Intelligent monitoring system and method for large-span steel structure construction

Abstract

The invention relates to the technical field of steel structure construction, in particular to an intelligent monitoring system and method for large-span steel structure construction, and the method comprises the steps: deploying a multi-source sensing network integrated with an electromagnetic eddy current sensor, a gyroscope and a photonic crystal strain gauge on the surface of a steel structure, and collecting the micro lattice distortion amount, the low-frequency vibration phase difference and the optical path strain gradient; a dynamic holographic twinborn model is constructed through a holographic interference calculation module, a structural deformation field is extracted by using a quantum annealing algorithm, a critical buckling load correction, a welding residual stress compensation coefficient and a hoisting path curvature optimization function are predicted based on a quantum convolutional neural network, and a construction regulation instruction is injected into construction equipment through a quantum state regulation module. According to the method, multi-source sensing, quantum intelligent calculation and structure safety control are integrated, and the state sensing precision, the regulation and control response speed and the safety guarantee capability in the construction process are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of steel structure construction, and in particular to an intelligent monitoring system and method for large-span steel structure construction. Background Art

[0002] During the construction of large-span steel structures, the stress on structural components is complex, the strain state is changeable, and factors such as local instability, buckling deformation and welding residual stress can easily lead to a decrease in the dynamic stability of the overall structure, which in turn causes safety hazards. In recent years, intelligent monitoring systems based on multi-sensor deployment have gradually been developed. At the same time, with the development of quantum computing, holographic simulation and multi-scale fusion modeling technology, the demand for micro-nanoscale deformation field perception, quantum computing-assisted prediction and real-time optimization of construction paths in the field of structural engineering continues to rise, promoting the evolution of intelligent construction control technology towards deep integration and high resolution.

[0003] However, there are three prominent problems in the existing technology: first, the perception dimension is limited, and it is impossible to simultaneously obtain the dynamic change characteristics of the structure at the microscopic lattice level, vibration coherence level and strain gradient level, resulting in fragmented perception results and difficulty in forming a systematic model; second, there is a lack of effective cross-scale computing mechanism at the data processing level. Conventional convolutional neural networks have limited ability to identify complex nonlinear evolution trends of structures, and it is difficult to achieve multi-target linkage prediction of critical buckling, welding residual stress and lifting path interference; third, there is a general lack of feedback closed-loop mechanism between construction control and structural status, and the effectiveness of the execution of control measures cannot be verified in real time, resulting in delayed system response and insufficient accuracy of control instructions. Summary of the invention

[0004] The present invention provides an intelligent monitoring system and method for the construction of large-span steel structures, and constructs an intelligent monitoring method that integrates high-precision perception, quantum intelligent prediction, and feedback adaptive control to meet the multiple requirements of safety, accuracy, and dynamics in the construction of large-span steel structures.

[0005] The intelligent monitoring method for large-span steel structure construction includes the following steps:

[0006] S1: Deploy a sensor network on the surface of the steel structure, wherein the sensor network includes an electromagnetic eddy current array sensor, a gyroscope and a photonic crystal strain gauge to synchronously collect key data in the steel structure construction, including microscopic lattice distortion, low-frequency vibration phase difference and optical path strain gradient;

[0007] S2: Input the collected key data into the holographic interferometric calculation module, solve the structural deformation field under the spatiotemporal entangled state through the quantum annealing algorithm, and construct a dynamic holographic twin model, which includes a nanoscale lattice displacement cloud map, a vibration phase coherence map, and a non-uniform strain field vortex feature;

[0008] S3: Based on the dynamic holographic twin model, use a quantum convolutional neural network to predict the evolution trend of the structural form, and output a construction control instruction set, which includes: the critical buckling load correction amount, the welding residual stress compensation coefficient, and the hoisting path curvature optimization function;

[0009] S4: Inject the construction control instruction set into the construction equipment control system in real time through the quantum state control module to drive dynamic compensation behaviors, which include:

[0010] implant strain field feedback in the laser welding robot to adjust the laser power in real time;

[0011] reconstruct the spatial configuration of the hoisting rigging based on the curvature optimization function;

[0012] Trigger the prestress redistribution mechanism when the critical buckling load correction amount reaches 90% of the material yield strength.

