnetic field of the missed robot when it is moving to calculate the position of the missed robot. At the same time, use the nine-axis IMU sensor as the main system to unify the timestamps of the nine-axis IMU sensor, laser scanner, and sonar speed meter to ensure that the data of different sensors are accurately matched at the same time to achieve synchronous positioning;

[0021] S32, using the fixed spacing between pipeline welds to assist positioning, inferring the pipeline position of the missed inspection robot based on the number of pipeline welds scanned by the laser scanner, and correcting the accumulated error generated by the inertial navigation nine-axis IMU sensor;

[0022] S33. Use a sonar speed meter to draw a speed curve of the missed detection robot during its movement, and calculate the travel distance of the missed detection robot and assist in positioning based on the relationship between speed and time.

[0023] S34. Utilize the particle filter data fusion algorithm to integrate the data position information of the nine-axis IMU sensor, laser scanner and sonar velocimeter to obtain the positioning result of the missed detection robot, and realize the correspondence between the position and time of the missed detection robot in the target detection section of the pipeline to be inspected.

[0024] Furthermore, in the step S24, the specific process of identifying and reconstructing the signal is: first, the acoustic signals under three working conditions of normal pipeline, leaking pipeline and ultrasonic assisted positioning signal are collected by the missed detection robot, and then processed by steps S21, S22 and S23 to form a training set, and then the training set is trained by the extreme learning machine method, and finally processed by steps S21, S22 and S23 to perform denoising and identification of multiple working conditions in the pipeline, and determine the travel time of the missed detection robot when the pipeline leaks.

[0025] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0026] 1. The present invention can achieve more accurate position and motion state estimation by fusing data from three different types of sensors: IMU, laser scanner and sonar velocimeter. Each sensor makes up for the limitations of other sensors, thereby significantly improving the measurement accuracy of the overall system, effectively reducing the impact of environmental changes on positioning accuracy, and increasing the robustness and reliability of the system.

[0027] 2. The present invention uses laser scanning technology to measure the geometric features of the inner wall of the pipeline, records the number of welds passed by the missed detection robot, and infers the position inside the pipeline based on this, effectively corrects the cumulative error generated by the inertial navigation IMU, and achieves precise positioning.

[0028] 3. The present invention realizes real-time speed detection of the missed-detection robot by using sonar speed measurement technology, calculates the travel distance of the missed-detection robot and assists in positioning.

[0029] 4. The present invention overcomes the disadvantages of traditional detection in water supply pipeline leakage detection, and can perform leakage detection without damaging the pipeline, changing the external environment, and without the need for personnel follow-up.

[0030] 5. The present invention utilizes intelligent parameter optimization to realize variational mode decomposition of pipeline signals, which can remove the interference of noise signals to the greatest extent and effectively retain the target signal.

[0031] 6. The present invention performs feature extraction on the reconstructed signal after filtering, constructs the feature limit of the reconstructed signal, makes up for the problem of insufficient information extraction of the leakage sound signal, can comprehensively reflect the acoustic characteristics of the pipeline detected during the movement of the missed detection robot, and realizes the accurate capture and identification of the pipeline leakage signal.

[0032] 7. The present invention adopts a self-floating pipeline detection device, which can realize rapid inspection of pipelines under the action of water flow, efficiently perform leakage detection, and greatly improve detection efficiency.

[0033] 8. The modular design of the present invention makes the system easy to maintain and upgrade, and each component can be upgraded or replaced individually according to technological development, thereby extending the service life of the system and maintaining its technological advancement. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solution in this embodiment, the following is a brief introduction to the drawings required for use in the description of the embodiment. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 It is a schematic diagram of a leakage detection system based on a city water supply pipeline leakage detection robot provided according to a preferred embodiment of the present invention;

[0036] Figure 2 It is a schematic diagram of a leakage detection robot transceiver device based on a city water supply pipeline leakage detection robot provided in accordance with a preferred embodiment of the present invention;

[0037] Figure 3 It is a schematic diagram of the structure of a city water supply pipeline leakage detection robot provided according to a preferred embodiment of the present invention;

[0038] Figure 4 It is a process flow chart of a leakage detection method based on a city water supply pipeline leakage detection robot provided in accordance with a preferred embodiment of the present invention;

[0039] Figure 5 It is a flow chart of adaptive variational modal signal denoising adopted in the leakage detection method based on the urban water supply pipeline leakage detection robot provided in the preferred embodiment of the present invention;

[0040] Explanation of the reference numerals: 1. transceiver of missed detection robot; 101. robot delivery claw; 102. robot recovery claw; 103. robot capture net; 2. missed detection robot; 201. waterproof housing; 202. anti-collision wheel; 203. laser scanner; 204. hydrophone; 205. sonar velocimeter; 206. nine-axis IMU sensor; 207. data acquisition main control board; 208. storage USB flash drive; 3. data processing computer; 4. pipeline to be detected; 5. pipeline weld; 6. pipeline leakage point. DETAILED DESCRIPTION

[0041] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following further describes how the present invention is implemented in conjunction with the accompanying drawings and specific implementation methods.

