Assembly type enamel tank body state detection system

Abstract

The invention relates to the technical field of industrial container detection, and discloses an assembled enamel tank body state detection system, which comprises a three-dimensional sensor group provided with a three-dimensional heterogeneous sensing network, the three-dimensional sensor group is arranged on the surface of an enamel tank body through pasting to form a biological neural network sensing system, and an energy supply unit comprises thermoelectric power generation and wireless charging embedded magnetic attraction power supply. According to inner wall corrosion diagnosis, codes at the inner wall corrosion position are decoded through a code coating protection film, a decoding result is positioned, full-time and full-domain monitoring of the state of the enamel tank is achieved, and the intelligent operation and maintenance level of industrial equipment is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial container detection, and discloses a detection system for the state of an assembled enamel tank body. Background Art

[0002] Existing technologies mostly rely on single sensors or manual visual inspection, making it difficult to identify complex defects under multi-physical field coupling. In particular, the detection sensitivity for concealed defects such as microcracks and early corrosion is insufficient, and it is impossible to construct a three-dimensional stereoscopic perception network for the state of the tank body. The diagnosis of inner wall corrosion relies on off-line sampling inspection, and it is impossible to achieve dynamic positioning and non-contact real-time monitoring of the corrosion location. The curved surface structure of the tank body leads to uneven layout of the sensor array, obvious signal attenuation in the edge area, and it is difficult for traditional rigid sensors to fit the curved surface, resulting in detection blind spots. Moreover, there is a lack of a signal enhancement mechanism designed based on bionics, and edge defects are easily missed. Summary of the Invention

[0003] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0004] To solve the above technical problems, the main purpose of the present invention is to provide a detection system for the state of an assembled enamel tank body, including:

[0005] A three-dimensional heterogeneous sensing network is set up by the three-dimensional sensor group, which is placed on the surface of the enamel tank body by sticking to form a biological neural network perception system;

[0006] The power supply unit includes thermoelectric power generation and wireless charging embedded magnetic adsorption power supply to ensure power supply in a dual way;

[0007] The diagnosis of inner wall corrosion decodes the code of the inner wall corrosion part by encoding and coating a protective film, and locates the decoding result.

[0008] As a preferred solution of the detection system for the state of the assembled enamel tank body of the present invention, wherein:

[0009] The three-dimensional heterogeneous sensing network includes a three-layer stacked structure, and the three-layer stacked structure includes a surface layer, an intermediate layer, and a bottom layer;

[0010] The surface layer is a piezoelectric film array, which is densely paved on the surface of the enamel tank body in a regular polygon;

[0011] The intermediate layer is a fiber Bragg grating, which is wound in a three-dimensional helix, and the pitch of the three-dimensional helix is determined by the regular polygon;

[0012] The underlying layer is a differential eddy current probe composed of miniaturized electromagnetic coils.

[0013] As a preferred embodiment of the assembled enamel tank state detection system of the present invention, wherein:

[0014] The biological neural network perception system includes a synaptic network, and the synaptic network constructs a hierarchical conductive network based on a flexible substrate;

[0015] The hierarchical conductive network includes a primary main trunk and secondary branches. The primary main trunk is used to connect multiple sensor clusters, and the secondary branches are used to connect individual sensing units;

[0016] The working method of the biological neural network includes: the piezoelectric film array generates a time pulse sequence, the fiber Bragg grating outputs a frequency modulation signal, the electromagnetic data is converted into pulse intensity, and the edge node performs pulse timing correlation analysis.

[0017] As a preferred embodiment of the assembled enamel tank state detection system of the present invention, wherein:

[0018] The sensing network realizes multi-modal data fusion through the cooperative working mechanism of a three-layer stacked structure, specifically including:

[0019] The surface piezoelectric film array detects the dynamic stress wave on the surface of the enamel tank through a regular polygon close-packing configuration and converts mechanical vibration into an electrical pulse signal;

[0020] The intermediate layer fiber Bragg grating is based on a three-dimensional helical winding, and the pitch of the three-dimensional helical winding is determined by a regular polygon to monitor the shear strain distribution of the tank body in real time;

[0021] The underlying differential eddy current probe extracts the electromagnetic characteristic parameters of the metal matrix inside the tank through the complementary layout of miniaturized electromagnetic coils;

[0022] The three-layer data is fused through a spatio-temporal registration algorithm to generate a three-dimensional damage topology map for locating the surface crack propagation path and the internal corrosion area.

[0023] As a preferred embodiment of the assembled enamel tank state detection system of the present invention, wherein:

[0024] The primary main trunk is composed of a cross-linked structure and extends along the curvature direction of the enamel tank to form a closed-loop path;

[0025] The secondary branches bifurcate from the primary main trunk at a certain angle, and the branch density is determined by the spatial distribution of the sensor clusters;

[0026] A capacitive coupling node is provided at the junction of the primary main trunk and the secondary branches to achieve low-loss transmission of pulse signals through impedance matching;

[0027] The fractal network has self-healing properties. When a local fracture occurs, the conductive path is reconstructed through the electrochemical migration of silver nanoparticles.

[0028] As a preferred solution of the assembled enamel tank body state detection system of the present invention, wherein:

[0029] The thermoelectric power generation adopts a curved surface adaptive thermoelectric structure, including: a flexible thermoelectric arm array composed of a Bi2Te3 / Sb2Te3 heterostructure forms a serpentine folding unit, forms a closed-loop circuit along the circumferential direction of the tank body, the hot end is coupled with the inner wall of the tank body through a graphene heat conduction film, and the cold end is connected to a radiation-type micro heat dissipation fin;

[0030] The phase change energy storage leveling layer fills a paraffin-carbon nanotube composite phase change material between the hot and cold ends to buffer temperature fluctuations through solid-liquid phase changes;

[0031] The contact pressure self-regulating mechanism adopts a shape memory alloy spring to dynamically adjust the hot end contact pressure according to the surface curvature of the tank body.

