Electrical synchronous control system and control method for steel corridor hydraulic pump source

Abstract

The present application relates to the technical field of electrical automatic control, and discloses an electrical synchronous control system and control method for a steel corridor hydraulic pump source, comprising: collecting the surface temperature field of the pipe network to construct a pipe network viscosity field representing fluid resistance, and calculating the delay amount and delay gradient of the instruction transmission according to the pipe network viscosity field, so as to expand the control optimization space; then, extracting the corridor torsion and the viscous dissipation rate in the pipe to evaluate the particle solution, and intervening the algorithm optimization energy according to the local and global viscosity proportion; finally, eliminating the asynchronous deviation by aligning the historical time sequence, extracting the convergent optimal solution, and fusing the delay change rate to reconstruct the control duty cycle of the variable frequency motor. The present application effectively overcomes the interference of complex temperature difference on the consistency of hydraulic transmission, and significantly improves the synchronization accuracy of multi-pump collaborative lifting.

Description

Technical Field

[0001] This invention relates to the field of electrical automatic control technology, and more specifically, to an electrical synchronous control system and control method for a hydraulic pump source for a steel connecting corridor. Background Technology

[0002] In the construction and operation of modern large-scale buildings, a cluster of multiple hydraulic pumps and jacks is typically used for synchronous lifting or sliding operations. Existing hydraulic synchronization systems mostly rely on fixed preset parameters for closed-loop feedback regulation. However, when working in suspended environments at high altitudes, hydraulic pipelines are highly susceptible to asymmetrical sunlight and ambient temperature differences, leading to extreme temperature gradients within the pipeline network and causing the viscous resistance of the hydraulic oil to exhibit a spatially non-uniform distribution. This abrupt change in rheological resistance causes nonlinear delays and misalignments in the response of each actuator to control commands. Traditional control schemes cannot isolate and compensate for the timing deviations caused by uneven fluid viscosity in real time, easily leading to multi-point synchronous motion misalignment, which in turn causes serious hidden dangers such as torsion deformation, stress imbalance, and even fatigue damage to the connecting corridor itself. Summary of the Invention

[0003] This invention provides an electrical synchronous control system and control method for the hydraulic pump power source of steel connecting corridors, which solves the technical problems mentioned in the background art.

[0004] Firstly, an electrical synchronization control method for the hydraulic pump source of a steel connecting corridor is applied to an electrical synchronization control system. This system includes a fiber optic sensor array, a position sensor, a hybrid hydraulic pipeline network, a hydraulic pump source group, hydraulic jacks, and a variable frequency motor, comprising: The surface temperature field of the hybrid hydraulic pipeline network is collected by the fiber optic sensing array, and combined with preset oil parameters, a pipeline viscosity field characterizing the fluid transmission resistance is constructed. Based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network, the fluid delay in fluid transmission to each hydraulic jack is calculated. Calculate the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronous control. The torsion of the connecting corridor is obtained by the posture sensor, and the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network is combined to calculate the fitness value for evaluating the optimization particles. Based on the viscosity ratio characteristics of local pipelines and global pipelines, an inertial weight is constructed to suppress premature convergence of the algorithm and to intervene in the optimization speed of the optimization particles. Based on the fluid delay corresponding to each hydraulic pump source, the historical evaluation state is time-aligned to eliminate physical asynchronous deviations and update the global optimal solution. Extract the converged solution at the end of the iteration, combine it with the delay change rate of the fluid delay to generate feedforward compensation, and reconstruct the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

[0005] Secondly, the electrical synchronization control system for the hydraulic pump source of the steel connecting corridor, in executing the electrical synchronization control method for the hydraulic pump source of the steel connecting corridor as described in any one of the claims, includes: The pipeline viscosity field construction module collects the surface temperature field of the hybrid hydraulic pipeline through the fiber optic sensing array, and constructs a pipeline viscosity field characterizing the fluid transmission resistance by combining it with preset oil parameters. The fluid delay calculation module calculates the fluid delay transmitted to each hydraulic jack based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network. The particle swarm initialization module calculates the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronization control. The fitness evaluation module obtains the torsion of the connecting corridor through the pose sensor and calculates the fitness value for evaluating the optimization particles by combining the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network. The optimization control module constructs an inertial weight to suppress premature convergence of the algorithm based on the viscosity ratio characteristics of local pipelines and global pipelines, thereby intervening in the optimization speed of the optimization particles. The iterative update module performs time-series alignment of historical evaluation states based on the fluid delay amount corresponding to each hydraulic pump source, eliminating physical asynchrony deviations to update the global optimal solution; The duty cycle calculation module extracts the converged solution after the iteration ends, generates feedforward compensation by combining the delay change rate of the fluid delay, and reconstructs the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

[0006] The beneficial effects of this invention are as follows: By collecting pipeline temperature and reconstructing viscosity distribution, this scheme effectively quantifies the non-uniform delay disturbance caused by rheological resistance to command transmission; simultaneously, by extracting the torsional magnitude of the connecting corridor and the fluid energy dissipation rate for joint optimization, and introducing adaptive attenuation intervention and timing alignment mechanisms, it effectively eliminates misalignment deviations during asynchronous collaborative execution of multiple nodes. This invention significantly reduces synchronization errors caused by differences in solar radiation temperature or uneven flow resistance in the pipeline network, avoids the risk of stress imbalance or resonance tearing of the connecting corridor body, and effectively improves the control accuracy and response consistency of the hydraulic jack cluster. Attached Figure Description

[0007] Figure 1 This is a flowchart of the electrical synchronization control method for the hydraulic pump source of the steel connecting corridor according to the present invention. Detailed Implementation

[0008] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0009] Example 1: As Figure 1 As shown, an electrical synchronization control method for the hydraulic pump source of a steel connecting corridor is applied to an electrical synchronization control system. The system includes a fiber optic sensor array, a position sensor, a hybrid hydraulic pipeline network, a hydraulic pump source group, hydraulic jacks, and a variable frequency motor. The surface temperature field of the hybrid hydraulic pipeline network is collected by the fiber optic sensing array, and combined with preset oil parameters, a pipeline viscosity field characterizing the fluid transmission resistance is constructed. Based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network, the fluid delay in fluid transmission to each hydraulic jack is calculated. Calculate the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronous control. The torsion of the connecting corridor is obtained by the posture sensor, and the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network is combined to calculate the fitness value for evaluating the optimization particles. Based on the viscosity ratio characteristics of local pipelines and global pipelines, an inertial weight is constructed to suppress premature convergence of the algorithm and to intervene in the optimization speed of the optimization particles. Based on the fluid delay corresponding to each hydraulic pump source, the historical evaluation state is time-aligned to eliminate physical asynchronous deviations and update the global optimal solution. Extract the converged solution at the end of the iteration, combine it with the delay change rate of the fluid delay to generate feedforward compensation, and reconstruct the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

[0010] This embodiment is applied to an electrical synchronous control system that includes a fiber optic sensor array, a posture sensor, a hybrid hydraulic pipeline network, a hydraulic pump source group, hydraulic jacks, and a variable frequency motor. The fiber optic sensor array is installed to cover the entire area along the layout path of the hybrid hydraulic pipeline network. The posture sensor is deployed at the key structural nodes of the steel connecting corridor and at each jacking point. Each hydraulic pump source in the hydraulic pump source group is sequentially connected to the corresponding variable frequency motor, the hybrid hydraulic pipeline network, and the hydraulic jacks. The jacking end of each hydraulic jack is rigidly connected to the corresponding jacking point of the steel connecting corridor.