[0013] Optionally, the S1 includes:

[0014] S11: Deploy an electromagnetic eddy current array sensor on the surface of the steel structure, collect the electromagnetic signals on the surface of the steel structure through the electromagnetic eddy current array sensor. The electromagnetic eddy current array sensor uses the principle of electromagnetic field induction to detect the tiny deformations on the surface and inside of the steel structure. Based on the collected electromagnetic signals on the surface of the steel structure, use the electromagnetic eddy current response model to convert the electromagnetic signals into microscopic lattice distortion amounts. The microscopic lattice distortion amount represents the distortion degree of the microscopic lattice on the surface of the steel structure and reflects the microscopic deformation of the structure;

[0015] S12: Deploy gyroscopes at key positions of the steel structure, collect the angular velocity data of the steel structure through the gyroscopes. Based on the collected angular velocity data, use the integral calculation method to deduce the vibration displacement and phase information of each position point of the steel structure. By calculating the phase difference between different position points, obtain the low-frequency vibration phase difference. The low-frequency vibration phase difference reflects the phase change caused by vibration during the construction process of the steel structure and characterizes the amplitude and phase difference of the vibration;

[0016] S13: Install a photonic crystal strain gauge on the surface of the steel structure, collect the optical path data on the surface of the steel structure through the photonic crystal strain gauge. Based on the collected optical path data, calculate the optical path change amount by comparing the optical path differences at different positions. Based on the optical path change amount, use the optical path strain gradient calculation method to obtain the optical path strain gradient on the surface of the steel structure, and generate an optical path strain gradient field based on the optical path strain gradient. The optical path strain gradient represents the strain change rate between different positions and reflects the local strain distribution of the steel structure during the construction process;

[0017] S14: The collected microscopic lattice distortion, low-frequency vibration phase difference and optical path strain gradient are synchronously processed by a synchronous data acquisition unit to generate a multi-source data set.

[0018] Optionally, S2 includes:

[0019] S21: Input multi-source data sets into the holographic interferometry calculation module to construct the Hamiltonian model of the spatiotemporal entangled state, which describes the interaction between various physical quantities in the system. The Hamiltonian model combines the cross-scale coupling relationship between lattice displacement, vibration phase difference and optical path strain gradient to better simulate the deformation behavior of the structure;

[0020] S22: Use quantum annealing algorithm to solve Hamiltonian model, including discretizing Hamiltonian model into quantum bit-encoding Ising model, which describes the interaction between quantum bits in the system. By solving the Ising model, the distribution of ground state energy and related physical quantities is obtained.

[0021] S23: Extract the probability amplitude distribution from the output of the quantum annealing process to generate the structural deformation field, including Fourier-Bessel transform of the output probability amplitude to obtain the nanometer-scale lattice displacement cloud map. By calculating the coherence map of the vibration phase, the relationship between the vibration mode and the phase difference is analyzed. These maps help to understand the dynamic behavior of the structure under different conditions.

[0022] Optionally, in the quantum annealing process, the annealing time and temperature are set, and the lowest energy state is sought through the annealing operation until the required accuracy is achieved.

[0023] Optionally, S2 further includes:

[0024] S24: Extract vortex features in non-uniform strain fields based on the optical path strain gradient field, including locating the core of the vortex by analyzing the rotation behavior of the strain field, and judging the effectiveness of the vortex. The vortex is caused by the inhomogeneity in the strain field. Marking effective vortices is crucial for understanding the local deformation and possible failure areas of the structure.

[0025] S25: The obtained lattice displacement cloud map, vibration phase coherence map and inhomogeneous strain field vortex characteristics are integrated to construct a dynamic holographic twin model, which can present tiny changes in the structure with extremely high resolution, ensuring reliability and accuracy in engineering applications.

[0026] Optionally, the S3 includes:

[0027] S31: Input the nanoscale lattice displacement cloud map, vibration phase coherence map and inhomogeneous strain field vortex features in the dynamic holographic twin model into the quantum convolutional neural network for feature encoding and fusion;

[0028] The quantum convolutional neural network contains multiple quantum convolutional layers and traditional convolutional layers. The quantum convolutional layer maps the lattice displacement data into a quantum state by means of qubit encoding, tensors the vibration phase coherence map and the vortex core position, and forms a fused feature tensor, which is then input into the traditional convolutional layer for further processing. The training process of the quantum convolutional neural network is based on an optimized loss function to ensure that the network can accurately extract features from the input data;

[0029] S32: After the training of the quantum convolutional neural network is completed, the quantum convolutional neural network is used to predict the critical buckling load correction amount, including performing quantum measurement on the quantum state to obtain the probability distribution of the displacement gradient, and calculating the critical buckling load correction amount by analyzing the peak value of the probability distribution. This process helps to predict the behavior of the structure under different loads and ensures timely adjustment during the construction process to avoid structural instability;

[0030] S33: Based on the optical path strain gradient field, calculate the welding residual stress compensation coefficient. By analyzing the optical path strain gradient on the welding path, find the value of the maximum strain gradient and calculate the corresponding compensation coefficient. This compensation coefficient is used to correct the residual stress that may be generated during the welding process, thereby reducing the impact on the structural stability;

[0031] S34: After calculating the welding residual stress compensation coefficient, use the vortex characteristics of the non-uniform strain field to generate an optimized function for the curvature of the lifting path. By analyzing the position of the vortex core, fit an optimized lifting path curve to reduce the structural deviation during the lifting process. The optimized path will help improve the accuracy and safety of the lifting process;

[0032] S35: Integrate the critical buckling load correction amount, the welding residual stress compensation coefficient, and the optimized function for the curvature of the lifting path to generate a construction control instruction set.

