Multi-parameter adjustment control system for civil engineering construction equipment

Abstract

The invention relates to the technical field of control and adjustment, in particular to a civil engineering construction equipment multi-parameter adjustment control system, which comprises a data acquisition module for capturing multi-dimensional physical states and geological environment data of key nodes of equipment; the data analysis module adopts a quantum annealing optimizer to analyze a multi-parameter coupling relationship to generate a decoupling scheme, and combines a multi-target attention controller to allocate resource priorities of an inner ring, a middle ring and an outer ring to form a coding control strategy; and the optimization module identifies chaotic features based on the Lyapunov exponent for executing the feedback data, quantifies performance indexes through a three-dimensional evaluation space, updates a parameter library by using a quantum migration engine and outputs a correction instruction to form a closed-loop optimization path. According to the invention, the problem of insufficient cooperative control precision under coupling interference of multiple execution mechanisms is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of control and regulation, and in particular to a multi-parameter regulation control system for civil engineering construction equipment. Background Art

[0002] Civil engineering construction equipment refers to a wide range of mechanized equipment used in civil and construction engineering. By replacing manual labor in high-load, high-risk operations, it significantly improves construction efficiency and ensures worker safety. Given the increasing scale of modern building structures and infrastructure, construction equipment such as cranes, excavators, and pile drivers are deployed to precisely handle critical processes such as earth excavation, material handling, and foundation construction.

[0003] In the actual application of automated control of civil engineering construction equipment, the collaborative control accuracy under the coupling interference of multiple actuators is insufficient. Specifically, this is manifested in the strong nonlinear coupling and external disturbances between physical quantities such as hydraulic pressure, load displacement, and motion trajectory during operation. As a result, existing single-parameter adjustment methods are unable to effectively maintain the overall stability and operation accuracy of the system. For example, during the construction of pile-driving machinery, when the drill pipe rotation speed, downforce, and cooling water flow need to change in coordination to adapt to different geological conditions, the single-parameter PID closed-loop control algorithm is often unable to dynamically decouple and compensate for the mutual influence of various physical quantities in real time. This can cause the drill pipe to experience attitude deviation or abnormal vibration of the power head when encountering sudden changes in the formation, thereby affecting the verticality of the hole and increasing equipment loss. This phenomenon is particularly prominent when there is time delay and noise interference in the feedback signals of multiple sensors. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a multi-parameter adjustment and control system for civil engineering construction equipment to solve the problem of insufficient collaborative control accuracy under coupling interference of multiple actuators in the automated control of civil engineering construction equipment.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows: The multi-parameter adjustment and control system for civil engineering construction equipment provided by the present invention comprises: Data acquisition module, which captures multi-dimensional physical state data and geological environment data at key execution nodes of the equipment; A data processing module inputs the multi-dimensional physical state data and geological environment data, generates prediction results in parallel through a physical mechanism model, a deep learning model, and a Bayesian probability model, dynamically evaluates the confidence of each model based on the actual sensor measurement value, and fuses and outputs optimization instructions; A data analysis module analyzes the multi-parameter coupling relationship in the optimization instruction and generates a coding control strategy; The collaborative module responds to the coding control strategy, processes sudden changes in hydraulic pressure through the inner-loop millisecond-level sliding mode compensation mechanism, matches load and speed parameters through the middle-loop second-level ant colony optimization mechanism, and adjusts the mechanism stiffness through the synergistic effect of the outer-loop working condition adaptive adjustment mechanism, and outputs a drive signal to the actuator; The optimization module receives the execution feedback data after the actuator executes the driving signal, analyzes the timeliness, energy consumption and stability indicators of the feedback data, and outputs the correction parameters to the data processing module and the data analysis module to form a dynamic closed-loop optimization path.

[0006] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data acquisition module includes: An omnidirectional piezoelectric sensor matrix is ​​used to capture the directional characteristics of pressure waves and is arranged in a matrix on the surface of the hydraulic cylinder; A fiber Bragg grating deformation chain for inverting three-dimensional bending deformation is implanted along the drill pipe axis; A multispectral environmental sensing unit for monitoring geological structure and thermal distribution, integrating millimeter-wave radar and infrared imaging; Among them, the hydraulic pressure wave characteristics output by the omnidirectional piezoelectric sensing matrix, the drill pipe deformation characteristics generated by the fiber Bragg grating deformation chain, and the geological thermal distribution characteristics obtained by the multispectral environmental perception unit are input into the data processing module.

[0007] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data processing module is also used to: Receiving measured value data from the data acquisition module; Calculate the deviation between the predicted values ​​of the physical mechanism model, the deep learning model, and the Bayesian probability model and the measured values ​​through KL divergence; Using the inverse of the deviation as the basis for weight distribution, dynamically distribute the weight of each model in the fusion optimization instruction; When the continuous prediction deviation of a single model exceeds a threshold, the incremental training procedure is activated to reconstruct the model parameters.

[0008] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data analysis module is also used to: receiving optimization instructions from the data processing module; Analyzing the multi-parameter coupling relationship in the optimization instruction through a quantum annealing optimizer to generate a quadratic unconstrained binary optimization problem; Solving the quadratic unconstrained binary optimization problem to obtain a multi-parameter decoupling solution; According to the multi-parameter decoupling scheme and the geological risk level output by the data processing module, the control resource priorities of the inner loop, the middle loop and the outer loop are allocated by the multi-objective attention controller; Output the coded control strategy to the collaborative module.

[0009] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the collaborative module is also used to: Receive the coding control strategy output by the data analysis module; According to the resource priority allocation result in the coding control strategy: The inner ring millisecond sliding mode compensation mechanism processes the hydraulic pressure mutation within 10ms and generates the sliding mode compensation amount; Zhonghuan's second-level ant colony optimization mechanism searches for the optimal matching point between load and speed every 200ms and outputs the ant colony optimization parameters; The outer ring working condition adaptive adjustment mechanism adjusts the drill pipe stiffness through the Bayesian optimization algorithm to generate a stiffness adjustment value; Synchronously integrating the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value; The output driving signal is sent to the synaptic pulse coding interface to drive the actuator array.