[0032] As a preferred solution of the assembled enamel tank body state detection system of the present invention, wherein:

[0033] The wireless charging embedded magnetic attraction power supply adopts a magnetoelectric composite topology;

[0034] The magnetoelectric composite topology includes a domain-divided magnetic attraction array, dynamic magnetic force compensation, and thermal-electric collaborative management;

[0035] The domain-divided magnetic attraction array includes an outer ring and a central area. The outer ring area is arranged with neodymium iron boron permanent magnets to form a gradient magnetic field. The central area is embedded with a planar spiral coil to form a resonant coupling with an external charging base;

[0036] The dynamic magnetic force compensation uses a Hall sensor array to monitor the magnetic attraction gap in real time, and the electromagnetic auxiliary coil dynamically compensates the magnetic force according to the gap value;

[0037] The thermal-electric collaborative management integrates a micro heat pipe on the back plate of the wireless charging module to direct the heat generated by the coil to the cold end of the thermoelectric power generation, forming an energy recycling chain.

[0038] As a preferred solution of the assembled enamel tank body state detection system of the present invention, wherein:

[0039] The inner wall corrosion diagnosis adopts segmented color gradient coding, divides the inner wall surface into several equally spaced annular regions, sets a unique color code formed by mixing two or more basic colors in each region, and the color codes of adjacent regions are connected by a gradient color transition;

[0040] The mixing ratio of the color codes is associated with the axial position coordinates of the annular region, so that when corrosion occurs, the color codes produce recognizable color peeling features due to the breakage of the film layer.

[0041] As a preferred embodiment of the state detection system for the assembled enamel tank body of the present invention, wherein:

[0042] The inner wall corrosion diagnosis further includes: collecting the reflection spectrum data of the corrosion area by multi-spectral imaging, and performing differential comparison between the detected abnormal color codes and the original coding data by establishing a chromaticity-position mapping database;

[0043] For the corrosion boundary of the gradient color transition section, an edge fusion algorithm is used to compensate and calculate the attenuation gradient of adjacent color codes, and the topological association rules between annular regions are introduced to map discrete corrosion points into a three-dimensional polar coordinate system, and finally the corrosion area is output.

[0044] As a preferred embodiment of the state detection system for the assembled enamel tank body of the present invention, wherein:

[0045] The coding film is prefabricated at the interface between the metal matrix and the enamel layer by screen printing, so that the underlying color codes are directly exposed after the corrosion penetrates the enamel layer;

[0046] In the detection stage, a telescopic guide rail is used to carry a linear array sensor to perform spiral scanning along the axial direction of the tank body, and the data of the pressure sensor in the tank is synchronously combined. When local color code anomalies are detected, a wall thickness ultrasonic probe is automatically triggered for secondary verification.

[0047] The beneficial effects of the present invention:

[0048] Through the regular polygon tiling and spiral winding layout of the three-layer heterogeneous sensing network, the full-dimensional coverage of the surface stress, strain, and electromagnetic signals of the tank body is realized. Combining with the fractal conductive network constructed by the synaptic network, simulating the pulse timing analysis ability of the biological neural network, significantly improving the spatio-temporal resolution and feature extraction efficiency of defect signals, and accurately and quickly identifying micro-cracks and micron-sized corrosion pits on the enamel tank body;

[0049] The segmented color gradient coding film is integrated at the interface between the metal matrix and the enamel layer. After the corrosion penetrates, the color code peeling is triggered. Combining with multi-spectral imaging and edge fusion algorithm, the three-dimensional polar coordinate accurate positioning of the corrosion position is realized. Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0051] Figure 1 It is the tank topology diagram of the state detection system for the prefabricated enamel tank body of the present invention;

[0052] Figure 2 It is the flow chart of the inner wall corrosion diagnosis method for the state detection system of the prefabricated enamel tank body of the present invention;

[0053] Figure 3 It is the schematic diagram of the outer surface structure of the enamel tube of the state detection system for the prefabricated enamel tank body of the present invention.

[0054] Reference numerals: 1, piezoelectric film array; 2, regular polygon piezoelectric film; 3, fiber Bragg grating; 4, differential eddy current probe; 5, fiber Bragg grating array; 6, electromagnetic eddy current array. Specific embodiments

[0055] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0056] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0057] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it a separate or selectively exclusive embodiment from other embodiments.

[0058] Embodiment 1

[0059] As Figure 1 shown, the state detection system for the prefabricated enamel tank body includes:

[0060] The three-dimensional heterogeneous sensor group sets up a three-dimensional heterogeneous sensing network, which is placed on the surface of the enamel tank body by sticking to form a biological neural network perception system;

[0061] The three-dimensional heterogeneous sensing network includes a three-layer stacked structure, and the three-layer stacked structure includes a surface layer, an intermediate layer and a bottom layer;

[0062] [[ID=4K]]A specific implementation method of a three-layer heterogeneous sensing network includes:

[0063] Integrate a chip-level atomic clock at the edge node and distribute timestamps through each sensor sub-node; the piezoelectric film array uses an event-driven ADC to trigger an interrupt when the stress wave threshold is detected and record the atomic clock timestamp at the trigger moment; the demodulator of the fiber Bragg grating samples at a fixed frequency, and the start moment of each scan cycle is aligned with the atomic clock; the eddy current probe obtains differential signals and continuously samples through a high-speed ADC to form an eddy current impedance spectrum, and each frame of data is appended with the starting timestamp of the atomic clock.

[0064] Divide the tank body into three-dimensional grid cells. Each grid cell receives surface layer, middle layer, and bottom layer data.

