Guardian robot for gas well bore and method of detecting and handling obstacles

Abstract

The application provides a protection robot for a gas well wellbore, which comprises a body, the body comprising a first main body and a second main body connected to each other, a detection mechanism arranged on the first main body, the detection mechanism comprising a sensor module for detecting the state of the inner wall of the gas well, and a working mechanism connected with the sensor module and used for removing the obstacles on the inner wall of the gas well, the working mechanism being arranged on the second main body.

Description

Technical Field

[0001] This invention relates to the field of defect detection and maintenance inside gas well shafts, specifically to a protective robot for gas well shafts, and a method for detecting and handling obstacles. Background Technology

[0002] Currently, gas well casings are crucial facilities for natural gas extraction and transportation, and their safety and stability directly impact the production efficiency and economic benefits of gas wells. Because gas well casings operate under high temperature, high pressure, and high-speed oil and gas flow environments, they are susceptible to corrosion, erosion, and wear, leading to a decrease in structural strength and even cracks or holes. This can result in serious consequences such as leaks, blockages, and even explosions. Therefore, regular defect inspections of gas well casings, timely application of protective films to address corrosion and erosion, and removal of foreign matter such as pressure-backed fluid, proppant, and formation sand particles adhering to the well wall are essential measures to ensure the safe and stable operation of gas wells.

[0003] However, existing technologies for detecting and removing obstacles in oil and gas wells cannot meet the needs of oil and gas well inspection and maintenance, especially for oil and gas wells of different diameters and depths, and for complex obstacles, where there is a lack of effective means of detection and removal.

[0004] Therefore, it is desirable in the art to provide a protective robot for gas well shafts to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to propose a protective robot for gas well shafts, which can not only identify the location, shape and type of obstacles through a data processing system, but also effectively remove obstacles through a working mechanism.

[0006] According to a first aspect of the present invention, a protective robot for a gas well shaft is provided, comprising a body, the body including a first main body and a second main body interconnected with each other.

[0007] A detection mechanism is mounted on the first main body, the detection mechanism including a sensor module for detecting the condition of the inner wall of the gas well, and

[0008] A working mechanism connected to the sensor module and used to clear obstacles from the inner wall of the gas well is mounted on the second main body.

[0009] In one embodiment, the sensor module includes a plurality of acoustic wave transmitters disposed at an end of the first body and an acoustic wave receiver for receiving acoustic wave signals.

[0010] In one embodiment, the sensor module further includes a data processing system connected to the acoustic receiver.

[0011] The data processing system is configured to use the characteristics of the acoustic signal as parameters and establish a Gaussian mixture model using the K-means algorithm, thereby identifying the location, shape, and type of obstacles.

[0012] In one embodiment, the obstacle type includes loose obstacles and viscous obstacles.

[0013] The working mechanism includes a cleaning module that outputs fluid from an injector to remove the loose obstacles, and a spraying module that outputs solvent from a cleaner to remove the sticky obstacles.

[0014] In one embodiment, the obstacle type further includes rigid obstacles, and the working mechanism further includes a cutting module for cutting the rigid obstacle into obstacle fragments by a cutter.

[0015] The end of the fuselage is provided with mechanical components for removing the obstacle fragments from the gas well shaft.

[0016] In one embodiment, the protective robot further includes wheel sets disposed on both sides of the body, the wheel sets including telescopic portions disposed on both sides of the first body and the second body, and rollers disposed on the telescopic portions away from the body and for abutting against the inner wall of the gas well shaft.

[0017] In one embodiment, the detection mechanism further includes a camera module for capturing images and videos of the inner wall of the gas well shaft.

[0018] In one embodiment, the detection mechanism further includes a wireless communication module connected to the camera module and the sensor module, wherein the wireless communication module is capable of transmitting the data collected by the camera module and the sensor module to a ground control station.

[0019] In one embodiment, the detection mechanism further includes a navigation module connected to the sensor module.

[0020] The navigation module is configured to determine whether the protective robot can pass through the narrowed area of ​​the gas well shaft based on the condition of the inner wall of the gas well.

[0021] According to a second aspect of the present invention, a method for detecting and handling obstacles using a protective robot as described above is provided, comprising the following steps:

[0022] S1. Acquire the sound wave signal through the sound wave transmitter and the sound wave receiver;

[0023] S2. The data processing system uses the characteristics of the acoustic signal as parameters and establishes a Gaussian mixture model using the K-means algorithm to identify the location, shape, and type of obstacles.

