A speed control method and system for a vacuum tube high-speed train

Abstract

The application provides a speed control method and system for a vacuum tube high-speed train, the method comprising: determining the real-time running speed and actual position of a target train through a positioning speed measurement unit, and receiving a speed reference curve sent by a ground control center; predicting the speed tracking error between the train running speed and the speed reference curve within a future preset time period through a speed tracking controller based on the speed reference curve, the real-time running speed and the actual position; optimizing the restraining force of the target train according to the speed tracking error, determining the target restraining force that minimizes the speed tracking error; wherein the restraining force comprises a traction force and a braking force; and controlling the running speed of the target train according to the target restraining force. Through the prediction of the speed tracking error within the future preset time period and the prospective optimization of the restraining force, the application enables the train to respond to the changes in the speed curve in front in advance during high-speed running, and effectively prevents the accident risks caused by insufficient safety distance.

Description

Technical Field

[0001] This application relates to the field of rail transit control technology, and in particular to a speed control method and system for a vacuum tube high-speed train. Background Technology

[0002] With the rapid advancement of global urbanization and the deepening of regional economic integration, existing high-speed rail and civil aviation transportation systems are gradually facing bottlenecks in terms of speed, energy consumption, and punctuality. Evacuated Tube Transport (ETT), as an emerging ultra-high-speed ground transportation mode, creates a low-pressure or near-vacuum environment within long-distance tubes, reducing air resistance to less than one percent of that at atmospheric pressure, thus allowing trains to operate at speeds exceeding 1000 kilometers per hour.

[0003] Existing speed control methods for high-speed railways or maglev trains typically employ proportional-integral-derivative (PID) control based on current errors or automatic train protection based on fixed braking curves. These methods struggle to handle multiple types of hard constraints while predicting future states. When a train approaches the preceding train at high speed and communication experiences random delays, existing methods cannot anticipate and proactively adjust the speed before constraint conflicts occur, often leading to delayed emergency braking or overshooting of braking force, resulting in insufficient safe distances between trains. Summary of the Invention

[0004] This application provides a speed control method and system for high-speed trains in vacuum tubes, which solves the problem that related technologies cannot predict and actively adjust the speed before constraint conflicts occur, resulting in insufficient safe distance between trains.

[0005] The first aspect of this application provides a speed control method for a vacuum tube high-speed train, the speed control method for the vacuum tube high-speed train comprising: The real-time operating speed and actual position of the target train are determined by the positioning and speed measurement unit, and the speed reference curve sent by the ground control center is received. Based on the speed reference curve, the real-time operating speed, and the actual position, the speed tracking controller predicts the speed tracking error between the train's operating speed and the speed reference curve within a preset time period in the future. The traction force of the target train is optimized based on the speed tracking error to determine the target traction force that minimizes the speed tracking error; wherein, the traction force includes traction force and braking force. The speed of the target train is controlled based on the target restraining force.

[0006] Optionally, in a first implementation of the first aspect of this application, the step of predicting the speed tracking error between the train's running speed and the speed reference curve within a future preset time period based on the speed reference curve, the real-time operating speed, and the actual position via a speed tracking controller includes: A position deviation is generated based on the difference between the actual position and the corresponding reference position in the speed reference curve; and a speed deviation is generated based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve. The position deviation and the speed deviation are combined into a first predicted state variable, and the first predicted state variable is processed according to a preset state transition matrix to generate the predicted state variable for the next sampling period; wherein, the state transition matrix is ​​determined by the sampling period, the train mass and the linearized damping coefficient. The predicted state variable of the next sampling period is used as the second predicted state variable and combined with the preset state transition matrix and recursively to generate the predicted state variables corresponding to the next N consecutive sampling periods. The speed deviation component is extracted from the predicted state variables corresponding to N consecutive sampling periods in the future to obtain the vehicle speed tracking error corresponding to N consecutive sampling periods in the future.

[0007] Optionally, in the second implementation of the first aspect of this application, the predicted state variable for the next sampling period is obtained through the following state update model: , in, , m For train quality, b The linearized damping coefficient, The sampling period is k At the current sampling time, For model perturbation, As the first predicted state variable, To control the amount of restraint force, This is the second predicted state variable.