[0033] Optionally, if the critical buckling load correction amount reaches a certain threshold in S35, a warning will be triggered and the construction instructions will be frozen, and the coupling effect between the residual stress compensation coefficient and the optimized function for the curvature of the lifting path will be verified, and the structural safety margin will be confirmed to meet the specified requirements through finite element simulation.

[0034] Optionally, S4 includes:

[0035] S41: Inject the welding residual stress compensation coefficient into the control system of the laser welding robot and adjust the laser power in real time through the quantum state control module;

[0036] S42: Based on the curvature optimization function of the hoisting path, reconstruct the spatial configuration of the hoisting tackle, including generating a control point sequence of the hoisting path by analyzing the curvature components in the curvature optimization function to ensure stability and accuracy during the hoisting process. The hydraulic servo system adjusts the length and angle of the hoisting tackle according to the control point sequence to ensure that the geometric shape of the tackle meets the hoisting requirements;

[0037] S43: If the critical buckling load correction reaches 90% of the material yield strength, trigger the prestress redistribution mechanism. In the buckling risk area, apply reverse stress through shape memory alloy actuators to adjust the stress distribution and reduce the local buckling risk. At the same time, verify the nanoscale lattice displacement nephogram, calculate the strain energy density, and ensure that it is within the safe range to prevent structural failure caused by excessive strain energy accumulation;

[0038] S44: Verify the structural state after dynamic compensation behavior through a quantum channel, including encoding the laser power, tackle configuration, and reverse stress into quantum states, comparing the quantum states with the dynamic holographic twin model, verifying the fidelity of the quantum states, ensuring that the structural state conforms to the expectation. If the fidelity does not reach the preset standard, trigger the sensing network to re-collect key data to ensure that all key data is monitored and adjusted in real time.

[0039] Optionally, during the process of the hydraulic servo system in S42 adjusting the length and angle of the hoisting tackle according to the control point sequence, the tension of the tackle is monitored in real time. If the tension exceeds the safe range, trigger the smoothing filter of the path curvature to ensure the safety of the hoisting process.

[0040] An intelligent monitoring system for large-span steel structure construction, used to implement the above-mentioned intelligent monitoring method for large-span steel structure construction, includes the following modules:

[0041] Data acquisition module: Used to deploy electromagnetic eddy current array sensors, gyroscopes, and photonic crystal strain gauges on the surface of the steel structure to collect the microscopic lattice distortion amount, low-frequency vibration phase difference, and optical path strain gradient in real time;

[0042] Holographic interference calculation module: Used to input the collected data into the quantum annealing algorithm calculation process, solve the deformation field in the structural space-time entanglement state, and construct a dynamic holographic twin model;

[0043] Intelligent prediction and decision-making module: Includes a quantum convolutional neural network, used to perform feature fusion and trend prediction on the deformation characteristics in the holographic twin model, and output the critical buckling load correction amount, welding residual stress compensation coefficient, and hoisting path curvature optimization function;

[0044] Quantum state regulation module: Used to dynamically adjust the control parameters of construction equipment according to the prediction results, including laser power adjustment, hoisting tackle configuration reconstruction, and prestress reverse application control;

[0045] State verification feedback module: used to verify the fidelity of the structural state after the execution of the construction regulation instruction through a quantum channel. If the verification fails to meet the standard, it triggers the re-acquisition of the full-link data and the regulation correction.

[0046] Advantages of the present invention:

[0047] In the present invention, a composite sensing network including an electromagnetic eddy current array sensor, a gyroscope, and a photonic crystal strain gauge is deployed on the steel structure surface to collect multi-source key data such as microscopic lattice distortion amount, low-frequency vibration phase difference, and optical path strain gradient, and a dynamic holographic twin model is synchronously constructed. The quantum annealing algorithm is used to solve the structural deformation field in the space-time entanglement state, realizing the accurate restoration of the deformation characteristics from the nanoscale to the structural scale, forming a virtual-real fusion model with high resolution and high dynamic responsiveness, and significantly improving the ability to identify instability signs such as structural buckling, vibration, and local strain in advance.

