Water pile foundation construction parameter self-adaptive adjusting system based on multi-sensor fusion

Abstract

The present application relates to the technical field of pile foundation engineering, and discloses a water pile foundation construction parameter self-adaptive adjustment system based on multi-sensor fusion, which comprises a hybrid sensor and excitation array module, a geomechanics parameter inversion module, a physical model self-correction module, a model predictive control module and a construction parameter execution module. The hybrid sensor and excitation array module is used for emitting coded acoustic wave signals to the geological body below the pile foundation. The geomechanics parameter inversion module is used for being based on collected echo data and a corrected pile-soil coupling system physical model. The physical model self-correction module is used for being based on collected pile body dynamic response data and inverted geomechanics parameters. The model predictive control module is used for being based on the output pile front geomechanics parameters. The construction parameter execution module is used for receiving and executing the current optimal construction control instruction in the optimal construction control parameter sequence. Through bidirectional collaborative closed loop of geological inversion and model self-correction, the accuracy of geological perception and model benchmark is improved, the model predictive control is used to realize the predictability adjustment of the front geological disturbance, and the problems of control blindness and hysteresis are solved.

Description

Technical Field

[0001] This invention relates to the field of pile foundation engineering technology, specifically to an adaptive adjustment system for underwater pile foundation construction parameters based on multi-sensor fusion. Background Technology

[0002] Offshore pile foundation construction is a critical geotechnical engineering step, such as in the construction of large-scale infrastructure projects like cross-sea bridges and offshore wind farms. The complex offshore construction environment, especially the complex, variable, and unseen underwater geological conditions beneath the pile foundations, poses a significant challenge to the precise control of construction parameters.

[0003] Current pile foundation construction processes largely rely on geological survey reports obtained before construction and the on-site operational experience of construction personnel. However, the data provided in geological survey reports is usually sparse and low-precision, failing to accurately reveal potential localized and sudden geological anomalies along the pile penetration path, such as isolated boulders, karst caves, or weak interlayers. This "blindness" regarding the geological conditions ahead leads to significant lags in construction control. When the pile foundation encounters abnormal geological bodies, the control system or operators often only passively adjust construction parameters after detecting delayed feedback signals such as a surge in hammer blows, a spike in drill torque, or a halt in penetration. This "reactive" control strategy not only results in low construction efficiency and high equipment energy consumption but also easily leads to engineering safety hazards such as pile deviation, tilting, and even structural damage.

[0004] To improve automation levels, some technologies attempt to introduce model-based control methods. However, the pile-soil coupling physical models upon which these systems rely are typically simplified or static. In complex actual construction processes, key model parameters such as the side friction characteristics of the pile-soil interface and the system's equivalent damping dynamically change with increasing penetration depth and soil disturbance. Existing technologies lack a mechanism to utilize real-time dynamic response data generated during construction to calibrate and correct the physical model itself, leading to a gradual disconnect between the numerical model and physical reality. This severely limits the accuracy and robustness of control strategies, preventing the achievement of truly adaptive optimization control. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an adaptive adjustment system for construction parameters of underwater pile foundations based on multi-sensor fusion. This system solves the problems of lagging adjustment of construction parameters and low control accuracy caused by the lack of predictability of the geological conditions ahead and the inability to self-correct its physical model in real time in existing pile foundation construction control.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an adaptive adjustment system for underwater pile foundation construction parameters based on multi-sensor fusion, comprising;

[0007] The hybrid sensing and excitation array module is used to transmit coded acoustic signals to the geological body below the pile foundation and to collect the echo data of the acoustic signals and the dynamic response data of the pile body generated during the construction process.

[0008] The geomechanical parameter inversion module, which is connected to the hybrid sensing and excitation array module, is used to invert the geomechanical parameters in front of the pile based on the acquired echo data and a calibrated pile-soil coupling system physical model.

[0009] The physical model self-calibration module, which is connected to the hybrid sensing and excitation array module and the geomechanical parameter inversion module, is used to self-calibrate the internal pile-soil coupling system physical model based on the collected pile dynamic response data and the inverted geomechanical parameters, and provide the calibrated physical model to the geomechanical parameter inversion module.

[0010] The model prediction and control module, which is connected to the geomechanical parameter inversion module, is used to continuously optimize and solve for the optimal construction control parameter sequence within a finite time window based on the output geomechanical parameters before the pile.

