Multi-cylinder synchronous accurate control method for hydraulic jacking system of tower crane

Abstract

The invention provides a multi-cylinder synchronous accurate control method for a hydraulic jacking system of a tower crane, and belongs to the technical field of tower cranes. A pressure sensor and a displacement sensor are mounted on each hydraulic cylinder of a hydraulic jacking system of the tower crane, and a force sensor is arranged at a key position of a tower body, so that a multi-parameter real-time monitoring network is established; an improved Gaussian kernel function nonlinear matrix mapping algorithm is adopted to calculate a pressure deviation coefficient and a displacement synchronization metric value between hydraulic cylinders, the pressure adjustment processing frequency is dynamically adjusted according to the pressure deviation coefficient value, and self-adaptive load balancing control is achieved through tower body stress unbalance threshold value calculation and stress adjustment gain calculation. And finally, the flow distribution of each hydraulic cylinder is adjusted through a proportional valve, and the convergence of the system is ensured by utilizing Lyapunov stability analysis, so that the technical problem that the multi-cylinder synchronous control of the hydraulic jacking system of the tower crane is not accurate enough is solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of tower cranes, and in particular relates to a multi-cylinder synchronous precise control method for a tower crane hydraulic jacking system. BACKGROUND

[0002] The tower crane hydraulic jacking system is the core mechanism for realizing the vertical height adjustment of the tower crane. The traditional technology mainly adopts a centralized control method, which uniformly adjusts the pressure and flow of each hydraulic cylinder through a central controller, cooperates with the pressure sensor and displacement sensor of the foundation to monitor the state, and uses a proportional regulating valve to control the working parameters of each hydraulic cylinder to realize the synchronous jacking movement of the tower body. This technology is widely used in large tower cranes in the fields of construction, port loading and unloading, industrial production, etc. The traditional hydraulic jacking control system generally has the defect of insufficient synchronous control precision. The lack of effective coordination mechanism between each hydraulic cylinder leads to difficulty in timely compensation of pressure fluctuation and displacement deviation. The multi-cylinder coupling nonlinear characteristics make it difficult for the traditional linear control method to accurately handle the complex dynamic response relationship. The centralized control architecture has high computational complexity and slow response speed when dealing with multi-variable coordination problems, and cannot meet the technical requirements of high-precision real-time synchronous control. In the current tower crane hydraulic jacking operation, the pressure difference and displacement inconsistency between each hydraulic cylinder cannot be accurately controlled, which leads to the tilting and twisting of the tower body during jacking, seriously affecting the operation precision and equipment safety, especially under heavy load working conditions and complex environmental conditions. The problem of insufficient multi-cylinder synchronous control precision is more prominent. That is, the existing technology has the technical problem of insufficient multi-cylinder synchronous control precision of the tower crane hydraulic jacking system. SUMMARY

[0003] Therefore, the present application provides a multi-cylinder synchronous precise control method for a tower crane hydraulic jacking system, which can solve the technical problem of insufficient multi-cylinder synchronous control precision of the tower crane hydraulic jacking system in the prior art.

[0004] The application is implemented in the following manner: the application provides a tower crane hydraulic jacking system multi-cylinder synchronous precise control method, which collects hydraulic cylinder pressure values, displacement amounts and tower body stress distribution data; a hydraulic cylinder network node communication topology is constructed based on a distributed consistency coordination control algorithm, the information interaction relationship between cylinders is described through an adjacency matrix, and a decentralized coordination control architecture is established; a multi-scale matrix mapping hierarchical processing system is established, a sparse matrix compression sensing reconstruction mechanism is constructed, and an n-dimensional observation matrix is reconstructed into an m-dimensional original state matrix; a nonlinear matrix mapping algorithm based on an improved Gaussian kernel function is used to calculate pressure deviation coefficient values and displacement synchronization measurement values between the hydraulic cylinders; when the pressure deviation coefficient value is in the interval [0, 0.05], the pressure regulation processing frequency is reduced to 60% of the original frequency; when the pressure deviation coefficient value is in the interval (0.05, 0.15], the current pressure regulation processing frequency is maintained; when the pressure deviation coefficient value is in the interval (0.15, 0.30], the pressure regulation processing frequency is increased to 150% of the original frequency; a stress imbalance strength threshold value is calculated through a tower body stress imbalance threshold calculation equation, if the tower body stress imbalance signal strength is greater than the stress imbalance strength threshold value, adaptive load balancing control is enabled; a stress regulation gain range is calculated through a stress regulation gain calculation equation, if the stress imbalance strength is in the stress regulation gain range, the output force gain of each cylinder is adjusted; if the stress imbalance strength is less than the standard control threshold value, a standard synchronous control mode is adopted; the flow distribution of each hydraulic cylinder is adjusted through a proportional valve, the target synchronous state is calculated based on the distributed consistency coordination control algorithm, the system convergence is ensured through Lyapunov stability analysis, and the precise synchronous control of the pressure and displacement of each hydraulic cylinder is realized.

[0005] Among them, the multi-parameter real-time monitoring network is specifically a data acquisition network composed of sensors distributed at key positions of the hydraulic jacking system, and is used for real-time acquisition of system operation state parameters; the multi-scale matrix mapping hierarchical processing system is specifically a step-by-step mapping from a single-cylinder scale to a system scale to realize multi-level extraction of pressure and displacement information.

[0006] Among them, the distributed consistency coordination control algorithm is specifically a multi-agent coordination control method based on graph theory, each hydraulic cylinder is regarded as a network node, and global coordination control is realized through information interaction between adjacent nodes.

[0007] Among them, the distributed consistency coordination control algorithm describes the communication topology structure between the hydraulic cylinders through the construction of a Laplacian matrix, and designs a distributed control law based on a consistency protocol, so that the pressure and displacement states of each cylinder gradually tend to be consistent.

[0008] The distributed consistent coordination control algorithm adopts the pressure value, displacement, and speed state parameter of each hydraulic cylinder as input, calculates the consistency error of each node through Laplacian matrix operation, obtains the control gain parameter combined with stability analysis, and outputs the target pressure and flow control instruction of each cylinder.

[0009] The adjacency matrix is a mathematical matrix used to describe the communication connection relationship between hydraulic cylinders, and the matrix elements represent whether there is an information exchange channel between the corresponding cylinders.

[0010] The multi-scale matrix mapping is a kind of information processing method, which realizes multi-level feature extraction and fusion of data through matrix transformation of different scale levels.

[0011] The compressed sensing reconstruction mechanism is a data reconstruction method based on signal sparsity theory, which can recover the original high-dimensional signal from a small amount of observation data.

[0012] The tower body stress imbalance threshold calculation equation is used to calculate the stress imbalance strength judgment threshold according to the tower body structure parameters and working conditions, and the input includes the tower body weight, wind load, and hydraulic cylinder stress distribution data, and the output is the stress imbalance strength threshold.

[0013] The stress adjustment gain calculation equation is used to calculate the gain range of the output force adjustment of each hydraulic cylinder, and the input includes the current output force of each cylinder, the stress imbalance degree of the tower body, and the system safety factor, and the output is the stress adjustment gain range.

[0014] The improved Gaussian kernel function expression is Where x i and x j are the state vectors of the i th and j th hydraulic cylinders, p i and p j are the pressure values of the i th and j th hydraulic cylinders, d i and d j are the displacements of the i th and j th hydraulic cylinders. The improved Gaussian kernel function can more accurately describe the similarity relationship between hydraulic cylinders in high-dimensional feature space by introducing pressure difference items and displacement difference items, enhancing the sensitivity of the system to pressure fluctuations and displacement deviations, and improving the precision of multi-cylinder synchronous control.