[0024] Compared with the prior art, the advantages of the present invention are as follows:

[0025] Firstly, this invention not only identifies the location, shape, and type of obstacles through a data processing system, but also effectively removes obstacles through a working mechanism. Furthermore, the working mechanism can employ different working modules to target different types of obstacles, further improving the quality and efficiency of obstacle removal.

[0026] Secondly, the navigation module of the present invention can determine whether there is a narrowing area or obstacle inside the gas well based on the condition of the inner wall of the gas well, which helps to plan the path for the protection robot.

[0027] In other words, when a narrowing area is detected within the gas well shaft, it can be further determined whether the protective robot can pass through the narrowing area; when an obstacle is detected within the gas well shaft, the obstacle can be removed by the working mechanism. Attached Figure Description

[0028] The invention will now be described in detail with reference to the accompanying drawings, in which:

[0029] Figure 1 The schematic diagram illustrates the structure of a protective robot for a gas well shaft according to the present invention.

[0030] In the accompanying drawings, the same parts use the same reference numerals. The drawings are not drawn to scale. Detailed Implementation

[0031] To make the technical solutions and advantages of the present invention clearer, exemplary embodiments of the present invention will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not an exhaustive list of all embodiments. Furthermore, without conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0032] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0033] In this invention, unless otherwise explicitly specified and limited, the terms "installation", "connection", "linking", "fixing", etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components.

[0034] Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0035] The invention will now be further described with reference to the accompanying drawings.

[0036] Figure 1 The schematic diagram shows the structure of a protective robot 100 for a gas well shaft according to the present invention.

[0037] like Figure 1 As shown, according to a first aspect of the present invention, a protective robot 100 for a gas wellbore is provided, comprising a body consisting of a first body 11 and a second body 12 connected to each other. Preferably, both the first body 11 and the second body 12 are constructed as cylindrical structures, and the diameter of the first body 11 is larger than the diameter of the second body 12, thereby enabling easier axial movement within the oil and gas well.

[0038] Preferably, the body structure is articulated, and the length and width of the body can be adjusted according to the diameter of the gas well casing, thereby allowing the protective robot 100 to walk along the well wall inside the gas well casing, and also allowing the protective robot 100 to turn or detour when encountering obstacles.

[0039] In this invention, the first body 11 and the second body 12 are further covered with waterproof, corrosion-resistant, and high-temperature resistant materials (such as silane coatings, ceramic coatings, etc.), thereby enabling the protective robot 100 used for gas well casings to more easily adapt to the harsh environment of oil and gas wells.

[0040] In one embodiment, such as Figure 1 As shown, the protective robot 100 for gas well shafts also includes a detection mechanism mounted on the first main body 11 and a working mechanism mounted on the second main body 12. The detection mechanism includes a sensor module 21. Preferably, the present invention can not only detect the state of the gas well inner wall and accurately identify the location, shape, and type of obstacles through the sensor module 21, but also effectively remove obstacles through the working mechanism, as detailed below.

[0041] According to the present invention, such as Figure 1As shown, the sensor module 21 includes a plurality of acoustic wave transmitters 111 disposed at the end of the first body 11, an acoustic wave receiver 112 for receiving acoustic wave signals, and a data processing system connected to the acoustic wave receiver 112. Preferably, the sensor module 21 further includes a laser sensor.

[0042] In this invention, multiple acoustic transmitters 111 can send high-frequency acoustic waves to the inner wall of the gas well, and the acoustic receiver 112 can capture the reflected acoustic signals. The acoustic signals are then converted into a data format that can be analyzed by a data processing system.

[0043] In one embodiment of the present invention, the data processing system is configured to use the characteristics of the acoustic signal as parameters and determine the type of obstacle using the K-means algorithm, determine the location of the obstacle using the acoustic echo, and use the dot matrix data returned by the laser to identify the location and shape of the obstacle using a Gaussian mixture model.

[0044] Preferably, the characteristics of the acoustic signal include, but are not limited to, frequency (f). i ), amplitude (A i ) and phase (φ) i The characteristics of laser dot matrix data include, but are not limited to, data density, normals, and curvature.