[0011] The construction parameter execution module, which is connected to the model prediction and control module, is used to receive and execute the current optimal construction control command in the optimal construction control parameter sequence.

[0012] Preferably, the hybrid sensing and excitation array module includes:

[0013] An active excitation unit installed at the end of the pile foundation structure is used to transmit the coded acoustic signal;

[0014] Distributed passive sensing units are laid along the pile body to collect echo data and pile dynamic response data.

[0015] Preferably, the distributed passive sensing unit is composed of a multi-channel fiber Bragg grating sensor array;

[0016] The echo data is the weak reflection signal generated by the acoustic pulse emitted by the active excitation unit on the geological discontinuity surface;

[0017] The dynamic response data of the pile body is the forced vibration signal generated under the action of hammering or drilling construction load.

[0018] Preferably, the geomechanical parameter inversion module contains a wave equation forward modeling operator. ;

[0019] The geomechanical parameter inversion module solves an objective function. The inversion is achieved by minimizing the objective function, which quantifies the mismatch between the observed echo data and the theoretical echo data calculated based on the wave equation forward modeling operator, the corrected pile-soil coupled system physical model, and the geomechanical parameters to be inverted.

[0020] The objective function It can be represented as:

[0021] :

[0022] In the formula,

[0023] Let be the objective function to be minimized;

[0024] The vector represents the geomechanical parameters to be solved.

[0025] For the received observation echo data;

[0026] In order to obtain the geological parameters and the provided calibrated model parameters Below is the theoretical echo data calculated using forward modeling;

[0027] The square of the L2 norm is used to quantify the degree of mismatch between theoretical and observed echoes, forming a data mismatch term.

[0028] is the regularization coefficient, a positive scalar used to balance the weights of data mismatch terms and regularization terms;

[0029] This is a regularization term.

[0030] Preferably, the physical model self-calibration module has a forced vibration response operator embedded within it. ;

[0031] The physical model self-calibration module solves the self-calibration objective function. The self-calibration is achieved by minimizing the problem. The self-calibration objective function is used to quantify the degree of mismatch between the measured dynamic response data of the pile body and the theoretical forced vibration response calculated based on the forced vibration response operator, the geomechanical parameters obtained by inversion, and the physical model parameters of the system to be calibrated.

[0032] Furthermore, the self-calibration objective function It can be represented as:

[0033] :

[0034] In the formula, Let be the self-calibration objective function to be minimized. The parameter vector of the physical model of the system to be solved. The measured forced vibration response data received from the hybrid sensing and excitation array module, To use the known geological parameters provided by the geomechanical parameter inversion module Model parameters to be determined and known construction inputs Below, the theoretical forced vibration response calculated by forward modeling, The regularization coefficient is . This is a regularization term for the model parameters, used to ensure the physical rationality of parameter adjustments.

[0035] Preferably, the physical model parameters of the system to be corrected include the side friction coefficient of the pile-soil interface, the equivalent material damping coefficient of the system, or the equivalent spring stiffness of the boundary conditions of the model calculation domain.

[0036] Preferably, the model prediction and control module has a discrete-time state-space prediction model for the pile foundation construction process.

[0037] The model prediction control module is used to convert the front-piling geomechanical parameters output by the geomechanical parameter inversion module into the external disturbance sequence in the state space prediction model for the next N time steps.

[0038] Preferably, the model predictive control module, in each control cycle, plans the optimal construction control parameter sequence by solving a constrained finite-time domain optimization problem based on the generated future external disturbance sequence and the current system state.

[0039] Preferably, the constraints of the constrained finite-time optimization problem include: state constraints for ensuring system safety, such as the maximum allowable stress or maximum allowable tilt angle of the pile body; and control constraints for limiting the physical capabilities of the construction equipment.

[0040] The model prediction and control module has a discrete-time state-space prediction model for the pile foundation construction process. The prediction model can be expressed as follows:

[0041] :

[0042] In the formula, For discrete time steps, for The system state vector at time t. for The control vector applied at any given time may include the single hammer energy during hammering operations, the drill pressure during drilling operations, and the torque of the power head. for The external disturbance vector encountered by the time-based system mainly characterizes the hindering effect of geological conditions on construction. This is the state transition function.