[0048] In the present invention, the data output by the dynamic holographic twin model is encoded with quantum characteristics and input into a multi-layer quantum-conventional hybrid convolutional network to realize the accurate prediction of the critical buckling load correction amount, the welding residual stress compensation coefficient, and the hoisting path curvature optimization function. The multi-dimensional strain states during the laser welding and hoisting processes are fused to construct an instruction set for key construction behaviors. On the basis of maintaining dynamic adjustability, the structural safety margin is verified through finite element simulation to ensure the precise control and risk avoidance capabilities of the construction operation in high-complexity scenarios.

[0049] In the present invention, the regulation parameters are injected into the construction equipment control system through the quantum state regulation module, and the laser power and the configuration of the hoisting tackle are adjusted in real time. At the same time, combined with the strain energy density criterion and the prestress redistribution mechanism, reverse stress is actively applied in high-risk areas to suppress the buckling expansion. The structural state fidelity verification of the dynamic compensation behavior is carried out through the quantum channel. Once it deviates from the expected state, the re-acquisition of the full data link of the sensor network is triggered, realizing the closed-loop adaptation between the regulation parameters and the structural response, and improving the stability and fault tolerance ability of the system under complex working conditions. Description of the drawings

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

[0051] Figure 1 It is a schematic flowchart of the method according to the embodiment of the present invention;

[0052] Figure 2Schematic diagram of the system process according to an embodiment of the present invention. Detailed implementation manners

[0053] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative ways to implement them; moreover, the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0054] It should be pointed out that in the specification, when referring to "an embodiment", "embodiment", "exemplary embodiment", "some embodiments", etc., it indicates that the described embodiment may include specific features, structures or characteristics, but not necessarily every embodiment includes the specific feature, structure or characteristic. In addition, when combining an embodiment to describe a specific feature, structure or characteristic, implementing such a feature, structure or characteristic in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0055] Generally, terms can be understood at least in part from their use in the context. For example, at least in part depending on the context, the term "one or more" used herein can be used to describe any feature, structure or characteristic in a singular sense, or can be used to describe a combination of features, structures or characteristics in a plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but rather, at least in part depending on the context, can allow for the existence of other factors that are not necessarily explicitly described.

[0056] As Figure 1 shown, an intelligent monitoring method for large-span steel structure construction includes the following steps:

[0057] S1: Deploy a sensing network on the surface of the steel structure. The sensing network includes an electromagnetic eddy current array sensor, a gyroscope, and a photonic crystal strain gauge, and synchronously collect key data during the construction of the steel structure. The key data includes the microscopic lattice distortion amount, the low-frequency vibration phase difference, and the optical path strain gradient;

[0058] S2: Input the collected key data into a holographic interference calculation module, solve the structural deformation field in the space-time entanglement state through a quantum annealing algorithm, and construct a dynamic holographic twin model. The dynamic holographic twin model includes a nanoscale lattice displacement cloud map, a vibration phase coherence map, and a non-uniform strain field vortex feature;

[0059] S3: Based on the dynamic holographic twin model, use a quantum convolutional neural network to predict the evolution trend of the structural form, and output a construction regulation instruction set. The construction regulation instruction set includes: the critical buckling load correction amount, the welding residual stress compensation coefficient, and the hoisting path curvature optimization function;

[0060] S4: Inject the construction control instruction set into the construction equipment control system in real time through the quantum state control module to drive the dynamic compensation behavior, where the dynamic compensation behavior includes:

[0061] implant strain field feedback in the laser welding robot to adjust the laser power in real time;

[0062] reconstruct the spatial configuration of the hoisting tackle based on the curvature optimization function;

[0063] trigger the prestress redistribution mechanism when the critical buckling load correction reaches 90% of the material yield strength.

[0064] S1 includes:

[0065] S11: Deploy electromagnetic eddy current array sensors on the surface of the steel structure, and collect the electromagnetic signals on the surface of the steel structure through the electromagnetic eddy current array sensors. Specifically, the electromagnetic eddy current array sensors are deployed according to the coordinates of the extreme points of the critical buckling mode, and the sensor node spacing d satisfies:

[0066]

[0067] where L is the span of the steel structure, σ s is the material yield strength, and E is the elastic modulus;

[0068] Collect electromagnetic signals based on the electromagnetic field induction principle, input the electromagnetic signals into the electromagnetic eddy current response model, and output the microscopic lattice distortion Δl i (i = 1, 2,..., N), and the model expression is:

[0069]

[0070] where ΔZ i is the change in sensor impedance, Z0 is the initial impedance, and K1 is the calibration coefficient;

[0071] S12: Deploy gyroscopes at key positions of the steel structure, and collect the angular velocity data of the steel structure through the gyroscopes. Specifically, the key positions are in the finite element modal analysis Perform double integral calculation on the collected angular velocity data ω j (t) (j = 1, 2,..., M) to obtain the vibration displacement u j (t):