[0065] Among them, the surface layer data is the piezoelectric pulse density, reflecting the intensity of the surface stress wave; the middle layer is the fiber strain gradient data, characterizing the shear deformation; the bottom layer is the eddy current conductivity change data, indicating subsurface defects; calculate the damage probability of each grid through a weighted scoring mechanism to generate three-dimensional damage.

[0066] Furthermore, all sensors are connected to a synchronous controller with an atomic clock, and each sensor data packet is appended with four groups of time tags: the acquisition start time, the transmission delay compensation time, the edge node reception time, and the global unified timestamp.

[0067] The spatial mapping step first preprocesses the data, then performs spatial coordinate transformation on the processed data, and finally conducts grid-based fusion.

[0068] Furthermore, data preprocessing includes filtering environmental vibration noise from piezoelectric signals, compensating temperature drift for fiber strain data, and calibrating the probe lift-off effect for eddy current data.

[0069] Furthermore, the piezoelectric unit is mapped to a pre-set polar coordinate system according to the pre-stored position table, the fiber strain points are converted through the guide groove scale and mapped to polar coordinates, and the coordinates of the eddy current probe are fed back in real time by the robotic arm.

[0070] Furthermore, grid-based fusion divides the tank body into three-dimensional grids, and each grid comprehensively scores the data of the three sensors.

[0071] When multiple piezoelectric sensors detect stress waves generated by the same crack, the wave source position is deduced by the time difference of arrival of the pulses, and combined with the sound wave propagation speed in the tank body material, a time difference - spatial distance conversion model is established. The time difference - spatial distance conversion model is a common physical calculation method for distance and time difference, used to calculate the sound wave propagation distance within a period of time difference.

[0072] Intercept multi-modal data with a 1ms period. The piezoelectric data corresponds to the surface stress distribution, the fiber data corresponds to the three-dimensional strain field reconstruction, and the eddy current data corresponds to the subsurface defect mapping.

[0073] The method for implementing defect mapping rules includes a spatial mapping method and a quantization correlation formula;

[0074] Furthermore, the spatial mapping method includes axial mapping (Z-axis) with the crack midpoint as the center, extending up and down as the detection interval; circumferential mapping (θ-axis) with the circumferential range where the crack is located as the correlation area; radial mapping from the surface crack to the eddy current detection layer to obtain surface crack data by the fiber optic strain layer.

[0075] Furthermore, the quantization correlation formula includes crack projection overlap degree = (eddy current abnormal area θ-Z projection ∩ surface crack θ-Z projection) / surface crack projection area; strain gradient calculation = (crack tip strain value - crack middle strain value) / distance.

[0076] The surface layer is a piezoelectric film array, densely paved on the surface of the enamel tank in a regular polygon;

[0077] As Figure 3 shown, it includes a piezoelectric film array 1, a fiber Bragg grating array 5, and an electromagnetic eddy current array 6, constituting a three-dimensional three-layer structure.

[0078] Furthermore, regular polygon piezoelectric films 2 are tiled on the outer surface of the enamel tank to form a piezoelectric film array 1;

[0079] Fiber Bragg gratings 3 are wound in a three-dimensional helix to form a fiber Bragg grating array 5.

[0080] Miniature electromagnetic coils form differential eddy current probes 4, and multiple differential eddy current probes 4 form an electromagnetic eddy current array 6.

[0081] The piezoelectric film array is directly attached to the outer surface of the enamel tank, and the surface is a sputtered interdigital electrode, which can adopt a polygonal honeycomb dense paving structure.

[0082] Furthermore, the piezoelectric film array is used to capture surface mechanical vibrations in real time and detect the resonance frequency shift of the tank structure.

[0083] In this application, a preferred piezoelectric film array can use a flexible piezoelectric material with an anti-corrosion insulating layer coated on the surface to adapt to the high-temperature and humid environment of the enamel tank.

[0084] Furthermore, each regular polygon unit covers the tank surface by spiral arrangement to eliminate detection blind spots.

[0085] Furthermore, the upper electrode of the piezoelectric film array can use a serpentine silver nanowire grid or other grid materials to reduce mechanical stress concentration, and the lower electrode completely covers the aluminum foil substrate to form a uniform electric field distribution.

[0086] Each piezoelectric unit is connected to the edge node through a silver paste wire, and the wire path extends along the vertices of the regular polygon, using fractal routing to reduce electromagnetic interference.

[0087] The working principle of the piezoelectric film array includes:

[0088] When mechanical vibrations occur on the surface of the tank body due to corrosion, cracks or external impacts, the crystal structure of the piezoelectric film deforms, resulting in the redistribution of internal polarized charges.

[0089] Each piezoelectric unit is connected to a comparator circuit, and a dynamic threshold is preset. Only when the signal amplitude exceeds the threshold, a pulse is triggered. The moment when the pulse appears corresponds to the arrival time of the stress wave, and the timestamp is recorded by the edge node clock.

[0090] Through the combination of material optimization, topological layout and pulse coding technology, the piezoelectric film array realizes high-resolution perception of the dynamic stress on the surface of the enamel tank body. Its time pulse sequence not only carries the spatio-temporal information of the damage, but also provides the core data source for the localization of early corrosion and cracks through the pulse timing correlation analysis inspired by the biological neural network.

[0091] The intermediate layer is a fiber Bragg grating, which is wound in a three-dimensional helix, and the pitch of the three-dimensional helix is determined by a regular polygon;

[0092] The intermediate layer is a fiber Bragg grating array, which is embedded in the intermediate layer of the flexible substrate. Further, a single fiber of the fiber Bragg grating array is connected in series with multiple sensors, and is wound along a helical path to cover the strains in the X, Y, and Z axes.

