An automatic control method for the fracturing platform of a fracturing truck

By collecting multi-source data for timestamp synchronization and constructing a working condition feature map, and combining fuzzy controllers and nonlinear programming, the synchronization deviation and response lag problems of fracturing trucks under deep and complex working conditions were solved, thus achieving wellhead pressure stability and construction continuity.

CN123109616APending Publication Date: 2026-11-17SHENYANG INST OF ENG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611618435.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The existing fracturing truck control system suffers from problems such as synchronization deviation, response lag, insufficient equipment status monitoring, and unreasonable power distribution under deep and complex working conditions, resulting in wellhead pressure fluctuations and construction interruptions.

Method used

By collecting data from multiple sources and synchronizing timestamps, a working condition feature map is constructed. Fuzzy controllers and nonlinear programming optimization are used, combined with equipment health and hysteresis characteristics, to achieve dynamic control parameter adjustment and load balancing, predict working condition transitions, and optimize power distribution.

Benefits of technology

It improved the accuracy of load distribution among multiple vehicles, reduced wellhead pressure fluctuations, enhanced equipment fault tolerance, and ensured the continuity and stability of construction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123109616A_ABST
    Figure CN123109616A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of fracturing platform control technology and discloses an automatic control method for fracturing trucks. It collects equipment operating parameters, process parameters, formation parameters, and historical operating data of the fracturing truck group. Millisecond-level timestamp synchronization and alignment of each acquisition node is achieved using BeiDou time synchronization PPS pulse signals. Combining real-time displacement, sand ratio, fracturing fluid viscosity, and plunger pump inlet and outlet pressure difference, the method distinguishes between Newtonian fluid and power-law fluid segments, calculating pipeline friction and local resistance segment by segment. The pressure data collected at the wellhead is converted into bottomhole flowing pressure, constructing a unified spatiotemporal benchmark operating condition dataset. A feature map covering multiple typical operating conditions is pre-constructed. Real-time operating condition modes are identified through weighted cosine similarity matching. The operating condition migration trend is predicted through a sliding time window. With bottomhole target pressure and total platform displacement as constraints, and considering equipment health and hysteresis characteristics, weights are set, and the displacement allocation target for each fracturing truck is solved through nonlinear programming.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of fracturing platform control technology, specifically an automatic control method for a fracturing platform of a fracturing truck. Background Technology

[0002] As deep shale gas fracturing evolves towards higher displacement, higher sand ratio, and higher pump pressure, the complexity and control difficulty of fracturing truck control systems have increased significantly. Existing fracturing truck control technologies mostly employ a master-slave architecture combined with conventional PID closed-loop control. While this can achieve basic pressure and flow regulation and multi-truck synchronization, it presents the following technical challenges under complex deep-seated conditions: Existing synchronization strategies only compensate for fixed communication transmission delays, without considering the discreteness of response lag at the actuator end caused by pump wear, pipeline pressure drop, and aging of different fracturing trucks. During long-cycle operations, as equipment temperature rises and performance dynamically degrades, the response lag of each fracturing truck will shift nonlinearly, leading to a gradual increase in synchronization deviation and causing severe fluctuations in wellhead pressure.

[0003] Pressure-flow loops generally use fixed-parameter PID control, which cannot adapt to the step changes in fracturing fluid viscosity and proppant sand ratio during fracturing. At the switching nodes of working conditions such as sand addition, displacement, and flow rate increase, pressure overshoot or response lag is prone to occur, frequently triggering system pressure protection shutdown and causing construction interruption.

[0004] Equipment status monitoring relies on fixed threshold alarms, without taking into account the performance degradation patterns throughout the equipment's lifecycle; moreover, the early warning mechanism only outputs alarm signals independently and does not form a linkage and fault tolerance with the control loop. When fault symptoms appear, the control strategy cannot be automatically adjusted, and manual intervention to shut down and troubleshoot is the only option, making it difficult to ensure operational continuity.

[0005] The existing power distribution strategy only makes passive adjustments based on the current pressure feedback, without combining forward information on construction procedures for power pre-allocation; when switching between working conditions such as increased displacement and increased sand ratio, the power system response lags, causing pressure drop. Summary of the Invention

[0006] The purpose of this invention is to provide an automatic control method for the fracturing platform of a fracturing truck, so as to solve one or more problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an automatic control method for the fracturing platform of a fracturing truck, comprising the following specific steps: Collect engine operating parameters, gearbox status, plunger pump inlet and outlet pressure and displacement, and lubricating oil temperature of the fracturing truck; collect process parameters including sand ratio, viscosity, and density of the fracturing fluid; connect the formation parameters and construction sequence data of the fracturing design; and summarize the historical operating data of the equipment. The PPS pulse signal based on BeiDou time synchronization is used to synchronize and align the timestamps of the data collected by each acquisition node, eliminating clock drift between different acquisition units. By adopting a recursive fitting method, combined with real-time displacement, sand ratio, fracturing fluid viscosity and plunger pump inlet and outlet pressure difference, the friction resistance and local resistance of the pipeline are calculated. The bottom hole flow pressure conversion method is corrected by combining the rheological characteristics of fracturing fluid, and the pressure data collected at the wellhead is converted into the bottom hole flow pressure. This realizes the spatial dimension correspondence between wellhead equipment data and formation conditions, forming a condition dataset under a unified spatiotemporal reference.

[0008] Based on the aforementioned working condition dataset, a fracturing working condition feature map is pre-constructed, which includes different sand ratios, displacement rates, fracturing fluid types, and formation physical property parameters. The real-time working condition data is compared with the fracturing working condition feature map to identify the working condition mode in which the current operation is located. By analyzing time series data through a sliding time window, the probability and direction of the shift in operating conditions can be predicted, and the characteristic interval of the next operating condition can be identified. Using the bottom-hole target pressure and the platform's total displacement as constraints, and combining the equipment health and response hysteresis characteristics of each fracturing truck with weights, a sequential quadratic programming method is used for nonlinear programming solution. The optimization objective function is a weighted combination of maximizing the overall health margin of the unit and minimizing the pressure response hysteresis dispersion, with weight coefficients of 0.6 and 0.4, respectively. The displacement allocation target of each truck is obtained by solving the problem, and a dynamic setpoint sequence of the main control loop of a single truck is generated. The lower the health of the truck, the higher the constraint weight of its target displacement upper limit and pressure response gain.