[0043] Preferably, the construction parameter execution module includes:

[0044] The communication interface unit is used to receive the current optimal construction control command via an industrial fieldbus;

[0045] The execution drive unit converts the digitized current optimal construction control commands into physical drive signals applied to the proportional servo valve of the hydraulic hammer or the motor inverter of the drilling rig power head.

[0046] This invention provides an adaptive adjustment system for underwater pile foundation construction parameters based on multi-sensor fusion. It has the following beneficial effects:

[0047] 1. This invention constructs a cognitive closed loop by setting up a geomechanical parameter inversion module and a physical model self-calibration module, and enabling them to work collaboratively. The physical model self-calibration module uses forced vibration response data to calibrate the physical model of the pile-soil coupling system and provides the calibrated physical model to the geomechanical parameter inversion module; the geomechanical parameter inversion module then inverts geological parameters based on the calibrated physical model and provides the inversion results to the physical model self-calibration module as boundary conditions. This bidirectional enhancement mechanism significantly improves the joint estimation accuracy of physical model parameters and geomechanical parameters.

[0048] 2. This invention, through its model prediction control module, can utilize the geomechanical parameters output by the geomechanical parameter inversion module to transform them into a sequence of external disturbances within a finite future time window in the state-space prediction model. Based on the prediction of future geological disturbances, the model prediction control module can continuously optimize and solve for the optimal construction control parameters, realizing a shift from passive feedback control to proactive predictive control. This allows for the early avoidance of the impact of severe geological conditions, improving the safety and efficiency of construction.

[0049] 3. This invention employs a hybrid sensing and excitation array module, which, through the dual data acquisition function of its distributed passive sensing unit, achieves optimized matching of data sources. It uses high signal-to-noise ratio forced vibration response data to correct the physical model and uses high-sensitivity weak echo data to invert geological parameters. By using data with different characteristics for the design of different optimization objectives, it ensures the robustness of model correction and the accuracy of geological inversion, thereby improving the reliability of the data-driven foundation of the entire system. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the system framework of the present invention;

[0051] Figure 2 This is a schematic diagram of the hybrid sensing and excitation array module architecture of the present invention;

[0052] Figure 3 This is a schematic diagram of the internal workflow of the geomechanical parameter inversion module of the present invention;

[0053] Figure 4 This is a schematic diagram of the internal workflow of the physical model self-calibration module of the present invention;

[0054] Figure 5 This is a schematic diagram of the internal unit architecture of the construction parameter execution module of the present invention. Detailed Implementation

[0055] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Please see the appendix Figure 1 -Appendix Figure 5 This invention provides an adaptive adjustment system for underwater pile foundation construction parameters based on multi-sensor fusion, comprising:

[0057] The hybrid sensing and excitation array module is used to transmit coded acoustic signals to the geological body below the pile foundation and to collect the echo data of the acoustic signals and the dynamic response data of the pile body generated during the construction process.

[0058] Specifically, the hybrid sensing and excitation array module in this embodiment functions to actively transmit detection signals and acquire multi-physics response data of the pile-soil system. The hybrid sensing and excitation array module specifically includes an active excitation unit, a distributed passive sensing unit, and an attitude and position sensing unit.

[0059] The active excitation unit is responsible for emitting controllable, specifically encoded acoustic pulses into the geological medium beneath the pile foundation structure. In one specific embodiment, the active excitation unit consists of one or more sets of piezoelectric transducers, preferably broadband types, integrated in a ring array at the end of the pile foundation structure, such as above the pile shoe or near the drill bit of the drill rod. The active excitation unit is electrically connected to a programmable high-power signal generator, which generates encoded waveforms with good autocorrelation characteristics, such as linear frequency modulation (LFM) signals. It can be represented as;

[0060] ;

[0061] In the formula, For instantaneous signal transmission, For time, The signal amplitude, The duration of the pulse. For rectangular window functions, The starting frequency of the signal. By using linear frequency modulation and transmitting signals with specific codes, it helps to improve the signal-to-noise ratio and range resolution in subsequent signal processing through pulse compression techniques.