[0093] Specifically, the core of the fiber Bragg grating is a section of specially treated optical fiber with a periodic grating inscribed inside. When the external environment changes, the physical properties of the grating change, resulting in a shift in the wavelength of the reflected light. By measuring this wavelength shift, the strain or temperature change experienced by the optical fiber can be obtained.

[0094] Further, the optical fiber is wound around the tank body in a three-dimensional helical path, covering the axial, circumferential and radial directions. This helical structure enables a single optical fiber to simultaneously sense deformations in different directions, including the extension or contraction of the tank body along the length direction when it is subjected to internal pressure, the torsion of the tank body caused by external loads, and the expansion or collapse of the tank wall due to the pressure difference inside and outside the tank wall.

[0095] Further, the pitch of the helix is determined by the size of the regular polygon unit of the surface piezoelectric film. As a preferred example, if the side length of the regular polygon is A mm, the pitch is set to B mm to ensure that the helical path is aligned with the surface sensors, forming a spatially matched detection network to avoid data misalignment.

[0096] Multiple micro gratings are connected in series on a single optical fiber to form dense detection nodes. Each grating is assigned a unique reflection wavelength, and data at different positions are distinguished by the wavelength.

[0097] The intermediate layer is used to monitor the distribution of the deformation gradient of the tank body, detect the shear stress at the interface between the enamel and the substrate, and compensate for the interference of temperature on the piezoelectric signal.

[0098] The underlying layer is a differential eddy current probe composed of miniaturized electromagnetic coils.

[0099] The underlying layer is an electromagnetic eddy current array, which is installed on the surface of the tank body and isolated by a flexible substrate at the same time.

[0100] Furthermore, the eddy current probe adopts a differential coil structure.

[0101] Furthermore, the underlying layer is used to detect the corrosion of the metal matrix, identify the hidden cracks under the enamel layer, and evaluate the change of the coating thickness.

[0102] The three-layer data generates a three-dimensional damage topology map through spatial mapping, which is used to locate the propagation path of surface cracks and the internal corrosion area.

[0103] Specifically, the specific implementation method of the three-dimensional damage topology map includes:

[0104] The three-layer data usually corresponds to three types of detection data with different dimensions:

[0105] The first layer outputs the surface point cloud coordinates (X, Y, Z) and geometric parameters such as crack width and orientation through the surface topography data from the piezoelectric film array.

[0106] The second layer detects the subsurface defects through the subsurface defect data from the eddy current sensor and outputs the Z-axis offset of the defect depth, the cross-sectional size and the expansion angle.

[0107] The third layer outputs a volume data matrix through the internal corrosion data from the fiber Bragg grating, reflecting the spatial position and morphology of the corrosion area.

[0108] Furthermore, the specific implementation method of spatial mapping includes:

[0109] Taking the geometric center of the target carrier as the origin, a cylindrical coordinate system is established.

[0110] Taking the CAD model of the enamel tank body as the reference, a global three-dimensional coordinate system (X, Y, Z) is established.

[0111] The surface layer is the piezoelectric film point cloud data (X, Y, Z), which is registered with the three-dimensional surface of the tank body through the iterative closest point.

[0112] The subsurface layer is the eddy current detection data, which is mapped to the global coordinate system by interpolation based on the corresponding point coordinates on the surface.

[0113] The internal layer is the fiber Bragg grating strain data, which is converted into a three-dimensional strain field through the meshing of the tank body and superimposed on the global coordinate system.

[0114] Spatio-temporal synchronization embeds all sensor data with a unified timestamp.

[0115] The modeling of the damage condition of the enamel tank includes surface crack modeling, subsurface defect modeling, and internal corrosion modeling.

[0116] Among them, the specific implementation methods of surface crack modeling, subsurface defect modeling, and internal corrosion modeling are as follows:

[0117] Surface crack modeling involves segmenting the crack point cloud of the piezoelectric film array to extract the crack contour point set; fitting the crack trend based on the crack contour point set, calculating the width, comparing with historical point cloud data, and tracking the crack tip displacement through the optical flow method to generate the crack propagation vector field.

[0118] Subsurface defect modeling uses the finite element method to solve the Maxwell equation, converts the change in eddy current impedance into the equivalent geometric parameters of subsurface defects, and the geometric parameters include depth, the major axis / minor axis of the cross-sectional ellipse. The defect geometric parameters are converted into a three-dimensional ellipsoid and embedded in the global coordinate system.

[0119] Internal corrosion modeling establishes a strain characteristic library for the corrosion area, including material loss caused by corrosion, and local strain showing an alternating distribution of tension and compression. By setting a classifier, the fiber Bragg grating strain field data is segmented into a normal area, a corrosion transition area, and a corrosion core area.

[0120] In this application, a preferred specific implementation method for establishing spatial association rules and constructing a three-dimensional topological map can be selected: The construction of spatial association rules includes surface-subsurface association, subsurface-internal association, and dynamic weight assignment.

[0121] In the surface-subsurface association, if there is a subsurface defect voxel under the surface crack tip, it is determined that the crack propagates inward, and the predicted value of the crack depth is corrected.

[0122] In the subsurface-internal association, when the subsurface defect voxel overlaps with the corrosion transition area voxel in space, it is determined that corrosion induces subsurface peeling, and the risk level of this area is increased.

[0123] Dynamic weight assignment dynamically adjusts the weights of each layer of data in the fusion according to the sensor confidence.

[0124] The method for establishing a three-dimensional topological map includes storing the damage information in the global coordinate system, and each node contains: surface crack probability, subsurface defect size, corrosion grade, and comprehensive risk score.

[0125] Crack path positioning predicts the crack path of the enamel tank by combining the current crack trend and the internal corrosion distribution. The future crack trend tends to be in the position with severe internal corrosion.