[0009] For the pressure and displacement coupled control loop of a single fracturing truck, the dynamic setpoint sequence is used as the control target, and the fuzzy controller is used to self-tune the PID control parameters online. The fuzzy controller takes pressure deviation, deviation change rate, sand ratio change, and fracturing fluid viscosity change rate as inputs, and dynamically adjusts the proportional coefficient, integral coefficient, and derivative coefficient by combining the fuzzy control rules set based on the power-law rheological properties of the fracturing fluid. Combining pipeline resistance and rheological parameters, the pressure disturbance ΔP caused by the step change in sand ratio and fracturing fluid viscosity is calculated in advance based on the power-law fluid friction resistance formula. Then, it is converted into a displacement compensation value ΔQ=ΔP / Kq through the plunger pump displacement-pressure sensitivity coefficient Kq, which serves as the feedforward compensation amount. Kq is obtained by calibrating the plunger pump under rated operating conditions, with units of m³ / (min·MPa). At the operating condition switching node, the feedforward amount is output in a gradient of gradual increase and decrease. Two threshold ranges are set for the pressure deviation change rate and sand ratio change rate to construct a transitional operating condition range. Under steady-state conditions, both indicators are below their respective lower thresholds, and conventional PID control is used. During the transition period between operating conditions, if either indicator is in the transition range and neither exceeds the upper threshold, fuzzy PID control and conventional PID control are used in a weighted mixed output. Under strong disturbance conditions, if either indicator exceeds the upper threshold, fuzzy PID control and feedforward compensation control are switched to full. Integral amplitude limiting is also used to weaken the integral effect when the pressure deviation is too large.

[0010] Using the control command of the main fracturing truck as the synchronization benchmark, and combined with the timing of the displacement control command, the command reception time, controller calculation time, actuator response time, pump output time, and pipeline pressure propagation time to the wellhead are collected in real time for each slave truck. The total hysteresis of the entire control link is decomposed into four parts: communication transmission hysteresis, controller operation hysteresis, actuator dead zone hysteresis, and pipeline pressure propagation hysteresis. The pump displacement amplitude deviation is calculated in combination with the pump body wear degree and plunger volumetric efficiency. It is not included in the hysteresis category through displacement inner loop closed-loop correction. The hysteresis of each part is determined one by one through timestamp comparison, mechanism characteristic calibration, and fluid dynamics calculation. Formula for calculating total hysteresis across the entire control link: This represents the total hysteresis of the entire control chain for a single fracturing truck, in milliseconds (ms). This represents the communication transmission delay, which is the time difference between the time the main fracturing truck sends the command and the time the slave truck receives the command. It is obtained by comparing and statistically analyzing timestamps, and the unit is ms. This represents the controller's operational hysteresis, which is the operation cycle from the completion of input acquisition to the output of the control command. It is obtained through controller characteristic calibration and is measured in milliseconds (ms). This represents the dead zone hysteresis of the actuator, which is the dead zone time from when the actuator receives a command to when the pump body establishes a stable output pressure. It is negatively correlated with the health of the equipment and is obtained through mechanism characteristic calibration, with the unit being milliseconds (ms). This represents the pressure propagation hysteresis in the pipeline, which is the time it takes for the pressure wave to propagate from the plunger pump outlet to the wellhead. It is calculated from the total length of the fracturing string and the propagation velocity of the pressure wave in the fracturing fluid, and is expressed in milliseconds (ms).

[0011] The dead zone hysteresis of the actuator and the volumetric efficiency hysteresis of the pump body correspond to the equipment health in real time. The lower the health, the greater the hysteresis of the corresponding part. The advance compensation amount of the control command is generated based on the total hysteresis of the entire link, and the corresponding compensation gain is set according to the working condition. The compensation gain is increased under the working condition of large displacement and high sand ratio, and the compensation gain is decreased under the working condition of small displacement and replacement extrusion. A two-dimensional load weight matrix of hysteresis and health is constructed, and the load of each vehicle displacement is allocated according to the principle that the smaller the hysteresis and the higher the health. The load adjustment adopts a gradient smooth transition.

[0012] By combining load distribution results and operating condition migration trends, the timing nodes and parameter change trends of fracturing construction procedures are analyzed. The rate and duration of change of displacement and sand ratio in the next stage are predicted by the sliding time window. Combined with pipeline characteristics and pump efficiency parameters, the power demand change curve is calculated in advance, and a power preload command is output to the power system. Based on the power change rate, the output distribution between the diesel engine and the electric drive unit is matched in three levels. Under the steady-state pumping condition with a low power change rate, the diesel engine undertakes the main basic load, while the electric drive unit is in hot standby and only participates in minor power peak shaving. Under the condition of slow displacement and sand ratio increase with a medium power change rate, the diesel engine undertakes the basic power, while the electric drive unit undertakes the dynamic fluctuation component. During the step switching process of the condition with a high power change rate, the electric drive unit prioritizes and quickly makes up for the power gap, while the diesel engine steadily increases the output power according to the preset slope. The power reserve coefficient is adjusted according to the health status of each vehicle's equipment. For vehicles with a health status below the warning value, the power reserve coefficient is increased, and their maximum output power is limited.

[0013] By combining real-time operating parameters and power output data, and based on historical operating data of the fracturing truck, the performance degradation trends of the engine, transmission, and plunger pump are recursively fitted, and the equipment health index and degradation rate are calculated. Based on the absolute value of health and the degradation rate, a progressive three-level fault-tolerance strategy is set up. Each level is linked to the control loop. The health ranges of each level are mutually exclusive and include control measures at each level: The fault isolation level corresponds to a health of <60%. The load of the corresponding vehicle is transferred to the redundant vehicle in a gradient manner through the multi-vehicle load balancing loop, triggering a shutdown warning. The function degradation level corresponds to a health of 60% ≤ health < 80%. The load ratio of the corresponding vehicle is reduced in a gradient manner, and its pressure response speed requirement and power output limit are reduced simultaneously. At the same time, the parameter adjustment measures of the warning level are implemented. The warning level corresponds to a health of ≥80% and a degradation rate exceeding the preset threshold of degradation rate. The integral gain of the main control loop of the corresponding vehicle is reduced, its power reserve coefficient is increased, and the parameter monitoring frequency is increased. Based on the performance degradation trend, predict the remaining time required for the health status to drop to the failure threshold, and plan the load transfer sequence and the timing of backup vehicle intervention in advance.

[0014] A full-process simulation model of the fracturing platform is constructed, and real-time operating data and equipment status data are forward-synchronized to the simulation model. At the same time, the control commands and fault-tolerant strategies output by the simulation are reverse-synchronized to the actual control system, so as to visualize and simulate the fracturing operation process and verify the parameters. When each fracturing stage reaches 1 / 3 or 2 / 3 of its progress, parameters including formation filtration coefficient and formation fracture pressure gradient are retrieved through real-time data inversion, and the fracturing condition characteristic map and feedforward compensation coefficient are corrected. After each fracturing operation is completed, physical property parameters including formation fracturing pressure and filtration coefficient are retrieved based on the construction data. The fracturing condition characteristic map and control parameter library are then corrected. At the same time, the characteristic parameters of equipment performance degradation are updated, and the matching parameters of each control link are adjusted synchronously, including the influence coefficient of health on hysteresis, the health weight of power distribution, and the feedforward gain coefficient of fuzzy PID control. A four-dimensional evaluation index system was established, consisting of pressure fluctuation amplitude, multi-vehicle synchronization deviation, power response time, and operating condition identification accuracy. The weights of the four indicators were 0.3, 0.25, 0.25, and 0.2, respectively. Each indicator was scored normally from 0 to 100 points according to a preset benchmark range, and the overall operating score was obtained by comprehensive weighting. When the score was ≥90, the optimization step size was 0.5 times the benchmark value; when the score was 70 ≤ score < 90, the benchmark value was used; and when the score was < 70, the benchmark value was used. The optimization step size for the next iteration was automatically adjusted based on the evaluation results, and the optimized control strategy was applied to the next fracturing stage.