[0042] See also Figure 1 A preferred example of the present invention provides a leakage detection system based on a leakage detection robot for urban water supply pipelines, comprising a leakage detection robot transceiver 1, a leakage detection robot 2 and a data processing computer 3. The leakage detection robot transceiver 1 is arranged at both ends of a pipeline 4 to be detected when in use, the leakage detection robot 2 is arranged inside the pipeline 4 to be detected when in use, and is used to detect the location where the leakage occurs in the pipeline 4 to be detected, and the data processing computer 3 is arranged outside the pipeline 4 to be detected, and is used to receive an offline signal recorded by the leakage detection robot 2 during its movement inside the pipeline 4 to be detected.

[0043] Specifically, see Figure 2 The missed detection robot transceiver 1 comprises a robot delivery claw 101, a robot recovery claw 102 and a robot capture net 103; the robot delivery claw 101 is used to deliver the missed detection robot 2 into the pipeline 4 to be inspected, the robot capture net 103 is used to capture the missed detection robot 2, and the robot recovery claw 102 is used to recover the missed detection robot (2); when in use, the missed detection robot 2 is delivered into the pipeline 4 to be inspected by the robot delivery claw 101, and after the inspection is completed, the missed detection robot 2 is captured by the robot capture net 103, and then the missed detection robot 2 is recovered by the robot recovery claw 102.

[0044] Specifically, see Figure 3 The missed detection robot 2 includes a waterproof shell 201, an anti-collision wheel 202 and a laser scanner 203 arranged on the top of the waterproof shell 201, and a hydrophone 204, a sonar speed meter 205, a nine-axis IMU sensor 206, a data acquisition main control board 207 and a storage USB disk 208 arranged inside the waterproof shell 201. The laser scanner 203, the hydrophone 204, the sonar speed meter 205 and the nine-axis IMU sensor 206 are all electrically connected to the data acquisition main control board 207, and the data acquisition main control board 207 is electrically connected to the storage USB disk 208.

[0045] Specifically, in the preferred embodiment of the present invention, the hydrophone 204 is preferably a BRC-50A hydrophone of Guangzhou Banrichao Information Technology Co., Ltd., and the applicable frequency range is 1 Hz to 50 KHz. The function of the hydrophone 204 is to convert the collected sound signal into an electrical signal and then store it in the storage U disk 208 in the form of a digital signal through the data acquisition main control board 207.

[0046] Specifically, in the preferred example of the present invention, the nine-axis IMU sensor 206 is preferably the Lunqu Technology N100 inertial navigation, and the output frequency is 400 Hz.

[0047] See also Figure 4 The preferred embodiment of the present invention provides a leakage detection method based on a city water supply pipeline leakage detection robot, which is a detection method implemented based on the preferred embodiment of the leakage detection system based on the city water supply pipeline leakage detection robot, and specifically comprises the following steps:

[0048] S1, putting the missed detection robot 2 into the target detection section of the pipeline 4 to be detected, and driving the missed detection robot 2 to move inside the target detection section of the pipeline 4 to be detected. During the movement of the missed detection robot 2, the detection device is used to collect acoustic signals and position information of the missed detection robot. When the missed detection robot 2 moves to the detection end point of the target detection section of the pipeline 4 to be detected, the missed detection robot 2 is taken out and the data processing computer 3 is used to receive the offline signal recorded during its movement. Among them, the detection device is specifically composed of a missed detection robot transceiver 1, a laser scanner 203, a hydrophone 204, a sonar velocimeter 205, a nine-axis IMU sensor 206, a data acquisition main control board 207 and a storage U disk 206;

[0049] S2, collecting the acoustic signals collected in step S1 from the leakage detection robot 2 through the data processing computer 3, and processing and identifying the collected acoustic signals to determine the specific travel time of the leakage detection robot 2 when the leakage occurs;

[0050] S3, using the position information of the missed detection robot, combined with the position layout information of the target detection section of the pipeline 4 to be detected and the size information of the pipeline 4 to be detected, constructing the corresponding relationship between the position and time of the missed detection robot 2 in the target detection section of the pipeline 4 to be detected;

[0051] S4. Using the correspondence between the position and time of the leakage detection robot 2 in the target detection section of the pipeline 4 to be detected constructed in step S2, combined with the specific moving time of the leakage detection robot 2 when the leakage occurs obtained in step S2, the position where the leakage occurs in the pipeline 4 to be detected is determined, thereby realizing leakage detection.