[0010] Furthermore, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention further includes: The inner ring millisecond sliding mode compensation mechanism is used to receive real-time pressure data from the piezoelectric sensor matrix on the surface of the hydraulic cylinder, set the output torque saturation threshold to prevent actuator overload, and output the sliding mode compensation amount to the performance analysis unit of the optimization module; The Zhonghuan second-level ant colony optimization mechanism is used to dynamically adjust the pheromone volatilization rate according to the load change rate based on the resource priority allocation result, and output the ant colony optimization parameters to the synaptic pulse coding interface; The outer ring working condition adaptive adjustment mechanism is used to detect the geological mutation signal of the multispectral environment perception unit, update the drill pipe stiffness value through the Bayesian optimization algorithm, and input the geological mutation characteristics into the parameter iteration library of the data processing module.

[0011] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the optimization module is also used to: receiving the execution vibration data output by the collaborative module; Extract the Lyapunov exponent of the vibration signal to identify the chaotic characteristics and mark them as the basis for determining abnormal events; Establish a stability index for the quantitative driving signal in a three-dimensional evaluation space; Update the cross-scenario optimal control strategy to the parameter iteration library of the data processing module through the quantum migration engine; When the risk of drill pipe deviation is identified, a stiffness compensation instruction is output to the outer ring working condition adaptive adjustment mechanism.

[0012] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the optimization module is also used to: Abnormal vibration events are determined based on chaotic characteristics, and responses to the abnormal vibration events are executed, the associated fault paths in the knowledge graph are cut off, gain instructions are sent to the collaborative module, the compensation strength of the inner-loop millisecond-level sliding mode compensation mechanism is enhanced, and the optimization constraint condition boundaries of the data analysis module are reset.

[0013] Furthermore, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention further includes: The data acquisition module captures the drill pipe deformation data and transmits it to the data processing module; The data processing module improves the weight of the physical mechanism model and outputs optimization instructions to the data analysis module; The data analysis module strengthens the inner loop control priority according to the optimization instructions and generates a coding control strategy input to the collaborative module; The collaborative module outputs the execution vibration data to the optimization module; The optimization module updates the sliding mode compensation parameters according to the execution vibration data, and outputs correction instructions to the coordination module. The coordination module integrates the correction instructions to suppress the deviation of the drill rod.

[0014] Furthermore, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the quasi-synaptic pulse coding interface is also used for: Receive the sliding mode compensation, ant colony optimization parameters and stiffness adjustment value synchronously integrated by the collaborative module; Encoding the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value into a phase modulated pulse group; Drive the hydraulic actuator array to perform the action of suppressing drill rod deviation.

[0015] Beneficial effects of the present invention: The present invention uses a multi-source heterogeneous sensor network to capture the multi-dimensional physical state data and environmental parameters of key nodes of the equipment in real time, and utilizes the parallel calculation and dynamic weight distribution mechanism of the physical mechanism model, deep learning model and Bayesian probability model to effectively integrate the advantages of multi-model prediction and reduce the prediction deviation caused by environmental interference; based on the quantum optimization algorithm, the strong coupling relationship of multiple parameters is analyzed to generate a decoupling control scheme, and the control resources of the inner loop millisecond-level sliding mode compensation, the middle loop second-level ant colony optimization and the outer loop working condition adaptive adjustment are dynamically allocated in combination with the multi-objective attention mechanism to achieve the coordinated optimization of hydraulic pressure mutation suppression, load and speed dynamic matching and mechanism stiffness; by extracting the chaotic characteristics of the execution feedback data to construct a three-dimensional evaluation system, and combining the quantum migration engine to realize the knowledge transfer of cross-scenario control strategies, forming a dynamic closed-loop optimization path including abnormal diagnosis, parameter correction and constraint boundary reset, and ultimately improving the collaborative control accuracy of civil engineering construction equipment under the coupling interference of multiple actuators. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0017] Figure 1 This is a system architecture diagram of a multi-parameter adjustment and control system for civil engineering construction equipment provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. The technical solutions provided by each embodiment of the present invention are described in detail below in conjunction with the drawings. In order to better understand the purpose of the present invention, the present invention is further described in detail below.

[0019] See also Figure 1 The multi-parameter adjustment and control system for civil engineering construction equipment provided by the present invention includes: Data acquisition module, which captures multi-dimensional physical state data and geological environment data at key execution nodes of the equipment; A data processing module inputs the multi-dimensional physical state data and geological environment data, generates prediction results in parallel through a physical mechanism model, a deep learning model, and a Bayesian probability model, dynamically evaluates the confidence of each model based on the actual sensor measurement value, and fuses and outputs optimization instructions; A data analysis module analyzes the multi-parameter coupling relationship in the optimization instruction and generates a coding control strategy; The collaborative module responds to the coding control strategy, processes sudden changes in hydraulic pressure through the inner-loop millisecond-level sliding mode compensation mechanism, matches load and speed parameters through the middle-loop second-level ant colony optimization mechanism, and adjusts the mechanism stiffness through the synergistic effect of the outer-loop working condition adaptive adjustment mechanism, and outputs a drive signal to the actuator; The optimization module receives the execution feedback data after the actuator executes the driving signal, analyzes the timeliness, energy consumption and stability indicators of the feedback data, and outputs the correction parameters to the data processing module and the data analysis module to form a dynamic closed-loop optimization path.

[0020] Data Acquisition Module: An omnidirectional piezoelectric sensor array is arranged in a matrix on the surface of the equipment's hydraulic cylinder. The pressure wave propagation direction vector is reconstructed by measuring the phase difference of the charge change of each unit. Fiber Bragg Grating (FBG) deformation chains are implanted at 20 cm intervals along the drill pipe axis. The three-dimensional curvature of the drill pipe is inverted by demodulating the reflected wavelength drift. The stratum density distribution data scanned by millimeter-wave radar and the drill bit temperature field data captured by an infrared thermal imager are integrated. After spatiotemporal registration, the hydraulic pressure wave characteristics, drill pipe deformation characteristics, and geological thermal distribution characteristics are output to the data processing module. This module uses differential pulse code modulation technology to convert analog signals into noise-resistant digital sequences. The sampling frequency is dynamically switched according to the equipment's operating conditions, completing a closed-loop sensing loop from physical-layer perception to digital signal conversion.

[0021] The data processing module receives the three types of feature data transmitted by the data acquisition module and uses the Navier-Stokes equations to establish a hydraulic circuit pressure-flow transfer function as a physical mechanism model. Simultaneously, a deep learning model is constructed using a temporal convolutional network to predict displacement deviations over the next five seconds based on historical vibration spectra. A Bayesian risk model is then developed by combining geological parameters with the probability transfer matrix of drilling resistance. After each model executes predictions in parallel, sensor measurements are collected in real time to calculate the KL divergence quantification of model prediction deviations. The inverse of the deviation is used to dynamically assign weights to each model in the fusion optimization instructions. If a single model's prediction deviation exceeds a threshold for ten consecutive times, an online incremental training program is activated to reconstruct the model's parameter library.