[0045] It is worth noting that K-means and Gaussian Mixture Models (GMMs) are both existing statistical techniques widely used in machine learning, especially in cluster analysis. K-means is a simple and efficient clustering method that iteratively optimizes the cluster centers to minimize the distance from each sample point to its cluster center. Gaussian Mixture Models, on the other hand, are a probability-based clustering method that assumes the data consists of multiple Gaussian distributions, each corresponding to a cluster.

[0046] I. K-means clustering to determine obstacle types.

[0047] In this invention, acoustic reflection data of common downhole obstacles are collected on the surface. The type and material of the obstacles are identified using a K-means algorithm, such as corrosion product residues caused by the addition of corrosion inhibitors, sulfur deposits, formation rock cuttings, and organic matter. K values ​​are determined according to the common types of downhole obstacles, and then the K-means algorithm is used to identify the type of obstacle.

[0048] First, the distance function is calculated using the K-means algorithm to classify obstacles:

[0049] (1) Suppose we have a set of data points (X = x1, x2, ..., x) collected through sound wave reflection. n ), where each data point (x) i) is a feature vector that contains the frequency (f) of the sound wave signal. i ), amplitude (A i ), and phase (φ) i Preferably, in the K-means algorithm, we can include these features in the distance calculation:

[0050] (2) Randomly select one cluster point and calculate the distance function for each collected data point: For each data point (x i ), calculate its relationship with each cluster center μ k The distance is calculated and assigned to the nearest cluster center. In this invention, this distance is typically the square of the Euclidean distance.

[0051] Calculate the sum of squares of all distances, i.e. Corresponding to sound signals, that is:

[0052]

[0053] Among them, (μ fk ), (μ Ak ), and (μ φk ) are the mean values ​​of frequency, amplitude, and phase corresponding to the cluster centers, respectively.

[0054] For the frequency cluster center mean (μ) fk The calculation method is as follows:

[0055]

[0056] The cluster mean (\mu_{Ak}) of the amplitude is calculated as follows:

[0057]

[0058] The cluster mean (\mu_{\phi k}) of the phase is calculated as follows:

[0059]

[0060] Among them, (r ik ) is an indicator variable, if the data point (x) i If ) is assigned to cluster (k), then (r) ik =1), otherwise (r) ik =0).

[0061] For each data point (x) i ), calculate the probability of selecting it as the next cluster center (P(x)). i The probability is (x) i The square of the distance from (x) to the nearest cluster center (D(x)i ) 2 Divide by the sum of the squares of all distances.

[0062]

[0063] The next cluster center is randomly selected based on this probability distribution. Then, the distribution is as follows: for each data point that is not selected as a cluster center, the square of its distance from the already selected cluster centers is calculated, and this square is used as the probability of being selected as the next cluster center. This process is repeated until (K) cluster centers are selected, thus determining the obstacle type. Preferably, by combining the amplitude, frequency, and phase angle features of the obstacle that were not extracted in the ground experiment, the clustering results can be compared to determine the obstacle type.

[0064] II. Determining the Location of Obstacles

[0065] By receiving the reflected ultrasonic signals emitted by itself, and based on the time difference between the emission and reception of the ultrasonic waves and the propagation speed, the propagation distance can be calculated, and the distance between the obstacle and the robot can be further obtained.

[0066] (1) Set the robot's travel speed v inside the wellbore;

[0067] (2) The predicted position of the robot is y = vt + x0.

[0068] Preferably, x0 is the coordinate matrix of the previous position. Therefore, the position update equation is set as: Y = cx + v, where y is the ultrasonic echo time.

[0069] Preferably, assuming the robot moves at a speed (v), and the time from the emission of a sound wave from the ultrasonic sensor to the receipt of the reflected wave is (t), and the speed of the sound wave is (c) (approximately 343 m / s in air), then the distance (d) between the robot and the obstacle can be calculated using the following formula:

[0070]

[0071] Third, the shape of the obstacle is determined by the dot matrix collected by the laser sensor and by Gaussian clustering.

[0072] 1. Assume the laser sensor acquires dot matrix data (X = {x1, x2, ..., x...}). n}).

[0073] 2. Calculate the mean and variance of the data.