[0015] The pressure deviation coefficient is a quantitative index that measures the degree of pressure difference between hydraulic cylinders, which is calculated by statistical analysis of the variance and mean of the pressure values of each cylinder.

[0016] The displacement synchronization measurement value is a numerical index that evaluates the consistency of the displacement of each hydraulic cylinder, reflecting the level of synchronization control accuracy of the system.

[0017] Specifically, the adaptive load balancing control is a control strategy that adjusts the output force of each hydraulic cylinder in real time according to the force distribution of the tower body to ensure that the tower structure is subjected to uniform force.

[0018] Additionally, pressure and displacement sensors are installed on each hydraulic cylinder of the tower crane's hydraulic jacking system, while force sensors are arranged at key locations in the tower structure to establish a multi-parameter real-time monitoring network. These key locations are specifically important parts of the tower structure that bear the main loads and stress concentrations, including the connection between the tower and the jacking mechanism, the main stress nodes of the tower, and the deformation-sensitive areas of the tower.

[0019] This invention solves the technical problem of insufficient synchronization accuracy in traditional centralized control methods by constructing a multi-cylinder precise synchronization control system based on a distributed consensus coordination control algorithm, establishing a multi-parameter real-time monitoring network and a multi-scale matrix mapping processing mechanism. This invention employs an improved Gaussian kernel function nonlinear matrix mapping algorithm to accurately calculate the pressure deviation coefficient and displacement synchronization metric between each hydraulic cylinder. Through an adaptive load balancing control strategy, it dynamically adjusts the output force gain of each cylinder, effectively eliminating pressure fluctuations and displacement deviations between hydraulic cylinders in traditional technologies. This significantly improves the accuracy and consistency of multi-cylinder coordinated control and avoids adverse phenomena such as tower tilting and torsion during the lifting process. In summary, this invention solves the technical problem of insufficient precision in multi-cylinder synchronization control of tower crane hydraulic lifting systems mentioned in the background art. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention.

[0021] Figure 2 This is a graph showing the synchronous control process of hydraulic cylinder pressure and displacement in the embodiment.

[0022] Figure 3 This is a schematic diagram illustrating the changes in the intensity of the unbalanced stress on the tower body and the switching of the control mode in the embodiment.

[0023] Figure 4 This is the feature space distribution diagram of the improved Gaussian kernel function mapping in the embodiment. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0025] like Figure 1 The diagram shown is a flowchart of a multi-cylinder synchronous precision control method for a tower crane hydraulic jacking system provided by the present invention. This method includes the following steps:

[0026] S01. Install pressure sensors and displacement sensors on each hydraulic cylinder of the tower crane's hydraulic jacking system, and at the same time arrange force sensors at key locations in the tower structure to establish a multi-parameter real-time monitoring network and collect data on the pressure value, displacement, and force distribution of each hydraulic cylinder.

[0027] S02. Based on the distributed consensus coordination control algorithm, construct the communication topology of hydraulic cylinder network nodes, treat each hydraulic cylinder as an independent control node, describe the information interaction relationship between cylinders through the adjacency matrix, and establish a decentralized coordination control architecture.

[0028] S03. Establish a hierarchical processing system for multi-scale matrix mapping, realize multi-level extraction of pressure and displacement information by mapping from single-cylinder scale to system scale, construct a sparse matrix compression sensing reconstruction mechanism, and reconstruct the m-dimensional original state matrix from the n-dimensional observation matrix using the assumption of system state sparsity.

[0029] S04. A nonlinear matrix mapping algorithm based on an improved Gaussian kernel function maps the original pressure-displacement matrix to a high-dimensional feature space for processing using the improved Gaussian kernel function, and calculates the pressure deviation coefficient and displacement synchronization metric between each hydraulic cylinder.

[0030] S05. When the pressure deviation coefficient value is in the range [0, 0.05], reduce the pressure regulation processing frequency to 60% of the original frequency; when the pressure deviation coefficient value is in the range (0.05, 0.15], maintain the current pressure regulation processing frequency; when the pressure deviation coefficient value is in the range (0.15, 0.30], increase the pressure regulation processing frequency to 150% of the original frequency.

[0031] S06. Calculate the unbalanced load strength threshold using the unbalanced load strength threshold calculation equation. If the unbalanced load signal strength is greater than the unbalanced load strength threshold, then enable adaptive load balancing control. Calculate the force adjustment gain range using the force adjustment gain calculation equation. If the unbalanced load strength is within the force adjustment gain range, then adjust the output force gain of each cylinder. If the unbalanced load strength is less than the standard control threshold, then adopt the standard synchronous control mode.

[0032] S07. The flow distribution of each hydraulic cylinder is adjusted by a proportional valve, the target synchronization state is calculated based on a distributed consensus coordination control algorithm, and Lyapunov stability analysis is used to ensure system convergence, thereby achieving precise synchronous control of the pressure and displacement of each hydraulic cylinder.

[0033] Among them, the multi-parameter real-time monitoring network refers to a data acquisition network composed of sensors distributed at various key locations in the hydraulic jacking system, which is used to acquire system operating status parameters in real time.

[0034] The distributed consensus coordination control algorithm is a graph-based multi-agent coordination control method that treats each hydraulic cylinder as a network node and achieves global coordination control through information interaction between neighboring nodes. The algorithm demonstrates significant effectiveness in this hydraulic jacking system, enabling multi-cylinder coordination and synchronization without a central controller, thus significantly improving system reliability and fault tolerance. The algorithm constructs a Laplace matrix to describe the communication topology between hydraulic cylinders and designs a distributed control law based on a consensus protocol, gradually bringing the pressure and displacement states of each cylinder towards uniformity. The algorithm is suitable for tower crane hydraulic jacking systems because the system has a multi-cylinder distributed characteristic, with coupling relationships between cylinders and extremely high requirements for synchronization accuracy. Traditional centralized control methods are susceptible to single-point failures, while distributed control can fully utilize the distributed characteristics of the system and achieve globally optimal control through local information interaction. The algorithm uses the pressure, displacement, and speed of each hydraulic cylinder as input, calculates the consistency error of each node through Laplace matrix operations, obtains control gain parameters through stability analysis, and finally outputs the target pressure and flow control commands for each cylinder, achieving asymptotic convergence of the system state.

[0035] The adjacency matrix is ​​a mathematical matrix used to describe the communication connection between hydraulic cylinders. The matrix elements indicate whether there is an information exchange channel between the corresponding cylinders.

[0036] Multi-scale matrix mapping is an information processing method that uses matrix transformations at different scale levels to extract and fuse multi-level features of data.

[0037] Among them, compressed sensing reconstruction mechanism is a data reconstruction method based on signal sparsity theory, which can recover the original high-dimensional signal from a small amount of observation data.

[0038] The improved Gaussian kernel function is a kernel function that combines the hydraulic cylinder pressure and displacement parameters with the standard Gaussian kernel function, and its expression is: Where x i and x j Let p be the state vector of the i-th and j-th hydraulic cylinders, respectively. i and p j d represents the pressure values ​​of the i-th and j-th hydraulic cylinders, respectively. i and d jLet σ, α, and β represent the displacements of the i-th and j-th hydraulic cylinders, respectively, and let σ, α, and β be kernel function parameters. The pressure value originates from the data collected by the pressure sensor in step S01, and the displacement value originates from the data collected by the displacement sensor in step S01. The improved Gaussian kernel function, by introducing pressure and displacement difference terms, can more accurately characterize the similarity relationship between hydraulic cylinders in a high-dimensional feature space, effectively enhancing the system's sensitivity to pressure fluctuations and displacement deviations, and significantly improving the accuracy of multi-cylinder synchronous control. Compared to the traditional Gaussian kernel function, which only considers the single dimension of state vector distance, the improved kernel function can simultaneously consider the two key control objectives of pressure consistency and displacement consistency, exhibiting stronger adaptability in handling multi-cylinder coupled nonlinear problems. The improved kernel function, through the exponential decay characteristic of the pressure difference term, gives hydraulic cylinders with similar pressures a higher similarity weight in the feature space, which is beneficial for forming a pressure-balanced control strategy; through the constraint effect of the displacement difference term, it ensures that displacement synchronization is fully reflected in the mapping process, thereby achieving a coordinated unity of pressure balance and displacement synchronization.