[0011] The fiber optic sensing array uses a distributed fiber Bragg grating temperature measurement array, with a temperature measurement accuracy of no less than [missing information]. Response time no greater than The sampling wavelength range is to Temperature measurement range is to The spacing of the fiber optic sensor array is positively correlated with the pipe length, pipe diameter, and ambient temperature gradient. The spacing range is as follows: to The spacing between pipe bends, reducers, valves, and other throttling elements is reduced to the original spacing. This allows for precise acquisition of local temperature fields. Each sensing unit in the fiber optic sensing array corresponds to a unique pipeline spatial node, and the spatial coordinates of each sensing unit are mapped one-to-one with the three-dimensional coordinates of the corresponding pipeline node. The mapping relationship is completed through on-site calibration.

[0012] The pose sensor employs a six-degree-of-freedom inertial measurement unit in conjunction with a GNSS differential positioning module, achieving a displacement measurement accuracy of no less than [a certain level]. The angle measurement accuracy is no less than The sampling frequency is not lower than One set of posture sensors is installed at each jacking point of the steel connecting corridor, and two sets are installed at the mid-span and end supports of the steel connecting corridor. The coordinate system of all posture sensors is mapped to the overall structural coordinate system of the steel connecting corridor. The origin of the coordinate system is defined as the lower left corner of one end support of the steel connecting corridor. The X-axis is along the length of the steel connecting corridor, the Y-axis is along the width of the steel connecting corridor, and the Z-axis is along the vertical jacking direction.

[0013] The fiber optic sensor array, posture sensor, and variable frequency motor drive system adopt the IEEE 1588PTP precision time protocol for hard synchronization. The master clock is deployed on the system control host, and the clock synchronization error of all slave devices does not exceed [a certain value]. The synchronization trigger signal uses hardware level triggering to ensure strict alignment of the timestamps of all acquired data and control commands. The system control cycle range is [value missing]. to The sampling period is consistent with the control period, and the entire process execution time of data acquisition, algorithm calculation, and control output does not exceed the control period. This is to meet the requirements of strong real-time control.

[0014] The mathematical model of the controlled object of the system comprises two parts: a hydraulic system model and a dynamic model of the steel connecting corridor structure. The hydraulic system model includes the flow characteristic model of the variable frequency motor and hydraulic pump source, the fluid dynamics model of the hydraulic pipeline, and the load-displacement characteristic model of the hydraulic jack. The mapping relationship between the speed of the variable frequency motor and the output flow rate of the hydraulic pump source is as follows:

[0015] In the formula, Let be the output flow rate of the hydraulic pump source at time t, in units of . ; Let t be the speed of the variable frequency motor at time t, in units of ; This refers to the nominal displacement of the hydraulic pump, in units of... ; The volumetric efficiency of the hydraulic pump, with a value range of [value missing]. to .

[0016] The fluid dynamics model of the hydraulic pipeline is constructed based on the one-dimensional unsteady flow equation, considering the compressibility of the fluid and the elastic deformation of the pipeline. The pressure and flow rate transmission relationship of the fluid in the pipeline is as follows:

[0017]

[0018] In the formula, This represents the pressure of the fluid inside the pipeline, expressed in Pa. The path coordinates are along the pipeline axis, in units of ; The rated density of the hydraulic oil, in units of... The range of values ​​is to ; The axial velocity of the fluid within the pipeline, in units of... ; Physical time, in units of ; The dynamic viscosity of hydraulic oil, in units of... ; This refers to the inner diameter of the pipe, in units of... ; The equivalent bulk modulus of hydraulic oil, in units of... The range of values ​​is to .

[0019] The load-displacement characteristic model of a hydraulic jack is as follows:

[0020]

[0021] In the formula, Let t be the piston displacement of the hydraulic jack at time t, in meters (m). The flow rate of the hydraulic jack input at time t, in units of... ; The effective working area of ​​the hydraulic jack piston, in units of... ; The output thrust of the hydraulic jack at time t is expressed in N. The pressure in the rodless chamber of the hydraulic jack at time t is expressed in Pa. The equivalent load mass borne by the hydraulic jack, expressed in kg; The viscous damping coefficient of the load, in units of . ; The stiffness coefficient of the load, in units of ; This represents the static friction force of the hydraulic jack, measured in N.

[0022] The dynamic model of the steel connecting corridor structure is constructed using multi-degree-of-freedom dynamic equations obtained through finite element discretization. The equations are in the following form:

[0023] In the formula, The mass matrix of the steel connecting corridor has the following dimensions. N is the number of finite element nodes in the steel connecting corridor structure; The damping matrix of the steel connecting corridor has the following dimensions. ; Here is the stiffness matrix of the steel connecting corridor, with dimension . ; Let be the displacement vector of each node of the steel connecting corridor, with dimension . ; The velocity vector has a dimension of ; Let be the acceleration vector, with dimension . ; The lifting force vector applied to each hydraulic jack, with dimension . .

[0024] The system software adopts a modular design, including a data acquisition module, a data preprocessing module, a core algorithm module, a control output module, and a fault diagnosis and protection module. Each module is implemented using a deterministic state machine architecture. The state machine includes six core states: initialization state, data acquisition state, parameter calculation state, algorithm iteration state, control output state, and fault protection state. The switching trigger conditions and execution logic of each state are as follows: The initialization state is the first execution state after the system is powered on. It completes hardware device initialization, parameter loading, clock synchronization, sensor calibration, and historical buffer sequence initialization. After initialization is completed, it automatically switches to the data acquisition state. In the data acquisition state, the temperature data of the fiber optic sensor array, the pose data of the pose sensor, the speed data of the variable frequency motor, and the pressure and flow data of the hydraulic system are collected simultaneously. After the data acquisition is completed, the system switches to the data preprocessing state. In the data preprocessing state, filtering, noise reduction, clock alignment, and outlier removal of the collected data are completed. After preprocessing is completed, the system switches to the parameter calculation state. In the parameter calculation state, the pipe network viscosity field, fluid delay, and delay gradient are calculated. After the parameter calculation is completed, the algorithm iteration state is switched. In the algorithm iteration state, particle initialization, fitness value calculation, particle velocity and position update, and global optimal solution update are completed. After the iteration converges or the maximum number of iterations is reached, the system switches to the control output state. In the control output state, the control duty cycle is calculated, feedforward compensation and limiting are completed, and the drive signal is output to the variable frequency motor. After the control output is completed, the system switches to the data acquisition state and enters the next control cycle. The fault protection state is triggered when a fault signal is detected in any state. It executes a graded protection strategy according to the fault level, including alarm prompt, speed reduction operation, smooth shutdown, and emergency shutdown. After the fault is cleared, it switches to the initialization state.