[0052] Specifically, in step S1 , the position information of the missed detection robot includes parameter information of the nine-axis IMU sensor 206 , the number of pipeline welds 5 crossed by the missed detection robot 2 recorded by the laser scanner 203 , and the speed curve of the missed detection robot collected by the sonar speed meter 205 .

[0053] Specifically, in step S2, the collected acoustic signal is processed and identified by using an adaptive variational modal decomposition method to achieve denoising of the multi-condition signals in the pipeline and distinguish the multi-condition signals, so as to determine the specific moving time of the missed detection robot 2 when the leak occurs. The acoustic signal processing flow is as follows: firstly, an adaptive variational modal decomposition algorithm is applied to reasonably decompose it into multiple modal components in the frequency domain according to the sound source information, and then some noise domain modal components are screened out to achieve denoising of the pipeline acoustic signal, and then the denoised signal is feature extracted, and finally the pipeline leak and normal signal are distinguished according to the extracted features.

[0054] More specifically, the specific steps of processing and identifying the acoustic signal collected in the above step S2 using the adaptive variational mode decomposition method are as follows:

[0055] S21, setting the number of decomposition layers and penalty factors of the variational mode decomposition method, and decomposing the acoustic signal into multiple decomposition signals according to different decomposition parameters;

[0056] S22, calculating the mutual information entropy of each decomposition signal, and using multi-objective optimization iterative calculation to obtain the optimal decomposition parameter corresponding to the minimum mutual information entropy and the optimal decomposition signal corresponding to the optimal decomposition parameter; wherein the multi-objective optimization specifically adopts the frost ice optimization (RIME) algorithm;

[0057] S23, performing Fourier transform and screening denoising on the best decomposed signal to obtain a reconstructed signal, calculating the bubble entropy value of the reconstructed signal, and thereby constructing a characteristic vector of pipeline leakage;

[0058] S24. Using an extreme learning machine (ELM) algorithm to identify and reconstruct the signal, obtain a signal corresponding to the pipeline leakage and a time corresponding to the occurrence of the leakage.

[0059] Specifically, in step S24, the specific process of identifying and reconstructing the signal is as follows: first, the acoustic signals under three working conditions, namely, normal pipeline, leaking pipeline and ultrasonic assisted positioning signal, are collected by the leakage detection robot 2, and then processed by steps S21, S22 and S23 to form a training set. Then, the training set is trained by the extreme learning machine method, and processed by steps S21, S22 and S23 to perform denoising and identification of multiple working conditions in the pipeline, and determine the travel time of the leakage detection robot when the pipeline leaks.

[0060] More specifically, see Figure 5, the specific process of denoising and identifying the multi-condition signals in the pipeline and obtaining the travel time of the leakage detection robot when the pipeline leaks is as follows:

[0061] A1. Using the randomly generated [K, α] as a set of initial combination parameters, the mutual information entropy value between the IMFs of each order generated by VMD decomposition is used as the fitness function value in the parameter optimization of the frost algorithm, where the mutual information entropy can be expressed as:

[0062]

[0063] Among them, α and β are two random variables, their joint distribution probability is A(α,β), and their marginal distribution probabilities are A(α) and A(β) respectively;

[0064] A2. Use the soft frost search strategy to conduct extensive global search and perform random search in the solution space to increase the diversity of solutions. The formula is expressed as:

[0065] R new =R best +r1·cos(h)·(b·h·(U b -L b )+L b )(2)

[0066] Among them, R new is the new position; R best is the best solution in the current population; r1 is a random number in the range [-1,1]; h is an adjustment coefficient related to the number of iterations; b is an environmental factor; U b and L b are the upper and lower bounds of the solution space respectively;

[0067] A3. Apply the frost puncture mechanism to further refine the search, simulate the formation of frost, and optimize the currently found optimal solution through a more targeted local search;

[0068] R new =R best if r3<F norm (S i ) (3)

[0069] Among them, R new is the updated position; R best is the position of the best individual in the population; r3 is a random number in the range [0,1]; F norm (S i ) is the normalized value of individual fitness.