[0015] The beneficial effects of this invention are as follows: 1. This invention collects multi-source data on equipment operation parameters, process parameters, formation parameters, and historical operation data of fracturing trucks. It uses BeiDou time synchronization PPS pulse signals to synchronize and align millisecond-level timestamps across each acquisition node, eliminating clock drift errors between different acquisition units. Combining real-time displacement, sand ratio, fracturing fluid viscosity, and plunger pump inlet and outlet pressure difference, it distinguishes between Newtonian fluid and power-law fluid segments, calculating pipeline friction and local resistance segment by segment. The pressure data collected at the wellhead is converted into bottomhole flowing pressure, achieving spatial correspondence between wellhead data and formation conditions, and constructing a unified spatiotemporal benchmark dataset. A feature map covering multiple typical operating conditions is pre-constructed. Weighted cosine similarity matching identifies real-time operating condition modes, and a sliding time window predicts the trend of operating condition migration. With bottomhole target pressure and total platform displacement as constraints, and considering equipment health and hysteresis characteristics, weights are set, and nonlinear programming is used to solve for the displacement allocation target of each fracturing truck. This improves the accuracy of multi-truck load allocation and operating condition adaptability, and reduces wellhead pressure fluctuation amplitude.

[0016] 2. This invention addresses the cascaded coupling control loop of the outer pressure loop and inner displacement loop of a single fracturing truck. It employs a fuzzy controller with two-dimensional core input and three-dimensional output to achieve online self-tuning of PID parameters. The fuzzy controller uses pressure deviation and pressure deviation change rate as core inputs, and sand ratio change and fracturing fluid viscosity change rate as feedforward correction inputs. Combining the power-law rheological characteristics of fracturing fluid with field commissioning experience, 49 fuzzy control rules are constructed to dynamically adjust proportional, integral, and derivative coefficients. The pressure disturbance caused by step changes in sand ratio and viscosity is pre-calculated based on pipeline rheological parameters, generating feedforward compensation quantities and pre-investing them in a gradient before switching operating conditions. A dual-index, dual-threshold range is set to achieve smooth switching between three control modes. Integral limiting and hysteresis protection mechanisms are used to avoid frequent boundary switching; pressure overshoot and response lag are suppressed at operating condition switching nodes such as sand addition and displacement increase, reducing the number of system protection shutdowns.

[0017] 3. This invention uses the control commands of the main fracturing truck as the synchronization benchmark, decomposing the total hysteresis of the entire control link into four components: communication transmission, controller calculation, actuator dead zone, and pipeline pressure propagation. The hysteresis of each component is calibrated one by one through timestamp comparison, characteristic calibration, and fluid dynamics calculation. The actuator dead zone hysteresis value is dynamically corrected based on equipment health, and a corresponding control command advance compensation amount is generated based on the total hysteresis. A two-dimensional load weight matrix of hysteresis and health is constructed, and the displacement load ratio of each fracturing truck is adjusted according to the gradient smooth transition principle. The power demand change curve is calculated in advance based on the predicted parameter change trend of the construction process sequence, and the output distribution of diesel engines and electric drive units is matched in three levels according to the power change rate. The power reserve coefficient is adjusted based on equipment health. This eliminates the synchronization deviation dispersion problem caused by equipment wear and pipeline propagation, achieving advance response of the power system and dynamic load balance, and simultaneously improving the equipment's fault tolerance. Attached Figure Description

[0018] Figure 1 This is the overall flowchart of the automatic control system for the fracturing platform of this invention; Figure 2 This is a flowchart of the fuzzy PID self-tuning and feedforward control sub-flowchar for a single fracturing truck according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The parameter values ​​involved in the embodiments are exemplary configurations for this scenario. Those skilled in the art can obtain the corresponding parameters through on-site calibration or historical data fitting based on the actual well depth, well type, equipment configuration and process requirements of fracturing operations, and make adaptive adjustments. This does not constitute a limitation on the scope of protection of the present invention.

[0021] like Figures 1-2 As shown, this embodiment uses the fracturing operation of a deep tight oil well in the Shenyang peripheral block of Liaohe Oilfield as a fixed application scenario. The well has a vertical depth of 3800m and a horizontal section length of 1000m. A combined operation unit consisting of 6 Type 2500 diesel-electric fracturing trucks is used. The fracturing operation cycle for a single section is approximately 28 minutes. The fracturing fluid used is a guar gum system. The designed bottom hole target pressure is 62MPa, and the maximum operation flow rate is 12m³ / min. The specific implementation process of this method is as follows: Before construction begins, the deployment and benchmark calibration of all data acquisition nodes are completed. The data acquisition nodes are divided into four categories: the vehicle-mounted power unit collects operating parameters such as engine speed, torque, lubricating oil temperature, and transmission gear position, with a sampling period of 100ms; the plunger pump unit collects pump inlet and outlet pressure, real-time displacement, and plunger volumetric efficiency, with a sampling period of 50ms; the wellhead pressure detection unit collects wellhead oil pressure and casing pressure, with a sampling period of 200ms; and the downhole parameter detection unit receives formation parameters such as formation fracturing pressure and filtration coefficient obtained from logging while drilling, while also importing design data such as the construction sequence, stage sand ratio, and stage displacement for this project.

[0022] Each type of acquisition node is independently connected to the BeiDou timing module. The sampling clock of each node is calibrated at the second level by hardware triggering the PPS pulse signal output by BeiDou. The calibration cycle is consistent with the BeiDou timing update cycle. The timestamps of all operating data are uniformly set to millisecond precision.

[0023] For nodes with different sampling frequencies, using the BeiDou unified time reference as the axis, a linear interpolation method is adopted to complete the timestamp resampling of asynchronous sampling data, thereby eliminating clock drift errors between different acquisition units.