[0025] The system employs a priority-based preemptive scheduling strategy for real-time multi-task scheduling. Task priorities, from highest to lowest, are: fault diagnosis and protection, clock synchronization, data acquisition, control output, core algorithm, data preprocessing, and parameter storage. The execution cycle of each task matches the control cycle. High-priority tasks can interrupt the execution of low-priority tasks. Shared data is protected using a mutex lock mechanism to prevent data contention and inconsistency issues.

[0026] The historical buffer sequence is implemented using a circular queue structure. The queue length is no less than twice the number of sampling points corresponding to the maximum fluid delay. Each storage unit in the queue corresponds to a temporal feature dataset at a sampling moment. The dataset contains temperature field data, viscosity field data, pose data, fluid delay, control command data, and fitness value data at that moment. All data has a corresponding timestamp. The queue is updated using a first-in-first-out (FIFO) approach.

[0027] The local transient temperature of each pipeline node is acquired using a fiber optic sensor array. Each sensing unit of the fiber optic sensor array corresponds to a unique pipeline node. Temperature data of each node at time t is acquired through synchronous triggering. The local transient temperature is denoted as […]. ,in is a spatial coordinate vector used to characterize the three-dimensional spatial position of the corresponding sensing unit on the hybrid hydraulic pipeline, with units of meters (m). The physical sampling time is expressed in seconds. The unit is K.

[0028] To eliminate interference from solar radiation and environmental background, a global average temperature baseline for the entire pipeline network is calculated. This global average temperature baseline is used to eliminate the interference of overall environmental temperature fluctuations on the pipeline temperature acquisition results, and to isolate the global temperature drift caused by asymmetric factors such as solar radiation and environmental airflow. The calculation method involves averaging the local transient temperatures of all pipeline nodes at the same sampling time using a global spatial integration. The calculation formula is as follows:

[0029] In the formula, This refers to the entire area corresponding to the hybrid hydraulic pipeline network, with units of... ; A volume element within the entire spatial domain, with units of . ; The average temperature baseline of the global pipeline network at time t is given in K.

[0030] The temperature difference fluctuation, representing the deviation of local transient temperature from the average temperature benchmark, is substituted into a pre-defined oil viscosity-temperature characteristic model to calculate the fluid viscosity at each node. This data is then aggregated to generate a network viscosity field reflecting the overall rheological resistance. The pre-defined oil viscosity-temperature characteristic model is constructed based on the standard viscosity-temperature characteristics of hydraulic oil and is used to characterize the nonlinear relationship between hydraulic oil viscosity and temperature. The temperature difference fluctuation represents the difference between the local transient temperature of the corresponding pipeline node and the global network average temperature benchmark at the same sampling time, representing the degree to which the temperature at that node deviates from the overall average level. After substituting the temperature difference fluctuation into the viscosity-temperature characteristic model, the real-time dynamic viscosity of the hydraulic oil at the corresponding node can be calculated. The viscosity data from all nodes are aggregated to form a viscosity field covering the entire hybrid hydraulic network. The calculation formula is as follows:

[0031] In the formula, The viscosity field of the pipeline at time t and spatial location r is given by units of . ; The reference dynamic viscosity corresponds to the hydraulic oil at the standard temperature. The rated dynamic viscosity is determined by the grade of the selected hydraulic oil, and the unit is 1000 liters. ; This is the viscosity-temperature sensitive scalar coefficient of the oil, with units of... The range of values ​​is to The viscosity-temperature characteristic curve of the selected hydraulic oil is obtained by fitting it using the least squares method, and is used to characterize the sensitivity of the hydraulic oil viscosity to temperature changes.

[0032] Obtain the pipeline geometry set, which includes the inner diameter and length of the pipes along the route. The pipeline geometry set is a collection of geometric parameters for the pipe routing paths from each hydraulic pump source to the corresponding hydraulic jack in the hybrid hydraulic network. It includes the inner diameter, length, direction, location of throttling elements such as elbows, valves, and reducers, as well as the local resistance coefficient for each segment of the routing path. All parameters are determined based on the actual as-built drawings and on-site measured data of the pipeline network. Each pipe routing path from a hydraulic pump source to a corresponding hydraulic jack corresponds to a unique pipeline geometry set. The discretization of the pipeline geometry set uses a constant step size discretization method, with the step size ranging from [value missing]. to Each discretized path segment corresponds to a unique path coordinate, pipe inner diameter, and local resistance coefficient.

[0033] To quantify the hysteresis response caused by the throttling effect of pipe inner diameter, the flow resistance parameters along the pipeline corresponding to the pipeline geometry set are calculated using the Poiseuille resistance principle. The Poiseuille resistance principle describes the friction loss of viscous fluid flowing in a circular pipeline under laminar flow conditions. The throttling effect of pipe inner diameter refers to the influence of changes in pipe inner diameter on fluid flow resistance; the smaller the pipe inner diameter, the greater the friction loss, and the more pronounced the hysteresis response in command transmission. The flow resistance parameters along the pipeline are derived based on the Poiseuille resistance principle and are used to characterize the resistance to fluid flow per unit length of pipeline along the pipeline path. Under laminar flow conditions, the Reynolds number of the fluid in the pipeline is calculated using the following formula:

[0034] In the formula, Let be the Reynolds number of the fluid in the pipeline, and be a dimensionless parameter. The average axial velocity of the fluid in the pipeline, in units of ; This refers to the inner diameter of the pipe, in meters (m). When the Reynolds number is less than... When the flow is classified as laminar, the Poiseuille drag formula is used to calculate the friction loss; when the Reynolds number is greater than or equal to... When the flow is classified as turbulent, the friction loss is calculated using the Brassius formula. The formula for calculating the friction loss parameter is as follows:

[0035] In the formula, Let be the friction coefficient at time t and path position s, which is a dimensionless parameter; This represents the inner diameter of the pipe at location s along the corresponding path, in meters (m).