[0022] The Data Analysis Module analyzes the multi-parameter coupling relationships within the optimization instructions output by the Data Processing Module, mapping the 12 control parameters involved in the triple closed-loop into a 128-dimensional quadratic unconstrained binary optimization matrix. A quantum annealing optimizer iteratively solves for the lowest-energy solution set to generate a multi-parameter decoupling scheme. Simultaneously, based on the geological risk level output by the Data Processing Module, a multi-objective attention controller dynamically allocates control resource priority weights for the inner, middle, and outer loops. The decoupling scheme and priority weights are encoded into a pulse phase modulation strategy, which is then input into the collaboration module via a low-latency transmission protocol.

[0023] The collaborative module responds to the coded control strategy of the data analysis module. The inner-loop sliding mode compensation mechanism compensates for sudden torque changes within 10ms based on the hydraulic pressure error sliding surface and sets an output torque saturation threshold to prevent actuator overload. The middle-loop ant colony optimization mechanism dynamically adjusts the pheromone volatilization rate every 200ms, using hydraulic power consumption as a cost function, to search for the optimal match between load and speed. The outer-loop working condition adjustment mechanism updates the drill pipe stiffness parameters through Gaussian process regression. The three-loop mechanism collaboratively outputs sliding mode compensation, ant colony optimization parameters, and stiffness adjustment values ​​on a millisecond / second / working condition timescale, driving the actuator array via a synaptic pulse coding interface.

[0024] The optimization module receives vibration feedback data from the collaborative module, extracts the Lyapunov exponent of the vibration signal to identify chaotic characteristics, and establishes a three-dimensional assessment space to quantify system stability based on energy consumption, time delay, and accuracy. When drill pipe deviation risk is detected, the quantum migration engine injects the cross-scenario optimal control strategy into the model parameter library of the data processing module. Stiffness compensation instructions are simultaneously output to the collaborative module's outer loop mechanism. Associated fault paths in the knowledge graph are disconnected, the inner loop's sliding mode compensation strength is enhanced, and the quantum optimization constraint boundaries of the data analysis module are reset, completing the self-evolutionary cycle from fault diagnosis to parameter correction.

[0025] Specifically, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data acquisition module includes: An omnidirectional piezoelectric sensor matrix is ​​used to capture the directional characteristics of pressure waves and is arranged in a matrix on the surface of the hydraulic cylinder; A fiber Bragg grating deformation chain for inverting three-dimensional bending deformation is implanted along the drill pipe axis; A multispectral environmental sensing unit for monitoring geological structure and thermal distribution, integrating millimeter-wave radar and infrared imaging; Among them, the hydraulic pressure wave characteristics output by the omnidirectional piezoelectric sensing matrix, the drill pipe deformation characteristics generated by the fiber Bragg grating deformation chain, and the geological thermal distribution characteristics obtained by the multispectral environmental perception unit are input into the data processing module.

[0026] An omnidirectional piezoelectric sensing matrix utilizes a 3×3 array of piezoelectric ceramic elements on the surface of the hydraulic cylinder. The pressure wave propagation direction vector is calculated by measuring the phase difference of the charge change generated by each element under pressure. Each piezoelectric element incorporates a charge amplifier circuit to eliminate signal attenuation caused by electromagnetic interference. The pressure wave's directional characteristics are reconstructed by measuring the phase delay between adjacent elements. This directional vector information is used to infer the location of the hydraulic system's impact force source. This sensing mechanism captures sudden pressure events in the hydraulic circuit in real time, providing positioning data for subsequent mechanism anti-disturbance control.

[0027] Fiber Bragg Grating (FBG) deformation chain: A set of fiber Bragg gratings (FBGs) is implanted every 20 cm along the drill pipe axis, with three sets of FBGs arranged in a spatially orthogonal pattern. Drill pipe bending causes the grating period to shift, and the wavelength shift of the reflected light is linearly related to the curvature. The three-dimensional bending deformation field of the drill pipe is reconstructed by demodulating the wavelength offset. Bending characteristics include axial deflection and torsional deformation components. This deformation data can identify abnormal contact points between the drill pipe and the rock formation. This distributed fiber optic sensing architecture overcomes the limited spatial resolution of traditional strain gauges.

[0028] The multispectral environmental perception unit uses a millimeter-wave radar to transmit frequency-modulated continuous waves, generating point cloud data on the formation density distribution based on echo time delay and Doppler shift. An infrared thermal imager collects heat radiation from friction between the drill bit and the rock formation, outputting a temperature gradient field. The millimeter-wave point cloud and thermal imager data are aligned to a coordinate reference using a spatiotemporal registration unit, and then fused to generate a geological thermal distribution feature map. This feature map includes identification of thermal anomalies in rock fractures and 3D annotations of interfaces between soft and hard geology, providing geological information for adjusting the stiffness of the equipment mechanism.

[0029] Data collaboration mechanism: The pressure direction vector output by the piezoelectric matrix, the bending deformation gradient inverted by the fiber Bragg grating (FBG), and the geological thermal distribution feature map generated by the multispectral unit are encapsulated into a unified data frame using a time-division multiplexing protocol. The data frame header is embedded with a timestamp and spatial coordinate tags and transmitted via the RS-485 industrial bus to the feature decoder in the data processing module. This mechanism prevents temporal and spatial misalignment among the three data streams during transmission, ensuring the reliability of the subsequent model's collaborative analysis of the physical state.

[0030] Specifically, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data processing module is further used to: Receiving measured value data from the data acquisition module; Calculate the deviation between the predicted values ​​of the physical mechanism model, the deep learning model, and the Bayesian probability model and the measured values ​​through KL divergence; Using the inverse of the deviation as the basis for weight distribution, dynamically distribute the weight of each model in the fusion optimization instruction; When the continuous prediction deviation of a single model exceeds a threshold, the incremental training procedure is activated to reconstruct the model parameters.

[0031] The measured data reception mechanism specifically uses an industrial bus protocol to receive three characteristic data streams transmitted by the data acquisition module in real time. These data streams include the hydraulic pressure direction vector, the three-dimensional bending gradient of the drill pipe, and the geological thermal distribution map. A sliding time window is used to align the three data streams in time and space. The window length matches the equipment operation cycle to eliminate phase errors caused by sensor sampling delays. The aligned data packets are input into the model prediction and comparison unit as the measured physical state values.