[0072] u j (t) = ∫∫ω j (t)dt 2 ;

[0073] And extract the phase information φ j (t): φ j (t) = arg[uj (t);

[0074] Calculate the low-frequency vibration phase difference Δφ between adjacent nodes jk : Δφ jk = φ j (t) - φ k (t), filter out the phase difference components with frequency ≤ 10 Hz and store them as the low-frequency vibration phase difference tensor ΔΦ;

[0075] S13: Install a photonic crystal strain gauge on the surface of the steel structure, and collect the optical path data on the surface of the steel structure through the photonic crystal strain gauge, specifically including:

[0076] The lattice constant a of the photonic crystal strain gauge is 0.98a0 ± 0.02 nm, which matches the lattice constant a0 of the steel structure; calculate the Bragg wavelength shift Δλ k (k = 1, 2,..., K) through the optical path strain gradient calculation formula:

[0077] Obtain the optical path strain gradient where α = 2.35 × 10 3 is the sensitivity coefficient and λ0 is the initial wavelength;

[0078] S14: Synchronize the collected microscopic lattice distortion amount Δl i , low-frequency vibration phase difference Δφ jk and optical path strain gradient to generate a multi-source data set Specifically including:

[0079] Map Δl i to the global coordinate system according to the sensor coordinates (x i , y i , z i ) to generate a microscopic lattice distortion amount spatial matrix: ΔL = [Δl1, Δl2,..., Δl N T ;

[0080] Encode Δφ jk as the low-frequency vibration phase difference tensor ΔΦ and append the timestamp t n (n = 1, 2,..., T);

[0081] Interpolate to the same spatial grid as ΔL to generate an optical path strain gradient field

[0082] Verify that the spatial correlation coefficient between ΔL and is ≥ 0.85, and if not up to standard, trigger the recalibration of the electromagnetic eddy current array sensor.​

[0083] S2 includes:

[0084] S21: Input the multi-source data set into the holographic interference calculation module to construct the spatio-temporal entanglement state Hamiltonian model H(x, y, z, t). Among them, the expression of the Hamiltonian model H is:

[0085]

[0086] Among them, is the reduced Planck constant, m is the equivalent mass, and λ, μ are the cross-scale coupling coefficients;

[0087] S22: Solve the Hamiltonian model H(x, y, z, t) through the quantum annealing algorithm, specifically including:

[0088] Discretize H into the Ising model that can be encoded by qubits:

[0089]

[0090] Among them, J ij is determined by the spatial gradient correlation matrix of ΔL and and h i is determined by the phase coherence of ΔΦ;

[0091] S23: Extract the probability amplitude distribution ψ(x, y, z, t) output by the quantum annealing to generate the structural deformation field, specifically including:

[0092] Perform a Fourier-Bessel transform on ψ to obtain the nanoscale lattice displacement cloud map U(x, y, z) = Re[ψ]·δ, where δ = 0.1 nm is the displacement resolution;

[0093] Calculate the vibration phase coherence map C φ , Among them represents the tensor convolution;

[0094] Set the annealing time T a = 20 μs and the temperature T = 10 mK in the quantum annealing processor, and perform the annealing operation until the ground state energy difference ΔE < 1×10 -5 eV;

[0095] S2 also includes:

[0096] S24: Extract the vortex characteristics of the non-uniform strain field based on the optical path strain gradient field Specifically including:

[0097] Calculate the vortex core position

[0098] When the vorticity tensor norm is reached, it is marked as a valid vortex;

[0099] S25: Fuse the nanoscale lattice displacement cloud map U, the vibration phase coherence map C φ and the vortex characteristics of the non-uniform strain field to construct a dynamic holographic twin model, satisfying:

[0100] The resolution of the nanoscale lattice displacement cloud map ≤ 0.1 nm;

[0101] The frequency domain sampling interval of the vibration phase coherence map ≤ 0.1 Hz;

[0102] The vortex characteristic positioning error ≤ 3 mm.