[0062] The distributed passive sensing unit is responsible for collecting the dynamic response of the pile structure under different working conditions. In one specific embodiment, the distributed passive sensing unit consists of a multi-channel fiber Bragg grating (FBG) sensor array laid spirally or longitudinally along the entire length of the pile at preset intervals. The FBG sensor array is connected to a high-speed fiber Bragg grating demodulator via an optical fiber link. The demodulator is used to demodulate the wavelength drift signal fed back by the FBG sensor in real time and convert it into high-fidelity strain or vibration data.

[0063] The distributed passive sensing unit has dual data acquisition capabilities:

[0064] Weak echo signal acquisition:

[0065] This is used to capture, with high sensitivity, the reflected or scattered echo signals generated by the acoustic pulses emitted by the active excitation unit at geological discontinuities (such as rock strata or weak interlayers) in front of the pile. The echo signals are the main data input for the geomechanical parameter inversion module to perform inversion calculations.

[0066] Forced vibration response monitoring:

[0067] This data is used to record, with high fidelity and in real-time, strain wave propagation and structural vibration response at various measuring points on the pile body under heavy loads during hammering or drilling operations. The forced vibration response data is the primary data input for the physical model self-calibration module to correct model parameters.

[0068] The attitude and position sensing unit is responsible for providing real-time spatial attitude information of the pile foundation structure during construction. In one specific embodiment, the attitude and position sensing unit includes a Global Navigation Satellite System Real-Time Measurement (GNSS-RTK) receiver and a high-precision dual-axis inclinometer installed on the pile top or construction platform. The GNSS-RTK receiver is used to acquire the three-dimensional coordinates of the pile top in the global geodetic coordinate system, and the dual-axis inclinometer is used to measure the pitch and roll angles of the pile body relative to the vertical line of gravity.

[0069] The geomechanical parameter inversion module, which is connected to the hybrid sensing and excitation array module, is used to invert the geomechanical parameters in front of the pile based on the acquired echo data and a calibrated pile-soil coupling system physical model.

[0070] Specifically, in this embodiment, the geomechanical parameter inversion module consists of one or more industrial-grade servers equipped with high-performance graphics processing units (GPUs), which are used to accelerate the numerical inversion calculations described later.

[0071] The geomechanical parameter inversion module is used to receive weak echo signals collected by the distributed passive sensing unit of the hybrid sensing and excitation array module, and the signals are denoted as observation echo data. The geomechanical parameter inversion module is also used to receive the corrected physical model parameters of the pile-soil coupling system from the physical model self-calibration module; these parameters are denoted as the corrected model parameters. .

[0072] The function of the geomechanical parameter inversion module is to, based on the observed echo data and the corrected model parameters By solving a nonlinear optimization problem, the geomechanical parameters of the unconstructed area in front of the pile foundation structure can be obtained.

[0073] Specifically, the geomechanical parameter inversion module contains a forward modeling operator for the wave equation of a pile-soil coupled system, denoted as... The forward operand operator This is a numerical implementation of the finite-difference time-domain (FDTD) or finite-element method (FEM) of the elastic dynamics wave equations of the pile-soil coupled system (as described above). The forward modeling operator... Able to invert based on a given set of geomechanical parameters and a given set of system physical model parameters The theoretical echo response was calculated using forward simulation. .

[0074] The geomechanical parameters to be inverted It is a vector that contains the geological medium in front of the pile at a series of discrete depth nodes. Physical properties, such as elastic modulus ,density Compared to Poisson .

[0075] The geomechanical parameter inversion module solves an objective function. The inversion is achieved by minimizing the objective function. In one specific embodiment, the objective function is... The structure is as follows:

[0076] ;

[0077] In the formula,

[0078] Let be the objective function to be minimized;

[0079] The vector represents the geomechanical parameters to be solved.

[0080] For the received observation echo data;

[0081] In order to obtain the geological parameters and the provided calibrated model parameters Below is the theoretical echo data calculated using forward modeling;

[0082] The square of the L2 norm is used to quantify the degree of mismatch between theoretical and observed echoes, forming a data mismatch term.

[0083] is the regularization coefficient, a positive scalar used to balance the weights of data mismatch terms and regularization terms;

[0084] This is a regularization term.

[0085] The regularization term This is used to introduce prior geological information (e.g., stratigraphic structures are typically layered or smoothly varied) into the inversion problem to overcome the ill-conditioned nature of the inverse problem and ensure the stability and physical plausibility of the solution. In different embodiments, the regularization term can employ Tikhonov regularization (i.e., L2 norm) or Total Variation (TV) regularization (i.e. (e.g., L1 norm of gradient).