[0074] (1) Mean:

[0075] (2) Variance:

[0076] (3) Calculate the standard deviation:

[0077] (4) Calculate the Gaussian normal distribution function: The Gaussian normal distribution function (also known as the normal probability density function) is:

[0078]

[0079] (5) Determine the 97% confidence interval: For a normal distribution, 97% of the data lies within ±2.17 standard deviations of the mean. Therefore, we can calculate the upper and lower bounds of this interval:

[0080] Lower bound=μ-2.17; Upper bound=μ+2.17

[0081] 3. Noise reduction of dot matrix data: This means retaining data points within this range and removing data points outside this range.

[0082] 4. Extract features from the dot matrix data: local density, surface normal, and curvature.

[0083] (1) Local density: Local density can be estimated by calculating the average distance between a point and its neighbors. First, for each point (x... i You need to find the set of its (k) nearest neighbors (N(x)). i Then, the local density (ρ(x)) i The following formula can be used to calculate it:

[0084]

[0085] Among them, (dist(x) i ,x j )) is a point (x i ) and (x j The Euclidean distance between them.

[0086] Where, point (x) i ) and (x j The formula for calculating the Euclidean distance between two points is:

[0087]

[0088] Among them, (x ik ) and (x jk ) are points (x) i ) and (x j The coordinates of point (k) in the (k)th dimension are given by (d), where (d) is the dimension of the space. This formula essentially calculates the "straight-line" distance between two points.

[0089] (2) Surface normal: The surface normal passes through the point (x) i This is obtained by fitting a local surface at point (x) and calculating the normal vector of that surface. Typically, this can be achieved by calculating the normal vector of the surface at point (x). i The covariance matrix of the (k) nearest neighbors of is obtained and principal component analysis (PCA) is performed on it. Surface normal (n(x) i )) is the eigenvector corresponding to the smallest eigenvalue of the covariance matrix:

[0090]

[0091] in, It is (N(x) i The mean of the midpoints.

[0092] (3) Curvature: Curvature is determined by evaluating the point (x) i The curvature is estimated by the local surface change rate at point (k). It can be calculated using the change in the surface normal or the eigenvalues ​​of the covariance matrix of the (k) nearest neighbor. The curvature estimation formula here is:

[0093]

[0094] Wherein, (λ min ,λ mid ,λ max ) is the covariance matrix (C(x) i The eigenvalues ​​of the eigenvalues ​​are sorted by size.

[0095] 5. Standardize the extracted features: For each feature (f), calculate its mean (μ). f ) and standard deviation (σ) f Then, standardize it using the following formula:

[0096]

[0097] 6. Substitute the standardized eigenvalues ​​into the GMM (Geometry Model) and cluster the points in the feature space, with each cluster representing a potential obstacle. (Clustering is performed using the Python database GaussianMixture.) Preferably, the covariance of the three eigenvalues ​​is calculated:

[0098]

[0099] Where, μ ρ and μ n k X These are the mean values ​​of ρ, n, and k, respectively.

[0100] Preferably, the mixing coefficients of the Gaussian mixture model are calculated by expectation maximization, which is the weight of the (k)th Gaussian distribution in the whole model.

[0101] (1) Initialization: Substitute the mean (μ) k ), covariance matrix (Σ k Given equal initial mixing coefficients (π) k The sum of these numbers equals 1.

[0102] (2) Expectation Step (E): Calculate each data point (x) i The posterior probability of belonging to each Gaussian distribution (k):

[0103]

[0104] (3) M-step (Maximization Step): Use these posterior probabilities to update the parameters of each Gaussian distribution: ① Update the mixing coefficients: ② Update the mean and covariance matrices.

[0105] (4) Iteration: Repeat the E step and M step until the parameters converge or the predetermined number of iterations is reached.

[0106] In this process, (γ(z) ik )) is the data point (x) i Let π be the posterior probability from the (k)th Gaussian distribution, and (n) be the total number of data points. Through iteration of the EM algorithm, we can obtain the optimal mixing coefficient (π) for each Gaussian distribution. k ).

[0107] Let represent the probability density function of a Gaussian distribution, where (x) is a random variable, (μ) is the mean, and (Σ) is the covariance matrix. For the normal distribution of the three characteristic variables of the prior boundary in this invention, the probability density function is:

[0108]

[0109] Where (x) is a (3)-dimensional random vector, (μ) is the mean vector, (Σ) is a (k×k) covariance matrix, and (|Σ|) is the determinant of the covariance matrix.