[0039] The pressure deviation coefficient is a quantitative indicator that measures the degree of pressure difference between hydraulic cylinders. It is calculated by statistically analyzing the variance and mean of the pressure values ​​of each cylinder. The pressure values ​​are derived from the data collected by the pressure sensor in step S01.

[0040] Among them, the displacement synchronization metric is a numerical index for evaluating the consistency of displacement of each hydraulic cylinder, reflecting the level of synchronization control accuracy of the system. The displacement consistency is calculated based on the displacement data collected by the displacement sensor in step S01.

[0041] The calculation equation for the unbalanced stress threshold of the tower body is used to calculate the unbalanced stress intensity judgment threshold based on the tower body structural parameters and working conditions. The inputs include the tower body weight, wind load, and force distribution data of each hydraulic cylinder. The output is the unbalanced stress intensity threshold. The force distribution data of the tower body comes from the force sensor data collected in step S01.

[0042] The force adjustment gain calculation equation is used to calculate the gain range of each hydraulic cylinder's output force adjustment. The inputs include the current output force of each cylinder, the degree of force imbalance of the tower body, and the system safety factor. The output is the force adjustment gain range. The current output force of each cylinder is calculated based on the pressure value collected by the pressure sensor in step S01 and the piston area of ​​the hydraulic cylinder.

[0043] Among them, the unbalanced stress intensity threshold is the critical value for determining the degree of unbalanced stress on the tower body. It is calculated by the unbalanced stress threshold calculation equation and is used as the judgment condition for enabling adaptive load balancing control in step S06.

[0044] Among them, the force adjustment gain range is the gain variation range of each hydraulic cylinder's output force adjustment, which is calculated by the force adjustment gain calculation equation and is used to adjust the parameter range of each cylinder's output force gain in step S06.

[0045] The standard control threshold is the force imbalance intensity judgment value using the standard synchronous control mode, which is set to 50% of the force imbalance intensity threshold and is used as the activation condition for the standard synchronous control mode in step S06.

[0046] Among them, adaptive load balancing control is a control strategy that adjusts the output force of each hydraulic cylinder in real time according to the force distribution of the tower body to ensure that the tower body structure is subjected to uniform force. The force distribution of the tower body is determined based on the force distribution data of the tower body collected by the force sensor in step S01.

[0047] The target synchronization state is the ideal pressure and displacement state of each hydraulic cylinder when they are running synchronously. It is calculated based on the pressure value and displacement of each hydraulic cylinder by a distributed consensus coordination control algorithm.

[0048] Lyapunov stability analysis is a mathematical analysis method used to prove the stability and convergence of a control system. It verifies the asymptotic stability of the system by constructing a Lyapunov function.

[0049] The specific implementation methods of the above steps are described in detail below.

[0050] The specific implementation of step S01 involves constructing a hierarchical sensor network architecture to achieve real-time monitoring of multiple parameters. The core objective of this step is to establish a data acquisition infrastructure that comprehensively covers key locations of the hydraulic jacking system. First, pressure sensors and displacement sensors are installed at the cylinder body and piston rod of each hydraulic cylinder. The pressure sensors are strain gauge pressure transmitters with a measurement accuracy of 0.25% of full scale, a response time of less than 1 millisecond, and a measurement range set from 0 to 40 MPa. The displacement sensors are magnetostrictive linear displacement sensors with a measurement accuracy of 0.01 mm, and the stroke range is determined based on the maximum stroke of the hydraulic cylinder. Then, force sensors are arranged at key load-bearing nodes of the tower structure, including the foundation connection at the bottom of the tower, the connecting flanges of each section of the tower, and the connection position at the top of the tower. The force sensors are axial strain gauge force sensors with a measurement accuracy of 0.5% of full scale. A data communication network based on fieldbus technology is established, using the Controller Area Network (CLAN) bus protocol to achieve real-time data exchange between the sensors and the control system. The data acquisition frequency is set to 100 Hz to ensure the real-time performance and accuracy of the system status information. The analog signal output from the sensor is converted into a digital signal by an analog-to-digital converter, and a filtering algorithm is used to eliminate signal interference. This filtering algorithm is based on the Kalman filtering principle and can effectively suppress the impact of sensor noise and environmental interference on data quality.

[0051] The specific implementation of step S02 is based on graph theory to construct a distributed control network topology. The purpose of this step is to establish a decentralized multi-cylinder coordinated control architecture. Each hydraulic cylinder is defined as an independent control node in the network, with each node possessing independent data processing and decision-making capabilities. Nodes share information and coordinate control through communication links. An adjacency matrix is ​​constructed to describe the communication connections between hydraulic cylinders. The matrix has an n×n dimension, where n is the number of hydraulic cylinders. A matrix element value of 1 indicates a direct communication connection between the corresponding two hydraulic cylinders, and an element value of 0 indicates no direct connection. A degree matrix is ​​calculated based on the adjacency matrix. The diagonal elements of the degree matrix represent the connection degree of each node, i.e., the number of other nodes directly connected to that node. The Laplace matrix is ​​further calculated; this matrix is ​​the difference between the degree matrix and the adjacency matrix. The eigenvalue distribution of the Laplace matrix directly affects the convergence performance of the distributed consensus coordinated control algorithm. A distributed control protocol is designed, where each hydraulic cylinder node calculates the control input based on the state differences with its neighboring nodes, achieving the global coordination goal through local information interaction. Compared with traditional centralized control, this distributed architecture has stronger fault tolerance and scalability; the failure of a single node will not affect the normal operation of the entire system.

[0052] The specific implementation of step S03 involves establishing a multi-level data processing and feature extraction mechanism. The purpose of this step is to achieve a step-by-step mapping and fusion from local information of a single cylinder to the global state of the system. First, a single-cylinder-scale data preprocessing layer is established to normalize and standardize the raw sensor data from step S01, eliminating dimensional and numerical range differences between different sensors. Then, an inter-cylinder-scale feature extraction layer is established, extracting the main feature components of the state parameters of each hydraulic cylinder using principal component analysis (PCA). This algorithm, based on the eigenvalue decomposition principle of the covariance matrix, effectively reduces data dimensionality while retaining key information. Next, a system-scale global state reconstruction layer is established, employing a compressed sensing reconstruction algorithm based on sparse representation theory. This algorithm utilizes the sparsity of the hydraulic system state to reconstruct complete system state information from partial observation data. This reconstruction algorithm is based on the optimization principle of minimizing the L1 norm, achieving accurate recovery of the original high-dimensional state information through iteratively solving the sparse coefficient vector. Finally, a sparse dictionary matrix is ​​constructed to describe the sparse representation basis of the system state, with each atom in the dictionary corresponding to a typical system operating mode. The sparse decomposition problem is solved using an orthogonal matching pursuit algorithm. This algorithm employs a greedy search strategy to progressively select the best-matching dictionary atoms until the reconstruction accuracy requirement is met. A reconstruction error threshold of 5% of the original signal energy is set to ensure that the reconstruction quality meets the control accuracy requirements.