[0008] Optionally, in a third implementation of the first aspect of this application, the step of optimizing the train's traction force based on the speed tracking error and determining the target traction force that minimizes the speed tracking error includes: Based on the vehicle speed tracking error in the next N consecutive sampling periods and the preset speed tracking error weight, a speed deviation evaluation value is generated, and based on the candidate restraining force in the next N consecutive sampling periods and the preset control input penalty weight, a restraining force fluctuation evaluation value is generated. The cumulative evaluation value for the next N consecutive sampling periods is generated based on the speed deviation evaluation value and the restraining force fluctuation evaluation value. Based on the predicted state variables for the next N consecutive sampling periods and the preset terminal cost matrix, a terminal evaluation is generated, and a comprehensive evaluation value is generated by combining the accumulated evaluation value. Under preset constraints, the candidate restraining force corresponding to the minimum comprehensive evaluation value is determined as the target restraining force.

[0009] Optionally, in the fourth implementation of the first aspect of this application, the expression for the target restraining force that minimizes the vehicle speed tracking error is: , in, To minimize vehicle speed tracking error, This represents the step offset at the current sampling time. As the speed tracking error weight, To control the input penalty weights, For the terminal cost matrix, This is the reference speed for the speed baseline curve.

[0010] Optionally, in the fifth implementation of the first aspect of this application, after the step of optimizing the traction force of the target train based on the speed tracking error and determining the target traction force that minimizes the speed tracking error, the method further includes: Based on the target restraint force, predict the target speed and target predicted position of the target train within a future preset time period; When the distance between the predicted target position and the predicted position of the preceding vehicle is less than the safe distance at any time within the preset future time period, a spacing constraint violation instruction is generated. Based on the spacing constraint violation instruction, obtain the candidate restraint force sequence generated in the current optimization cycle, and identify the number of times the upper or lower restraint force threshold is reached in the candidate restraint force sequence. When the number of times the limit is exceeded exceeds a preset threshold, the reference speed in the speed reference curve is successively reduced according to the preset speed reduction step size to generate a reduced speed reference value, and the traction force of the target train is optimized a second time based on the reduced speed reference value.

[0011] A second aspect of this application provides a speed control system for a vacuum tube high-speed train. The speed control system is used to implement a speed control method for the vacuum tube high-speed train. The speed control system includes: The determination module is used to determine the real-time running speed and actual position of the target train through the positioning and speed measurement unit, and to receive the speed reference curve sent by the ground control center; The prediction module is used to predict the speed tracking error between the train's running speed and the speed reference curve within a future preset time period, based on the speed reference curve, the real-time running speed, and the actual position, through the vehicle speed tracking controller. An optimization module is used to optimize the traction force of the target train based on the speed tracking error, and determine the target traction force that minimizes the speed tracking error; wherein the traction force includes traction force and braking force. The control module is used to control the running speed of the target train according to the target restraining force.

[0012] Optionally, the prediction module includes: The generation unit is used to generate a position deviation based on the difference between the actual position and the corresponding reference position in the speed reference curve; and to generate a speed deviation based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve. The processing unit is configured to combine the position deviation and the speed deviation into a first predicted state variable, and process the first predicted state variable according to a preset state transition matrix to generate a predicted state variable for the next sampling period; wherein, the state transition matrix is ​​jointly determined by the sampling period, the train mass, and the linearized damping coefficient; the predicted state variable for the next sampling period is used as a second predicted state variable and combined with the preset state transition matrix and recursively generated to generate predicted state variables corresponding to the next N consecutive sampling periods; The extraction unit is used to extract the speed deviation component from the predicted state variables corresponding to N consecutive sampling periods in the future, and obtain the vehicle speed tracking error corresponding to N consecutive sampling periods in the future.

[0013] A third aspect of this application provides an electronic device, including a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory, and when the processor executes the computer program, it implements the steps of the speed control method for a vacuum tube high-speed train provided in the first aspect of this application.

[0014] The fourth aspect of this application provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps of the speed control method for a vacuum tube high-speed train provided in the first aspect of this application.