[0032] The collaborative calculation of multi-model deviations involves constructing a hydraulic system transfer function based on the Navier-Stokes equations in the physical mechanism model, outputting coupled pressure-flow predictions. A deep learning model uses a time-convolutional network architecture to process historical vibration spectrum sequences and generate displacement trajectory predictions. Finally, a Bayesian probability model calculates the probability distribution of geological mutation risk using the Markov Chain Monte Carlo method. After the three models output predictions in parallel, the KL divergence is calculated with the measured values ​​of the corresponding physical quantities to quantify the degree of deviation from each model's predictions under real-time operating conditions.

[0033] The dynamic weight allocation strategy normalizes the KL divergence deviations of each model, using the inverse of the deviation as the base value for the weight allocation coefficient. A proportional constraint mechanism is introduced to prevent the weight of a single model from exceeding a set threshold. A linearly weighted fusion optimization command is generated. This command includes a three-dimensional control vector for pressure compensation, trajectory correction, and geological risk level. The weight ratio of each dimension of the vector reflects the current confidence level of the model in real time. The optimization command is updated every 200ms and output to the data analysis module.

[0034] The model parameter reconstruction mechanism specifically triggers incremental training when the KL divergence of a specific model exceeds a preset threshold for 10 consecutive control cycles. The model's participation in weight allocation is frozen, and the 500 most recently acquired sensor data sets are loaded as training samples. The fully connected layer parameters are fine-tuned while maintaining the network depth. After training is complete, the model's prediction error decay rate on the test set is verified. If it meets the design requirements, the model is reactivated to participate in weight allocation, the parameter library is updated, and the data is synchronized to the edge computing unit.

[0035] Specifically, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the data analysis module is further used to: receiving optimization instructions from the data processing module; Analyzing the multi-parameter coupling relationship in the optimization instruction through a quantum annealing optimizer to generate a quadratic unconstrained binary optimization problem; Solving the quadratic unconstrained binary optimization problem to obtain a multi-parameter decoupling solution; According to the multi-parameter decoupling scheme and the geological risk level output by the data processing module, the control resource priorities of the inner loop, the middle loop and the outer loop are allocated by the multi-objective attention controller; Output the coded control strategy to the collaborative module.

[0036] The optimization instruction parsing mechanism specifically receives a fused optimization instruction package output by the data processing module, which contains a three-dimensional vector of pressure compensation, trajectory correction, and geological risk level. A quantum annealing optimizer deconstructs the nonlinear coupling relationships between these vectors and maps the 12 control parameters involved in the triple closed loop into a 128-bit quadratic unconstrained binary optimization matrix. The diagonal elements of the matrix represent the independent influencing factors of the parameters, while the off-diagonal elements quantify the strength of the mutual interference between the parameters. This mapping process transforms the engineering control problem into an energy minimization problem that can be handled by quantum bits.

[0037] The quantum solution and decoupling solution generation process involves a quantum annealing optimizer iteratively calculating the ground-state energy solutions of the optimization matrix on a quantum processor. Each solution corresponds to a multi-parameter equilibrium state. The three solutions with the smallest Hamming distance are selected to form the eigenvalue decoupling solution, including the hydraulic system pressure stabilization point, the load motion path, and the optimal range of drill pipe stiffness. The decoupling solution output is accompanied by a confidence probability parameter, which is determined by the strength of the tunneling effect during the quantum annealing process.

[0038] To prioritize multi-objective resources, the multi-objective attention controller receives the decoupling solution while simultaneously acquiring real-time geological risk level data from the data processing module. A three-layer attention weight network is constructed: the first layer calculates the criticality of inner-loop sliding mode compensation to pressure stability; the second layer assesses the energy sensitivity of middle-loop path optimization; and the third layer weighs the response priority of outer-loop stiffness adjustment to geological risk. The three-layer network weights are dynamically scaled based on risk level, outputting millisecond-level resource allocation instructions for the inner, middle, and outer loops.

[0039] Control strategy encoding and output involves encoding the intrinsic control points and multi-objective resource allocation instructions in the decoupled scheme into a phase-modulated pulse sequence. The pulse width corresponds to the control value amplitude, and the pulse interval represents the frequency of parameter adjustment. A cyclic redundancy check (CRC) is added to the encoded data packet and transmitted via an anti-interference bus to the instruction cache of the collaborative module. The transmission protocol supports policy updates within 50 μs, meeting the execution response latency requirements of the triple closed-loop control system.

[0040] Specifically, in the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the collaborative module is further used to: Receive the coding control strategy output by the data analysis module; According to the resource priority allocation result in the coding control strategy: The inner ring millisecond sliding mode compensation mechanism processes the hydraulic pressure mutation within 10ms and generates the sliding mode compensation amount; Zhonghuan's second-level ant colony optimization mechanism searches for the optimal matching point between load and speed every 200ms and outputs the ant colony optimization parameters; The outer ring working condition adaptive adjustment mechanism adjusts the drill pipe stiffness through the Bayesian optimization algorithm to generate a stiffness adjustment value; Synchronously integrating the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value; The output driving signal is sent to the synaptic pulse coding interface to drive the actuator array.

[0041] The encoded control strategy parsing process involves receiving the phase-modulated pulse sequence transmitted by the data analysis module and converting it into a triple control instruction set through a decoder. Resource priority allocation indicates that the inner-loop sliding mode compensation accounts for 60% of the computing resources, the middle-loop ant colony optimization accounts for 25%, and the outer-loop stiffness adjustment accounts for 15%. This ratio fluctuates dynamically based on the real-time geological risk level. The control strategy parsing thread runs on a real-time operating system, ensuring instruction unpacking and task distribution within 1ms.

[0042] The inner loop's millisecond-level sliding mode compensation involves a sliding mode compensation mechanism that establishes a dynamic sliding surface based on the sudden gradient of hydraulic pressure changes and employs an exponential convergence law to shorten the system's convergence time. The compensation mechanism monitors data from the main oil circuit pressure sensor. When the detected pressure deviation exceeds a threshold, it calculates a compensation current based on the deviation derivative to drive the proportional valve. The compensation value passes through a saturation function limiter before output to prevent actuator overload. The entire process completes within 10ms, and the generated pressure compensation data packet is timestamped.