[0086] The geomechanical parameter inversion module solves the objective function based on an iterative optimization algorithm using gradients. The minimum value of the objective function. To efficiently calculate the objective function. Relative to the geomechanical parameters to be inverted The gradients of each component are preferably calculated using the adjoint-state method, which significantly reduces the computational complexity of gradient solving. After the iterative optimization algorithm converges, the optimal geomechanical parameter vector is denoted as... .

[0087] The physical model self-calibration module, which is connected to the hybrid sensing and excitation array module and the geomechanical parameter inversion module, is used to self-calibrate the internal pile-soil coupling system physical model based on the collected dynamic response data and the inverted geomechanical parameters, and provide the calibrated physical model to the geomechanical parameter inversion module.

[0088] Specifically, in this embodiment, the physical model self-calibration module is used to receive high signal-to-noise ratio dynamic response data of the pile body collected by the distributed passive sensing units of the hybrid sensing and excitation array module during construction (e.g., a single hammer blow or a drilling stroke). This data is denoted as forced vibration response data. The physical model self-calibration module is also used to receive the inverted geomechanical parameters from the geomechanical parameter inversion module. .

[0089] The core function of the physical model self-calibration module is to utilize high-energy forced vibration response data. The equivalent physical parameters of the internally solidified pile-soil coupling system physical model are corrected in reverse. The physical model parameters are the benchmark for the inversion calculation of the geomechanical parameter inversion module, and their accuracy directly determines the reliability of the inversion results.

[0090] Specifically, the physical model self-calibration module contains a forced vibration response operator. Forced vibration response operator This is a numerical simulation of the dynamic response of a pile-soil coupled system under known construction loads (such as hammering force), using a forced vibration response operator. Able to base on a given set of geomechanical parameters Given the physical model parameters of the system to be corrected And a known construction input The theoretical forced vibration response of the pile was calculated using forward simulation. .

[0091] System physical model parameters to be corrected It is a vector that contains equivalent parameters that are difficult to pre-determine theoretically in FDTD or FEM models, such as the side friction coefficient of the pile-soil interface, the equivalent material damping coefficient of the system, and the equivalent spring stiffness of the boundary conditions of the model's computational domain.

[0092] The physical model self-calibration module solves an objective function. The self-calibration of model parameters is achieved by minimizing the objective function. In one specific embodiment, the objective function is... The structure is as follows:

[0093] :

[0094] In the formula, Let be the self-calibration objective function to be minimized. The parameter vector of the physical model of the system to be solved. The measured forced vibration response data received from the hybrid sensing and excitation array module, To use the known geological parameters provided by the geomechanical parameter inversion module Model parameters to be determined and known construction inputs Below, the theoretical forced vibration response calculated by forward modeling, The regularization coefficient is . This is a regularization term for the model parameters, used to ensure the physical rationality of parameter adjustments.

[0095] The physical model self-calibration module uses gradient-based optimization algorithms (such as L-BFGS) to solve the objective function. By finding the minimum value, a set of optimal, corrected system physical model parameters is obtained. .

[0096] The physical model self-calibration module has a data output function, used to output the calibrated physical model parameters of the pile-soil coupling system. Feedback is output to the geomechanical parameter inversion module.

[0097] The physical model self-calibration module outputs the calibrated model parameters. Provided to the geomechanical parameter inversion module, the geomechanical parameter inversion module in its objective function Use As a known parameter (i.e.) ), a more accurate Significantly improved the forward modeling operator The accuracy makes it possible to base on weak echo signals The inversion calculation benchmark is closer to physical reality, thus significantly improving its inversion results. The accuracy and reliability of [the system / mechanism].

[0098] The geomechanical parameter inversion module outputs improved accuracy geomechanical parameters from the pile front. Provided to the physics model self-calibration module. The physics model self-calibration module in its objective function Use As a known parameter (i.e.) ), a more accurate For forced vibration response operator It provides more precise external boundary conditions. This makes it possible to utilize measured data. For model parameters When performing inverse correction, the residual between the theoretical and measured responses can be attributed more accurately to... Inaccuracy, not The uncertainty of the correction results More precise.