[0036] By combining fluid density and the viscosity field of the pipeline network, path integration is performed along the actual pipeline route using the flow resistance parameters to obtain the fluid delay reflecting the actual physical time difference between the arrival of commands from each pump source at each execution end. The fluid density is the rated density of the selected hydraulic oil, a constant, and the unit is... The path integral is the actual piping path from the corresponding hydraulic pump source to the corresponding hydraulic jack. During integration, the pipe network viscosity field value, pipe inner diameter parameter, friction coefficient, and local resistance coefficient are substituted segment by segment along the piping path. The final integral result is the physical time delay of the control command issued by the hydraulic pump source being transmitted to the corresponding hydraulic jack actuator through the hydraulic fluid. The calculation formula is as follows:

[0037] In the formula, The fluid delay along the pipe routing path at time t is expressed in seconds. This is the actual piping path from the hydraulic pump source to the corresponding hydraulic jack, in meters; This represents the cross-sectional area of ​​the pipe at location s along the corresponding path, in units of... , ; The rated output flow rate of the hydraulic pump source, in units of... ; This represents the total number of local resistance elements along the pipe routing path. Let be the local resistance coefficient of the i-th local resistance element, which is a dimensionless parameter; This represents the path position corresponding to the i-th local resistance element, in meters. Let be the Dirac function, used to characterize the locational characteristics of local resistance. For the discretized pipeline geometry set, the path integral is implemented using the trapezoidal integral method, and the discretized calculation formula is:

[0038] In the formula, The total number of segments after discretization of the pipe routing path; The cross-sectional area of ​​the pipe in the k-th segment is given by [value]. ; Let be the friction coefficient of the k-th segment at time t, which is a dimensionless parameter; The inner diameter of the pipe in the kth segment is in meters. Let be the local drag coefficient of the k-th segment, and be a dimensionless parameter; is the length of the k-th segment, in meters.

[0039] The delay gradient is obtained by spatially differentiating the fluid delay. The spatial derivative is applied to the fluid delay corresponding to each pipe path within the entire hybrid hydraulic network. The derivative result characterizes the rate of change of the fluid delay in the spatial distribution of the network, reflecting the degree of abrupt change in fluid delay within local pipe sections. The formula for calculating the delay gradient is:

[0040] In the formula, Spatial gradient operator, unit: ; Let be the delay gradient of the fluid delay at time t in the global space, and let be a vector parameter with units of . The magnitude of the delay gradient corresponds to the magnitude of the delay change, and the direction corresponds to the spatial direction in which the delay changes the most rapidly. For the discretized spatial distribution of the pipeline network, the delay gradient is discretized using the central difference method, and the boundary position is calculated using the forward difference or backward difference method.

[0041] To avoid the algorithm getting stuck in a dead zone due to local viscosity abrupt changes, the optimization space is dynamically stretched based on the extreme value of the delay gradient. A local viscosity abrupt change refers to a sudden increase in fluid viscosity caused by a rapid temperature change in a local pipe section, leading to a sudden change in fluid delay. This abrupt change can cause a steep local region in the algorithm's optimization space, trapping the search process in a local optimum dead zone. The extreme value of the delay gradient is the maximum value of the delay gradient magnitude across the entire space at the same sampling time. Dynamically mapping the stretching factor of the optimization space based on this extreme value allows for adaptive stretching of the optimization space in the region of delayed abrupt changes, broadening the algorithm's search range and preventing the search process from getting stuck in a dead zone. The stretching factor is positively correlated with the extreme value of the delay gradient; the larger the extreme value of the delay gradient, the higher the stretching factor, and the greater the expansion of the optimization space. The formula for calculating the stretching factor is:

[0042] In the formula, Let be the stretching factor of the optimization space at time t, and be a dimensionless parameter; The maximum value of the delay gradient magnitude in the global space at time t, in units of ; The average value of the delay gradient magnitude in the global space at time t, in units of The stretching factor is limited to a range of 1 to 5 to avoid overstretching, which could lead to a divergence in the optimization space.

[0043] The control commands from the previous steady-state cycle are superimposed with a stretching factor to generate initial compensation commands with topology adaptation capabilities, completing the particle initialization of the unbounded optimization space. The control commands from the previous steady-state cycle are the control command matrix output to each variable frequency motor after the system is in a stable synchronous state following the end of the previous synchronization control cycle. This matrix contains the speed control parameters of each hydraulic pump source corresponding to the variable frequency motor, which can be used as the initial benchmark for this iterative optimization. The population size of the particle swarm ranges from [value missing]. to The population size is positively correlated with the number of nodes in the hydraulic pump source group. Each particle corresponds to a set of control compensation commands to be optimized. The position vector dimension of the particle is consistent with the number of nodes in the hydraulic pump source group, and each dimension corresponds to a control parameter of a hydraulic pump source. During particle initialization, the control command matrix of the previous steady-state cycle and the stretching factor are performed element-wise to generate an initial particle swarm covering the entire optimization space. The distribution of the initial particle swarm can be adaptively adjusted according to the distribution of the pipeline delay gradient, possessing pipeline topology adaptive capability. The calculation formula is as follows:

[0044] In the formula, The number of the optimal particle is determined by a value of [value]. to positive integers, The population size of the particle swarm; The initial position vector of each particle, i.e., the initial compensation command, has a dimension of , This represents the total number of nodes in the hydraulic pump source group. This is the control command matrix for the previous steady-state cycle, with dimension [missing information]. ; For the whole Matrix, dimension ; This is a matrix element-level multiplication operation, also known as the Hadamard product. For the first The random number matrix corresponding to each particle has a dimension of . The elements in the matrix are intervals. A random number that is uniformly distributed within the range.

[0045] The initial velocity vectors of all particles are initialized to... The vector has the same dimensions as the position vector, and the upper and lower bounds of the particle's position are:

[0046] In the formula, Let be the lower bound vector of the particle position, with dimension . The value is taken as the rated speed of the hydraulic pump source. ; Let be the upper bound vector of the particle's position, with dimension . The value is taken as the rated speed of the hydraulic pump source. ; This represents the current iteration number of the algorithm, and its value is a non-negative integer. When a particle goes out of bounds, the boundary absorption method is used to handle it; elements exceeding the upper and lower limits are directly assigned the corresponding upper and lower limit values.

[0047] The torsional displacement of the connecting corridor, characterizing the asynchronous deformation deviation of the structure, is extracted using a pose sensor. The pose sensor synchronously collects the three-dimensional spatial coordinates and attitude angles of each node of the steel connecting corridor, and maps them uniformly to the overall structural coordinate system of the steel connecting corridor through coordinate transformation, obtaining the vertical displacement data of each jacking point. The torsional displacement of the connecting corridor is the second-order differential result of the elevation deviation at different jacking points of the steel connecting corridor, used to characterize the structural torsion and deformation caused by the asynchronous jacking of the hydraulic jacks. The larger the torsional displacement value, the more severe the asynchronous deformation deviation of the steel connecting corridor. The formula for calculating the torsional displacement of the connecting corridor is:

[0048] In the formula, In the first The first particle Driven by iterative instructions, the predicted pose vectors of each lifting point of the steel connecting corridor have the following dimensions: The unit is meters (m). For the first The first particle The output flow vectors of each hydraulic pump source corresponding to the iteration command have a dimension of The unit is ; This is the inverse matrix of the effective working area of ​​the hydraulic jack, with dimension . ; It is a second-order Laplace operator, with units of . ; This is a 2-norm operation, and the result is a dimensionless parameter.