[0015] Compared with the prior art, this application has the following beneficial effects: By receiving the speed reference curve sent by the ground control center and combining it with the real-time operating speed and actual position obtained by the positioning and speed measurement unit, the train speed tracking controller predicts the speed tracking error within a preset time period in the future, and optimizes and determines the target traction force accordingly, ultimately controlling the train's operating speed. Through the prediction of the speed tracking error within a preset time period and the forward-looking optimization of the traction force, the train can respond in advance to changes in the speed curve ahead during high-speed operation, effectively preventing the risk of accidents caused by insufficient safety distance. Attached Figure Description

[0016] Figure 1 A schematic flowchart illustrating the speed control method for a vacuum tube high-speed train provided in this application embodiment; Figure 2 A schematic diagram of the program module for the speed control system of a vacuum tube high-speed train provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0017] To make the inventive objectives, features, and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] To address the problem that related technologies cannot predict and proactively adjust train speed before constraint conflicts occur, resulting in insufficient safe distances between trains, this application provides a speed control method for high-speed trains using vacuum tubes. Figure 1 This is a flowchart illustrating the speed control method for a vacuum tube high-speed train provided in this embodiment. The speed control method for the vacuum tube high-speed train includes the following steps: Step 110: Determine the real-time running speed and actual position of the target train through the positioning and speed measurement unit, and receive the speed reference curve sent by the ground control center.

[0019] Specifically, the ground control center pre-calculates and issues a speed control reference curve based on track conditions (curves, turnouts, gradients, etc.), temporary speed limits, and the position of the train ahead. This curve characterizes the target operating speed of the train at different locations. The real-time operating speed reflects the train's current actual motion state, while the actual position determines the current track section. By matching the real-time operating status with the speed reference curve, the relationship between the train's current operating state and the target operating state can be determined. This provides fundamental data support for predicting speed tracking errors and ensures that the control process always revolves around the predetermined operating target, thereby improving the accuracy and controllability of train operation.

[0020] Step 120: Based on the speed reference curve, real-time operating speed, and actual position, predict the speed tracking error between the train's operating speed and the speed reference curve within a preset time period using the speed tracking controller.

[0021] Specifically, the speed tracking controller constructs state variables based on the current operating state and establishes a predictive model in conjunction with the train's motion characteristics. It then recursively calculates the operating state at multiple future sampling times to obtain the potential speed deviation in the future time domain. Since the prediction results reflect not only the current error but also its future development trend, the control system can anticipate speed changes in advance, avoiding lag caused by controlling solely based on the current state and improving tracking accuracy and control stability under high-speed operating conditions.

[0022] Step 130: Optimize the traction force of the target train based on the speed tracking error, and determine the target traction force that minimizes the speed tracking error.

[0023] Specifically, during the optimization process, the speed tracking error within a preset time period is used as the evaluation criterion. A comprehensive evaluation model is established by combining this with changes in the traction force. By calculating and comparing different candidate traction forces, the optimal traction force is determined to achieve the best overall evaluation result. Since the traction force includes both traction and braking force, when the traction force is greater than 0, it is traction force, which controls train acceleration; when the traction force is less than 0, it is braking force, which controls train deceleration. Therefore, both train acceleration and deceleration control can be achieved, allowing the train speed to continuously approach the target speed and reducing the impact of speed fluctuations on operational quality.

[0024] Step 140: Control the running speed of the target train according to the target restraining force.

[0025] Specifically, once the target traction force is determined, the corresponding control command is sent to the train actuator, which then outputs the appropriate traction or braking force to act on the train. As the train's operating status changes, the positioning and speed measurement unit continuously acquires the latest speed and position information and re-executes the speed tracking error prediction and traction force optimization process, enabling the control system to adjust the control output in real time based on the latest operating status. By continuously cyclically executing the above process, the train's operating speed can stably track the speed reference curve, thereby improving operating accuracy, operational stability, and operational safety.

[0026] In one optional implementation of this embodiment, a position deviation is generated based on the difference between the actual position and the corresponding reference position in the speed reference curve; and a speed deviation is generated based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve; the position deviation and speed deviation are combined into a first predicted state variable, and the first predicted state variable is processed according to a preset state transition matrix to generate a predicted state variable for the next sampling period; wherein, the state transition matrix is ​​jointly determined by the sampling period, train mass, and linearized damping coefficient; the predicted state variable for the next sampling period is used as a second predicted state variable and combined with the preset state transition matrix and recursively generated to generate predicted state variables corresponding to the next N consecutive sampling periods; the speed deviation component is extracted from the predicted state variables corresponding to the next N consecutive sampling periods to obtain the vehicle speed tracking error corresponding to the next N consecutive sampling periods.