[0043] Zhonghuan's second-level ant colony optimization involves the ant colony optimization engine initializing a set of virtual path points, with each ant representing a load-speed combination. The optimization mechanism adjusts the pheromone volatility coefficient based on the load change rate, and the path selection probability is negatively correlated with the hydraulic system's energy consumption. The ant population is updated every 200ms, employing an elite retention strategy to avoid local optimality. The power and speed values ​​of the current optimal matching point are output as optimization parameters. The optimization process uses a separate memory pool to prevent task blocking.

[0044] Adaptive adjustment of the outer working condition loop constructs a drill pipe stiffness objective function based on a Bayesian optimization framework, with the geological risk level and the drill pipe deformation gradient as inputs. The stiffness adjustment mechanism uses Gaussian process regression to fit the objective function and samples the optimal stiffness value using an expectation lifting algorithm. The drill pipe stiffness adjustment cycle matches the outer working condition assessment rhythm, generating a stiffness adjustment value with an additional confidence assessment parameter, which is then used in weight calculation for subsequent strategy fusion.

[0045] Multi-parameter fusion and actuation involves inputting the three-loop output parameter set (sliding mode compensation, optimization parameters, and stiffness) into a pulse fusion unit. The compensation is converted into a pulse-width modulated signal, the optimization parameters are encoded as pulse density, and the stiffness controls the carrier frequency. During the generation of the fused signal, timestamp synchronization is verified, and interpolation is performed to compensate for parameter lag. The final drive signal is transmitted via a low-latency bus to a synaptic pulse encoding interface, which controls the piston displacement of the hydraulic actuator using charge.

[0046] Specifically, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention further includes: The inner ring millisecond sliding mode compensation mechanism is used to receive real-time pressure data from the piezoelectric sensor matrix on the surface of the hydraulic cylinder, set the output torque saturation threshold to prevent actuator overload, and output the sliding mode compensation amount to the performance analysis unit of the optimization module; The Zhonghuan second-level ant colony optimization mechanism is used to dynamically adjust the pheromone volatilization rate according to the load change rate based on the resource priority allocation result, and output the ant colony optimization parameters to the synaptic pulse coding interface; The outer ring working condition adaptive adjustment mechanism is used to detect the geological mutation signal of the multispectral environment perception unit, update the drill pipe stiffness value through the Bayesian optimization algorithm, and input the geological mutation characteristics into the parameter iteration library of the data processing module.

[0047] The inner-loop sliding mode compensation mechanism acquires real-time pressure direction vector data from the piezoelectric sensor matrix on the hydraulic cylinder surface and calculates the impact force source location information based on the phase delay difference between adjacent piezoelectric units. Based on this position information, the compensation mechanism dynamically constructs the sliding mode surface equation. When the hydraulic pressure gradient exceeds a threshold, exponential reaching control is triggered, generating a compensation current to drive the proportional valve. The compensation value is then clipped using a saturation function to limit the output torque to a set range to prevent actuator overload. The compensation data packet, along with a timestamp, is then transmitted to the performance analysis unit of the optimization module. Zhonghuan's ant colony optimization mechanism analyzes the resource priorities assigned by the data analysis module and initializes a virtual ant population to search for load-speed matching paths. The real-time load change rate is calculated based on the hydraulic motor current value, and the pheromone volatility coefficient is dynamically adjusted to control the swarm's exploration capability. The path selection probability function is linked to the cylinder power consumption index. After each generation of swarm iteration, the elite solution set is retained, and the speed and pressure combination that minimizes energy consumption is output as the optimization parameter. This is directly written into the parameter buffer of the synaptic pulse encoding interface via a high-speed serial interface.

[0048] The outer ring condition adjustment mechanism continuously monitors the geological thermodynamic map generated by the multispectral unit and flags geological mutation signals when abnormal temperature gradients are detected in rock fractures. A stiffness objective function is constructed based on a Bayesian optimization framework, and Gaussian process regression is used to sample optimal stiffness values ​​within a confidence interval. The adjustment mechanism generates stiffness parameters with associated confidence probabilities. After spatial and temporal alignment, the geological spatial positioning information and the stiffness adjustment values ​​are synchronously entered into the parameter iteration library index area of ​​the data processing module. Timestamp features are used to match the model reconstruction task. Specifically, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the optimization module is further used to: receiving the execution vibration data output by the collaborative module; Extract the Lyapunov exponent of the vibration signal to identify the chaotic characteristics and mark them as the basis for determining abnormal events; Establish a stability index for the quantitative driving signal in a three-dimensional evaluation space; Update the cross-scenario optimal control strategy to the parameter iteration library of the data processing module through the quantum migration engine; When the risk of drill pipe deviation is identified, a stiffness compensation instruction is output to the outer ring working condition adaptive adjustment mechanism.

[0049] The optimization module receives the equipment execution vibration time-domain signal output by the collaboration module via a parallel data channel and decomposes the vibration signal into a spectral sequence using a short-time Fourier transform. The phase-space reconstruction trajectory of adjacent sampling windows in the spectral sequence is calculated, and the maximum Lyapunov exponent is extracted using the Wolf algorithm. When the exponent value consistently exceeds the chaos determination threshold, an abnormal event is marked. Event determination is based on the additional equipment location code and operating condition timestamp, forming a spatiotemporal coordinate system benchmark for abnormality diagnosis. A three-dimensional evaluation space was constructed using control command delay time, hydraulic power consumption rate, and drill pipe offset angle variance as axes. The three-axis data collected at each sampling moment was mapped to spatial points. The drive signal stability fluctuation spectrum was generated by calculating the rate of change of the Euclidean distance between adjacent points. A radial basis function network was used to cluster the fluctuation spectrum, and a stability quantification factor was output and recorded in the system performance log. The quantum migration engine retrieves optimal cross-scenario parameters from a historical database, including high-frequency vibration suppression strategies for bridge pile foundation construction and low-speed, high-torque control curves for tunnel excavation. Using the principle of quantum teleportation, the strategy feature vector is decomposed into classical bits and quantum entangled bits. The classical bits are then orthogonally projected onto the storage structure of the parameter iteration library. The quantum bits then drive the reorthogonalization of the parameter weight matrix, enabling the gradual integration and update of the new strategy into the parameter iteration library. When the three-dimensional assessment space detects that the drill pipe offset angle variance exceeds the safe range for three consecutive sampling periods, an offset risk flag is triggered. The optimization module uses a Bayesian inference model to generate a stiffness compensation triplet instruction: the axial stiffness compensation amount is calculated based on the offset gradient, the radial stiffness compensation frequency matches the geological resonance spectrum, and the compensation duration is correlated with the equipment's operating inertia. This instruction is transmitted via a high-priority interrupt channel to the real-time control thread of the outer loop's adaptive operating condition adjustment mechanism.