[0049] To compensate for system energy loss caused by internal shear friction of the fluid, the viscous dissipation rate of the entire system is calculated based on the viscosity field of the pipeline network and the velocity gradient within the pipe. The viscous dissipation rate characterizes the rate of energy loss due to interlayer shear friction of hydraulic oil flowing within the pipeline; a higher viscous dissipation rate indicates greater useless energy consumption in the system. The calculation of the viscous dissipation rate is based on the internal friction energy consumption theory of fluid mechanics and is positively correlated with the spatial gradient of fluid viscosity and velocity. By spatially integrating the viscous dissipation rate of the entire pipeline network, the total viscous energy consumption of the entire hydraulic system can be obtained. The calculation formula is as follows:

[0050] In the formula, In the first The first particle The total global viscous dissipation power of the system under the drive of iteration instructions, in W; In the first The first particle Driven by iterative instructions, the microscopic velocity matrix of the fluid within the pipeline, in units of... ; The spatial gradient of fluid velocity, in units of .

[0051] The torsional moment of the connecting corridor and the viscous dissipation rate are weighted and fused to calculate a fitness value with minimizing deformation and energy consumption as the joint evaluation objective. The fitness value is used to evaluate the quality of the control command corresponding to the optimization particle; the smaller the fitness value, the smaller the deformation of the steel connecting corridor and the lower the system energy consumption, indicating better control performance. In the weighted fusion process, the torsional moment of the connecting corridor and the viscous dissipation rate are first dimensionlessized, and then preset weight coefficients are set for the two dimensionless indicators. The values ​​of the weight coefficients can be adjusted according to the requirements of synchronization accuracy and energy consumption control at the engineering site. The sum of the two weight coefficients is 1. The calculation formula is as follows:

[0052] In the formula, For the first The particle in the first The fitness value during iteration is a dimensionless parameter; , is the baseline value of the torsion of the connecting corridor, is the calculation result of the second norm of the torsion under the steady state of the system under no-load, and is a dimensionless parameter; The reference value for viscous dissipation power is the total viscous dissipation power of the system under no-load rated flow, in W. and These are preset weighting coefficients, all of which are greater than [a certain value]. , and satisfy the dimensionless parameter. , The value range is from 0.6 to 0.9. The range of values ​​is to .

[0053] Calculate the integral proportion characteristic of the local viscosity at the current node versus the overall viscosity over the entire network. The local viscosity at the current node is the viscosity field value of the pipe network at the corresponding node. The overall viscosity over the entire network is the total viscosity value obtained by spatial integration of the viscosity field. The integral proportion characteristic is the ratio of the volume integral corresponding to the local viscosity at the current node to the overall viscosity over the entire network. It characterizes the proportion of the viscosity at that node in the overall viscosity distribution; a higher proportion indicates a greater impact of the viscous drag at that node on the system. The formula for calculating the integral proportion characteristic is:

[0054] In the formula, Let be the proportion of local viscosity integral at time t and spatial location r, where is a dimensionless parameter with a value range of . to .

[0055] To avoid algorithmic search stalling in high-viscosity pipe sections, inertial weights are constructed based on the integral percentage characteristics for dynamically adjusting damping. High-viscosity pipe sections are defined as those with integral percentage characteristics exceeding a preset threshold. In pipe sections with high fluid flow resistance and high command transmission delay, the optimization step size of the algorithm is easily reduced, leading to search stagnation and premature convergence. Inertia weights are used to adjust the degree to which particles inherit the iteration velocity of the previous generation during algorithm iteration. This dynamically adjusts the algorithm's global search and local exploitation capabilities. Inertia weights constructed based on integral proportion characteristics can increase the particle's optimization step size in high-viscosity pipe sections, avoiding search stagnation. The calculation formula is:

[0056] In the formula, Let be the inertia weight at time t, and be a dimensionless parameter with a value range of . to The higher the integral percentage feature, the smaller the inertia weight, the larger the particle's optimization step size, and the stronger the global search capability. For the discretized particle position vector, the inertia weight corresponding to each dimension is calculated from the average integral percentage feature of the hydraulic pump source pipe routing path corresponding to that dimension.

[0057] Multiplying the inertia weight by the current iteration velocity decays and preserves kinetic energy. Then, the individual particle cognitive bias and the global social bias are superimposed to output the updated next-generation optimization velocity to guide the optimization direction. The current iteration velocity is the particle's optimization velocity vector in the current iteration generation. Multiplying the inertia weight by the current iteration velocity yields the velocity component inherited by the particle from the previous iteration, used to preserve the particle's optimization kinetic energy. The individual particle cognitive bias is the difference between the particle's own historical best position and its current position, used to guide the particle to search towards its own historical best solution; the global social bias is the difference between the particle swarm's global best position and its current position, used to guide the particle to search towards the swarm's best solution. Superimposing these three velocity components yields the particle's optimization velocity for the next iteration, calculated using the following formula:

[0058] In the formula, For the first The first particle The next-generation optimization speed of the iteration is dimensional. ; For the first The first particle The current iteration speed of the iteration, the dimension is ; , The learning factors are all positive dimensionless constants, and their values ​​range from [value range missing]. to , used to adjust the weight of individual cognition and global social weight; , For interval The uniformly distributed random numbers within the range are dimensionless parameters used to increase the randomness of the algorithm's search. For the first The individual historical best position vector of each particle, with dimension . , which is the position corresponding to the minimum fitness value in the iteration history of this particle; The vector representing the current global optimal position of the particle swarm has a dimension of . , which is the position corresponding to the minimum fitness value in the entire particle swarm iteration history.

[0059] The upper and lower limits of particle velocity are constrained as follows:

[0060] In the formula, Let be the lower bound vector of particle velocity, with dimension . The value is ; The upper bound vector of particle velocity, with dimension . The value is When a particle's velocity exceeds the limit, the boundary absorption method is used to handle it; elements exceeding the upper and lower limits are directly assigned the corresponding upper and lower limit values.

[0061] The formula for updating particle positions is:

[0062] After the particle positions are updated, the fitness value of each particle is recalculated, and the individual's historical best position and the global best position are updated. The algorithm terminates iteration when either of these conditions is met: the number of iterations reaches a preset maximum number of iterations, or the change in the global best fitness value over a consecutive preset number of iterations is less than a convergence threshold. The algorithm then terminates iteration and outputs the current global best position as the converged solution. The maximum number of iterations can be set within a certain range. to The convergence threshold ranges from 100 to 100. to .