[0099] Through information interaction and iterative enhancement between the geomechanical parameter inversion module and the physical model self-calibration module, this invention achieves model ( ) and data ( The dual optimization of ( ) forms an adaptive closed loop in which cognitive accuracy continuously improves itself.

[0100] The model prediction and control module, which is connected to the geomechanical parameter inversion module, is used to continuously optimize and solve for the optimal construction control parameter sequence within a finite time window based on the geomechanical parameters it outputs.

[0101] Specifically, in this embodiment, the model prediction and control module establishes a discrete-time state-space prediction model for the pile foundation construction process. This model describes the dynamic behavior of the pile foundation system under construction control and external geological disturbances, and its mathematical form can be expressed as:

[0102] :

[0103] In the formula, For discrete time steps, for The system state vector at time t. for The control vector applied at any given time may include the single hammer energy during hammering operations, the drill pressure during drilling operations, and the torque of the power head. for The external disturbance vector encountered by the time-based system mainly characterizes the hindering effect of geological conditions on construction. This is the state transition function.

[0104] The function of the predictive control module is to convert the geomechanical parameters output by the geomechanical parameter inversion module into the pre-pile geomechanical parameters. Transformed into the future in the prediction model External disturbance sequence within each time step (i.e., the prediction time domain) . Characterized by depth Varying physical parameters of soil and rock (such as elastic modulus) The model prediction control module uses its internal model to map the physical parameters that vary with spatial depth to parameters that vary with future time steps. The changing equivalent construction resistance or system disturbance, wherein the equivalent construction resistance constitutes the external disturbance sequence. .

[0105] The model predictive control module in each control cycle Based on the generated future perturbation sequence and the current system state obtained from the hybrid sensing and excitation array module. By solving a finite-time constrained optimization problem, the optimal control sequence is planned. In a specific embodiment, the objective function of the optimization problem is... It can be represented as:

[0106] ;

[0107] ;

[0108] In the formula, Describe the objective function. This represents the control input vector (column vector), which controls the decision variables of the problem. Indicates time The state vector describes the state at time step [0, 1]. All dynamic variables, Indicates a reference state. Let L2 be the weighted L2 norm of the vector, and let L2 be the weighted penalty for state deviations. It is a weighted matrix. The weighted 2-norm represents the different penalty weights assigned to each state component, specifically... ,in It is a diagonal matrix or a symmetric matrix, and the diagonal elements are usually chosen to be positive definite real numbers. Let L2 represent the weighted L2 norm of the control input, and L3 represent the penalty applied to the control input. The control input is defined at each time step. Control quantity related, matrix Used to represent the weights of different control inputs. Specifically: ,in It is a positive definite weighting matrix used to represent different degrees of penalty for the control signal. Indicates the control time during the prediction phase. The state at that time, This represents a combination of time steps. It is the current moment. It is the increment of the time step.

[0109] , This represents the state weighting matrix, used for the second penalty term.

[0110] The optimization problem is subject to the following constraints when it is solved:

[0111] State transition constraints: That is, the system dynamics must follow the predictive model;

[0112] State constraints: State constraints are used to ensure system safety, such as the maximum allowable stress of the pile body and the maximum allowable tilt angle.

[0113] Control constraints: Control constraints are the physical capacity limitations of construction equipment, such as the minimum / maximum output energy range of a hydraulic hammer and the maximum torque of a power head.

[0114] The model predictive control module uses numerical optimization algorithms (such as Sequence Quadratic Programming (SQP) or interior-point methods) to solve the constrained optimization problem and obtain an optimal control sequence. .

[0115] The model predictive control module employs a rolling optimization strategy, only optimizing the optimal control sequence. The first element in the equation, i.e., the current time. Optimal control command .

[0116] The construction parameter execution module, which is connected to the model prediction control module, is used to receive and execute the current optimal construction control instruction in the optimal construction control parameter sequence;

[0117] Specifically, in this embodiment, the function of the construction parameter execution module is to receive control commands from the model prediction control module and convert them into physical driving actions for the pile foundation construction equipment, so as to achieve precise adjustment of construction parameters.

[0118] The construction parameter execution module includes a communication interface unit, which is used to receive the current time output from the model prediction control module via an industrial fieldbus. Optimal control command To ensure the real-time transmission of control commands.