[0050] Specifically, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the optimization module is further used to: Abnormal vibration events are determined based on chaotic characteristics, and responses to the abnormal vibration events are executed, the associated fault paths in the knowledge graph are cut off, gain instructions are sent to the collaborative module, the compensation strength of the inner-loop millisecond-level sliding mode compensation mechanism is enhanced, and the optimization constraint condition boundaries of the data analysis module are reset.

[0051] When the Lyapunov exponent identified by the optimization module consistently exceeds the chaos threshold, the abnormal vibration event determination logic is triggered. The event response thread searches the fault knowledge graph for historical nodes with identical spatiotemporal markers and calculates the cosine similarity between the current fault signature and the graph nodes using a graph neural network. Directed pruning is initiated for association paths with excessive similarity, severing the fault propagation path from the equipment's hydraulic system to the actuator vibration and generating instructions for updating the fault graph topology. A disturbance-resistant gain control vector is simultaneously constructed and a gain instruction is sent to the inner-loop sliding mode compensation mechanism of the collaborative module. The instruction contains a two-tuple consisting of a compensation intensity proportional coefficient and a boost duration. The proportional coefficient is linearly adjusted based on the deviation amplitude of the vibration chaos index. The gain data packet is transmitted via the real-time kernel message queue, triggering the sliding mode compensation mechanism to dynamically adjust the approach rate of the control law and amplify the compensation output amplitude to the preset safety margin. The reset task invokes the optimization constraint boundary manager of the data analysis module. The fault characteristic frequency component is extracted from the spatiotemporal labels of the knowledge graph pruning events. Based on this frequency characteristic, the temperature threshold and jump amplitude boundary of the quantum annealing optimizer are reset. The boundary reset instruction is written to the constraint register of the decision module, overwriting the original parameters to generate a disturbance-resistant optimization strategy.

[0052] Specifically, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention further includes: The data acquisition module captures the drill pipe deformation data and transmits it to the data processing module; The data processing module improves the weight of the physical mechanism model and outputs optimization instructions to the data analysis module; The data analysis module strengthens the inner loop control priority according to the optimization instructions and generates a coding control strategy input to the collaborative module; The collaborative module outputs the execution vibration data to the optimization module; The optimization module updates the sliding mode compensation parameters according to the execution vibration data, and outputs correction instructions to the coordination module. The coordination module integrates the correction instructions to suppress the deviation of the drill rod.

[0053] The fiber Bragg grating (FBG) deformation chain in the data acquisition module collects the wavelength drift signal of drill pipe bending strain in real time and reconstructs the three-dimensional deformation field of the drill pipe using a curvature inversion algorithm. After spatiotemporal registration, the deformation field data is encapsulated into feature data frames and transmitted via a deterministic Ethernet protocol to the input buffer queue of the competitive data processing module.

[0054] The competitive optimization data processing module uses a physical mechanism model to process drill pipe deformation characteristics, increasing its weight to 60% of the current fusion weight. The physical model calculates the deformation stress distribution based on continuum mechanics equations and outputs a trajectory correction optimization vector. Combining the vibration predictions from the deep learning model with Bayesian risk probabilities, it generates a three-dimensional optimization command containing a pressure compensation gradient, drilling speed correction, and risk level. This command is then transmitted to the data analysis module via a real-time bus. The data analysis module analyzes the trajectory-pressure coupling relationship in the optimization instructions and increases the resource allocation priority of the inner-loop sliding mode control to 70%. Using a quantum annealing optimizer, the 12 coupling parameters are mapped into a quadratic unconstrained binary optimization matrix to generate a drill pipe attitude decoupling solution. This solution is encoded into a pulse-width modulated control strategy frame and written into the collaboration module's strategy buffer. When the collaboration module executes inner-loop sliding mode compensation, the equipment body vibrates. The vibration acceleration signal is converted into an execution vibration data packet after anti-aliasing filtering. The data packet is appended with a spatial orientation tag and a microsecond timestamp and transmitted to the optimization module's input stream processing engine via a shared ring memory area. The optimization module extracts the Lyapunov exponent spectrum of the vibration signal and identifies chaotic resonance characteristics in specific frequency bands. Based on the chaos intensity, it dynamically updates the reaching law coefficients of the inner-loop sliding mode compensation and generates sliding mode compensation parameter correction instructions. These correction instructions, which carry a binary tuple consisting of the weight adjustment ratio and the duration of the action, are transmitted via a priority interrupt channel to the control law update port of the collaborative module. The execution center then integrates the correction parameters and reconfigures the sliding mode controller, increasing the compensation strength for sudden pressure changes by 35% and effectively suppressing drill pipe deviation.

[0055] Specifically, the multi-parameter adjustment and control system for civil engineering construction equipment of the present invention, the synaptic-like pulse coding interface is also used for: Receive the sliding mode compensation, ant colony optimization parameters and stiffness adjustment value synchronously integrated by the collaborative module; Encoding the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value into a phase modulated pulse group; Drive the hydraulic actuator array to perform the action of suppressing drill rod deviation.

[0056] The synaptic pulse encoding interface accesses three parameter sets output by the collaborative module through a shared memory area: the sliding mode compensation is translated into a scalar current compensation value, the ant colony optimization parameters are unpacked into a power density vector at the optimal load speed point, and the stiffness adjustment value includes the axial stiffness coefficient and radial frequency. The timestamps of these three parameters are synchronized and verified before being sent to the encoding buffer. The verification mechanism performs linear interpolation compensation for parameters with excessive time deviations to ensure data timing consistency. The phase-modulated pulse generator maps the current compensation value to a pulse width, which is positively correlated with the compensation current amplitude. The load power density vector is encoded as a pulse density modulation sequence, where the number of pulses per unit time corresponds to the power density amplitude. The stiffness coefficient and frequency value jointly modulate the phase angle of the carrier signal, generating a reference oscillation waveform with a phase-shift characteristic. The three modulated signals are superimposed in the time domain to form a composite pulse group, the pulse group envelope of which reflects the multi-dimensional drive requirements of the hydraulic actuator. The drive engine analyzes the envelope characteristics of the composite pulse train and converts them into four differential drive current signals. These current signals are fed into the proportional solenoid valve coils of the hydraulic actuator array, where piston displacement varies linearly with the current value. Axial compensation current drives the vertical cylinder to suppress drill pipe pitch deflection, while radial compensation current controls the lateral cylinder to suppress lateral swing. Actuator displacement feedback signals are transmitted in real time to the inner loop of the collaborative module, forming a closed loop to suppress drill pipe attitude deflection.