[0063] For each command node in the hydraulic pump source group, its unique fluid delay is extracted. Each hydraulic pump source in the hydraulic pump source group corresponds to an independent command node. The piping path from each command node to the corresponding hydraulic jack is different, resulting in different fluid delays. Each command node corresponds to a unique, exclusive fluid delay, denoted as . ,in This is the number of the command node in the hydraulic pump source group, with a value of [value to be filled in]. to positive integers, This represents the total number of nodes in the hydraulic pump source group.

[0064] To eliminate the misleading effect of asynchronous physical command transmission on algorithm evaluation results, the current evaluation moment is traced back to the historical buffer sequence according to a specific fluid delay, and the aligned temporal feature dataset is extracted. The historical buffer sequence consists of temporal feature data of all sampling moments within a preset duration stored by the system, with the data stored in the order of sampling timestamps. Reverse tracing refers to tracing the current evaluation moment... Subtract the dedicated fluid delay of the corresponding command node at the backtracking moment to obtain the historical backtracking moment. The control command issued at that moment will happen at the current evaluation moment. The data reaches the corresponding execution end. By extracting the timing feature data of each commanding node at the corresponding backtracking moment from the historical buffer sequence, the timing features of different commanding nodes can be aligned, eliminating evaluation bias caused by asynchronous instruction transmission. If the backtracking moment is not a sampling point in the historical buffer sequence, linear interpolation is used to calculate the timing feature data of the corresponding moment; if the backtracking moment exceeds the storage range of the historical buffer sequence, the timing feature data of the earliest moment in the buffer sequence is used.

[0065] The local fitness evaluation is reconstructed in the temporal feature dataset, and the parameter combination that minimizes the comprehensive bias is selected to align and update the global optimum. The local fitness evaluation is the fitness evaluation result calculated for a single command node based on the aligned temporal feature data, and the comprehensive bias is the cumulative value of the local fitness evaluation results of all command nodes. By traversing all candidate particle parameter combinations, the corresponding comprehensive bias is calculated, and the parameter combination that minimizes the comprehensive bias is selected as the updated global optimum. The calculation formula is as follows:

[0066] In the formula, To align with the updated global optimal solution; In order to target the The local fitness evaluation function for each node is calculated using the following formula:

[0067] In the formula, For the corresponding number The backtracking time of each node; For the first The predicted pose corresponding to each node; For the first The pipeline viscous dissipation power corresponding to each node; , The first The reference values ​​for torsional amount and dissipated power for each node.

[0068] The convergent solution output at the end of the optimization iteration is transformed into basic frequency conversion commands. After the optimization iteration ends, the current global optimal solution is output as the convergent solution. The basic frequency conversion commands are the basic control signals for driving the frequency conversion motor, obtained based on the convergent solution. The conversion process uses the space vector pulse width modulation method to convert the speed control parameters in the convergent solution into the corresponding reference drive signals for the frequency conversion motor. The calculation formula of the space vector separation modulation function is as follows:

[0069] In the formula, The basic frequency conversion command is a dimensionless parameter with a value range of [value range missing]. to ; The rated speed of the variable frequency motor, in units of .

[0070] To counteract the dynamic hysteresis disturbance caused by drastic temperature changes in the pipeline network, the time derivative of the fluid delay is calculated to obtain the rate of change of delay, which is then used to map the dynamic feedforward compensation. Dramatic temperature changes in the pipeline network cause rapid changes in fluid viscosity, leading to dynamic changes in the fluid delay over time, generating dynamic hysteresis disturbances that affect the accuracy of synchronous control. The rate of change of delay is the first derivative of the fluid delay with respect to time, characterizing the rate of change of the fluid delay over time and reflecting the intensity of the dynamic hysteresis disturbance. The rate of change of delay is calculated using a discrete-time difference operator, and the formula is as follows:

[0071] In the formula, The rate of change of time delay at time t, in units of ; Let be the system sampling period, in seconds. To suppress the high-frequency noise amplification characteristics of the differentiating element, after the delay rate of change is calculated, a first-order low-pass digital filter is used for filtering. The filter's transfer function is:

[0072] In the formula, Here are the filter coefficients, and is a dimensionless parameter with a value range of . to ; This is the Z-transform operator.

[0073] By using a preset feedforward gain coefficient matrix, the filtered rate of change of delay is mapped to the corresponding dynamic feedforward compensation amount. This allows for feedforward compensation of dynamic hysteresis disturbances, offsetting the impact of disturbances on control accuracy. The formula for calculating the dynamic feedforward compensation amount is as follows:

[0074] In the formula, Let be the dynamic feedforward compensation amount at time t, which is a dimensionless parameter with a value range of . to ; Here is the feedforward gain coefficient matrix, and is a dimensionless parameter with a value range of . to This can be achieved through on-site debugging and adjustment.

[0075] The dynamic feedforward compensation is merged and fused with the basic frequency converter command to generate a control duty cycle that compensates for asynchronous hysteresis, thus precisely driving the variable frequency motor. The merging process involves element-wise superposition of the basic frequency converter command and the dynamic feedforward compensation. The resulting signal is the final control duty cycle output to the variable frequency motor. This control duty cycle can simultaneously achieve synchronous control based on the optimal convergence solution, as well as compensation for asynchronous hysteresis and dynamic disturbances in fluid transmission. The calculation formula is as follows:

[0076] In the formula, The control duty cycle output to the variable frequency motor at time t is a dimensionless parameter. The control duty cycle needs to be limited before output; the limiting range is [range to be specified in the original text]. to Values ​​exceeding the limit range are directly assigned the corresponding upper and lower limits, while dead zone compensation is set. When the control duty cycle is less than... When, directly assign a value This avoids unstable operation of the variable frequency motor in the low-speed range.

[0077] The system's initial calibration and online self-calibration process includes: During the system power-on initialization phase, the initial calibration process is executed, including: temperature zero-point calibration of the fiber optic sensor array (placing all sensing units in a constant-temperature environment, collecting temperature data, and correcting zero-point deviation); pose zero-point calibration of the pose sensors (placing the steel connecting corridor in a horizontal, stationary state, collecting pose data at each point, and setting it as the zero point of the structural coordinate system); no-load flow calibration of the hydraulic system (controlling the variable frequency motor to run at rated speed, collecting the output flow of each hydraulic pump source, and correcting the parameters of the flow characteristic model); and on-site calibration of the oil viscosity-temperature characteristics (collecting hydraulic oil viscosity data at different oil temperatures, and correcting the reference viscosity and viscosity-temperature sensitivity coefficient of the viscosity-temperature characteristic model). During system operation, an online self-calibration is performed every hour to correct sensor zero-point drift and model parameter deviations. The calibration process does not affect the system's normal synchronous control.