[0027] In this embodiment, based on the cumulative running time of the train from the start time, a reference speed corresponding to the cumulative running time is extracted from the speed reference curve, and the reference speed is integrated over time to generate a corresponding reference position. A position deviation is generated based on the difference between the actual position and the reference position, and a speed deviation is generated based on the difference between the real-time running speed and the reference speed. The position deviation reflects the distance difference between the actual arrival position and the planned arrival position, while the speed deviation reflects the difference between the actual running speed and the target running speed. By simultaneously introducing position and speed deviations, the train's operating state can be described from two dimensions: spatial position and running speed. After obtaining the position and speed deviations, they are combined to form the first predicted state variable. The predicted state variable characterizes the train's comprehensive operating state at the current moment, reflecting both the degree of deviation from the target trajectory and the trend of the deviation. The first predicted state variable is processed by a state transition matrix to obtain the predicted state variable corresponding to the next sampling period. The state transition matrix describes the relationship between the current state and the future state, and its internal parameters are determined by the sampling period, train mass, and linearized damping coefficient. The sampling period represents the time interval between two adjacent state updates, determining the time scale of state changes. Train mass reflects the train's inertial characteristics; the greater the mass, the smaller the speed change under the same restraining force. The linearized damping coefficient characterizes the comprehensive impact on speed changes during operation, including residual air resistance, equipment mechanical resistance, and other consumption factors related to speed changes. By incorporating these parameters into the state transition matrix, a correspondence between the current operating state and the operating state at the next moment can be established. For example, in the aforementioned scenario, the train has a position lag of 20m and a speed lag of 5m / s. After the sampling period ends, due to the continuous influence of speed deviation on position changes, the position deviation at the next moment may expand to 20.1m, while the speed deviation may change to 4.8m / s depending on the restraining force, thus forming new predicted state variables.

[0028] After obtaining the predicted state variables for the next sampling period, these are used as the second predicted state variables and further correlated with the state transition matrix, continuously recursively generating predicted state variables for the next N consecutive sampling periods. The recursive process essentially involves continuously extrapolating the operating state at multiple future moments based on the current operating state. For example, if the train maintains its current control state, the changes in position and speed deviations for the 1st, 2nd, and up to the Nth sampling periods can be obtained sequentially, forming a state sequence covering a preset future duration. Since each predicted state variable includes both position and speed deviations, the resulting state sequence can fully describe the train's future operating trend. When the train is in an acceleration zone, the recursive result may show a gradual decrease in speed deviation, while when the train approaches a speed-limited zone, the recursive result may show a gradual increase in speed deviation, thus reflecting the future direction of the operating state.

[0029] After generating the predicted state variables for N consecutive sampling periods, the speed deviation component is extracted from each predicted state variable to obtain the train speed tracking error for those N consecutive sampling periods. The train speed tracking error represents the degree of deviation between the actual operating speed and the target operating speed at each future moment. This allows the train to identify potential future speed deviation accumulation trends in advance and proactively adjust the traction force output before the deviation widens, preventing the train from deviating from the predetermined operating state for extended periods. This enables the train to proactively adjust its speed to a safe range before approaching the preceding train or entering a speed-limited section, without relying on emergency braking, thus avoiding sudden stops or collisions caused by accumulated deviations.

[0030] It should be noted that the train moves within the vacuum tube and is subject to traction, mechanical resistance, and residual air resistance. This is a nonlinear dynamic relationship because air resistance is proportional to the square of the speed. While the nonlinear model is accurate, the computational load for online optimization of the speed tracking controller is substantial. To meet real-time requirements (e.g., solving within 20ms), the nonlinear model is linearized near the reference speed.

[0031] Take state variables ,in , Where T represents time. This serves as the reference position for the baseline curve. This is the reference velocity for the baseline curve. Assuming the reference velocity changes slowly, the linearized result is: , in, This represents the change in the deviation state. = - , Indicates the sampling time. For train quality, The linearized damping coefficient, To control force deviation, This represents the model disturbance term (including unmodeled drag, ramp, etc.). Discretization yields the standard form: ， in, , The sampling period is k At the current sampling time, For model perturbation terms ， As the first predicted state variable, To control the amount of restraint force, This is the second predicted state variable. , Electromagnetic traction force, This is electromagnetic resistance.