[0024] After completing the time alignment, the bottom hole flow pressure conversion is carried out to realize the spatial dimension correspondence between the wellhead data and the formation conditions. The conversion adopts a recursive fitting method, starting from the wellhead pipe end, using the measured pressure at the wellhead as the initial value, superimposing the hydrostatic pressure of the corresponding well section, and deducting the friction resistance and local resistance of that section. The pressure of each section is calculated segment by segment downward along the fracturing string with an inner diameter of 139.7mm. The outlet pressure of each section of the string is used as the input boundary condition for the next section of the string. For the i-th calculated section of the fracturing string, the flow pressure at its lower section is: This represents the flow pressure at the lower section of the i-th tubing segment, in Pa; for the uppermost wellhead segment, This is the actual pressure value measured at the wellhead; This represents the flow pressure at the upper section of the i-th tubing segment, i.e., the pressure at the lower end of the previous tubing segment, in Pa. This indicates the density of the fracturing fluid, expressed in kg / m³. This represents the acceleration due to gravity, measured in m / s². This represents the vertical depth increment corresponding to the i-th segment of the tubing, in meters (m). The friction loss of the i-th tubular segment is represented by the flow resistance model corresponding to the Newtonian fluid segment and the power-law fluid segment, and the unit is Pa. This represents the local resistance loss of the i-th segment of the tubing, including resistance caused by scaling on the inner wall of the tubing and throttling of downhole tools, in Pa.

[0025] Every 1-second calculation cycle, the flow resistance parameters of the surface manifold are fitted and corrected based on the measured value of the pressure difference between the inlet and outlet of the plunger pump. The flow resistance parameters of the downhole pipe section are pre-calibrated based on the benchmark value of the same block and well type construction data. Every 10-second calculation cycle, a small dynamic adjustment is made based on the correspondence between the real-time wellhead pressure and the discharge rate.

[0026] The conversion process involves performing flow resistance calculations for vertical pipe sections, inclined well sections, and horizontal sections, distinguishing between the Newtonian fluid characteristic section and the power-law fluid characteristic section of the fracturing fluid, and matching the corresponding flow resistance calculation logic accordingly. At the same time, it incorporates the local resistance correction coefficients for scaling on the inner wall of the tubing string and throttling of downhole tools, and finally converts the pressure data collected at the wellhead into the bottom hole flow pressure to form a working condition dataset under a unified spatiotemporal reference.

[0027] A pre-constructed fracturing condition feature map covering typical construction scenarios in the block was created. The map is based on the full-condition and full-time data of historical fracturing operations in the block. Using process experience calibration combined with K-means clustering, it is first divided into five typical working condition modes: displacement, low sand ratio pumping, high sand ratio pumping, increased discharge rate, and decreased discharge rate. Then, for each mode, the characteristic parameter ranges and typical temporal variation patterns under the corresponding process stage, sand ratio range, discharge rate range, and formation properties are extracted. The similarity matching threshold of 0.85 is the engineering experience value calibrated on-site in the block, which can be adjusted in the range of 0.8 to 0.9 according to the complexity of the working conditions. The characteristic mean, fluctuation amplitude and rate of change of bottom hole flow pressure, wellhead pressure, platform total displacement and fracturing fluid viscosity are extracted in each classification interval. Combined with the fracture pressure and filtration coefficient of the corresponding formation, the classification is completed. Each type of working condition mode corresponds to a set of characteristic parameter intervals and typical time series variation laws.

[0028] During real-time construction, five core parameters are extracted at the current moment: bottom hole flow pressure, total platform displacement, fracturing fluid sand ratio, fracturing fluid viscosity, and bottom hole pressure deviation change rate. These parameters are then used to construct a feature vector, which is then used to perform cosine similarity calculation with the fracturing condition feature map. Before similarity calculation, weighting coefficients are set for each parameter, with bottom hole flow pressure accounting for 40%, platform total displacement accounting for 25%, fracturing fluid sand ratio and viscosity each accounting for 15%, pressure fluctuation rate accounting for 5%, and corresponding appropriate weighting coefficients used for each process stage of pre-flush fluid, sand-carrying fluid, displacement fluid, and displacement extrusion.

[0029] When the similarity is greater than 0.85, the corresponding working condition mode is determined to match, and the current working condition label is output, including five typical modes: replacement extrusion, low sand ratio pumping, high sand ratio pumping, increased discharge rate, and decreased discharge rate. Among them, low sand ratio pumping and high sand ratio pumping correspond to the sand-carrying liquid stage, increased discharge rate and decreased discharge rate correspond to the transition section of each process, and replacement extrusion corresponds to the replacement extrusion process stage.

[0030] A sliding time window with a length of 20s and a step size of 2s is used to extract the slope and direction of change of the feature vector within the window, and predict the probability of migration to each adjacent working condition within the next 15s. The working condition corresponding to the maximum probability is the predicted feature interval of the next working condition.

[0031] With the target pressure at the bottom of the well (62 MPa) and the target total displacement of the platform as constraints, and the weights of the equipment health and response hysteresis characteristics of each fracturing truck as well, the displacement allocation target of each truck is solved by nonlinear programming. The power-law fluid pipeline flow resistance calculation model is incorporated into the solution, and the nonlinear mapping relationship between the bottom of the well pressure and the displacement of a single truck is used as a constraint. The nonlinear programming method uses the displacement setpoint of each fracturing truck as the optimization variable, with the constraints being that the total displacement of the platform meets the target value and the bottom hole pressure deviation is within the allowable range of ±0.5MPa. The optimization objectives are to maximize the overall health margin of the unit and minimize the pressure response hysteresis dispersion. The optimization objective function is constructed by combining the weight coefficients corresponding to equipment health and response hysteresis. At the same time, the upper and lower limits of the rated displacement of a single fracturing truck and the upper limit of pressure output are included as boundary constraints. The optimal displacement allocation value of each truck is obtained by solving the problem, and the dynamic setpoint sequence of the control loop of a single fracturing truck is generated. Among them, the lower the health of the truck, the higher the constraint weight of its target displacement upper limit and pressure response gain.

[0032] The pressure-displacement coupling control loop of a single fracturing truck adopts a cascade control structure of an outer pressure loop and an inner displacement loop. The outer pressure loop outputs the displacement setpoint to the inner displacement loop, and the inner displacement loop outputs the actuator control quantity. The inner displacement loop adopts fixed parameter PID control, and its response speed is higher than that of the outer pressure loop. The outer pressure ring sampling period is set to 1 second to match the update cycle of the bottom hole flow pressure, and the inner displacement ring sampling period is set to 50 ms to adapt to the response frequency of the plunger pump actuator. An auxiliary closed loop with a sampling period of 200 ms is added to the inner pressure ring at the wellhead to quickly suppress wellhead pressure fluctuations.

[0033] The outer pressure loop uses a fuzzy controller to achieve online self-tuning of PID parameters. The fuzzy controller uses pressure deviation and pressure deviation change rate as core inputs, and sand ratio change and fracturing fluid viscosity change rate as feedforward correction inputs. It adopts a fuzzy inference structure with two-dimensional core input and three-dimensional output, and consists of four modules: fuzzification interface, rule base, fuzzy inference engine, and defuzzification interface.