[0057] The present invention solves the problem of insufficient cooperative control accuracy under coupling interference of multiple actuators by the following methods: An omnidirectional piezoelectric sensor matrix, fiber Bragg grating (FBG) deformation chain, and multispectral environmental sensing units are deployed at key nodes of the equipment to simultaneously capture the directional characteristics of hydraulic pressure waves, the three-dimensional deformation gradient of the drill pipe, and the characteristics of geological thermal distribution. The data processing module processes multi-source data in parallel using physical mechanism models, deep learning models, and Bayesian probability models. It uses KL divergence to calculate the deviation between each model's predicted values ​​and the sensor's measured values ​​in real time. Model weights are dynamically assigned based on the inverse of the deviation and integrated to generate optimization instructions. When a single model's continuous prediction deviation exceeds the limit, an incremental training program is activated to reconstruct the model parameters, mitigating model prediction deviations caused by environmental interference at the data source end.

[0058] The data analysis module analyzes the multi-parameter coupling relationships within the optimization instructions. Using a quantum annealing optimizer, the control parameters are mapped into a quadratic unconstrained binary optimization problem, which is then solved to generate a multi-parameter decoupling solution. Based on the geological risk level, a multi-objective attention controller dynamically allocates control resource priorities for the inner (millisecond level), middle (second level), and outer (operating condition level) loops. The inner loop's sliding mode compensation mechanism generates compensation based on the hydraulic pressure gradient; the middle loop's ant colony optimization mechanism adjusts the pheromone volatilization rate based on the load change rate, outputting the optimal load-speed matching parameters; and the outer loop's operating condition adaptive adjustment mechanism updates the drill pipe stiffness using a Bayesian optimization algorithm. The outputs of these three loops are synchronously integrated via a synaptic-like pulse coding interface to drive the actuator array for dynamic decoupling control of multiple actuators.

[0059] The optimization module receives execution vibration data, extracts Lyapunov exponents, and identifies chaotic characteristics as a basis for anomaly determination. A three-dimensional evaluation space is established to quantify system performance: timeliness, energy consumption, and stability. The quantum migration engine then injects cross-scenario optimal strategies into the parameter iteration library of the data processing module. When drill pipe deviation risk is detected, stiffness compensation instructions are output to the collaborative module, disconnecting the associated fault path in the knowledge graph. Simultaneously, the inner-loop sliding mode compensation strength is enhanced and the optimization constraint boundaries are reset, forming a dynamic closed-loop optimization path that continuously suppresses the coupling interference of multiple actuators.

[0060] In pile foundation construction scenarios, hydraulically driven drill stem mechanisms face the coupled interference of sudden pressure changes, load fluctuations, and geological disturbances. During implementation, an omnidirectional piezoelectric sensor array is arranged in a matrix on the surface of the hydraulic cylinder. The pressure wave propagation direction vector is reconstructed through the phase difference of the charge changes of adjacent units to locate the hydraulic circuit impact source. A spatially orthogonal fiber Bragg grating chain is implanted along the drill stem axis, and the wavelength drift is demodulated to invert the three-dimensional bending deformation gradient of the drill stem, identifying the area of ​​abnormal contact between the drill stem and the rock formation. The millimeter-wave radar formation density point cloud and the infrared thermal imager drill bit temperature field are synchronously fused, and after spatiotemporal registration, the three-dimensional coordinates of the thermal anomaly area in the rock formation fracture are generated. These three types of characteristic data are transmitted via the industrial bus to the sliding time window alignment unit of the data processing module to eliminate the transmission delay differences of multi-source signals.

[0061] The data processing module uses a physical mechanism model to analyze the Navier-Stokes equations for hydraulic systems. A deep learning model uses a time-convolutional network architecture to process historical vibration spectra. A Bayesian model constructs a Markov probability transition matrix for geological mutations. Real-time pressure sensor measurements are collected, and the prediction deviations of each model are calculated using KL divergence. The inverse of the deviations is used to dynamically assign fusion weights and generate optimization instructions. If a single model's prediction deviations exceed the limit for consecutive times, the model's permissions are frozen, the latest sensor dataset is loaded, and incremental training is performed to update the fully connected layer parameters.

[0062] The data analysis module maps the pressure compensation, trajectory correction, and geological risk level vectors in the optimization instructions into a multi-parameter coupled quadratic unconstrained binary optimization matrix. A quantum annealing optimizer iteratively solves the matrix's ground state energy solution and outputs an intrinsic decoupling solution for the hydraulic pressure stabilization point, load stability path, and optimal stiffness range. A multi-objective attention controller dynamically allocates resource weights for inner-loop sliding mode compensation, middle-loop ant colony optimization, and outer-loop stiffness adjustment based on the geological risk level, forming a phase-modulated pulse coding strategy.

[0063] The collaborative modules respond to the coding strategy and distribute tasks across three loops. The inner loop's sliding mode compensation mechanism generates a compensation current based on the pressure error sliding surface. This current is then limited by a saturation function to drive a proportional valve to suppress sudden pressure changes. The middle loop's ant colony optimization mechanism initializes a set of virtual path points, adjusts the pheromone volatilization rate based on the load change rate, and outputs speed-pressure matching parameters with optimal energy consumption. The outer loop's working condition adjustment mechanism updates the drill pipe stiffness value through Gaussian process regression in response to thermal anomalies in the rock fractures. After timestamp verification, the sliding mode compensation value is converted into a pulse-width modulated signal. The ant colony parameters are encoded as pulse density, and the stiffness value modulates the carrier phase. This fusion drives the hydraulic actuator array.

[0064] The optimization module collects and executes vibration spectrum sequences, extracting the Lyapunov exponents of the phase space reconstruction trajectory. When the exponent values ​​consistently exceed limits, an abnormal event is flagged, disrupting the hydraulic vibration propagation path within the fault knowledge graph. A three-dimensional evaluation space for timeliness, energy consumption, and stability is established. Consistently exceeding limits for drill pipe offset angle variance triggers Bayesian inference to generate stiffness compensation instructions. The quantum migration engine decomposes the cross-scenario optimal strategy feature vectors and injects them into the parameter iteration library. This simultaneously enhances the inner-loop sliding mode compensation strength and resets the quantum annealing temperature threshold, forming a closed-loop optimization path that suppresses interference from multi-actuator coupling.