[0032] In one optional implementation, the step of optimizing the train's traction force based on the speed tracking error and determining the target traction force that minimizes the speed tracking error includes: generating a speed deviation evaluation value based on the speed tracking error over the next N consecutive sampling periods and a preset speed tracking error weight; generating a traction force fluctuation evaluation value based on candidate traction forces over the next N consecutive sampling periods and a preset control input penalty weight; generating an accumulated evaluation value over the next N consecutive sampling periods based on the speed deviation evaluation value and the traction force fluctuation evaluation value; generating a terminal evaluation based on the predicted state variables over the next N consecutive sampling periods and a preset terminal cost matrix, and generating a comprehensive evaluation value by combining the accumulated evaluation value; and determining the candidate traction force corresponding to the minimum comprehensive evaluation value as the target traction force under preset constraints.

[0033] In this embodiment, it can be understood that the speed tracking error reflects the deviation between the predicted operating speed and the reference speed in each future sampling period, while the speed tracking error weight is used to characterize the importance of different errors to the control target. The larger the weight value, the higher the proportion of the corresponding error in the subsequent evaluation process. In the specific processing, the vehicle speed tracking error corresponding to the next N consecutive sampling periods is obtained and weighted according to the corresponding weights, so that the moments with larger deviations or higher importance can have a more significant impact on the final evaluation result, thereby obtaining the speed deviation evaluation value. At the same time, the traction force fluctuation evaluation value is generated based on the candidate traction forces in the next N consecutive sampling periods and the preset control input penalty weight. The candidate traction force represents a set of traction or braking force sequences that can be selected during the control process. Different candidate traction forces correspond to different speed change trends. The control input penalty weight is used to constrain the traction force change amplitude, and its function is to avoid frequent or drastic changes in traction force that lead to a decrease in train running smoothness and an increase in the load on the actuator. During the processing, the changes in candidate restraint forces in each future sampling period are calculated and weighted by combining the control input penalty weights. When a candidate restraint force sequence changes significantly between adjacent sampling periods, the corresponding restraint force fluctuation evaluation value will increase significantly; when the candidate restraint force changes relatively smoothly, the corresponding evaluation value will be relatively small.

[0034] After obtaining the speed deviation assessment value and the restraining force fluctuation assessment value, the two are fused to generate a cumulative assessment value for the next N consecutive sampling periods. The cumulative assessment value is used to comprehensively reflect the balance between speed tracking effect and control stability. When the speed deviation is small but the restraining force fluctuation is large, the cumulative assessment value may still remain at a high level; when the restraining force changes smoothly but the speed deviation is large, the cumulative assessment value will not reach a better result. Therefore, only candidate restraining force sequences that take into account both speed tracking accuracy and control output stability can obtain a smaller cumulative assessment value.

[0035] Based on this, a terminal evaluation is generated using the predicted state variables from N consecutive sampling periods and a preset terminal cost matrix. The terminal cost matrix describes the importance of the state at the end of the prediction interval and is essentially a set of parameters for enhancing the evaluation of the predicted endpoint state. Since future predictions only cover a finite time range, additional constraints are needed on the operating state at the predicted endpoint to prevent the control process from focusing only on short-term effects within the current prediction interval while ignoring subsequent operating trends. For example, when a train is about to enter a speed-limited section, a candidate traction force may achieve a small speed deviation within the prediction interval, but if it maintains a high speed deviation at the predicted endpoint, its corresponding terminal evaluation value will be large; another candidate traction force may have a slightly higher speed deviation in the short term, but if it is close to the target state at the predicted endpoint, its corresponding terminal evaluation value will be small. Finally, the terminal evaluation is combined with the accumulated evaluation value to generate a comprehensive evaluation value. The comprehensive evaluation value is used to uniformly measure the control effect of the candidate traction force throughout the entire prediction interval and at the predicted endpoint. After completing the comprehensive evaluation, all candidate traction forces are compared under preset constraints. The preset constraints limit the allowable range of control output and operating state, including traction capacity limits, braking capacity limits, and operating safety limits. By screening the comprehensive evaluation values ​​corresponding to each group of candidate restraining forces that meet the constraints, the group of candidate restraining forces with the smallest comprehensive evaluation value is determined as the target restraining force.

[0036] It should be noted that the expression for the target restraining force that minimizes the vehicle speed tracking error is: , in, This represents the step offset at the current sampling time, with a value ranging from 0, 1, 2, ... N -1, As the speed tracking error weight, To control the input penalty weights, For the terminal cost matrix, This is the reference speed for the speed baseline curve. For speed tracking error term, For the restraint force smoothing term, This is the terminal cost term.