[0103] S3 includes:

[0104] S31: Input the nanoscale lattice displacement cloud map U(x, y, z), the vibration phase coherence map C φ and the vortex characteristics of the non-uniform strain field in the dynamic holographic twin model into the quantum convolutional neural network for feature encoding and fusion:

[0105] The quantum convolutional neural network contains 5 quantum convolutional layers and 3 traditional convolutional layers. The quantum convolutional layer uses the qubit encoding method to map U(x, y, z) to the quantum state |ψ U > = ∑ i U i |i>, where |i> is the position ground state;

[0106] Concatenate C φ with the vortex core position (x v , y v , z v ) to generate a fused feature tensor and input it into the traditional convolutional layer;

[0107] The loss function for training the quantum convolutional neural network is where α = 0.7, β = 0.3, and the training period ≥ 1000 epochs;

[0108] S32: Predict the critical buckling load correction ΔP c through the quantum convolutional neural network, specifically including:

[0109] Perform a quantum measurement on the quantum state |ψ U > to obtain the displacement gradient probability distribution

[0110] Calculate the critical buckling load correction based on the peak of the probability distribution:

[0111]

[0112] Among which, K p = 1.5×10 3 N / mm 2 is the load sensitivity coefficient;

[0113] S33: Calculate the welding residual stress compensation coefficient K based on the optical path strain gradient field Specifically, it includes: w Extract the maximum optical path strain gradient on the welding path

[0114] Extract the maximum optical path strain gradient on the welding path

[0115] Calculate the compensation coefficient K w :

[0116] Among which is the initial strain gradient, is the threshold;

[0117] S34: Generate the hoisting path curvature optimization function f(κ) according to the vortex characteristics of the non-uniform strain field. Specifically, it includes:

[0118] Taking the vortex core position (x v , y v , z v ) as the control point, fitting the B-spline curve path, and the curvature κ satisfies:

[0119]

[0120] Among which ∈ = 1×10 -7 / m 2 is a small constant to prevent division by zero;

[0121] S35: Integrate the critical buckling load correction ΔP c , the welding residual stress compensation coefficient K w and the hoisting path curvature optimization function f(κ) to generate the construction control instruction set.

[0122] When ΔP c ≥0.9σ s , trigger an alarm and freeze the instruction output;

[0123] Verify the coupling effect of K w and f(κ), and confirm that the structural safety margin ≥1.8 through finite element simulation.

[0124] S4 includes:

[0125] S41: Through the quantum state control module, the welding residual stress compensation coefficient K wThe injection laser welding robot control system adjusts the laser power P(t) in real time, specifically including:

[0126] According to the optical path strain gradient field Calculate the local strain gradient amplitude

[0127]

[0128] Dynamically adjust the laser power: Where P0 is the reference power and the adjustment frequency ≥ 100Hz;

[0129] Verify the welding molten pool temperature T w Satisfy T w ≤0.6T m (T m is the melting point of the material), otherwise trigger K w The attenuation coefficient η = 0.8 of is used for power suppression;

[0130] S42: Based on the hoisting path curvature optimization function f(κ), reconstruct the spatial configuration of the hoisting rigging, specifically including:

[0131] Analyze the curvature component κ in f(κ) v , generate the hoisting path control point sequence {(x v ,y v ,z v ,κ v )};

[0132] Drive the hydraulic servo system to adjust the length L of the hoisting rigging v and the elevation angle θ v :

[0133]

[0134] Where d v is the adjacent vortex core spacing;

[0135] Verify the rigging tension F v Satisfy F v ≤0.8F ult (F ult is the ultimate tensile force of the rigging), otherwise trigger path curvature smoothing filtering; that is, when the rigging tension exceeds the safety threshold during hoisting, smooth the curvature component κ v generated in the hoisting path curvature optimization function, so as to adjust the spatial change rate of the hoisting path control points and reduce the risk of abnormal tension caused by excessive local curvature of the hoisting system.

[0136] S43: When the critical buckling load correction ΔP c reaches the material yield strength σs When it reaches 90% of

[0137] Apply a reverse stress σ in the buckling risk area through a shape memory alloy actuator r =-0.3ΔP c ;

[0138] Verify the strain energy density based on the nanoscale lattice displacement cloud map U(x,y,z) Ensure that W≤0.5W yield (W yield is the yield strain energy);

[0139] S44: Verify the structural state after dynamic compensation behavior through a quantum channel, specifically including:

[0140] Encode the laser power P(t), the rigging configuration parameters (L v ,θ v ) and the reverse stress σ r into the quantum state |Ψ ctrl >;

[0141] Perform a Bell basis measurement with the dynamic holographic twin model to verify the quantum state fidelity If Then trigger the sensing network to collect key data.