[0126] Corrosion hot spot positioning performs clustering analysis on the corrosion voxels and extracts the central coordinates of the largest connected domain as the priority maintenance area;

[0127] Within the neighborhood radius of a certain point on the enamel tank, if there are enough corrosion points, this area is considered a high-density corrosion area and is extended into a cluster. A cluster expands through points connected by density, and points that cannot be attributed to any cluster are considered noise points for the output result.

[0128] The damage report includes a list of crack locations, lengths, depths, corrosion volumes, distributions, and risk levels.

[0129] The biological neural network perception system includes a synaptic network, and the synaptic network constructs a hierarchical conductive network based on a flexible substrate;

[0130] The hierarchical conductive network includes a first-level main trunk and second-level branches. The first-level main trunk is used to connect multiple sensor clusters, and the second-level branches are used to connect individual sensing units;

[0131] The working method of the biological neural network includes: the piezoelectric film array generates a time pulse sequence, the fiber Bragg grating outputs a frequency modulation signal, the electromagnetic data is converted into pulse intensity, and the edge node performs pulse timing correlation analysis.

[0132] The first-level main trunk is composed of a cross-linked structure of silver nanowire and polyurethane composite material, and extends along the curvature direction of the enamel tank body to form a closed-loop path;

[0133] Specifically, the first-level main trunk connects multiple sensor clusters, that is, a collection of multiple sensing units, and is used for cross-regional signal aggregation and distribution.

[0134] The first-level main trunk extends along the curvature direction of the target carrier such as the enamel tank body to form a closed-loop path, adapts to the complex curved surface environment, ensures the structural stability and signal transmission redundancy, and the closed-loop avoids global failure caused by a single-point break.

[0135] The first-level main trunk can be composed of a cross-linked structure of silver nanowire and polyurethane composite material. The flexibility of polyurethane is used to match the surface deformation, and at the same time, the high conductivity of silver nanowire ensures low-impedance signal transmission.

[0136] The first-level main trunk carries the mixed signals of multiple sensor clusters, including the time pulse of the piezoelectric film, the frequency modulation signal of the fiber Bragg grating, and the pulse intensity of the electromagnetic data. The closed-loop path allows for bidirectional signal flow.

[0137] The second-level branches bifurcate from the first-level main trunk at a certain angle, and the branch density is determined by the spatial distribution of the sensor clusters;

[0138] Specifically, the secondary branches branch off from the primary trunk and directly connect to a single sensing unit, such as piezoelectric film, fiber grating, etc., extending from the primary trunk at a certain angle. The angle range balances the signal transmission efficiency and spatial distribution density, preventing the angle from being too small to cause branch congestion, or too large to cause increased signal transmission distance loss. The branch density is dynamically adjusted by the spatial distribution of the sensor cluster, and the branch density is increased in high-sensitivity areas.

[0139] A capacitive coupling node is provided at the junction of the primary trunk and the secondary branch to achieve low-loss transmission of the pulse signal through impedance matching;

[0140] Furthermore, at the connection point between the primary trunk and the secondary branch, signal coupling is achieved through microstructure capacitors. The high-frequency and low-impedance characteristics of the capacitors are utilized to reduce ohmic losses in pulse signal transmission. By adjusting the ratio of the branch diameter to the trunk cross-sectional area, the branch input impedance is matched with the trunk characteristic impedance, thereby reducing signal reflection.

[0141] The fractal network has self-repairing properties and reconstructs the conductive path through the electrochemical migration of silver nanoparticles when it is partially broken.

[0142] Furthermore, when the branch or trunk breaks due to mechanical stress, the silver nanoparticles at the break reconstruct the conductive path through the dissolution-deposition process under the action of electrochemistry. If anodic dissolution occurs, Ag at the break end is oxidized to Ag. + Enter the electrolyte; if it is cathode deposition, then Ag + It is reduced to Ag element at the opposite end to form a nanobridge;

[0143] The primary trunk and secondary branches construct a bionic neural network with efficient transmission, environmental adaptability and self-maintenance capabilities through the design of typed conductive network structure, material function synergy, and biological-like signal mechanism. It is particularly suitable for intelligent monitoring scenarios of curved equipment such as enameled tanks, and realizes neural-like signal processing from local sensing units to edge computing nodes.

[0144] The energy supply unit includes thermoelectric power generation and wireless charging embedded in a magnetic structure, which doubles up the energy supply.

[0145] The thermoelectric power generation adopts a curved surface adaptive thermoelectric structure, including:

[0146] The flexible thermoelectric arm array is composed of Bi2Te3 / Sb2Te3 heterostructures folded into serpentine units, forming a closed loop along the circumference of the tank. The hot end is coupled to the inner wall of the tank through a graphene thermal conductive film, and the cold end is connected to radiant micro-heat dissipation fins, expanding the surface area by 8 times.

[0147] The phase change energy storage and leveling layer is filled with paraffin-carbon nanotube composite phase change material between the hot and cold ends, buffering temperature fluctuations through solid-liquid phase change.

[0148] The contact pressure self-adjusting mechanism adopts a shape memory alloy spring to dynamically adjust the hot end contact pressure according to the curvature of the tank surface to ensure stable thermal resistance of the temperature difference interface.

[0149] A specific implementation method of thermoelectric power generation includes:

[0150] The flexible thermoelectric arm array uses molecular beam epitaxy technology to grow Bi2Te3 / Sb2Te3 heterojunction films on a quartz substrate. The thickness of the film needs to be controlled, in which Bi2Te3 layers (n-type) and Sb2Te3 layers (p-type) are alternately stacked to form a thermoelectric unit.

[0151] The serpentine folding structure is achieved by cutting serpentine channels on the thermoelectric film, with individual folding units arranged circumferentially to form a closed loop; electrodes are evaporated at both ends of the serpentine arms, and flexible wires are used to connect the thermoelectric units in series to form a total thermoelectric electromotive force array.