[0070] A4. The minimum mutual information entropy value in each generation is recorded as the local minimum mutual information entropy value, and the corresponding IMF combination parameter is the best component combination of this decomposition, recorded as the local best component combination. In order to search for the global optimal combination, the current optimal solution is improved by updating the population and iteration, and finally terminated when the preset number of iterations is reached or the fitness threshold is met. The [K, α] combination corresponding to the global minimum mutual information entropy value is the final parameter optimization target;

[0071] A5. After obtaining the optimal VMD decomposition parameters, perform VMD decomposition on the signal, screen the effective modal components and reconstruct them to achieve signal denoising.

[0072] A6. Calculate the bubble entropy value of the pipeline signal after denoising to construct a feature vector, use the extreme learning machine algorithm for pattern recognition, and obtain the travel time of the missed detection robot when the pipeline leaks.

[0073] Specifically, in step S3, the method for establishing the corresponding relationship between the position and time of the missed detection robot 2 in the target detection section of the pipeline 4 to be detected is as follows:

[0074] S31, using the nine-axis IMU sensor 206 to collect the acceleration, angular velocity, and relative angle of the magnetic field of the missed detection robot 2 when it is moving to calculate the position of the missed detection robot 2, and at the same time using the nine-axis IMU sensor 206 as the main system, unifying the timestamps of the nine-axis IMU sensor 206, the laser scanner 203, and the sonar velocimeter 205 to ensure that the data of different sensors are accurately matched at the same time to achieve synchronous positioning;

[0075] S32, using the fixed spacing between the pipeline welds 5 to assist positioning, inferring the pipeline position where the missed detection robot 2 is located according to the number of pipeline welds 5 scanned by the laser scanner 203, and correcting the accumulated error generated by the inertial navigation nine-axis IMU sensor 206;

[0076] S33, using the sonar speed meter 25 to draw a speed curve of the missed detection robot 2 during the moving process, and according to the relationship between speed and time, calculate the travel distance of the missed detection robot 2 and assist in positioning.

[0077] S34, using the particle filter data fusion algorithm, the data position information of the nine-axis IMU sensor 206, the laser scanner 203 and the sonar velocimeter 205 are integrated to obtain the positioning result of the missed detection robot 2, and realize the correspondence between the position and time of the missed detection robot 2 in the target detection section of the pipeline 4 to be detected.

[0078] Finally, it should be noted that the above description is only an embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A leakage detection system based on a city water supply pipeline leakage detection robot, characterized in that: The invention comprises a missed detection robot transceiver (1), a missed detection robot (2) and a data processing computer (3); the missed detection robot transceiver (1) is arranged at both ends of a pipeline to be detected (4) and is used to launch and retrieve the missed detection robot (2); the missed detection robot (2) is arranged inside the pipeline to be detected (4) and is used to detect the location where a leak occurs in the pipeline to be detected (4); and the data processing computer (3) is arranged outside the pipeline to be detected (4) and is used to receive an offline signal recorded during the movement of the missed detection robot (2) inside the pipeline to be detected (4); The missed detection robot transceiver (1) comprises a robot delivery claw (101), a robot recovery claw (102) and a robot capture net (103); the robot delivery claw (101) is used to deliver the missed detection robot (2) into the interior of the pipeline to be detected (4); the robot capture net (103) is used to capture the missed detection robot (2); and the robot recovery claw (102) is used to recover the missed detection robot (2); The missed detection robot (2) comprises a waterproof shell (201), an anti-collision wheel (202) and a laser scanner (203) arranged on the top of the waterproof shell (201), and a hydrophone (204), a sonar velocimeter (205), a nine-axis IMU sensor (206), a data acquisition main control board (207) and a storage USB disk (208) arranged inside the waterproof shell (201); the laser scanner (203), the hydrophone (204), the sonar velocimeter (205) and the nine-axis IMU sensor (206) are all electrically connected to the data acquisition main control board (207), and the data acquisition main control board (207) is electrically connected to the storage USB disk (208).

2. A leakage detection method based on a city water supply pipeline leakage detection robot is a detection method implemented based on the leakage detection system based on the city water supply pipeline leakage detection robot, characterized in that: The following steps are included: S1, placing the missed detection robot (2) into the target detection section of the pipeline (4) to be detected, and driving the missed detection robot (2) to move in the target detection section of the pipeline (4) to be detected. During the movement of the missed detection robot (2), using a detection device to collect acoustic signals and position information of the missed detection robot. When the missed detection robot (2) moves to the detection end point of the target detection section of the pipeline (4) to be detected, the missed detection robot (2) is taken out and an offline signal recorded during its movement is received by a data processing computer (3); S2, processing and identifying the acoustic signal collected in step S1, and determining the specific moving time of the leakage detection robot (2) when the leakage occurs; S3, using the position information of the missed detection robot collected in step S1, combined with the position layout information of the target detection section of the pipeline to be detected (4) and the size information of the pipeline to be detected (4), constructing the corresponding relationship between the position and time of the missed detection robot (2) in the target detection section of the pipeline to be detected (4); S4. Using the corresponding relationship between the position and time of the leak detection robot (2) in the target detection section of the pipeline to be detected (4) constructed in step S2, combined with the specific moving time of the leak detection robot (2) when the leak occurs obtained in step S2, the position where the leak occurs in the pipeline to be detected (4) is determined, thereby realizing the detection of the leak.