[0110] 7. Interpretation of results: The Gaussian distribution to which each point is most likely to belong determines its cluster label, thereby enabling the identification of the shape of the obstacle.

[0111] 8. Post-clustering processing: Analyze the clustering results to determine the final result.

[0112] Preferably, this invention optimizes the data initialization process based on wellbore data, introducing a new iterative update strategy to better adapt to the data distribution within the wellbore. For example, this improved method utilizes different data characteristics to adjust the weights of the Gaussian distribution, increases the weights of sound frequencies based on wellbore characteristics, and uses a mathematical model to describe the relationships between data points. These improvements aim to enhance the model's adaptability within the wellbore, thereby achieving better clustering results downhole.

[0113] In this invention, the types of obstacles include loose obstacles, sticky obstacles, and hard obstacles. Loose obstacles include mud, sand, and stones, sticky obstacles include grease and glue, and hard obstacles include metal and plastic.

[0114] Preferably, the working mechanism consists of four working modules: a sweeping module 31, a spraying module 32, and a cutting module 34. Therefore, this invention can utilize different working modules to complete the removal work for different types of obstacles, as described below.

[0115] For loose obstructions: the cleaning module 31 includes an injector (not shown) for aiming at the loose obstruction. Preferably, the injector can spray a high-pressure fluid (such as water or air) to flush or blow away the loose obstruction to restore the accessibility of the oil and gas well.

[0116] For sticky obstacles: the spraying module 32 includes a cleaner (not shown) containing solvent or cleaning agent and a vacuum cleaner (not shown). Preferably, the cleaner sprays solvent or cleaning agent (such as oil cleaner, water-based cleaner, special chemical cleaner, etc.) to dissolve or clean the sticky obstacle, and then the vacuum cleaner sucks away or extracts and removes the dissolved sticky obstacle from the oil and gas well.

[0117] For rigid obstacles: the cutting module 34 includes a cutter (not shown), so that when facing a rigid obstacle, the cutter can be used to cut the rigid obstacle into smaller obstacle fragments by means of laser or electric arc, which will help with the subsequent removal work.

[0118] It is worth noting that a mechanical element for cooperating with a cutter is provided at the end of the second body 12. Preferably, the mechanical element is capable of dragging or pushing obstacle fragments to remove them from the oil and gas well. In this invention, the mechanical element includes a rotatable clamp disposed at the end of the second body 12 and a telescopic robotic arm disposed at the end of the first body 11, wherein the clamp can automatically adjust its opening and closing angle and force according to the size of the obstacle, thereby completing the work of dragging or pushing the obstacle.

[0119] According to the present invention, the protective robot 100 for gas well shafts further includes wheel sets disposed on both sides of the body. For example... Figure 1 As shown, preferably, the wheel assembly includes telescopic portions disposed on both sides of the first body 11 and the second body 12, and rollers 15 disposed at the end of the telescopic portions away from the body. Preferably, after the protective robot 100 enters the oil and gas well, the rollers 15 remain in contact with the inner wall of the gas well casing.

[0120] In this invention, the maximum outer diameter of the protective robot 100 is greater than the inner diameter of the gas well shaft, and because the telescopic part has good extensibility, when the protective robot 100 encounters an obstacle or a narrowing area, it can shrink its overall size to ensure stable movement within the gas well shaft.

[0121] According to the present invention, the detection mechanism further includes a navigation module connected to the sensor module 21. Preferably, the navigation module is configured to plan a path for the protection robot 100 based on the condition of the gas well's inner wall. In other words, when a narrowing area is detected within the gas well casing, it can be determined whether the protection robot 100 can pass through the narrowing area; when an obstacle is detected within the gas well casing, the obstacle can be cleared by the working mechanism. In this invention, the navigation module can receive real-time data collected from the data processing system.

[0122] Preferably, when an obstacle is detected inside the gas well shaft, the protective robot 100 can identify the location, shape, and type of the obstacle through a data processing system, and can complete the obstacle removal work through a working mechanism.