[0152] The serpentine folding structure allows the thermoelectric arms to bend freely on the surface, and combined with the dynamic pressure adjustment of the spring, it achieves a tight fit on the entire curved surface of the tank, avoiding the fitting gap of traditional rigid thermoelectric modules.

[0153] The wireless charging embedded magnetic power supply adopts magneto-electric composite topology;

[0154] The magneto-electric composite topology includes a domain-specific magnetic attraction array, dynamic magnetic compensation, and thermal-electric coordinated management;

[0155] Neodymium iron boron permanent magnets are arranged in the outer ring area to achieve rapid adsorption and positioning;

[0156] A planar spiral coil embedded in the central functional area forms a resonant coupling with the external charging base. Dynamic magnetic force compensation uses a Hall sensor array to monitor the magnetic gap in real time. The electromagnetic auxiliary coil dynamically compensates the magnetic force based on the gap value, maintaining adsorption stability while avoiding overvoltage damage to the enamel layer.

[0157] Thermal-electric collaborative management integrates micro heat pipes on the back panel of the wireless charging module to direct the heat generated by the coil to the cold end of the thermoelectric power generation, forming an energy recycling chain.

[0158] A wireless charging embedded magnetic attraction structure includes machining multiple cylindrical grooves on a substrate, fixing permanent magnets at the same time, and covering the surface with a flexible magnetic isolation layer to avoid magnetic short circuits and prevent scratches on the enamel layer.

[0159] The central functional area is a multi-layer stacked structure, including multiple single-layer coils. The single-layer coils match the parameters of the external base coils. The layers are isolated by thin films and the whole is encapsulated in a silicone protective layer.

[0160] A specific implementation method of dynamic magnetic compensation includes:

[0161] A number of high-precision Hall sensors are symmetrically arranged at the four corners of the bottom surface of the magnetic attraction module, which are used to collect the air-gap magnetic field intensity in real time. The average gap is calculated through trigonometric functions. The sensors are encapsulated in the micro-grooves on the edge of the magnetic attraction module, and the surface is covered with transparent epoxy resin to prevent dust from interfering with the magnetic field signal.

[0162] The control logic is that when the average gap shrinks, the control module increases the current of the auxiliary coil to enhance the adsorption force; when the surface pressure of the tank body increases beyond the dangerous threshold of the enamel layer, the current is automatically reduced to avoid cracking of the enamel layer due to pressure.

[0163] The wireless charging embedded magnetic attraction structure realizes the coordination of high-reliability adsorption, high-efficiency energy transmission and intelligent thermal management, provides a stable composite power supply solution for the enamel tank body detection system, and is applicable to harsh industrial environments such as humidity and vibration.

[0164] Embodiment 2

[0165] The assembled enamel tank body condition detection system further includes:

[0166] As Figure 2 shown, the inner wall corrosion diagnosis method includes:

[0167] The inner wall corrosion diagnosis decodes the codes of the inner wall corrosion areas through the coded coating protection film and locates the decoding results.

[0168] Specifically, the steps of the inner wall corrosion diagnosis coded coating protection film further include: adopting a segmented color gradient coding, dividing the inner wall surface into several equally spaced annular regions, and setting a unique color code composed of more than two basic colors in each region, and the color codes of adjacent regions are connected by a gradient color transition; the mixing ratio of the color codes is associated with the axial position coordinates of the annular regions, and the thickness of the color film layer is controlled so that when corrosion occurs, the color codes produce recognizable color peeling characteristics due to the damage of the film layer. This coding method can be recognized through the visible light band without adding additional fluorescent or magnetic markers.

[0169] Further, the positioning process of the decoding result further includes: using a multi-spectral imaging device to collect the reflection spectrum data of the corrosion area, establishing a chromaticity-position mapping database, and performing differential comparison between the detected abnormal color codes and the original coded data; for the corrosion boundary of the gradient color transition section, using an edge fusion algorithm to compensate and calculate the attenuation gradient of adjacent color codes, and introducing the topological association rules between annular regions to map the discrete corrosion points into a three-dimensional polar coordinate system, and finally outputting an accurate positioning grid map with the corrosion area ratio exceeding the threshold of 5%.

[0170] Further, before the enamel is sintered and formed, the coded film is prefabricated at the interface between the metal substrate and the enamel layer by screen printing, so that the underlying color code is directly exposed after the enamel layer is corroded and penetrated; during the detection stage, a telescopic guide rail is used to carry a linear array sensor to perform a spiral scan along the axial direction of the tank body, and the data of the pressure sensor in the tank is synchronously combined. When local color code anomalies are detected, a wall thickness ultrasonic probe is automatically triggered for secondary verification, and the positioning coordinates and the corrosion grade classification results are transmitted to an external monitoring terminal through a radio frequency module.

[0171] Further, two functional films are added between the metal substrate and the enamel layer: the original segmented color code layer at the bottom layer, and a zinc-based fluorescence-sensitive layer fluorescent dye is added to the middle layer.

[0172] The detection principle includes early corrosion detection and late corrosion detection.

[0173] In early corrosion detection, when the enamel is not penetrated and the corrosive medium penetrates the enamel microcracks, it reacts with ZnO to generate Zn 2+ , triggering fluorescence dye quenching, and local dark spots can be seen under ultraviolet light.

[0174] In late corrosion detection, when the enamel is penetrated and the color code layer is exposed, the color anomaly is identified through a multispectral camera.

[0175] An adaptive model of corrosion morphology is established, and the corrosion morphology of the enamel tank body is adaptively identified through a feature library and a dynamic matching algorithm.