[0078] The system's adaptive processing flow for variable load conditions includes: real-time acquisition of pressure data at each lifting point using pressure sensors in the rodless chamber of the hydraulic jack; calculation of the real-time load mass at each point; identification of dynamic load changes; correction of the load displacement characteristic model of the hydraulic jack and the structural dynamics model of the steel connecting corridor based on the real-time load data; and updating of the calculation parameters for predicted pose and predicted flow velocity; adaptive adjustment of the weighting coefficients in the fitness value calculation according to the load change amplitude, where the load change amplitude is greater than... When, increase The value of is used to improve the control priority of synchronization accuracy; at the same time, the learning factor and inertia weight range of the particle swarm algorithm are adaptively adjusted to accelerate the convergence speed of the algorithm and adapt to the dynamic changes of the load.

[0079] The system's fault diagnosis and fault-tolerant protection process includes: real-time monitoring of the operating status of the fiber optic sensor array, posture sensor, variable frequency motor, and hydraulic system; identification of fault types such as abnormal sensor data, equipment failure, and excessive synchronization deviation; for faults caused by damage to a single fiber optic sensor unit, linear interpolation of temperature data from adjacent sensor units is used to replace the fault, ensuring the continuity of temperature field acquisition; for faults caused by abnormal data from a single posture sensor, posture data from adjacent points and prediction results from the structural dynamics model are fused to replace the fault, ensuring the effectiveness of posture monitoring; for faults caused by a single variable frequency motor or hydraulic pump source, a redundant control strategy for adjacent jacking points is adopted to smoothly adjust the jacking force at each point, ensuring the stress balance of the steel connecting corridor; for faults caused by excessive synchronization deviation, a graded protection strategy is implemented, issuing an alarm and reducing the jacking speed when the deviation exceeds a preset warning threshold; when the deviation exceeds a preset shutdown threshold, a smooth shutdown procedure is executed, locking the position of all hydraulic jacks to prevent deformation or falling accidents of the steel connecting corridor.

[0080] The algorithm's premature convergence detection and secondary optimization mechanism are as follows: The algorithm is considered to have prematurely converged when both of the following conditions are met: continuous The relative change in the global optimal fitness value over iterations is less than ; The standard deviation of the fitness values ​​of all particles in the particle swarm is less than .

[0081] After premature convergence is detected, a secondary optimization mechanism is executed: the current globally optimal particle is retained, and the remaining particles are reinitialized according to the initial particle generation rules. Simultaneously, the lower bound of the inertia weight is adjusted to... Learning factor Adjusted to , Adjusted to To enhance the algorithm's global search capability and escape local optima, the maximum number of iterations for secondary optimization is set to a fraction of the original maximum number of iterations. .

[0082] Outlier removal and filtering of the collected data are as follows: In the data preprocessing stage, outlier removal is first performed on the temperature data of the fiber optic sensor array and the displacement data of the pose sensor. Guidelines: For a sliding window with a length of Given a continuously sampled data sequence, calculate the arithmetic mean of the sequence. with standard deviation If the current sampled data satisfies If the value is not found, it is considered an outlier and is replaced by the linear interpolation result of the two valid data points within the window.

[0083] After outlier removal, a first-order low-pass filter is applied to the data. The filter formula is as follows:

[0084] In the formula, The filtered data is at time t; This represents the original data collected at time t; The filter coefficients are dimensionless, and the filter coefficients for temperature data are set to values ​​of... The filter coefficients for the pose data are taken as follows: To balance the filtering effect and response speed.

[0085] The mechanism for correcting the delay in hydraulic oil compressibility is as follows: To address the impact of hydraulic oil compressibility on fluid delay under high-pressure conditions, delay correction is implemented: When the hydraulic system working pressure is greater than At that time, a compressibility correction factor is introduced to the fluid delay, and the correction formula is as follows:

[0086] In the formula, This is the corrected fluid delay, expressed in seconds (s). Let be the average working pressure of the hydraulic system at time t, in Pa; This is the equivalent bulk modulus of hydraulic oil, expressed in Pa. The corrected delay can more accurately match the actual command transmission delay under high-pressure conditions, improving the accuracy of timing alignment.

[0087] The adaptive compensation mechanism for multi-node load redistribution is as follows: During the jacking process of the steel connecting corridor, when load redistribution occurs due to structural deformation, adaptive compensation is implemented: The pressure in the rodless chamber of each hydraulic jack is collected in real time, and the load percentage at each point is calculated using the following formula:

[0088] In the formula, Let be the load percentage of the k-th jack at time t, which is dimensionless; Let be the rodless cavity pressure of the k-th jack at time t, in Pa.

[0089] When the difference in load percentage between two adjacent jacks exceeds At the same time, the control command weights of the corresponding pump sources are adaptively adjusted, increasing the flow command gain at low-load points and decreasing the flow command gain at high-load points. The gain adjustment range is positively correlated with the difference in load ratio, and the maximum adjustment range does not exceed the basic command weight. This is to prevent torsional deformation of the steel connecting corridor caused by uneven load.

[0090] Example 2: For the electrical synchronization control system of the hydraulic pump source for the steel connecting corridor, the execution of the electrical synchronization control method for the hydraulic pump source of the steel connecting corridor as described in any one of the embodiments includes: The pipeline viscosity field construction module collects the surface temperature field of the hybrid hydraulic pipeline through the fiber optic sensing array, and constructs a pipeline viscosity field characterizing the fluid transmission resistance by combining it with preset oil parameters. The fluid delay calculation module calculates the fluid delay transmitted to each hydraulic jack based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network. The particle swarm initialization module calculates the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronization control. The fitness evaluation module obtains the torsion of the connecting corridor through the pose sensor and calculates the fitness value for evaluating the optimization particles by combining the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network. The optimization control module constructs an inertial weight to suppress premature convergence of the algorithm based on the viscosity ratio characteristics of local pipelines and global pipelines, thereby intervening in the optimization speed of the optimization particles. The iterative update module performs time-series alignment of historical evaluation states based on the fluid delay amount corresponding to each hydraulic pump source, eliminating physical asynchrony deviations to update the global optimal solution; The duty cycle calculation module extracts the converged solution after the iteration ends, generates feedforward compensation by combining the delay change rate of the fluid delay, and reconstructs the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