[0123] Preferably, when a narrowing zone is detected within the gas wellbore, the protective robot 100 can calculate a path through the narrowing zone using its navigation module. This path can be calculated using the following formula:

[0124] Formula for calculating the narrowing path: P 缩径 =f(D 缩径 ,S 传感器 ,L 机器人 )

[0125] Among them, P 缩径 It is the path calculated by the robot to pass through the narrowed area;

[0126] D 缩径 It is the diameter of the narrowed region;

[0127] S 传感器 It is data collected by sensors;

[0128] L 机器人 That is the length of the robot.

[0129] In this invention, function f is an algorithm adjusted according to actual conditions, taking into account obstacle positions in sensor data, spatial layout of narrowed areas, and the robot's physical dimensions. For example, we can represent this function using a simple proportional relationship:

[0130] P 缩径 =α·D 缩径 +β·S 传感器 +γ·L 机器人

[0131] In this example, (α), (β), and (γ) are weighting coefficients that determine the degree of influence each variable has on the path calculation. This is just an example; the actual function f will be more complex and needs to be adjusted according to the specific robot navigation system.

[0132] To implement this formula, we typically need to perform the following steps:

[0133] 1. Sensor data processing: Analyze sensor data to identify the location and size of obstacles.

[0134] 2. Path planning: Based on sensor data and the robot's dimensions, plan a path to avoid obstacles.

[0135] 3. Path optimization: Optimize the path to make it as short as possible while meeting safety requirements.

[0136] After obtaining the weights through the above steps, a formula can be derived.

[0137] In the above situation, the navigation module can also determine whether the protective robot 100 will get stuck when passing through the narrowing area. Therefore, the navigation module can make this determination using the following formula:

[0138] P 缩径 =max(0,D 缩径 -W 机器人 )

[0139] Where: P 缩径 It is the robot's path length;

[0140] D 缩径 It is the diameter of the narrowed region;

[0141] W 机器人 That is the width of the robot.

[0142] Preferably, if D 缩径 Greater than W 机器人 If the path length is zero, then the protective robot 100 can pass; otherwise, the path length is zero, and the protective robot 100 can pass.

[0143] Preferably, the roller 15 is made of rubber with anti-slip texture on its surface, so it can move freely on the inner wall of the gas well shaft, adjusting its direction and speed. The wheel is connected to the machine body by a telescopic part (such as a spring), which can buffer vibration and impact.

[0144] Preferably, regarding degrees of freedom: it is assumed that the protective robot 100 has n degrees of freedom, typically (n≥6), to ensure omnidirectional movement in three-dimensional space. Regarding the working range: the tubing diameter is D, and it is assumed that the working range of the protective robot 100 should be at least πD to ensure coverage of the entire inner wall surface of the gas wellbore.

[0145] According to the present invention, such as Figure 1 As shown, the detection mechanism also includes a camera module for capturing images and videos of the inner wall of the gas well shaft, and a wireless communication module 22 connected to the camera module and sensor module 21. Preferably, the wireless communication module 22 is capable of transmitting the data collected by the camera module and sensor module 21 to the ground control station.

[0146] Preferably, the sensor module 21 further includes, but is not limited to, temperature sensors, pressure sensors, distance sensors, and vibration sensors; the camera module includes, but is not limited to, visible light cameras, infrared cameras, and ultraviolet cameras. In this way, the protective robot 100 can not only detect and photograph the condition of the gas wellbore's inner wall and the location, shape, and type of obstacles in real time, but also measure parameters such as temperature, pressure, distance, and vibration within the oil and gas well to determine if there are any abnormal or dangerous conditions within the well. Furthermore,

[0147] Preferably, the wireless communication module 22 includes, but is not limited to, a radio transmitter, a radio receiver, and an antenna. In this way, not only can the protective robot 100 ensure stable communication with the ground control station, but data collected by the sensor module and camera module can also be transmitted to the ground control station via radio signals, facilitating ground personnel to monitor and guide the operation of the protective robot 100.

[0148] In one embodiment, the robot body is also equipped with a power module that can provide power to itself. Preferably, this is a battery or other power source, which can be replaced or recharged according to the working time and load of the protective robot 100 to support its normal operation.