[0176] Among them, the feature library is constructed to pre-store the spectral attenuation curves of typical corrosion morphologies. Through experiments, it is measured that the attenuation of the spectral attenuation curve shows an exponential distribution and the linear attenuation of crevice corrosion is superimposed with Gaussian noise.

[0177] Further, after the dynamic matching algorithm detects the color code abnormal area, its reflection spectral curve is extracted, and the similarity is calculated with the feature library, and the best matching model is selected to reconstruct the corrosion boundary.

[0178] Further, the change of subsurface conductivity is detected by an eddy current probe, and the local enamel layer thickness is measured by a micro ultrasonic thickness gauge. When the color code abnormal area meets the following conditions at the same time, it is determined as effective corrosion.

[0179] A specific implementation method for inner wall corrosion diagnosis includes:

[0180] Segmented color gradient coding, the inner wall of the tank is equally spaced along the axial direction into N annular regions, and each region corresponds to the axial coordinate z_i = i×50, (i = 1, 2,..., N), and is evenly distributed circumferentially at 360°.

[0181] Using the mixing of RGB three primary colors, the color code of each region is uniquely determined by the axial coordinate z_i, and the color codes of adjacent regions are transitioned through gradient colors to form a continuous color gradient band.

[0182] The screen-printed pre-coded film uses laser engraving to produce a circular hollow pattern on a stainless steel screen plate.

[0183] Corrosion area decoding includes multi-spectral imaging data acquisition, chromaticity position database, and gradient zone corrosion boundary processing.

[0184] For multi-spectral imaging data acquisition, a telescopic guide rail is configured to carry a linear array CCD sensor to perform spiral scanning on the inner wall, and synchronously collect the reflected spectral data of the inner wall. Before scanning, a standard color card is used for white balance correction to establish the mapping relationship between spectral reflectance and RGB values.

[0185] The chromaticity position database collects 100 groups of spectral data for each circular area on the inner wall of the newly manufactured tank body, calculates the average RGB value and CIELAB color space coordinates, and establishes a three-dimensional database of "axial coordinates, circumferential angles, and standard color codes". If there is an abnormal color difference, it is marked as a corrosion-suspected area and a secondary verification is triggered.

[0186] In the gradient zone corrosion boundary processing, in the gradient transition zone between adjacent circular areas of the tank body, the corrosion boundary is not a clear straight line, but a blurred band with gradually changing colors as the corrosion depth varies.

[0187] If the color codes of two adjacent standard areas are "color of area 1" and "color of area 2" respectively, when corrosion occurs in the gradient zone between them, the actually detected color is a mixed color of the two. The shallower the corrosion, the closer the color is to "color of area 1"; the deeper the corrosion, the closer the color is to "color of area 2".

[0188] The corrosion depth is judged by ultrasonic thickness measurement data, and the depth is converted into a ratio value from 0 to 1. For example, as the corrosion depth increases, the depth ratio increases. By adjusting the mixing degree of the two standard colors according to the depth ratio, in a preferred scheme, if the ratio value is 0, it is completely "color of area 1", if the ratio value is 1, it is completely "color of area 2", and the intermediate value is mixed proportionally to restore the true color transition law in the gradient zone and eliminate the recognition error caused by the blurred boundary during detection.

[0189] The tank body has a cylindrical inner wall. In order to accurately locate the corrosion area, it is necessary to convert the curved surface into a coordinate system convenient for calculation; the tank body has a fixed radius r of its own, and the circumferential angle θ starts from 0° with a certain generatrix of the tank body as the starting point. When scanning clockwise and dividing an angular unit for detection, multiple discrete corrosion points will be obtained. Through the above polar coordinates, the circumferential angle and axial height of each point are corresponding to the specific position on the cylindrical curved surface. Then, according to the distribution law of these points, an algorithm is used to "connect" them into a continuous area to form a complete corrosion contour.

[0190] Within the fitted continuous region, the proportion of the corrosion area in the total area of the corresponding annular region is statistically calculated. Through precise coordinate positioning and contour fitting, a three-dimensional grid map containing the corrosion location and area is finally output, visually showing the severity of corrosion.

[0191] Importantly, it should be noted that the construction and arrangement of the present application shown in multiple different exemplary embodiments are merely illustrative. Although only two embodiments are described in detail in this disclosure, those skilled in the art who refer to this disclosure should easily understand that many modifications are possible without substantially departing from the novel teachings and advantages of the subject matter described in this application. For example, changes in the dimensions, scales, structures, shapes and proportions of various components, as well as parameter values (such as temperature, pressure, etc.), installation arrangements, use of materials, colors, orientations, etc. For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of the element may be inverted or otherwise changed, and the nature, number or position of discrete elements may be altered or changed. Therefore, all such modifications are intended to be included within the scope of the present invention. The order or sequence of any process or method steps may be changed or reordered according to alternative embodiments. Any "means-plus-function" clause is intended to cover the structures that perform the functions described herein, and not only structurally equivalent but also equivalent structures. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0192] In addition, in order to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the currently considered best mode of implementing the present invention or those that are not relevant to the implementation of the present invention).

[0193] It should be understood that in the development of any actual implementation, as in any engineering or design project, a large number of specific implementation decisions may be made. Such development efforts may be complex and time-consuming, but for those of ordinary skill in the art who benefit from this disclosure, without excessive experimentation, the development efforts will be a routine task of design, manufacturing and production.