[0037] In one optional implementation, the target running speed and target predicted position of the target train within a preset time period are predicted based on the target restraining force; when the distance between the predicted target position and the predicted position of the preceding train at any time within the preset time period is less than the safe distance, a spacing constraint violation instruction is generated; a candidate restraining force sequence generated in the current optimization cycle is obtained based on the spacing constraint violation instruction, and the number of times the upper or lower limit threshold of the restraining force is reached in the candidate restraining force sequence is identified; when the number of exceedances exceeds a preset threshold, a reduced speed reference value is generated by successively decreasing the reference speed in the speed reference curve according to the preset speed reduction step size, and the restraining force of the target train is optimized a second time based on the reduced speed reference value.

[0038] In this embodiment, predicting the target speed and predicted position of the target train within a preset time period based on the target traction force essentially involves extrapolating the train's future trajectory based on the current operating state. The target speed represents the predicted speed at each future moment under the continuous action of the target traction force, and the predicted position represents the possible location the train might reach at each future moment after continuing to operate according to the corresponding speed change trend. Since the target traction force comprehensively considers speed tracking requirements and operational constraints, using this traction force for state extrapolation allows for the prediction of the train's operating trend over a future period.

[0039] For example, in a scenario where a high-speed train in a vacuum tube is traveling at 300 m / s and another train is ahead, after determining the target braking force in the current control cycle, the changes in operating speed at each future sampling time can be calculated sequentially. Furthermore, the corresponding position changes can be calculated based on the speed changes, thus forming a predicted trajectory sequence composed of future speeds and future positions. This predicted trajectory not only reflects the future operating state of the train but also provides a basis for determining safe distances. After obtaining the target predicted position, it is correlated with the predicted position of the preceding train to determine whether there are any safety risks during future operation. The predicted position of the preceding train represents its predicted operating position within a preset time period in the future. This can be achieved by the preceding train sending its predicted position to the ground control center, which then transmits it to the target train. The safe distance represents the minimum allowable interval between the two trains. When the distance between the target predicted position and the preceding train's predicted position is less than the safe distance at any future time, it indicates that continuing operation according to the current control trend will cause the distance between the two trains to approach or even exceed the safe boundary, thus creating a potential collision risk. Therefore, the system generates a distance constraint violation command. Spacing constraint violation instructions are used to indicate that the current control results can no longer meet the requirements for safe operation between trains. In essence, they are a kind of safety alarm sign used to trigger safety control procedures.

[0040] After generating the spacing constraint violation instruction, the candidate restraint force sequence generated within the current optimization cycle is further obtained. The candidate restraint force sequence represents multiple combinations of traction or braking force changes evaluated during the control calculation process, with each sequence corresponding to a future operating trend. The candidate restraint force sequences are then analyzed to identify the number of times they exceed the upper or lower restraint force threshold. The upper restraint force threshold represents the maximum traction capacity that the actuator can provide, and the lower restraint force threshold represents the maximum braking capacity that the actuator can provide. When a candidate restraint force reaches its corresponding threshold, it indicates that the control system has utilized the actuator's capacity to its limit. The number of exceedances is used to statistically analyze the frequency of reaching the capacity boundary throughout the entire prediction interval.

[0041] For example, in the scenario where two trains are close together, to maintain a safe distance, the control calculation may continuously attempt to increase the braking force. If a large number of candidate braking forces reach the maximum braking force limit in multiple consecutive sampling periods, it indicates that the current speed target is approaching the limit of the actuator's adjustment capability, and it is difficult to further expand the safe distance by relying solely on the current control strategy. When the number of times the limit is exceeded exceeds a preset threshold, it indicates that the reference speed corresponding to the current speed reference curve can no longer continuously meet the safety requirements under the given constraints. Therefore, it is necessary to actively downgrade the operating target. The preset speed reduction step size represents the speed reduction value each time the reference speed is adjusted. By successively reducing the reference speed in the speed reference curve according to the speed reduction step size, a new downgraded speed reference value can be generated. After the downgraded speed reference value is formed, since the new reference speed is lower than the original operating target, the control output will gradually change from traction to braking, thereby causing the train speed to decrease in advance, increasing the distance between the train and the preceding train, and restoring the future predicted distance to within the safe range, thus achieving active suppression and safety control of operational risks.