3. A leakage detection method based on a city water supply pipeline leakage detection robot according to claim 2, characterized in that: In step S1, the detection device is composed of a missed detection robot transceiver (1), a laser scanner (203), a hydrophone (204), a sonar velocimeter (205), a nine-axis IMU sensor (206), a data acquisition main control board (207) and a storage USB disk (208).

4. A leakage detection method based on a city water supply pipeline leakage detection robot according to claim 2, characterized in that: In step S1, the position information of the missed detection robot includes parameter information of the nine-axis IMU sensor (206), the number of welds (5) crossed by the missed detection robot (2) recorded by the laser scanner (203), and the velocity curve of the missed detection robot (2) collected by the sonar speed meter (205).

5. A leakage detection method based on a city water supply pipeline leakage detection robot according to claim 2, characterized in that: In step S2, the collected acoustic signal is processed and identified by using an adaptive variational mode decomposition method, which specifically includes the following steps: S21, setting the number of decomposition layers and penalty factors of the variational mode decomposition method, and decomposing the acoustic signal into a plurality of decomposition signals according to different decomposition parameters; S22, calculating the mutual information entropy of each of the decomposed signals, and using multi-objective optimization iterative calculation to obtain the optimal decomposition parameter corresponding to the minimum mutual information entropy and the optimal decomposition signal corresponding to the optimal decomposition parameter; S23, performing Fourier transform and screening denoising on the optimal decomposition signal to obtain a reconstructed signal, and calculating the bubble entropy value of the reconstructed signal to construct a feature vector of pipeline leakage; S24. Using an extreme learning machine algorithm to identify the reconstructed signal, obtain a signal corresponding to the pipeline leakage and a time corresponding to the occurrence of the leakage.

6. A leakage detection method based on a city water supply pipeline leakage detection robot according to claim 2, characterized in that: In step S3, the method for establishing the corresponding relationship between the position and time of the missed detection robot (2) in the target detection section of the pipeline (4) to be detected is as follows: S31, using the nine-axis IMU sensor (206) to collect the acceleration, angular velocity, and relative angle of the magnetic field of the missed detection robot (2) when it is moving to calculate the position of the missed detection robot (2), and at the same time using the nine-axis IMU sensor (206) as the main system, unifying the timestamps of the nine-axis IMU sensor (206), the laser scanner (203), and the sonar velocimeter (205), ensuring that the data of different sensors are accurately matched at the same time, so as to achieve synchronous positioning; S32, using the fixed spacing between the pipeline welds (5) to assist positioning, inferring the pipeline position where the missed detection robot (2) is located according to the number of pipeline welds (5) scanned by the laser scanner (203), and correcting the accumulated error generated by the inertial navigation nine-axis IMU sensor (206); S33, using the sonar speed meter (25) to draw a speed curve of the missed detection robot (2) during its movement, and according to the relationship between speed and time, calculating the travel distance of the missed detection robot (2) and assisting in positioning. S34, using a particle filter data fusion algorithm, the data position information of the nine-axis IMU sensor (206), the laser scanner (203) and the sonar velocimeter (205) are integrated to obtain the positioning result of the missed detection robot (2), so as to achieve the correspondence between the position and time of the missed detection robot (2) in the target detection section of the pipeline to be detected (4).

7. A leakage detection method based on a city water supply pipeline leakage detection robot according to claim 5, characterized in that: In step S24, the specific process of identifying and reconstructing the signal is as follows: first, the acoustic signals under three working conditions, namely, the normal pipeline, the leaking pipeline and the ultrasonic assisted positioning signal, are collected by the leakage detection robot (2); then, the acoustic signals are processed by steps S21, S22 and S23 to form a training set; then, the training set is trained by the extreme learning machine method; finally, the multiple working conditions in the pipeline are processed by steps S21, S22 and S23 to perform denoising and identification, and determine the travel time of the leakage detection robot when the pipeline leaks.

Images