[0119] The construction parameter execution module includes one or more execution drive units. These execution drive units are electrically connected to the communication interface unit and physically connected to the specific actuators of the pile foundation construction equipment. The core function of each execution drive unit is to process the digitized optimal control commands received from the communication interface unit. It is converted into a physical drive signal, such as a voltage signal, a current signal, or a pulse width modulation (PWM) signal.

[0120] The optimal control command It is a digital control vector, whose components correspond to one or more specific construction parameters. The execution drive unit precisely controls the action of the actuator according to the value of the command component.

Claims

1. A multi-sensor fusion based pile construction parameter adaptive adjustment system, characterized in that, Comprise; A hybrid sensing and excitation array module for transmitting coded acoustic wave signals to the geological body under the pile foundation and collecting echo data of the acoustic wave signals and pile body dynamic response data generated during construction; A geomechanical parameter inversion module connected with the hybrid sensing and excitation array module, for inverting the geomechanical parameters in front of the pile based on the collected echo data and a corrected pile-soil coupling system physical model; A physical model self-correction module connected with the hybrid sensing and excitation array module and the geomechanical parameter inversion module, for self-correcting the pile-soil coupling system physical model in it based on the collected pile body dynamic response data and the inverted geomechanical parameters, and providing the corrected physical model to the geomechanical parameter inversion module; A model predictive control module connected with the geomechanical parameter inversion module, for rolling optimization to solve the optimal construction control parameter sequence in a limited time window in the future according to the geomechanical parameters in front of the pile output by it; A construction parameter execution module connected with the model predictive control module, for receiving and executing the current optimal construction control instruction in the optimal construction control parameter sequence.

2. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The hybrid sensing and excitation array module comprises: An active excitation unit mounted at the end of the pile foundation structure for transmitting the coded acoustic wave signals; A distributed passive sensing unit laid along the pile body, which is used to collect echo data and pile body dynamic response data.

3. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 2, characterized in that, The distributed passive sensing unit is composed of a multi-channel optical fiber Bragg grating sensor array; The echo data is the weak reflection signal of the acoustic wave pulse transmitted by the active excitation unit at the geological discontinuous surface; The pile body dynamic response data is the forced vibration signal generated under the action of hammering or drilling construction load.

4. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The geomechanical parameter inversion module internally solidifies a wave equation forward operator; The geomechanical parameter inversion module realizes inversion by solving the minimization problem of an objective function, which is used to quantify the mismatch between the observed echo data and the theoretical echo data calculated based on the wave equation forward operator, the corrected pile-soil coupling system physical model and the geomechanical parameters to be inverted.

5. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The physical model self-correction module internally solidifies a forced vibration response operator; The physical model self-correction module realizes self-correction by solving the minimization problem of a self-correction objective function, which is used to quantify the mismatch between the measured pile body dynamic response data and the theoretical forced vibration response calculated based on the forced vibration response operator, the inverted geomechanical parameters and the system physical model parameters to be corrected.

6. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 5, characterized in that, The system physical model parameters to be corrected include the side friction coefficient of the pile-soil interface, the equivalent material damping coefficient of the system or the boundary condition equivalent spring stiffness of the model calculation domain.

7. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The model predictive control module internally establishes a discrete time state space prediction model of the pile foundation construction process; The model predictive control module is used to convert the geomechanical parameters in front of the pile output by the geomechanical parameter inversion module into the external disturbance sequence in the future N time steps in the state space prediction model.

8. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The model predictive control module plans an optimal construction control parameter sequence by solving a constrained finite time horizon optimization problem based on the generated future external disturbance sequence and the current system state at each control cycle.

9. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 8, characterized in that, The constraint conditions of the constrained finite time horizon optimization problem include state quantity constraints for ensuring safety of the system, such as maximum allowable stress or maximum allowable inclination angle of the pile body, and control quantity constraints for limiting physical capabilities of the construction equipment.

10. The multi-sensor fusion based offshore pile foundation construction parameter adaptive regulation system according to claim 1, characterized in that, The construction parameter execution module comprises: a communication interface unit configured to receive the current optimal construction control instruction through an industrial field bus; an execution driving unit configured to convert the digital current optimal construction control instruction into a physical driving signal applied to a proportional servo valve of a hydraulic hammer or a motor frequency converter of a rig power head.

Images