[0149] According to a second aspect of the present invention, a method for detecting and handling obstacles using a protective robot 100 as described above is provided, comprising the following steps:

[0150] First, sound wave signals are acquired through sound wave transmitter 111 and sound wave receiver 112;

[0151] Then, the sound wave signal is converted into a data format that can be analyzed by the data processing system. That is, the characteristics of the sound wave signal are used as parameters, and the location, shape and type of the obstacle are identified by establishing a Gaussian mixture model through the K-means algorithm, so as to issue instructions to the working mechanism to handle the obstacle.

[0152] Finally, the protective robot 100 returned to the wellhead after completing the obstacle removal work inside the gas wellbore.

[0153] Compared with existing technologies, the advantages of this invention are:

[0154] Firstly, this invention not only identifies the location, shape, and type of obstacles through a data processing system, but also effectively removes obstacles through a working mechanism. Furthermore, the working mechanism can employ different working modules to target different types of obstacles, further improving the quality and efficiency of obstacle removal.

[0155] Secondly, the navigation module of the present invention can determine whether there is a narrowing area or obstacle inside the gas well based on the condition of the inner wall of the gas well, so as to help plan the path for the protective robot 100.

[0156] In other words, when a narrowing area is detected in the gas well shaft, it can be further determined whether the protective robot 100 can pass through the narrowing area; when an obstacle is detected in the gas well shaft, the obstacle can be removed by the working mechanism.

[0157] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art can easily make changes or modifications within the scope of the present invention, and such changes or modifications should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A protective robot for gas well shafts, comprising: The fuselage includes a first main body (11) and a second main body (12) connected to each other. A detection mechanism is disposed on the first body (11), the detection mechanism including a sensor module (21) for detecting the condition of the inner wall of the gas well, and A working mechanism connected to the sensor module (21) and used to remove obstacles from the inner wall of the gas well is mounted on the second body (12).

2. The protective robot for gas well shafts according to claim 1, characterized in that, The sensor module (21) includes a plurality of acoustic wave transmitters (111) disposed at the end of the first body (11) and an acoustic wave receiver (112) for receiving acoustic wave signals.

3. The protective robot for gas well shafts according to claim 2, characterized in that, The sensor module (21) also includes a data processing system connected to the acoustic receiver (112). The data processing system is configured to use the characteristics of the acoustic signal as parameters and establish a Gaussian mixture model using the K-means algorithm, thereby identifying the location, shape, and type of obstacles.

4. The protective robot for gas well shafts according to claim 3, characterized in that, The obstacle types include loose obstacles and sticky obstacles. The working mechanism includes a cleaning module (31) that outputs fluid from an injector to remove the loose obstruction, and a spraying module (32) that outputs solvent from a cleaner to remove the sticky obstruction.

5. The protective robot for gas well shafts according to claim 4, characterized in that, The obstacle type also includes rigid obstacles, and the working mechanism further includes a cutting module (34) that uses a cutter to cut the rigid obstacles into obstacle fragments. The end of the fuselage is provided with mechanical components for removing the obstacle fragments from the gas well shaft.

6. The protective robot for gas well shafts according to claim 4, characterized in that, The protective robot also includes wheel sets on both sides of the body. The wheel sets include telescopic parts on both sides of the first body (11) and the second body (12), and rollers (15) on the telescopic parts away from the body and used to abut against the inner wall of the gas well shaft.

7. The protective robot for gas well shafts according to any one of claims 1 to 6, characterized in that, The detection mechanism also includes a camera module for capturing images and videos of the inner wall of the gas well shaft.

8. The protective robot for gas well shafts according to any one of claims 1 to 6, characterized in that, The detection mechanism also includes a wireless communication module (22) connected to the camera module and the sensor module, wherein the wireless communication module (22) is capable of transmitting the data collected by the camera module and the sensor module (21) to the ground control station.

9. The protective robot for gas well shafts according to any one of claims 1 to 6, characterized in that, The detection mechanism also includes a navigation module connected to the sensor module. The navigation module is configured to determine whether the protective robot can pass through the narrowed area of ​​the gas well shaft based on the condition of the inner wall of the gas well.

10. A method for detecting and handling obstacles using a protective robot according to any one of claims 1 to 9, comprising the following steps: S1. Acquire the sound wave signal through the sound wave transmitter (111) and the sound wave receiver (112); S2. The data processing system uses the characteristics of the acoustic signal as parameters and establishes a Gaussian mixture model using the K-means algorithm to identify the location, shape, and type of obstacles.

Images