[0194] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An assembled enamel tank body condition detection system, characterized in that Comprising: A three-dimensional heterogeneous sensing network is set up by a three-dimensional sensor group, which is placed on the surface of the enamel tank body by pasting to form a biological neural network perception system; The three-dimensional heterogeneous sensing network includes a three-layer stacked structure, and the three-layer stacked structure includes a surface layer, an intermediate layer and a bottom layer; The surface layer is a piezoelectric film array, which is densely paved on the surface of the enamel tank body in a regular polygon; The intermediate layer is a fiber Bragg grating, which is wound in a three-dimensional helix, and the pitch of the three-dimensional helix is determined by a regular polygon; The bottom layer is a differential eddy current probe composed of miniaturized electromagnetic coils; The power supply unit includes thermoelectric power generation and wireless charging embedded magnetic adsorption power supply to double ensure power supply; Inner wall corrosion diagnosis decodes the code of the inner wall corrosion part by coding and coating a protective film, and locates the decoding result.

2. The assembled enamel tank body state detection system according to claim 1, characterized in that: The biological neural network perception system includes a synaptic network, and the synaptic network constructs a hierarchical conductive network on a flexible substrate; The hierarchical conductive network includes a primary main trunk and secondary branches. The primary main trunk is used to connect multiple sensor clusters, and the secondary branches are used to connect single sensing units; The working method of the biological neural network includes: the piezoelectric film array generates a time pulse sequence, the fiber Bragg grating outputs a frequency modulation signal, the electromagnetic data is converted into pulse intensity, and the edge node performs pulse timing correlation analysis.

3. The assembled enamel tank body state detection system according to claim 1, characterized in that: The sensing network realizes multi-modal data fusion through the cooperative working mechanism of the three-layer stacked structure, specifically including: The surface piezoelectric film array detects the dynamic stress wave on the surface of the enamel tank body through a regular polygon dense paving configuration, and converts mechanical vibration into an electrical pulse signal; The intermediate layer fiber Bragg grating is based on a three-dimensional helix, and the pitch of the three-dimensional helix is determined by a regular polygon, which is used to monitor the shear strain distribution of the tank body in real time; The bottom differential eddy current probe extracts the electromagnetic characteristic parameters of the metal matrix inside the tank body through the complementary layout of miniaturized electromagnetic coils; The three-layer data is fused through spatio-temporal registration to generate a three-dimensional damage topology map for locating the surface crack propagation path and the internal corrosion area.

4. The assembled enamel tank body state detection system according to claim 2, characterized in that: The primary main trunk is composed of a cross-linked structure of a composite material, and extends along the curvature direction of the enamel tank body to form a closed-loop path; The secondary branches branch off from the primary main trunk at an angle, and the branch density is determined by the spatial distribution of the sensor clusters; A capacitive coupling node is provided at the joint of the primary main trunk and the secondary branches to achieve low-loss transmission of pulse signals through impedance matching.

5. The assembled enamel tank body state detection system according to claim 4, characterized in that: The thermoelectric power generation adopts a curved surface adaptive thermoelectric structure, including: an array of flexible thermoelectric arms is composed of a Bi2Te3 / Sb2Te3 heterostructure into a serpentine folding unit, forms a closed-loop circuit along the circumferential direction of the tank body, the hot end is coupled with the inner wall of the tank body through a graphene heat conduction film, and the cold end is connected with a radiation-type micro heat dissipation fin; The phase change energy storage leveling layer fills paraffin-carbon nanotube composite phase change material between the hot and cold ends to buffer temperature fluctuations through solid-liquid phase change. The contact pressure self-regulating mechanism uses a shape memory alloy spring to dynamically adjust the hot end contact pressure according to the curvature of the tank body surface.

6. The assembled enamel tank body condition detection system according to claim 5, wherein: The wireless charging embedded magnetic attraction power supply adopts a magnetoelectric composite topology; The magnetoelectric composite topology includes a segmented magnetic attraction array, dynamic magnetic force compensation, and thermal-electric collaborative management; The segmented magnetic attraction array includes an outer ring and a central area. Neodymium iron boron permanent magnets are arranged in the outer ring area to form a gradient magnetic field. A planar spiral coil is embedded in the central area to form a resonant coupling with an external charging base; The dynamic magnetic force compensation uses a Hall sensor array to monitor the magnetic attraction gap in real time, and the electromagnetic auxiliary coil dynamically compensates the magnetic force according to the gap value; The thermal-electric collaborative management integrates a micro heat pipe on the back plate of the wireless charging module to direct the heat generated by the coil to the cold end of the thermoelectric power generation to form an energy recycling chain.

7. The assembled enamel tank body condition detection system according to claim 1, wherein: The inner wall corrosion diagnosis uses a segmented color gradient coding method, divides the inner wall surface into several equally spaced annular regions, sets a unique color code formed by mixing two or more basic colors in each region, and the color codes of adjacent regions are connected by a gradient color transition; The mixing ratio of the color codes is associated with the axial position coordinates of the annular region, so that when corrosion occurs, the color code produces a recognizable color peeling feature due to the film layer breakage.

8. The assembled enamel tank body condition detection system according to claim 7, wherein: The inner wall corrosion diagnosis further includes: collecting the reflection spectrum data of the corrosion area by multispectral imaging, and performing differential comparison between the detected abnormal color code and the original coding data by establishing a chromaticity-position mapping database; For the corrosion boundary of the gradient color transition section, an edge fusion algorithm is used to compensate and calculate the attenuation gradient of adjacent color codes, and the topological association rules between annular regions are introduced to map discrete corrosion points into a three-dimensional polar coordinate system, and finally the corrosion area is output.

9. The assembled enamel tank body condition detection system according to claim 8, wherein: The coding film is prefabricated at the interface between the metal matrix and the enamel layer by screen printing, so that the bottom color code is directly exposed after the corrosion penetrates the enamel layer; During the detection stage, a telescopic guide rail is used to carry a linear array sensor to perform spiral scanning along the axial direction of the tank body, and the data of the pressure sensor in the tank is synchronously combined. When a local color code anomaly is detected, a wall thickness ultrasonic probe is automatically triggered for secondary verification.

Images