Claims

1. Multi-parameter adjustment and control system for civil engineering construction equipment, characterized in that: include: Data acquisition module, which captures multi-dimensional physical state data and geological environment data at key execution nodes of the equipment; A data processing module inputs the multi-dimensional physical state data and geological environment data, generates prediction results in parallel through a physical mechanism model, a deep learning model, and a Bayesian probability model, dynamically evaluates the confidence of each model based on the actual sensor measurement value, and fuses and outputs optimization instructions; A data analysis module analyzes the multi-parameter coupling relationship in the optimization instruction and generates a coding control strategy; The collaborative module responds to the coding control strategy, processes sudden changes in hydraulic pressure through the inner-loop millisecond-level sliding mode compensation mechanism, matches load and speed parameters through the middle-loop second-level ant colony optimization mechanism, and adjusts the mechanism stiffness through the synergistic effect of the outer-loop working condition adaptive adjustment mechanism, and outputs a drive signal to the actuator; The optimization module receives the execution feedback data after the actuator executes the driving signal, analyzes the timeliness, energy consumption and stability indicators of the feedback data, and outputs the correction parameters to the data processing module and the data analysis module to form a dynamic closed-loop optimization path.

2. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 1, characterized in that: The data acquisition module includes: An omnidirectional piezoelectric sensor matrix is ​​used to capture the directional characteristics of pressure waves and is arranged in a matrix on the surface of the hydraulic cylinder; A fiber Bragg grating deformation chain for inverting three-dimensional bending deformation is implanted along the drill pipe axis; A multispectral environmental sensing unit for monitoring geological structure and thermal distribution, integrating millimeter-wave radar and infrared imaging; Among them, the hydraulic pressure wave characteristics output by the omnidirectional piezoelectric sensing matrix, the drill pipe deformation characteristics generated by the fiber Bragg grating deformation chain, and the geological thermal distribution characteristics obtained by the multispectral environmental perception unit are input into the data processing module.

3. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 2, characterized in that: The data processing module is further used to: Receiving measured value data from the data acquisition module; Calculate the deviation between the predicted values ​​of the physical mechanism model, the deep learning model, and the Bayesian probability model and the measured values ​​through KL divergence; Using the inverse of the deviation as the basis for weight distribution, dynamically distribute the weight of each model in the fusion optimization instruction; When the continuous prediction deviation of a single model exceeds a threshold, the incremental training procedure is activated to reconstruct the model parameters.

4. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 3 is characterized in that: The data analysis module is further used to: receiving optimization instructions from the data processing module; Analyzing the multi-parameter coupling relationship in the optimization instruction through a quantum annealing optimizer to generate a quadratic unconstrained binary optimization problem; Solving the quadratic unconstrained binary optimization problem to obtain a multi-parameter decoupling solution; According to the multi-parameter decoupling scheme and the geological risk level output by the data processing module, the control resource priorities of the inner loop, the middle loop and the outer loop are allocated by the multi-objective attention controller; Output the coded control strategy to the collaborative module.

5. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 4, characterized in that: The collaboration module is further configured to: Receive the coding control strategy output by the data analysis module; According to the resource priority allocation result in the coding control strategy: The inner ring millisecond sliding mode compensation mechanism processes the hydraulic pressure mutation within 10ms and generates the sliding mode compensation amount; Zhonghuan's second-level ant colony optimization mechanism searches for the optimal matching point between load and speed every 200ms and outputs the ant colony optimization parameters; The outer ring working condition adaptive adjustment mechanism adjusts the drill pipe stiffness through the Bayesian optimization algorithm to generate a stiffness adjustment value; Synchronously integrating the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value; The output driving signal is sent to the synaptic pulse coding interface to drive the actuator array.

6. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 5, characterized in that: Also includes: The inner ring millisecond sliding mode compensation mechanism is used to receive real-time pressure data from the piezoelectric sensor matrix on the surface of the hydraulic cylinder, set the output torque saturation threshold to prevent actuator overload, and output the sliding mode compensation amount to the performance analysis unit of the optimization module; The Zhonghuan second-level ant colony optimization mechanism is used to dynamically adjust the pheromone volatilization rate according to the load change rate based on the resource priority allocation result, and output the ant colony optimization parameters to the synaptic pulse coding interface; The outer ring working condition adaptive adjustment mechanism is used to detect the geological mutation signal of the multispectral environment perception unit, update the drill pipe stiffness value through the Bayesian optimization algorithm, and input the geological mutation characteristics into the parameter iteration library of the data processing module.

7. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 6, characterized in that: The optimization module is further used to: receiving the execution vibration data output by the collaborative module; Extract the Lyapunov exponent of the vibration signal to identify the chaotic characteristics and mark them as the basis for determining abnormal events; Establish a stability index for the quantitative driving signal in a three-dimensional evaluation space; Update the cross-scenario optimal control strategy to the parameter iteration library of the data processing module through the quantum migration engine; When the risk of drill pipe deviation is identified, a stiffness compensation instruction is output to the outer ring working condition adaptive adjustment mechanism.

8. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 7, characterized in that: The optimization module is further used to: Abnormal vibration events are determined based on chaotic characteristics, and responses to the abnormal vibration events are executed, the associated fault paths in the knowledge graph are cut off, gain instructions are sent to the collaborative module, the compensation strength of the inner-loop millisecond-level sliding mode compensation mechanism is enhanced, and the optimization constraint condition boundaries of the data analysis module are reset.

9. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 8, characterized in that: Also includes: The data acquisition module captures the drill pipe deformation data and transmits it to the data processing module; The data processing module improves the weight of the physical mechanism model and outputs optimization instructions to the data analysis module; The data analysis module strengthens the inner loop control priority according to the optimization instructions and generates a coding control strategy input to the collaborative module; The collaborative module outputs the execution vibration data to the optimization module; The optimization module updates the sliding mode compensation parameters according to the execution vibration data, and outputs correction instructions to the coordination module. The coordination module integrates the correction instructions to suppress the deviation of the drill rod.

10. The multi-parameter adjustment and control system for civil engineering construction equipment according to claim 9, characterized in that: The synaptic-like pulse coding interface is further used for: Receive the sliding mode compensation, ant colony optimization parameters and stiffness adjustment value synchronously integrated by the collaborative module; Encoding the sliding mode compensation amount, ant colony optimization parameters and stiffness adjustment value into a phase modulated pulse group; Drive the hydraulic actuator array to perform the action of suppressing drill rod deviation.

Images