[0053] The specific implementation of step S04 involves using an improved Gaussian kernel function to achieve nonlinear feature mapping and similarity measurement. The purpose of this step is to accurately characterize the relationship between hydraulic cylinders and calculate synchronization control parameters in a high-dimensional feature space. An improved Gaussian kernel function is established, which introduces pressure difference and displacement difference terms based on the standard Gaussian kernel function. Multidimensional similarity measurement is achieved through the product of three exponential functions. The first exponential term, based on the Euclidean distance of the state vectors, reflects the overall state similarity of the hydraulic cylinders. The kernel function parameter σ is set to 1.5 times the standard deviation of the state vectors. The second exponential term, based on the absolute value of the pressure difference, enhances the system's sensitivity to pressure fluctuations. The parameter α is set between 0.1 and 0.3, with the specific value adjusted according to the system's pressure fluctuation characteristics. The third exponential term, based on the absolute value of the displacement difference, ensures that displacement synchronization is fully reflected in the mapping process. The parameter β is set between 0.05 and 0.2. The similarity measurement value between each hydraulic cylinder is calculated using a kernel matrix. The kernel matrix is ​​a symmetric matrix with diagonal elements equal to 1, and off-diagonal elements reflecting the degree of similarity between the corresponding hydraulic cylinders. The pressure deviation coefficient is calculated based on the kernel matrix. This coefficient is determined by the ratio of the standard deviation to the mean of the pressure values ​​of each hydraulic cylinder, reflecting the uniformity of the system pressure distribution. A displacement synchronization metric is calculated, which is determined by the ratio of the variance of the displacement of each hydraulic cylinder to the target displacement; a smaller value indicates better synchronization performance. This nonlinear mapping method can more accurately handle the nonlinear coupling characteristics of hydraulic systems compared to linear methods.

[0054] The specific implementation of step S05 involves adaptively adjusting the processing frequency of the control system based on the pressure deviation coefficient. The purpose of this step is to optimize computational resource allocation and system response speed while ensuring control accuracy. A real-time calculation mechanism for the pressure deviation coefficient is established. A sliding window algorithm is used to statistically analyze the pressure data from the most recent 100 sampling periods, calculating the standard deviation and mean of the pressure values ​​of each hydraulic cylinder, thereby determining the current value of the pressure deviation coefficient. Three pressure deviation coefficient ranges are set, each corresponding to a different frequency adjustment strategy. When the pressure deviation coefficient is between 0 and 0.05, it indicates a relatively uniform system pressure distribution. In this case, the pressure regulation processing frequency is reduced to 60% of the original frequency, i.e., from 100 Hz to 60 Hz, to reduce unnecessary computational overhead. When the pressure deviation coefficient is between 0.05 and 0.15, it indicates that the system pressure deviation is within the normal range, and the current processing frequency of 100 Hz is maintained. When the pressure deviation coefficient is between 0.15 and 0.30, it indicates a large pressure deviation in the system, requiring stronger control. In this case, the pressure regulation processing frequency is increased to 150% of the original frequency, i.e., to 150 Hz. When the pressure deviation coefficient exceeds 0.30, the emergency control mode is activated, the processing frequency is increased to 200 Hz, and an alarm mechanism is triggered. This adaptive frequency adjustment mechanism is based on feedback control theory, dynamically adjusting control parameters by monitoring the system status in real time, thus ensuring both control accuracy and improving system efficiency. A hysteresis control strategy is employed to prevent frequent frequency switching, with a hysteresis interval set at a difference of 0.02 between the upper and lower switching thresholds.

[0055] The specific implementation of step S06 involves establishing an adaptive load balancing control mechanism based on tower body stress analysis. The purpose of this step is to ensure uniform stress distribution and avoid localized stress concentration during the tower structure's lifting process. First, a calculation model for the tower body stress imbalance threshold is established. This model uses tower weight, wind load, and the current output force of each hydraulic cylinder as input parameters, and calculates the stress imbalance intensity threshold using the torque balance principle. Tower weight includes the tower's self-weight and additional loads such as the jib. Wind load is calculated according to local wind speed and the tower's windward area based on building structure load codes. The stress imbalance intensity is determined by the ratio of the variance of the output force of each hydraulic cylinder to the average output force. When this ratio exceeds a set threshold, adaptive load balancing control is activated. The stress imbalance intensity threshold is set to 0.15, a value determined based on the tower crane's structural safety factor and engineering experience. A stress adjustment gain calculation model is then established, using the current output force of each cylinder, the degree of stress imbalance, and the system safety factor as inputs, and outputting the force adjustment gain range of each hydraulic cylinder. The force adjustment gain range is set to ±20% of the current output force to ensure that the adjustment process does not exceed the safe operating range of the hydraulic cylinder. The standard control threshold is set to 50% of the force imbalance intensity threshold, i.e., 0.075. When the force imbalance intensity is less than this value, the standard synchronous control mode is adopted. The adaptive load balancing control strategy uses a proportional-integral-derivative (PID) control algorithm, with the proportional gain set to 1.2, the integral gain set to 0.8, and the derivative gain set to 0.3. Dynamic balance of the tower body's force is achieved by adjusting the output force of each hydraulic cylinder in real time. This control strategy can effectively compensate for differences in the output force of each cylinder caused by manufacturing tolerances, installation errors, and other factors.

[0056] The specific implementation of step S07 is to achieve precise synchronous operation of each hydraulic cylinder through integrated control. The purpose of this step is to unify and coordinate the aforementioned control algorithms and strategies to achieve the final synchronous control goal. A proportional valve flow distribution control system is established. The required hydraulic oil flow rate is calculated based on the target output force and current state of each hydraulic cylinder, and the oil inlet flow rate of each hydraulic cylinder is precisely adjusted through an electro-proportional valve. The proportional valve control signal is generated using pulse width modulation, with a modulation frequency set to 1 kHz and a duty cycle range of 10% to 90%, corresponding to the working range of the hydraulic cylinder from minimum to maximum output force. The target synchronization state of each hydraulic cylinder is calculated based on a distributed consensus coordination control algorithm. This algorithm gradually converges the pressure and displacement states of each cylinder to a consistent state through iterative calculation. The convergence speed of the algorithm is determined by the second smallest eigenvalue of the Laplace matrix. The convergence performance can be improved by optimizing the network topology. A Lyapunov function is established to verify the stability of the control system. This function is defined as the quadratic form of the state error of each hydraulic cylinder. The asymptotic stability convergence of the system is ensured by proving the negative definiteness of the Lyapunov function. An observer is designed to estimate system state variables that cannot be directly measured. Based on Kalman filtering theory, the observer can accurately estimate the system state even in the presence of measurement noise. System performance evaluation indicators are established, including parameters such as synchronization accuracy, response time, and overshoot. Synchronization accuracy requires displacement deviation of each hydraulic cylinder to be less than 1 mm and pressure deviation to be less than 0.5 MPa. The system response time is required to reach steady state within 2 seconds, and the overshoot is controlled within 5%. Closed-loop feedback control continuously corrects the control parameters to ensure that the system maintains good synchronization control performance under various operating conditions.

[0057] It should be noted that the key technical ideas of this invention mainly include three aspects: distributed consensus coordination control architecture, improved Gaussian kernel function nonlinear mapping algorithm, and multi-parameter adaptive control strategy.

[0058] The distributed consensus coordination control architecture offers significant advantages over traditional centralized control methods. Based on graph theory, this architecture treats each hydraulic cylinder as an independent node, achieving global coordinated control through local information exchange. This fundamentally solves the single-point-of-failure risk and communication latency problems inherent in centralized control. Traditional centralized control requires aggregating all sensor data to a central controller for unified processing; when the controller fails, the entire system fails. In contrast, each node in the distributed architecture possesses independent decision-making capabilities, and the failure of a single node does not affect the normal operation of other nodes, significantly improving system reliability and fault tolerance. Simultaneously, the distributed architecture reduces data transmission distance and communication load, lowers system latency, and improves control response speed.