[0142] As Figure 2 shown, an intelligent monitoring system for long-span steel structure construction, used to implement the above intelligent monitoring method for long-span steel structure construction, includes the following modules:

[0143] Data acquisition module: Used to deploy electromagnetic eddy current array sensors, gyroscopes and photonic crystal strain gauges on the steel structure surface to collect the microscopic lattice distortion amount, low-frequency vibration phase difference and optical path strain gradient in real time;

[0144] Holographic interference calculation module: Used to input the collected data into the quantum annealing algorithm calculation process, solve the deformation field in the structural space-time entanglement state, and construct a dynamic holographic twin model;

[0145] Intelligent prediction and decision-making module: Includes a quantum convolutional neural network, used to perform feature fusion and trend prediction on the deformation characteristics in the holographic twin model, and output the critical buckling load correction amount, welding residual stress compensation coefficient and hoisting path curvature optimization function;

[0146] Quantum state regulation module: Used to dynamically adjust the construction equipment control parameters according to the prediction results, including laser power adjustment, rigging configuration reconstruction and prestress reverse application control;

[0147] Status verification feedback module: used to verify the fidelity of the structural status after the execution of the construction control instruction through a quantum channel. If the verification fails to meet the standard, it triggers the re - acquisition of all - link data and the adjustment correction.

[0148] This invention covers any alternatives, modifications, equivalent methods and solutions made on the essence and scope of this invention. To enable the public to have a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments of this invention. However, those skilled in the art can fully understand this invention even without the description of these details. In addition, well - known methods, processes, procedures, components and circuits, etc. are not described in detail to avoid unnecessary confusion to the essence of this invention.

[0149] The above - mentioned are only the preferred embodiments of this invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of this invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this invention.

Claims

1. An intelligent monitoring method for the construction of long-span steel structures, characterized in that It includes the following steps: S1: Deploy a sensing network on the steel structure surface. The sensing network includes an electromagnetic eddy current array sensor, a gyroscope, and a photonic crystal strain gauge, and synchronously collect key data during the construction of the steel structure. The key data includes the microscopic lattice distortion amount, the low-frequency vibration phase difference, and the optical path strain gradient; S2: Input the collected key data into the holographic interference calculation module, solve the structural deformation field in the space-time entanglement state through the quantum annealing algorithm, and construct a dynamic holographic twin model. The dynamic holographic twin model includes a nanoscale lattice displacement cloud map, a vibration phase coherence map, and a vortex feature of the non-uniform strain field; S3: Based on the dynamic holographic twin model, use a quantum convolutional neural network to predict the structural form evolution trend and output a construction control instruction set. The construction control instruction set includes: the critical buckling load correction amount, the welding residual stress compensation coefficient, and the hoisting path curvature optimization function; S4: Inject the construction control instruction set into the construction equipment control system in real time through the quantum state control module to drive dynamic compensation behaviors. The dynamic compensation behaviors include: Install a strain field feedback in the laser welding robot to adjust the laser power in real time; Reconstruct the spatial configuration of the hoisting rigging based on the curvature optimization function; Trigger the prestress redistribution mechanism when the critical buckling load correction amount reaches 90% of the material yield strength.

2. The intelligent monitoring method for long-span steel structure construction according to claim 1, wherein, The S1 includes: S11: Deploy an electromagnetic eddy current array sensor on the steel structure surface, collect the electromagnetic signals on the steel structure surface through the electromagnetic eddy current array sensor, and based on the collected electromagnetic signals on the steel structure surface, use the electromagnetic eddy current response model to convert the electromagnetic signals into the microscopic lattice distortion amount; S12: Deploy a gyroscope at the key positions of the steel structure, collect the angular velocity data of the steel structure through the gyroscope, based on the collected angular velocity data, calculate the vibration displacement and phase information of each position point of the steel structure through the integral calculation method, and obtain the low-frequency vibration phase difference by calculating the phase difference between different position points; S13: Install a photonic crystal strain gauge on the steel structure surface, collect the optical path data on the steel structure surface through the photonic crystal strain gauge, based on the collected optical path data, calculate the optical path change amount by comparing the optical path differences at different positions, based on the optical path change amount, apply the optical path strain gradient calculation method to obtain the optical path strain gradient on the steel structure surface, and generate an optical path strain gradient field based on the optical path strain gradient; S14: Synchronously process the collected microscopic lattice distortion amount, low-frequency vibration phase difference, and optical path strain gradient through the synchronous data acquisition unit to generate a multi-source data set.

3. The intelligent monitoring method for large-span steel structure construction according to claim 2, characterized in that, The S2 includes: S21: Input the multi-source data set into the holographic interference calculation module to construct a Hamiltonian model in the space-time entanglement state; S22: Use the quantum annealing algorithm to solve the Hamiltonian model, including discretizing the Hamiltonian model into an Ising model that can be encoded by qubits, and obtaining the ground state energy and the distribution of related physical quantities by solving the Ising model; S23: Extract the probability amplitude distribution from the output of the quantum annealing process to generate a structural deformation field, including performing a Fourier-Bessel transform on the output probability amplitude to obtain a nanoscale lattice displacement cloud map, and analyzing the relationship between the vibration mode and the phase difference by calculating the coherence map of the vibration phase.