[0034] The two core input quantities, pressure deviation and pressure deviation change rate, are each divided into seven fuzzy subsets: {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}. The membership function adopts a combination of triangular and Gaussian forms. The universe of discourse for pressure deviation is set to [-3MPa, 3MPa], and the universe of discourse for pressure deviation change rate is set to [-2MPa / s, 2MPa / s].

[0035] The changes in sand ratio and fracturing fluid viscosity are directly corrected by gain coefficients to modify the fuzzy inference output. The correction coefficients are calibrated based on the power-law rheological characteristics of the fracturing fluid. The fuzzy rule base is constructed based on the power-law rheological characteristics of the fracturing fluid and field commissioning experience. It forms 49 basic rules corresponding to the 7×7 input fuzzy subset division. The output is the correction increment of proportional coefficient, integral coefficient, and derivative coefficient. A typical rule is that when the pressure deviation is large positive and the pressure deviation change rate is large negative, the correction increment of proportional coefficient is +0.15, integral coefficient is +0.08, and derivative coefficient is -0.05 to quickly suppress the pressure drop trend. The centroid method is used for defuzzification. The output correction is superimposed on the initial PID reference parameters to achieve online self-tuning.

[0036] The fuzzy inference uses the minimum operation inference method, and the defuzzification uses the centroid method. Finally, the dynamic correction values ​​of the proportional coefficient, integral coefficient, and derivative coefficient are output and superimposed on the initial PID parameters to achieve online self-tuning.

[0037] The fuzzy control rules dynamically adjust the output weights based on the power law exponent and consistency coefficient of the fracturing fluid. The higher the viscosity of the fracturing fluid and the more significant the power law characteristics, the greater the adjustment weight of the proportional coefficient corresponding to the pressure deviation and the higher the feedforward compensation weight corresponding to the change in sand ratio. For the transition zone of fracturing fluid from Newtonian fluid to power-law fluid, a gradual range of regular weights is set to ensure that the adjustment range of control parameters is continuously matched with the changes in the flow characteristics of fracturing fluid.

[0038] By combining pipeline resistance and rheological parameters, the pressure disturbance caused by the step change in sand ratio and fracturing fluid viscosity is calculated in advance, and the corresponding displacement compensation value is generated as the feedforward compensation amount. The superposition node of the feedforward compensation amount is located at the output end of the pressure outer loop controller, and together with the output signal of the pressure PID controller, it is fed into the set value input end of the displacement inner loop.

[0039] Under steady-state conditions, the feedforward compensation remains at zero output. Based on the prediction results of the working condition transition, the pressure disturbance corresponding to the changes in sand ratio and viscosity is calculated in advance before the working condition switch is triggered. The feedforward compensation is generated and pre-invested according to the gradient. When the disturbance actually occurs, the full compensation is completed, and the disturbance is suppressed in advance. When entering a strong disturbance working condition, the feedforward compensation reaches full output.

[0040] For example, 3 seconds before the change of working condition when the sand ratio increases from 8% to 18% in the sand-carrying liquid stage, the feedforward compensation is gradually pre-invested according to the gradual increase gradient, and the full compensation is completed at the moment of working condition change to avoid pipeline pressure oscillation caused by step compensation.

[0041] The system sets two threshold ranges for the pressure deviation change rate and the sand ratio change rate to achieve three-level control mode switching: the lower limit of the pressure deviation change rate is 0.2MPa / s and the upper limit is 0.8MPa / s, and the lower limit of the sand ratio change rate is 1% / s and the upper limit is 3% / s.

[0042] Under steady-state conditions, both indicators are below the lower threshold, so conventional PID control is used. During the transition period between operating conditions, if either of the two indicators is in the transition range and neither exceeds the upper threshold, fuzzy PID control and conventional PID control are used in a weighted mixed output, with the mixed weight calculated by weighting the normalized values ​​of the two indicators. Under strong disturbance conditions, if either of the two indicators exceeds the upper threshold, fuzzy PID control and feedforward compensation control are switched to full.

[0043] The mode switching process is equipped with a hysteresis protection interval to avoid frequent back-and-forth switching at the operating condition boundary; the control weight within the transition interval is dynamically adjusted according to the values ​​of the two indicators. The closer to the upper threshold, the higher the proportion of fuzzy PID and feedforward compensation; the closer to the lower threshold, the higher the proportion of conventional PID.

[0044] The system is equipped with integral limiting processing, with three control modes corresponding to independent integral limiting thresholds. The steady-state condition corresponds to a wider limiting range, the transient condition corresponds to a narrower limiting range, and the strong disturbance condition corresponds to the narrowest limiting range. The integral gain is dynamically adjusted according to the pressure deviation amplitude. The deviation amplitude and the integral gain are inversely related. When the pressure deviation exceeds the preset range, the integral gain is reduced proportionally. The integral limit setting has a hysteresis range for triggering and resetting, which avoids frequent switching of the integral action when the deviation fluctuates near the limit threshold.

[0045] Using the control command of the main fracturing truck as the synchronization benchmark, and combined with the timing of the displacement control command, the command reception time, controller calculation time, actuator response time, pump output time, and pipeline pressure propagation time to the wellhead are collected in real time for each slave truck. The total hysteresis of the entire control link is broken down into four components, and the hysteresis of each component is calibrated. The communication transmission hysteresis is obtained by statistically analyzing the difference between the timestamp of the master vehicle command transmission and the timestamp of the slave vehicle command reception. In this scenario, the calibration value is about 20ms. The controller's operational hysteresis is determined by calibrating the controller's operation cycle from input sampling completion to command output, which is approximately 15ms under normal operational load; the actuator's dead time hysteresis is determined by calibrating the actuator's dead time under different load levels, which is approximately 30ms under rated load. The pressure propagation hysteresis in the pipeline is calculated based on the total length of the fracturing string and the propagation velocity of the pressure wave in the fracturing fluid, which is approximately 70 ms at this well depth.

[0046] The pump displacement amplitude deviation is calculated by combining the current wear level of the pump body, the volumetric efficiency of the plunger, and the displacement set value. Under normal wear, the deviation is about 3% to 5%. This deviation is corrected in real time through the closed-loop control of the inner displacement loop and is not included in the hysteresis calculation.

[0047] Among them, the dead zone hysteresis of the actuator has a piecewise linear mapping relationship with the equipment health; the pump displacement amplitude deviation also has a piecewise linear mapping relationship with the equipment health. It is corrected in real time through the inner loop closed-loop control of the displacement, and divided into multiple continuous intervals according to the health value. Each interval corresponds to a set of hysteresis correction coefficients. When the health level is above 90% (normal range), the hysteresis remains at the factory-calibrated baseline value; when the health level drops to the warning range of 60%~80%, the hysteresis increases proportionally as the health level decreases.