[0059] The improved Gaussian kernel function nonlinear mapping algorithm introduces pressure and displacement difference terms into the standard Gaussian kernel function, achieving accurate measurement of multi-dimensional similarity between hydraulic cylinders. Compared to traditional linear mapping methods, it can better handle the nonlinear coupling characteristics of hydraulic systems. Traditional methods typically only consider the Euclidean distance of the state vectors, failing to fully reflect the coupling relationship between pressure consistency and displacement consistency. The improved kernel function, through its exponential decay characteristic, gives hydraulic cylinders with similar pressures higher weights in the feature space, which is beneficial for forming a balanced control strategy. Simultaneously, the constraint of the displacement difference term ensures that synchronization is fully reflected in the mapping process, thereby achieving a coordinated unity of pressure balance and displacement synchronization.

[0060] Multi-parameter adaptive control strategies dynamically adjust control frequency and gain parameters by real-time monitoring of key parameters such as pressure deviation coefficient and tower stress distribution. Compared to fixed-parameter control methods, this approach offers greater adaptability and higher control accuracy. Traditional methods use fixed control parameters and processing frequencies, making dynamic optimization impossible based on system state changes. In contrast, adaptive strategies can reduce the control frequency to save computational resources when the system is in good condition, and increase control strength to ensure accuracy when deviations exist, achieving an optimal balance between performance and efficiency.

[0061] The synergistic effect of these three key technological approaches has yielded significant technical results. The distributed architecture provides a distributed computing platform for the improved kernel function algorithm and adaptive control strategy, enabling complex nonlinear mapping calculations to be executed in parallel across nodes, thus improving algorithm execution efficiency. The improved kernel function algorithm provides an accurate similarity measurement and deviation calculation basis for the adaptive control strategy, making the dynamic adjustment of control parameters more precise and timely. The adaptive control strategy optimizes the operating parameters of the distributed control network in real time based on the calculation results of the improved kernel function, forming a closed-loop optimization feedback mechanism. This synergy allows the entire system to possess the reliability advantages of distributed control, the accuracy advantages of nonlinear algorithms, and the flexibility advantages of adaptive control, resulting in significant improvements over traditional methods in synchronization accuracy, system reliability, response speed, and resource utilization efficiency.

[0062] It should be noted that this invention also solves the following technical problem: the lag in control response and poor real-time performance of traditional hydraulic jacking systems under dynamic conditions. Existing control systems typically employ fixed control parameters and periodic state update mechanisms. When faced with dynamic conditions such as load changes and environmental disturbances, they cannot quickly adjust control strategies, resulting in significant time lag in control response. This leads to a significant decrease in the effectiveness of multi-cylinder synchronous control during dynamic processes, especially under complex operating modes such as frequent start-stop and variable-speed jacking, where the response lag problem of traditional control methods is even more severe. This invention achieves continuous acquisition of system status through a multi-parameter real-time monitoring network. Combined with the parallel processing capabilities of a distributed control architecture, each hydraulic cylinder node can simultaneously perform state perception and control decision-making, significantly shortening the response time of the control loop. Through a compressed sensing reconstruction mechanism and the fast calculation characteristics of an improved Gaussian kernel function, millisecond-level control response speed is achieved, effectively solving the problem of control response lag in traditional technologies.

[0063] Furthermore, this invention also solves the technical problems of improper handling of multivariable coupling relationships and poor adaptability of control strategies in existing hydraulic control systems. Traditional control methods typically simplify the control problem of each hydraulic cylinder into independent single-variable control, ignoring the coupling relationships between cylinders and the overall characteristics of the system. When dealing with the coordinated control of multivariables such as pressure, displacement, and velocity, there is a lack of effective coupling relationship modeling and processing mechanisms, resulting in poor adaptability of the control strategy and the inability to automatically adjust control parameters according to changes in system state. This invention establishes a complete multivariable coupling relationship model through multi-scale matrix mapping, utilizes an improved Gaussian kernel function to handle complex nonlinear coupling relationships in a high-dimensional feature space, and dynamically adjusts control parameters according to real-time state through an adaptive load balancing control strategy, achieving adaptive optimization of multivariable coordinated control. This significantly improves the adaptability of the control system to different operating conditions and the consistency of control performance.

[0064] Specifically, the principle of this invention is as follows: The technical principle behind solving the problem of insufficient precision in multi-cylinder synchronous control lies in establishing a complete distributed coordinated control theoretical system and a precise nonlinear mapping algorithm. A multi-parameter real-time monitoring network, by installing high-precision pressure and displacement sensors on each hydraulic cylinder and deploying force sensors at key locations on the tower, achieves comprehensive real-time acquisition of system state parameters, providing a reliable data foundation for precise control and solving the control blind spot problem caused by insufficient sensor information in traditional technologies. The distributed consensus coordinated control algorithm treats each hydraulic cylinder as an independent intelligent node, constructing a Laplace matrix to describe the information interaction relationship between cylinders, establishing a decentralized coordinated control architecture, enabling each node to achieve globally optimal control based on local information interaction, avoiding the computational bottleneck and single-point failure risk of centralized control. Simultaneously, a consensus protocol ensures that the states of each cylinder gradually converge, achieving a high-precision synchronous control effect. The multi-scale matrix mapping hierarchical processing system, with its step-by-step mapping from the single-cylinder scale to the system scale, combined with a compressed sensing reconstruction mechanism utilizing the system state sparsity assumption, can accurately reconstruct complete system state information from limited observation data, significantly improving the information processing accuracy and computational efficiency of the control algorithm. The improved Gaussian kernel function, by introducing the exponential decay characteristics of pressure and displacement difference terms, accurately characterizes the similarity relationship between hydraulic cylinders in a high-dimensional feature space. This gives hydraulic cylinders with similar pressure and displacement higher coordination weights, forming a precise synchronization control strategy that effectively solves the problem that traditional linear control methods cannot handle the nonlinear characteristics of multi-cylinder coupling. The adaptive load balancing control strategy monitors the stress distribution on the tower body in real time and dynamically adjusts the output force gain of each hydraulic cylinder according to the degree of stress imbalance, ensuring the uniformity of stress on the tower structure and further improving the accuracy of multi-cylinder synchronization control. Lyapunov stability analysis provides a theoretical guarantee for the control system, ensuring the convergence of the algorithm and the stability of the control process, enabling the control scheme of this invention to achieve high-precision multi-cylinder synchronization control in both theory and practice.

[0065] The following provides a specific embodiment 1 of the present invention, and the specific implementation of each step in this embodiment 1 is described in detail below.

[0066] In this embodiment, the specific implementation of step S01 is the same as described above, and will not be repeated in detail here.

[0067] The specific implementation of step S02 is based on constructing a distributed control network topology based on graph theory, and establishing an adjacency matrix A to describe the communication connection relationship between hydraulic cylinders, as shown below:

[0068]

[0069] In the formula, A is an n×n dimensional adjacency matrix; aij is a matrix element, where a = 1 when there is a direct communication connection between the i-th hydraulic cylinder and the j-th hydraulic cylinder, otherwise a ij = 0; n is the total number of hydraulic cylinders. Calculate the degree matrix D based on the adjacency matrix, which is specifically expressed as follows: ij = 0; n is the total number of hydraulic cylinders. Calculate the degree matrix D based on the adjacency matrix, which is specifically expressed as follows:

[0070]

[0071] In the formula, D is an n×n dimensional matrix; is the degree of the i-th node. Further calculate the Laplacian matrix L, which is specifically expressed as follows:

[0072] L = D - A;

[0073] In the formula, L is an n×n dimensional Laplacian matrix, and the second smallest eigenvalue of this matrix determines the convergence speed of the distributed consensus algorithm.