4. The intelligent monitoring method for large-span steel structure construction according to claim 3, characterized in that, In the quantum annealing process, set the annealing time and temperature, and search for the lowest energy state through the annealing operation until the required accuracy is achieved.

5. The intelligent monitoring method for large-span steel structure construction according to claim 4, characterized in that The S2 further includes: S24: Based on the optical path strain gradient field, extract the vortex characteristics in the non-uniform strain field, including locating the core position of the vortex and judging the effectiveness of the vortex by analyzing the rotation behavior of the strain field; S25: Integrate the obtained lattice displacement cloud map, vibration phase coherence map, and non-uniform strain field vortex characteristics to construct a dynamic holographic twin model.

6. The intelligent monitoring method for long-span steel structure construction according to claim 5, characterized in that, The S3 includes: S31: Input the nanoscale lattice displacement cloud map, vibration phase coherence map, and non-uniform strain field vortex characteristics in the dynamic holographic twin model into the quantum convolutional neural network for feature encoding and fusion; S32: Predict the critical buckling load correction amount through the quantum convolutional neural network, including performing a quantum measurement on the quantum state to obtain the probability distribution of the displacement gradient, and calculating the critical buckling load correction amount by analyzing the peak of the probability distribution; S33: Calculate the welding residual stress compensation coefficient based on the optical path strain gradient field; S34: After calculating the welding residual stress compensation coefficient, use the non-uniform strain field vortex characteristics to generate a hoisting path curvature optimization function; S35: Integrate the critical buckling load correction amount, welding residual stress compensation coefficient, and hoisting path curvature optimization function to generate a construction control instruction set.

7. The intelligent monitoring method for large-span steel structure construction according to claim 6, wherein In S35, if the critical buckling load correction amount reaches a certain threshold, trigger an alarm and freeze the construction instruction, and verify the coupling effect between the residual stress compensation coefficient and the hoisting path curvature optimization function, and confirm that the structural safety margin meets the specified requirements through finite element simulation.

8. The intelligent monitoring method for large-span steel structure construction according to claim 7, characterized in that, The S4 includes: S41: Inject the welding residual stress compensation coefficient into the laser welding robot control system, and adjust the laser power in real time through the quantum state control module; S42: Based on the curvature optimization function of the hoisting path, reconstruct the spatial configuration of the hoisting rigging, including generating a control point sequence of the hoisting path by analyzing the curvature components in the curvature optimization function, and the hydraulic servo system adjusts the length and angle of the hoisting rigging according to the control point sequence; S43: If the critical buckling load correction amount reaches 90% of the material yield strength, trigger the prestress redistribution mechanism, including: In the buckling risk area, apply a reverse stress through the shape memory alloy actuator to adjust the stress distribution and reduce the local buckling risk; Calculate the strain energy density by verifying the nanoscale lattice displacement cloud map. S44: Verify the structural state after dynamic compensation behavior through a quantum channel, including encoding the laser power, rigging configuration, and reverse stress into quantum states, comparing the quantum states with the dynamic holographic twin model, verifying the fidelity of the quantum states, and if the fidelity does not meet the preset standard, triggering the sensing network to re-collect key data to ensure that all key data is monitored and adjusted in real time.

9. The intelligent monitoring method for large-span steel structure construction according to claim 8, characterized in that, During the process of the hydraulic servo system in S42 adjusting the length and angle of the lifting rigging according to the control point sequence, the tension of the rigging is monitored in real time. If the tension exceeds the safe range, smooth filtering of the path curvature is triggered.

10. An intelligent monitoring system for the construction of large-span steel structures, which is used to implement the intelligent monitoring method for the construction of large-span steel structures according to any one of claims 1-9, characterized in that, It includes the following modules: Data acquisition module: Used to deploy electromagnetic eddy current array sensors, gyroscopes, and photonic crystal strain gauges on the steel structure surface to collect the microscopic lattice distortion amount, low-frequency vibration phase difference, and optical path strain gradient in real time. Holographic interference calculation module: Used to input the collected data into the quantum annealing algorithm calculation process, solve the deformation field in the structural space-time entanglement state, and construct a dynamic holographic twin model. Intelligent prediction and decision-making module: Includes a quantum convolutional neural network, which is used to perform feature fusion and trend prediction on the deformation characteristics in the holographic twin model, and output the critical buckling load correction amount, welding residual stress compensation coefficient, and optimized function of the lifting path curvature. Quantum state regulation module: Used to dynamically adjust the control parameters of construction equipment according to the prediction results, including laser power adjustment, rigging configuration reconstruction, and prestress reverse application control. State verification and feedback module: Used to verify the fidelity of the structural state after the execution of the construction regulation instruction through a quantum channel. If the verification fails, trigger the re-collection and regulation correction of the full-link data.

Images