[0091] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. An electrical synchronous control method for steel corridor hydraulic pump sources, applied to an electrical synchronous control system, the system comprising an optical fiber sensing array, a pose sensor, a mixed hydraulic pipe network, a hydraulic pump source group, a hydraulic jack and a variable frequency motor, characterized in that, include: The surface temperature field of the hybrid hydraulic pipeline network is collected by the fiber optic sensing array, and combined with preset oil parameters, a pipeline viscosity field characterizing the fluid transmission resistance is constructed. Based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network, the fluid delay in fluid transmission to each hydraulic jack is calculated. Calculate the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronous control. The torsion of the connecting corridor is obtained by the posture sensor, and the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network is combined to calculate the fitness value for evaluating the optimization particles. Based on the viscosity ratio characteristics of local pipelines and global pipelines, an inertial weight is constructed to suppress premature convergence of the algorithm and to intervene in the optimization speed of the optimization particles. Based on the fluid delay corresponding to each hydraulic pump source, the historical evaluation state is time-aligned to eliminate physical asynchronous deviations and update the global optimal solution. Extract the converged solution at the end of the iteration, combine it with the delay change rate of the fluid delay to generate feedforward compensation, and reconstruct the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

2. The method for electrical synchronous control of steel gallery hydraulic pump source according to claim 1, characterized in that, The construction of the pipeline viscosity field characterizing fluid transport resistance includes: The local transient temperature of each pipeline node is obtained using the fiber optic sensing array. To isolate interference from sunlight and environmental background, the average temperature baseline of the entire pipe network is calculated; The temperature difference fluctuation between the local transient temperature and the average temperature reference is substituted into the preset oil viscosity-temperature characteristic model to calculate the fluid viscosity at each node, so as to aggregate and generate the pipeline viscosity field that reflects the rheological resistance of the whole domain.

3. The method for electrical synchronous control of steel gallery hydraulic pump sources as claimed in claim 2, wherein, The computational fluid delay amount for fluid transmission to each of the hydraulic jacks includes: Obtain the pipeline geometry set, which includes the pipeline inner diameter and the pipeline length along the line; To quantify the hysteresis response caused by the throttling effect of the pipeline inner diameter, the flow resistance parameters along the pipeline geometry set are calculated based on the Poiseuille resistance principle. By combining the fluid density and the viscosity field of the pipeline network, the flow resistance parameters along the actual pipeline path are integrated to obtain the fluid delay, which reflects the actual physical time difference between the arrival of each pump source command at each execution end.

4. The method for electrical synchronous control of steel gallery hydraulic pump sources as claimed in claim 3, wherein, The completion of particle initialization for synchronization control includes: The delay gradient is obtained by spatially differentiating the fluid delay. To avoid the algorithm getting stuck in a dead zone due to local viscous abrupt changes, the stretching factor of the optimization space is dynamically mapped based on the extreme value of the delay gradient. The control command from the previous steady-state cycle is superimposed with the stretching factor to generate an initial compensation command with topological adaptive capability, thereby completing the particle initialization of the unbounded optimization space.

5. The method for electrical synchronous control of steel gallery hydraulic pump sources as claimed in claim 4, wherein, The calculation used to evaluate the fitness value of the optimization particle includes: The torsion of the connecting corridor, which characterizes the asynchronous deformation deviation of the structure, is extracted using the posture sensor. To compensate for the system energy loss caused by internal shear friction of the fluid, the viscous dissipation rate of the whole domain is calculated based on the viscosity field of the pipeline network and the velocity gradient inside the pipe. The torsion of the connecting corridor and the viscous dissipation rate are weighted and fused to calculate the fitness value with minimizing deformation and energy consumption as the joint evaluation objective.

6. The electrical synchronization control method for the hydraulic pump source of a steel connecting corridor according to claim 5, characterized in that, The intervention in the optimization speed of the optimization particle includes: Calculate the integral ratio characteristics of the local viscosity of the current node and the overall viscosity of the whole domain; To avoid algorithmic search stalling in high-viscosity pipe sections, the inertial weights are constructed based on the integral proportion characteristics for dynamically adjusting the damping. The inertial weight is multiplied by the current iteration speed to decay and retain kinetic energy, and the individual particle cognitive bias and global social bias are superimposed to output the updated next-generation optimization speed to guide the optimization direction.

7. The electrical synchronization control method for the hydraulic pump source of a steel connecting corridor according to claim 6, characterized in that, The elimination of physical asynchrony bias to update the global optimal solution includes: For each command node in the hydraulic pump source group, extract its specific fluid delay amount; To eliminate the misleading effect of asynchronous physical command transmission on algorithm evaluation results, the current evaluation time is traced back to the historical buffer sequence according to the specific fluid delay amount, and the aligned time-series feature dataset is extracted. The local fitness evaluation is reconstructed in the time-series feature dataset, and the parameter combination that minimizes the overall bias is selected to align and update the global optimal solution.

8. The electrical synchronization control method for the hydraulic pump source of a steel connecting corridor according to claim 7, characterized in that, The reconfigured control duty cycle sent to the variable frequency motor includes: The converged solution output at the end of the optimization iteration is converted into a basic frequency conversion command; To counteract the dynamic hysteresis disturbance caused by drastic changes in pipeline temperature, the time derivative of the fluid delay is calculated to obtain the rate of change of delay, and this is used to map the dynamic feedforward compensation amount. The dynamic feedforward compensation amount is combined and fused with the basic frequency conversion command to generate the control duty cycle that compensates for asynchronous hysteresis in order to accurately drive the frequency conversion motor.

9. An electrical synchronization control system for a hydraulic pump source of a steel connecting corridor, wherein the electrical synchronization control method for a hydraulic pump source of a steel connecting corridor as described in any one of claims 1-8 is characterized in that, include: The pipeline viscosity field construction module collects the surface temperature field of the hybrid hydraulic pipeline through the fiber optic sensing array, and constructs a pipeline viscosity field characterizing the fluid transmission resistance by combining it with preset oil parameters. The fluid delay calculation module calculates the fluid delay transmitted to each hydraulic jack based on the pipeline geometry set and the viscosity field of the hybrid hydraulic pipeline network. The particle swarm initialization module calculates the delay gradient of the fluid delay in the spatial distribution, thereby expanding the optimization space of the algorithm and completing the particle initialization for synchronization control. The fitness evaluation module obtains the torsion of the connecting corridor through the pose sensor and calculates the fitness value for evaluating the optimization particles by combining the viscous dissipation rate inside the pipeline caused by the viscous field of the pipeline network. The optimization control module constructs an inertial weight to suppress premature convergence of the algorithm based on the viscosity ratio characteristics of local pipelines and global pipelines, thereby intervening in the optimization speed of the optimization particles. The iterative update module performs time-series alignment of historical evaluation states based on the fluid delay amount corresponding to each hydraulic pump source, eliminating physical asynchrony deviations to update the global optimal solution; The duty cycle calculation module extracts the converged solution after the iteration ends, generates feedforward compensation by combining the delay change rate of the fluid delay, and reconstructs the control duty cycle sent to the variable frequency motor to perform synchronous lifting.

Images