[0074] The specific implementation of step S03 is to establish a multi-level data processing mechanism and use the compressive sensing reconstruction algorithm to reconstruct the complete system state from partial observed data. The reconstruction model is specifically expressed as follows:

[0075] y = Φs + n;

[0076] In the formula, y is an m-dimensional observed vector, containing the state measurement values of some hydraulic cylinders; Φ is an m×k dimensional observation matrix, describing the mapping relationship between the observation and the original signal; s is a k-dimensional sparse signal vector, representing the representation of the system in the sparse domain; n is an m-dimensional noise vector, where m < k, m is the dimension of the actual observed data, and k is the dimension of the original signal. The sparse reconstruction optimization problem is specifically expressed as follows:

[0077] min s ||s||1 s.t. ||y - Φs||2 ≤ ∈;

[0078] In the formula, ||s||1 is the L1 norm, used for sparse constraint; ||y - Φs||2 is the L2 norm, representing the reconstruction error; ∈ is the reconstruction error threshold, set to 5% of the original signal energy, and is calculated by ∈ = 0.05×||s original ||2, where s original is the original signal vector.

[0079] The specific implementation of step S04 is to use an improved Gaussian kernel function to achieve non-linear feature mapping. The improved Gaussian kernel function is specifically expressed as follows:

[0080]

[0081] In the formula, K(x i , x j) represents the kernel function value between the i-th and j-th hydraulic cylinders; x i and x j These are the state vectors of the i-th and j-th hydraulic cylinders, respectively; p i and p j d represents the pressure values ​​of the i-th and j-th hydraulic cylinders, respectively, in MPa; i and d j σ represents the displacement of the i-th and j-th hydraulic cylinders, respectively, in mm; σ is the kernel function bandwidth parameter, set to 1.5 times the standard deviation of the state vector; α is the pressure difference weighting coefficient, ranging from 0.1 to 0.3; β is the displacement difference weighting coefficient, ranging from 0.05 to 0.2. The pressure deviation coefficient η is calculated based on the kernel function. p Specifically, it is expressed as follows:

[0082]

[0083] In the formula, η p This is the pressure deviation coefficient; The average pressure of each hydraulic cylinder; n is the total number of hydraulic cylinders. Displacement synchronization metric γ. d Specifically, it is expressed as follows:

[0084]

[0085] In the formula, γ d This is the displacement synchronization metric; d target The target displacement is expressed in mm; n is the total number of hydraulic cylinders.

[0086] The specific implementation method of step S05 is the same as described above, and will not be repeated in detail here.

[0087] The specific implementation of step S06 is to establish an adaptive load balancing control mechanism, where the output force F of each hydraulic cylinder is... i Based on the pressure and piston area, the calculations are as follows:

[0088] F i =p i ×S i ;

[0089] In the formula, F i p represents the output force of the i-th hydraulic cylinder, in kN. i S represents the pressure value of the i-th hydraulic cylinder, in MPa. i The piston area of ​​the i-th hydraulic cylinder is given in meters. 2 The specific formula for calculating the unbalanced strength ξ of the tower body is as follows:

[0090]

[0091] In the formula, ξ is the unbalanced strength, which is dimensionless; ξ represents the average output force of each hydraulic cylinder; n is the total number of hydraulic cylinders. The threshold value for force imbalance strength is ξ. th The specific calculation equation is expressed as follows:

[0092]

[0093] In the formula, ξ th k is the unbalanced stress strength threshold, dimensionless. s The safety factor, dimensionless, is 1.5 and is used to ensure the safety margin of the control system; W tower W represents the total weight of the tower, expressed in kN, obtained through design parameters and actual measurements. wind The wind load is expressed in kN and is calculated according to the building structure load code based on the local wind speed and the windward area of ​​the tower. F max This represents the maximum output force of a single hydraulic cylinder, expressed in kN. Standard control threshold ξ std The specific calculation formula is expressed as follows:

[0094] ξ std =0.5×ξ th ;

[0095] In the formula, ξ std The standard control threshold is dimensionless. Force-adjusted gain ΔG i The specific calculation equation is expressed as follows:

[0096]

[0097] In the formula, ΔG i k is the force adjustment gain of the i-th hydraulic cylinder, dimensionless; g This is the gain adjustment coefficient, dimensionless, with a value of 0.2, used to control the adjustment intensity.

[0098] The specific implementation of step S07 is to achieve precise synchronous control through a distributed consensus coordination control algorithm, and the consistency error e of each hydraulic cylinder is minimized. i The specific calculation formula is expressed as follows:

[0099] e i =x i -x sync ;

[0100] In the formula, e i Let x be the consistency error of the i-th hydraulic cylinder; i The current state of the i-th hydraulic cylinder; x sync This represents the system's target synchronization state. The distributed control law is specifically expressed as follows:

[0101]

[0102] In the formula, Let a be the rate of change of state of the i-th hydraulic cylinder; ij For elements of the adjacency matrix; u i This is the control input for the i-th hydraulic cylinder. Target synchronization state x sync The specific calculation formula is expressed as follows:

[0103]

[0104] In the formula, x sync Let n represent the system's target synchronization state; n is the total number of hydraulic cylinders. A Lyapunov function V is constructed to verify the system's stability, specifically represented as follows:

[0105]

[0106] In the formula, V is the Lyapunov function; when The system asymptotically converges. The time derivative of the Lyapunov function. Specifically, it is expressed as follows:

[0107]

[0108] In the formula, This is the time derivative of the Lyapunov function, and its non-positive derivative ensures system stability.

[0109] It's important to explain that the adjacency matrix construction principle is based on connected graph theory. It describes the communication topology between hydraulic cylinder nodes using binary matrix elements, improving system fault tolerance and information transmission efficiency compared to traditional star topologies. A failure in a single communication link will not paralyze the entire network. The Laplace matrix is ​​constructed based on the degree concept and adjacency relationships of a graph. The eigenvalue distribution of this matrix directly determines the convergence performance of the distributed consensus algorithm; a larger second-smallest eigenvalue indicates faster convergence. This eliminates the risk of single-point failures compared to traditional centralized control methods.

[0110] Compressed sensing reconstruction models, based on signal sparsity theory, utilize the sparsity characteristics of hydraulic system states in a certain transform domain to accurately reconstruct high-dimensional original signals from fewer observations. Compared to traditional full-sampling methods, this significantly reduces data transmission and storage requirements while maintaining reconstruction accuracy. The L1 norm minimization optimization problem, based on convex optimization theory, ensures the uniqueness and stability of the solution through sparsity constraints. Compared to L2 norm optimization, it better preserves the sparse structural characteristics of the signal.

[0111] The improved Gaussian kernel function achieves multi-dimensional similarity measurement through the product of three exponential terms. The first term, based on Euclidean distance, reflects the overall state similarity and is expressed as:

[0112]

[0113] The second and third terms are weighted based on the differences in pressure and displacement, respectively, and are expressed as follows:

[0114] exp(-α|p i -p j |)·exp(-β|d i -d j |);

[0115] Compared to the standard Gaussian kernel function, this method can simultaneously address both pressure and displacement consistency control objectives, significantly improving the accuracy of multi-cylinder synchronous control. The pressure deviation coefficient is defined mathematically as the coefficient of variation, quantifying the uniformity of pressure distribution by the ratio of the standard deviation to the mean. Its calculation formula is as follows:

[0116]

[0117] It has dimensionless characteristics and better comparability compared to the absolute deviation index.