[0048] In this scenario, the total latency of a single vehicle under normal conditions is approximately 135ms, while the total latency of a vehicle with severe wear and tear can reach 180~200ms.

[0049] The advance compensation amount of the control command is generated based on the total hysteresis of the entire link, and it is applied to the timing of the control command issuance in the form of a time offset. The main fracturing truck and each slave truck use the unified clock of Beidou time synchronization as the timing reference. The control command carries the target execution timestamp. The main fracturing truck sends the control command in advance according to the maximum value of the total hysteresis of the entire link of each slave truck. After receiving the command, each slave truck executes at the unified target time point according to the local synchronization clock of Beidou time synchronization, ensuring the synchronization of the actions of multiple trucks. The compensation gain is divided into three levels according to the working condition. The compensation gain is increased by 15%~20% under the high displacement and high sand ratio working condition, and decreased under the small displacement and displacement displacement working condition. The compensation gain for the sand-carrying liquid stage in this scenario is 1.15.

[0050] A two-dimensional load weight matrix of hysteresis and health is constructed, and the weight is calculated with health accounting for 60% and hysteresis accounting for 40%. The baseline weight coefficient corresponding to 100% health and 135ms baseline hysteresis is set to 1.0. For every 10% decrease in health, the corresponding weight coefficient decreases by 0.08. For every 100ms increase in hysteresis, the corresponding weight coefficient decreases by 0.06.

[0051] Hysteresis-Health 2D Load Weighting Coefficient Formula:

[0052] This represents the original weighting coefficient of the k-th fracturing truck; This represents the benchmark weighting coefficient, with a value of 1.0, corresponding to the ideal state of benchmark health and benchmark hysteresis. This represents the weight reduction for every 10 percentage point decrease in health score, with a value of 0.08, corresponding to the weight percentage of the health score dimension. This indicates the step size of the health status change, with a value of 10 (unit: percentage points). This represents the baseline health level, with a value of 100%. This represents the real-time health index of the k-th fracturing truck, expressed in % . This indicates the weight reduction for every 100ms increase in hysteresis, with a value of 0.06, corresponding to the weight percentage of the hysteresis dimension. This indicates the step size of the hysteresis change, with a value of 100ms. This represents the total baseline hysteresis, with a value of 135ms. This represents the total real-time end-to-end latency of the k-th fracturing truck, in milliseconds. This represents the normalized displacement load ratio of the k-th fracturing truck; This indicates the total number of fracturing trucks currently operating online; This represents the sum of the original weighting coefficients for all online fracturing trucks.

[0053] Taking the online operation of 6 fracturing trucks as an example, the actual weight coefficients of each truck are added together to obtain the total weight value. Then, the weight of each truck is divided by the total weight value to obtain the normalized displacement-load ratio. Combined with the platform's total displacement target, the displacement setting value of a single truck is calculated.

[0054] The load adjustment adopts a gradient smooth transition, with a single-step adjustment cycle of 2 seconds and a single-step adjustment amount not exceeding 5% of the current load, to achieve a smooth load switch; the master truck sends the load adjustment command and the synchronization timestamp to each slave truck, and each fracturing truck executes the load change synchronously according to a unified adjustment cycle and single-step adjustment step size.

[0055] During the adjustment process, the deviation between the total wellhead pressure and the output pressure of each vehicle is monitored in real time. When the deviation exceeds the allowable range of ±0.3MPa, the current gradient adjustment is paused, and the single-step adjustment step size of each vehicle is corrected according to the deviation value. The adjustment is continued after the deviation falls back to the allowable range. When a vehicle is taken out of operation due to a fault or a standby vehicle is put into operation, the weight recalculation and normalization process is automatically triggered.

[0056] The timing nodes of fracturing construction procedures are analyzed, the remaining execution time of the current procedure and the total parameter change of the next procedure are extracted, and the parameter change rate and actual switching trigger time are corrected by combining the procedure design timing and short-term working condition migration prediction results. By combining pipeline characteristics and pump efficiency parameters of a single pump, the change sequence of displacement and sand ratio is converted into a time sequence of power demand, and a power preload command is output to the power system. The output timing of the power preload command is ahead of the working condition switching time, and the lead time is matched with the total system hysteresis.

[0057] Based on the rate of power change, the output distribution between the diesel engine and the electric drive unit is matched in three levels: Under steady-state pumping conditions with a low rate of power change, the diesel engine undertakes 90% of the basic load, while the electric drive unit is in hot standby and only participates in minor power peak shaving; Under conditions of slow displacement and sand ratio increase with a medium rate of power change, the diesel engine undertakes the basic power, while the electric drive unit undertakes the dynamic fluctuation component; During the step switching process of conditions with a high rate of power change, the electric drive unit prioritizes to make up for the 80% power gap within 3 seconds, while the diesel engine steadily increases the output power at a preset slope of 5% / s.

[0058] The output switching between different operating conditions adopts a smooth transition method, and the diesel engine output adjustment rate is limited to a preset range to avoid large fluctuations in diesel engine speed; the output response speed of the electric drive unit prioritizes meeting the pressure control requirements, while adjusting the peak shaving depth in combination with the remaining power of the energy storage unit.

[0059] The power reserve coefficient is adjusted according to the health status of each fracturing truck. The power reserve coefficient is the proportion of reserved power to the rated power of the equipment. The power reserve coefficient of vehicles with normal health status is set at 10%, and the power reserve coefficient of vehicles with health status below the 60% warning value is increased to 20%, which limits their maximum output power. The upper limit of the actual continuous output power of a single vehicle is the rated power multiplied by (1-reserve coefficient). The reserved reserve power is used to cope with the temporary power demand caused by fluctuations in operating conditions.

[0060] Combining real-time operating parameters and power output data of the fracturing truck, and based on historical operating data of the fracturing truck, the recursive least squares method is used to perform polynomial fitting on two core degradation trends: engine output power attenuation and plunger pump volumetric efficiency reduction. The fitting input includes four types of time-series data for a single fracturing truck: historical running time, cumulative load rate, cumulative pumping volume, and cumulative start-stop count. The output includes performance degradation curves, degradation rate values, and remaining effective running time for each component. Based on rated performance parameters, the weighted average of the engine output power attenuation percentage and the plunger pump volumetric efficiency reduction percentage is taken as the comprehensive performance degradation rate, with each index having a weight of 0.5. A linear mapping formula is used: Health Index = 100 × (1 - Comprehensive Performance Degradation Rate) to obtain a health index ranging from 0 to 100, where 100 corresponds to the rated state of brand-new equipment, and 60 corresponds to the fault warning threshold.