[0118] The calculation of unbalanced stress strength is based on the concept of coefficient of variation in statistics. It quantifies the degree of unevenness in stress distribution by the ratio of the standard deviation to the mean of the output force of each cylinder. The calculation expression is as follows:

[0119]

[0120] This indicator provides a more comprehensive reflection of the overall stress state compared to the traditional maximum-minimum difference method. The stress imbalance strength threshold calculation considers the combined effects of tower weight, wind load, and hydraulic cylinder rated capacity, and ensures the reliability of the control strategy through a safety factor. Its expression is as follows:

[0121]

[0122] It has better adaptability compared to the fixed threshold method.

[0123] The distributed control law is based on the consensus theory of multi-agent systems. Each hydraulic cylinder node only needs to exchange information with its neighboring nodes to achieve global coordination. The control law expression is:

[0124]

[0125] Compared to centralized control methods, it offers greater scalability and fault tolerance. The Lyapunov function describes the system energy in quadratic form, and its expression is:

[0126]

[0127] By constructing a positive definite function and proving the non-positive nature of its derivative, the asymptotically stable convergence of the system is ensured, and its time derivative is:

[0128]

[0129] Compared to traditional stability analysis methods, this method provides rigorous mathematical proof and convergence guarantee. The negative definite derivative property of this function ensures the monotonically decreasing synchronization error of the multi-cylinder system, thereby achieving precise synchronization control.

[0130] To better understand and implement the present invention, the following is a specific application scenario of the present invention, Example 2: A technical team needs to design and implement a multi-cylinder synchronous precision control system for a QTZ80 tower crane. The crane is equipped with 6 hydraulic lifting cylinders, each with a working pressure of 16MPa and a stroke of 1200mm, for the section lifting operation of the tower body.

[0131] The technical team first installed pressure and displacement sensors on each hydraulic cylinder. The pressure sensors had a range of 0–25 MPa and an accuracy of ±0.1%, while the displacement sensors had a range of 0–1500 mm and an accuracy of ±0.05 mm. Simultaneously, strain gauge force sensors with a range of 0–500 kN and an accuracy of ±0.2% were placed at 12 key locations on the tower structure. A multi-parameter real-time monitoring network was established to collect data on the pressure values, displacements, and stress distribution of each hydraulic cylinder at a frequency of 100 Hz, enabling real-time monitoring of the system's operating status.

[0132] Based on a distributed consensus coordination control algorithm, the technical team constructed a 6×6 adjacency matrix to describe the communication topology between hydraulic cylinders. Each hydraulic cylinder acts as an independent control node, exchanging data via a CAN bus at a communication rate of 500kbps. The Laplace matrix was constructed using a symmetrical connection method to ensure bidirectional information transmission between nodes, forming a decentralized coordination control architecture.

[0133] In the hierarchical processing system of multi-scale matrix mapping, the technical team started with a 6-dimensional state vector at the single-cylinder scale and mapped it step by step to a 36-dimensional state matrix at the system scale. Using a compressed sensing reconstruction mechanism, a complete 36-dimensional original state matrix was reconstructed from a 24-dimensional observation matrix, with a reconstruction accuracy of 98.5%. The sparsity assumption is based on the physical characteristics of hydraulic systems, namely that under normal operating conditions, most state parameters change only slightly, with only a few key parameters showing significant changes.

[0134] The improved Gaussian kernel function parameters were set to σ = 0.8, α = 0.15, and β = 0.12, mapping the original pressure-displacement matrix to a 128-dimensional high-dimensional feature space for processing. Through kernel function transformation, the pressure deviation coefficient and displacement synchronization metric between each hydraulic cylinder were calculated. In actual operation, when the system is in a stable state, the pressure deviation coefficient typically remains around 0.03, and the displacement synchronization metric remains above 0.96.

[0135] Based on the real-time changes in the pressure deviation coefficient, the technical team implemented a dynamic frequency adjustment strategy. When the pressure deviation coefficient is in the range [0, 0.05], the system reduces the pressure regulation processing frequency from 50Hz to 30Hz to reduce unnecessary control intervention. When the pressure deviation coefficient is in the range (0.05, 0.15], the standard adjustment frequency of 50Hz is maintained. When the pressure deviation coefficient is in the range (0.15, 0.30], the adjustment frequency is increased to 75Hz to enhance the system's response speed and control accuracy.

[0136] The technical team calculated the unbalanced load threshold of the tower body using the tower weight (650kN), current wind load (85kN), and force distribution data for each hydraulic cylinder, determining the unbalanced load intensity threshold to be 45kN. When the detected unbalanced load signal intensity reaches 52kN, exceeding the set threshold, the system automatically activates the adaptive load balancing control mode. The force adjustment gain calculation equation determined the gain range to be 0.85–1.15. Since the current unbalanced load intensity of 52kN falls within the control range corresponding to this gain range, the system differentially adjusts the output force of each cylinder: cylinder 1 to 1.08, cylinder 2 to 0.92, cylinder 3 to 1.12, cylinder 4 to 0.88, cylinder 5 to 1.05, and cylinder 6 to 0.95.

[0137] In the flow distribution control stage, the technical team precisely adjusted the flow distribution of each hydraulic cylinder using proportional valves. The proportional valves used in the system have a response time of 15ms and a flow adjustment accuracy of ±2%. The distributed consensus coordination control algorithm calculated the target synchronization state as follows: the average target pressure of each cylinder is 15.8MPa, and the target displacement synchronization accuracy is ±1.2mm. Lyapunov stability analysis was used to verify the asymptotic convergence of the control system, ensuring that each hydraulic cylinder can reach synchronization within 45s.

[0138] In actual jacking operations, when a new standard section of the tower needed to be jacked, the technical team activated the multi-cylinder synchronous control program. Initially, the pressures of the six hydraulic cylinders were 15.2 MPa, 15.6 MPa, 15.9 MPa, 15.4 MPa, 15.7 MPa, and 15.3 MPa, with displacements of 856 mm, 851 mm, 863 mm, 849 mm, 858 mm, and 854 mm, respectively. Through real-time data acquisition via a multi-parameter monitoring network, the calculated pressure deviation coefficient was 0.087, and the displacement synchronization measurement was 0.943.

[0139] Since the pressure deviation coefficient is within the range of (0.05, 0.15), the system maintains a standard adjustment frequency of 50Hz. The distributed coordinated control algorithm starts working, and each hydraulic cylinder node exchanges status information through the communication topology defined by the adjacency matrix. After 23 seconds of coordinated control, the pressure of each cylinder gradually becomes consistent, eventually stabilizing within the range of 15.5 ± 0.03 MPa, with a displacement synchronization accuracy of ± 0.8 mm, meeting the control requirement of ± 1.2 mm. Figure 2 As shown, the pressure and displacement changes of each hydraulic cylinder exhibit good convergence characteristics throughout the entire synchronous control process.

[0140] like Figure 3 As shown, the stress distribution on the tower body remained relatively uniform throughout the entire lifting process, with the maximum stress imbalance intensity at 38kN, lower than the set threshold of 45kN. Therefore, the system adopted a standard synchronous control mode. The multi-scale matrix mapping system successfully extracted feature information at each level, and the accuracy of compressed sensing reconstruction remained above 98.2%, ensuring the accuracy of control decisions.

[0141] To verify the system's performance under complex operating conditions, the technical team conducted tests under level 4 wind load. When the wind load increased to 120kN, the tower's unbalanced stress intensity rose to 56kN, exceeding the 45kN threshold, and the system automatically switched to adaptive load balancing control mode. The gain range of the force adjustment gain calculation equation was 0.78–1.22, and the output force gains of each cylinder were adjusted to 1.18, 0.82, 1.21, 0.79, 1.15, and 0.85, respectively. Through precise gain adjustment, the tower's unbalanced stress intensity decreased to 42kN within 35 seconds, returning to the standard control threshold range.

[0142] The monitoring data statistics are shown in Table 1:

[0143] Table 1. Statistics of system control performance parameters under different operating conditions

[0144]

[0145] In the high-precision synchronous control test, the technical team set stricter control targets, requiring displacement synchronization accuracy to reach ±0.5mm and pressure consistency control within ±0.02MPa. By optimizing and improving the parameter settings of the Gaussian kernel function, adjusting α to 0.22 and β to 0.18, the system's sensitivity to subtle deviations was enhanced. Test results showed that under this parameter configuration, the system could achieve the required high-precision synchronous control target within 52s, reducing the pressure deviation coefficient to 0.021 and improving the displacement synchronization metric to 0.984. Figure 4 As shown, the improved Gaussian kernel function maps the state information of each hydraulic cylinder to a high-dimensional feature space, forming a clear cluster distribution, which effectively improves the accuracy of synchronous control.

[0146] The dynamic adjustment process of the hydraulic cylinder output force is shown in Table 2:

[0147] Table 2. Output force adjustment of each cylinder during adaptive load balancing control.

[0148]

[0149] In the system fault tolerance test, when the pressure sensor of hydraulic cylinder No. 2 malfunctioned, the distributed coordinated control algorithm automatically reconstructed the communication topology and estimated the state parameters of cylinder No. 2 through information exchange with neighboring nodes. After recalculating the Laplace matrix, the system regained synchronous control state within 38 seconds, demonstrating the excellent fault tolerance capability of the decentralized architecture.

[0150] Throughout the implementation process, the system demonstrated excellent energy consumption control. The precise flow distribution of the proportional valve reduced hydraulic pump power loss, resulting in an average energy consumption reduction of approximately 12% compared to traditional control methods. The hydraulic oil temperature was maintained within a reasonable range of 42–48°C, ensuring stable system thermal equilibrium. The successful implementation of multi-cylinder synchronous control ensured the safety and reliability of the tower lifting process, achieving millimeter-level lifting accuracy and meeting the stringent requirements of the project.

[0151] Compared to traditional centralized control methods, the distributed consensus coordination control algorithm employed in this invention fundamentally changes the control concept of multi-cylinder coordination. Traditional methods rely on a central controller for unified scheduling, which carries the risk of single-point failures and whose computational complexity increases exponentially with the number of cylinders. In contrast, the distributed algorithm achieves global optimization through local information interaction, possessing inherent fault tolerance and scalability. Compared to the standard kernel function, the improved Gaussian kernel function, by introducing pressure and displacement difference terms, can more accurately characterize the coupling relationship between hydraulic cylinders in a high-dimensional feature space, significantly improving the accuracy of nonlinear mapping. The multi-scale matrix mapping system overcomes the limitations of traditional single-scale processing, achieving progressive optimization from local to global through hierarchical information extraction and fusion, effectively solving the dimensionality curse problem of multi-cylinder systems. The compressed sensing reconstruction mechanism utilizes the sparse characteristics of the hydraulic system state, significantly reducing the number of sensors and data transmission volume required, reducing system complexity and cost while ensuring control accuracy. The adaptive load balancing control strategy dynamically adjusts the output of each cylinder according to the actual force distribution of the tower, overcoming the shortcomings of traditional uniform distribution methods that cannot adapt to structural asymmetry and external load changes, achieving truly intelligent coordinated control.

[0152] It should be noted that the variables involved in this invention are explained in detail in Table 3.

[0153] Table 3. Variable Explanation Table

[0154]

[0155]

[0156] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for precise synchronous control of multiple cylinders in a tower crane hydraulic jacking system, characterized in that, Data on pressure values, displacement, and tower stress distribution of each hydraulic cylinder are collected. A distributed consensus coordination control algorithm is used to construct the communication topology of the hydraulic cylinder network nodes. An adjacency matrix is ​​used to describe the information interaction between cylinders, establishing a decentralized coordination control architecture. A hierarchical processing system for multi-scale matrix mapping is established, and a compressed sensing reconstruction mechanism for sparse matrices is constructed to reconstruct the m-dimensional original state matrix from the n-dimensional observation matrix. Based on an improved Gaussian kernel function-based nonlinear matrix mapping algorithm, the pressure deviation coefficient and displacement synchronization metric between each hydraulic cylinder are calculated. When the pressure deviation coefficient value is in the interval [0, 0.05], the pressure regulation processing frequency is reduced to 60% of the original frequency. When the pressure deviation coefficient value is in the interval (0.05, 0.15], the current pressure regulation processing frequency is maintained. When the signal strength is within the range (0.15, 0.30), the pressure regulation processing frequency is increased to 150% of the original frequency. The unbalanced force threshold is calculated using the unbalanced force threshold calculation equation. If the unbalanced force signal strength is greater than the unbalanced force threshold, adaptive load balancing control is enabled. The unbalanced force gain range is calculated using the unbalanced force gain calculation equation. If the unbalanced force strength is within the unbalanced force gain range, the output force gain of each cylinder is adjusted. If the unbalanced force strength is less than the standard control threshold, the standard synchronous control mode is adopted. The flow distribution of each hydraulic cylinder is adjusted by the proportional valve. The target synchronization state is calculated based on the distributed consistency coordination control algorithm. Lyapunov stability analysis is used to ensure system convergence, thereby achieving precise synchronous control of the pressure and displacement of each hydraulic cylinder.

2. The multi-cylinder synchronous precision control method for the hydraulic jacking system of a tower crane according to claim 1, characterized in that, The multi-parameter real-time monitoring network is specifically a data acquisition network composed of sensors distributed at key locations in the hydraulic jacking system, used to acquire system operating status parameters in real time; the hierarchical processing system of multi-scale matrix mapping specifically achieves multi-level extraction of pressure and displacement information through step-by-step mapping from the single-cylinder scale to the system scale.

3. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 2, characterized in that, The distributed consensus coordination control algorithm is specifically a graph theory-based multi-agent coordination control method that treats each hydraulic cylinder as a network node and achieves global coordination control through information interaction between neighboring nodes.

4. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 3, characterized in that, The distributed consensus coordination control algorithm describes the communication topology between hydraulic cylinders by constructing a Laplace matrix and designs a distributed control law based on a consensus protocol, so that the pressure and displacement states of each cylinder gradually become consistent.

5. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 4, characterized in that, The distributed consensus coordination control algorithm uses the pressure value, displacement, and speed state parameters of each hydraulic cylinder as input, calculates the consistency error of each node through Laplace matrix operation, obtains the control gain parameter by combining stability analysis, and outputs the target pressure and flow control commands for each cylinder.

6. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 5, characterized in that, The adjacency matrix is ​​a mathematical matrix used to describe the communication connection relationship between hydraulic cylinders. The matrix elements indicate whether there is an information exchange channel between the corresponding cylinders.

7. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 6, characterized in that, The multi-scale matrix mapping is specifically an information processing method that uses matrix transformations at different scale levels to achieve multi-level feature extraction and fusion of data.

8. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 7, characterized in that, The compressed sensing reconstruction mechanism is a data reconstruction method based on signal sparsity theory.

9. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 8, characterized in that, The calculation equation for the unbalanced stress threshold of the tower body is used to calculate the threshold for determining the unbalanced stress intensity based on the tower body structural parameters and working conditions. The inputs include the tower body weight, wind load, and force distribution data of each hydraulic cylinder, and the output is the unbalanced stress intensity threshold.

10. The multi-cylinder synchronous precise control method for the hydraulic jacking system of a tower crane according to claim 9, characterized in that, The force adjustment gain calculation equation is used to calculate the gain range of each hydraulic cylinder's output force adjustment. The inputs include the current output force of each cylinder, the degree of force imbalance in the tower body, and the system safety factor. The output is the force adjustment gain range.

Images