[0061] A three-level linkage fault tolerance strategy is set based on the absolute value of health and the degradation rate. Each level is linked with the control loop in real time. The preset threshold for degradation rate is a critical value used to determine the abnormal accelerated degradation trend of equipment performance. The unit is % / hour. It is calibrated based on the 90th percentile value of the performance degradation rate in the historical operating data of the same type of fracturing truck. Exceeding this threshold indicates that the equipment has early failure potential.

[0062] Warning level: Health level is greater than 80% but the degradation rate exceeds the preset threshold, or health level is in the range of 60%~80% and the degradation rate exceeds the normal range. Reduce the integral gain of the corresponding vehicle main control circuit, increase its power reserve coefficient, and increase the frequency of parameter monitoring. Functional degradation level: The health status is in the range of 60%~80%. The load ratio of the corresponding vehicle is reduced in a gradient manner, and its pressure response speed requirement and power output limit are reduced simultaneously. Fault isolation level: When the health level is less than 60%, the entire load of the corresponding vehicle is transferred to the redundant fracturing vehicle through the multi-vehicle load balancing loop in a gradient manner, triggering a shutdown warning.

[0063] The redundant fracturing truck is normally in a low-load hot standby state, with the engine maintaining an idle speed and the plunger pump maintaining a low-pressure circulation state. When the health of the faulty vehicle drops to 65% and approaches the fault warning threshold, the redundant vehicle pre-pressurization process is triggered, gradually increasing the pump outlet pressure at a rate of 0.5MPa / s. After the outlet pressure stabilizes within ±0.5MPa of the system working pressure, the load transfer preparation stage begins.

[0064] The load transfer process is divided into three stages: unloading, transition, and cut-out. During the unloading stage, the load ratio of the faulty vehicle is gradually reduced within 10 seconds, while the load ratio of the redundant vehicle is increased simultaneously. During the transition stage, the faulty vehicle is kept running at a low load for 5 seconds. After confirming that the redundant vehicle is running stably, the cut-out stage begins. The entire transfer process controls the wellhead pressure fluctuation within ±0.5MPa to ensure that the fracturing operation is not interrupted.

[0065] Based on the performance degradation trend, predict the remaining time required for the health status to decline to the failure threshold, plan the load transfer sequence and the timing of backup vehicle switching in advance, and reserve sufficient buffer time.

[0066] A full-process simulation model of the fracturing platform is constructed. The model includes four core components: equipment dynamic simulation module, pipeline rheology simulation module, formation filtration simulation module, and control strategy simulation module. It can completely reproduce the entire process of fracturing operation from control command output, dynamic response, pipeline flow to bottom hole pressure response.

[0067] Real-time operating data and equipment status data are forward-synchronized to the simulation model, with the synchronization cycle consistent with the field data acquisition cycle; at the same time, the control commands and fault-tolerant strategies output by the simulation are reverse-synchronized to the actual control system to perform visual simulation and parameter verification of the fracturing construction process. The reverse control parameter synchronization adopts a combination of segmented correction and real-time fine-tuning: segmented correction is used for slowly changing parameters such as formation physical parameters, operating condition characteristic maps, and feedforward compensation coefficients, and batch updates are completed at the progress nodes of each fracturing segment; real-time fine-tuning is used for rapidly changing parameters such as PID control parameters and load distribution weights, and small-scale dynamic adjustments are made according to real-time operating deviations.

[0068] When the single-stage fracturing operation reaches the 1 / 3 and 2 / 3 progress nodes, the full amount of actual operation data of wellhead pressure, platform total displacement and fracturing fluid-sand ratio in the corresponding time period is extracted. Combined with the bottom hole pressure conversion results in the corresponding time period, the correspondence between formation filtration rate and bottom hole pressure response is fitted. The corrected formation filtration coefficient and formation fracture pressure gradient parameters are obtained through parameter inversion. The formation fracture moment is automatically identified by the sudden change in bottom hole pressure; then the corrected formation parameters are substituted into the corresponding formation interval of the fracturing condition characteristic map, and the pressure characteristic interval of the corresponding condition mode and the gain coefficient of the feedforward compensation are adjusted to complete the segmented correction of the characteristic map.

[0069] After a single-stage fracturing operation is completed, dynamic physical property parameters such as formation filtration coefficient and fracture conductivity are retrieved based on the full-stage construction data. The fracturing condition characteristic map and control parameter library are then corrected. At the same time, the characteristic parameters of equipment performance degradation are updated, and the matching parameters of each control link are adjusted synchronously, including the influence coefficient of health on hysteresis, the health weight of power distribution, and the feedforward gain coefficient of fuzzy PID control.

[0070] The control parameter library contains four core control parameters: fuzzy control rule benchmark parameters, feedforward compensation gain benchmark, hysteresis calibration benchmark, and dynamic distribution weight benchmark. When updating, the original parameters are iteratively corrected based on the actual operation data of this construction project, using a weighted average method. The correction weight is positively correlated with the confidence level of the current construction data.

[0071] A four-dimensional evaluation index system was established, which includes pressure fluctuation amplitude, multi-vehicle synchronization deviation, power response time, and working condition identification accuracy. Each index was normalized to a score of 0 to 100 points according to a preset benchmark range. Weight coefficients were set for each of the four indexes, and the overall operation score of the control strategy was obtained by comprehensive weighting. The optimization levels are divided according to the scoring results. Different levels correspond to different parameter adjustment steps. The higher the score, the smaller the step size, to ensure the stability of the optimization process; the lower the score, the larger the step size, to accelerate the convergence speed of the control strategy and apply the optimized control strategy to the next fracturing stage.

[0072] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0073] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An automatic control method for the fracturing platform of a fracturing truck, characterized in that, Includes the following steps: The equipment operation data, process parameters, formation parameters and historical operation data of the fracturing unit are collected, and after time alignment processing based on Beidou time synchronization and bottom hole flow pressure conversion processing, a working condition dataset with a unified spatiotemporal reference is constructed. The real-time operating data is matched with the pre-constructed fracturing operating condition feature map to identify the operating condition mode, predict the operating condition migration trend, and solve the pressure and displacement control target sequence of each fracturing truck by combining multiple constraints. For the pressure and displacement coupled control loop of a single fracturing truck, the pressure deviation, pressure deviation change rate, sand ratio change, and fracturing fluid viscosity change rate are used as inputs. The PID parameters are self-tuned online through preset fuzzy control rules, and feedforward compensation is generated in combination with pipeline rheological parameters. The control mode switching is realized by setting dual threshold ranges for key parameters, and integral amplitude limiting is adopted as a supporting measure. Using the main fracturing truck control command as the synchronization benchmark, the hysteresis of each link in the entire control chain is decomposed and the hysteresis is corrected in combination with the equipment health. Based on the total hysteresis, the control command advance compensation amount is generated, and a two-dimensional load weight matrix of hysteresis and health is constructed. Based on this matrix, the displacement load ratio of each fracturing truck is adjusted according to the gradient.

2. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, The equipment operation data includes engine operating parameters, transmission operating status parameters, plunger pump inlet and outlet pressures and displacement, and lubricating oil temperature; the collected process parameters include the sand ratio, viscosity, and density of the fracturing fluid; and the formation parameters of the fracturing design, construction sequence data, and historical operation data are also connected.

3. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, Based on the PPS pulse signal of BeiDou time synchronization, the data collected by each acquisition node is time-stamped and aligned to eliminate clock drift error between different acquisition units. By combining real-time displacement, sand ratio, fracturing fluid viscosity, and plunger pump inlet and outlet pressure difference, the pipeline friction resistance and local resistance are calculated. The calculation method for bottom hole flow pressure conversion is corrected by combining the rheological characteristics of fracturing fluid. The pressure data collected at the wellhead is converted into the bottom hole flow pressure, realizing the spatial dimension correspondence between wellhead data and formation conditions.

4. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, The fracturing condition characteristic map is pre-set with condition characteristic parameters corresponding to different sand ratios, displacement rates, fracturing fluid types and formation properties. The real-time operating data is compared with the fracturing operating condition feature map to identify the operating condition mode, and the operating condition migration trend is predicted by the sliding time window. Using the bottom-hole target pressure and the platform's total displacement as constraints, and combining the equipment health and response hysteresis characteristics to set weights, the pressure and displacement allocation targets of each fracturing truck are solved through nonlinear programming. The solution incorporates a power-law fluid pipeline flow resistance calculation model to generate a dynamic setpoint sequence for the control loop of a single fracturing truck.

5. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, A fuzzy controller is used to self-tune PID parameters online. The fuzzy controller takes the pressure deviation and the rate of change of pressure deviation as its core inputs, and the sand ratio change and the fracturing fluid viscosity change rate as feedforward correction inputs. It dynamically adjusts the proportional coefficient, integral coefficient, and derivative coefficient by combining the fuzzy control rules set based on the power-law rheological characteristics of the fracturing fluid. The sand ratio change and the fracturing fluid viscosity change rate are corrected and controlled by the rheological gain coefficient. The pressure disturbance caused by sand ratio and viscosity step is calculated by combining the pipeline rheological parameters, and the feedforward compensation amount is generated. Combined with the working condition migration prediction results, the feedforward compensation amount is generated in advance and pre-invested according to the gradient before the working condition switch. The system sets two threshold ranges for the pressure deviation change rate and sand ratio change rate to achieve three-level control mode switching, and adopts integral amplitude limiting processing.

6. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, The total hysteresis of the entire control link is decomposed into four components: communication transmission hysteresis, controller operation hysteresis, actuator dead zone hysteresis, and pipeline pressure propagation hysteresis. The pump displacement amplitude deviation is calculated by combining the pump body wear degree and plunger volumetric efficiency. It is corrected by the inner loop of the displacement and is not included in the hysteresis range. The hysteresis of each part is determined by timestamp comparison, characteristic calibration and fluid dynamics calculation. The dead zone hysteresis of the actuator is negatively correlated with the health of the equipment. The control command advance compensation amount is generated based on the total hysteresis, and the compensation gain is adjusted according to the operating condition type. Construct a two-dimensional load weight matrix of hysteresis and health, and allocate the displacement load ratio of each fracturing truck according to the gradient.

7. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, It also includes a power pre-allocation step: analyzing the timing nodes of the fracturing construction process, identifying the parameter change trend, combining the load allocation result with the working condition migration trend, predicting the rate of change and duration of the next stage of displacement and sand ratio through a sliding time window, calculating the power demand change curve by combining pipeline characteristics and pump efficiency parameters, and outputting a power pre-loading command. The output distribution ratio between diesel engines and electric drive units is matched in three levels according to the rate of power change. Under steady-state conditions, the diesel engine bears the basic load, while the electric drive unit is in hot standby and participates in peak shaving. Under gradually changing operating conditions, the diesel engine bears the basic load, while the electric drive unit bears the dynamic fluctuation component; Under step conditions, the electric drive unit prioritizes making up for the power shortfall, while the diesel engine steadily increases its output power at a preset slope. The power reserve coefficient is adjusted based on the equipment's health status to limit the maximum output power of fracturing trucks with low health status.

8. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, It also includes equipment health management steps: combining the real-time operating parameters and power output data of the fracturing truck, and recursively fitting the performance degradation trend of the engine, transmission, and plunger pump based on the historical operating data of the fracturing truck, to calculate the health index and degradation rate; Fault tolerance strategies are set according to the health index and the degradation rate, and control parameters and load distribution are adjusted in conjunction with each other to predict faults and plan load transfer timing.

9. The automatic control method for the fracturing platform of the fracturing truck according to claim 8, characterized in that, Based on the absolute value of the health index and the degradation rate, a three-level linkage fault tolerance strategy is set: at the early warning level, the integral gain of the corresponding fracturing truck control loop is reduced, the power reserve coefficient is increased, and the frequency of parameter monitoring is increased; The warning level corresponds to a health level greater than 80% but a degradation rate exceeding the preset threshold for degradation rate, or a health level between 60% and 80% and a degradation rate exceeding the normal range. The load ratio of the corresponding fracturing truck is reduced in a gradient manner by functional degradation, and the pressure response speed requirement and power output limit are reduced simultaneously. The fault isolation level uses a multi-vehicle load balancing loop to transfer the entire load of the corresponding fracturing truck to the redundant fracturing truck in a gradient manner, triggering a shutdown warning. Based on the predicted time remaining until the failure occurs according to the performance degradation trend, the load transfer sequence and the timing of the backup fracturing truck's entry are planned.

10. The automatic control method for the fracturing platform of the fracturing truck according to claim 1, characterized in that, It also includes a simulation correction step: constructing a full-process simulation model of the fracturing platform, synchronously mapping real-time operating data and control actions to the full-process simulation model of the fracturing platform, and carrying out parameter inversion in stages to correct the operating parameters and control strategies; The full-process simulation model of the fracturing platform can forwardly map the real-time working condition data and equipment status data to the simulation terminal, and backwardly synchronize the control commands and fault-tolerant strategies to the actual control system, realizing construction visualization simulation and parameter verification. When the single-stage fracturing operation reaches the 1 / 3 and 2 / 3 progress nodes, the formation filtration coefficient and formation fracture pressure gradient parameters are inverted by real-time data, and the fracturing condition characteristic map and feedforward compensation coefficient are corrected. After a single-stage fracturing operation is completed, the formation and equipment parameters are retrieved based on the construction data, and the control parameter library and performance degradation characteristic parameters are updated. Establish a four-dimensional evaluation index system, and automatically adjust the optimization step size of the next round of control strategy iteration based on the evaluation results.