[0042] In one optional implementation, the estimated time for the target train to enter the communication relay area is determined based on the target train's current location, operating speed, and the location of adjacent zone boundaries. When the time interval between the estimated time and the current time is less than a preset relay preparation threshold, an auxiliary communication link is established between the target train and the target zone's wireless base station, and the communication quality parameters corresponding to the auxiliary communication link are obtained. Based on the first communication quality parameter corresponding to the current communication link and the second communication quality parameter corresponding to the auxiliary communication link, a link switching evaluation result is generated. When the link switching evaluation result meets the switching conditions, the data transmission task of the target train is switched from the current communication link to the auxiliary communication link, and the auxiliary communication link is updated to the current communication link. After the link switching is completed, the communication resources corresponding to the original current communication link are released.

[0043] In this embodiment, the estimated time when the target train enters the communication relay area is determined based on the target train's current position, operating speed, and the position of the adjacent zone boundary. The communication relay area is the region where the wireless coverage of two adjacent zones overlaps, within which the target train can simultaneously receive wireless signals from the current zone's wireless base station and the next zone's wireless base station. The estimated time represents the predicted time when the target train will reach the corresponding position in the communication relay area after continuing to operate in the current state. Due to the high operating speed of the vacuum tube high-speed train, if a new communication connection is established only after the train reaches the zone boundary, control data interruption or increased communication delay may occur. Therefore, it is necessary to predict the time of approaching the boundary in advance. When the time interval between the estimated time and the current time is less than a preset relay preparation threshold, an auxiliary communication link is established between the target train and the target zone's wireless base station, and the communication quality parameters corresponding to the auxiliary communication link are obtained. The auxiliary communication link represents an additional communication connection established with the next zone's wireless base station while maintaining the current communication connection, thus forming a dual-link communication state. The communication quality parameters are used to characterize the reliability of the wireless communication link, and can be composed of indicators such as received signal strength, signal-to-noise power ratio, packet loss rate, and transmission delay. As the target train approaches the zone boundary, the wireless propagation environment continuously changes. Therefore, it is necessary to obtain the communication quality parameters of the auxiliary communication link in real time to provide a basis for link switching. For example, after the train enters the communication relay area, the signal quality corresponding to the current zone base station gradually decreases, while the signal quality corresponding to the next zone base station gradually improves. At this time, the auxiliary communication link is already capable of undertaking data transmission tasks.

[0044] Link handover evaluation results are generated based on the first communication quality parameter corresponding to the current communication link and the second communication quality parameter corresponding to the auxiliary communication link. These results reflect whether the next zone's communication link has met the conditions to replace the current link. During generation, not only are the communication quality values ​​at a single moment considered, but also the trend of communication quality changes, thus avoiding false handovers due to instantaneous wireless fluctuations. For example, if the communication quality corresponding to the next zone's base station is higher than that of the current zone's base station for several consecutive sampling periods, it indicates that the train is gradually entering the coverage area of ​​the next zone, and the link handover evaluation results will gradually approach the handover conditions. Conversely, if the communication quality only increases briefly within a single sampling period, the current communication state remains unchanged, thereby improving communication handover stability. When the link handover evaluation results meet the handover conditions, the data transmission task of the target train is switched from the current communication link to the auxiliary communication link, and the auxiliary communication link is updated to the current communication link. The data transmission task includes uploading train operation status, receiving speed control commands, and exchanging safety monitoring data. Since the auxiliary communication link has been established and verified before the handover, the data transmission task can be directly migrated to the new wireless base station, achieving uninterrupted data communication. For example, during the process of a train moving from zone A to zone B, if the communication quality of zone B is consistently better than that of zone A, the onboard communication unit will synchronously migrate the control command reception service and the operation data upload service to the wireless base station of zone B, thereby completing the communication relay.

[0045] After the handover is completed, the communication resources corresponding to the original current communication link are released, causing the original wireless base station to stop allocating communication resources to the target train, thereby reducing wireless resource consumption and improving the overall communication efficiency of the system. This forms a complete wireless vehicle-to-ground communication handover process from relay area prediction, dual link establishment, communication quality assessment, link handover, and resource release, thus ensuring the continuity and reliability of control data transmission during train cross-regional operation.

[0046] In one optional implementation, the train's operating state is determined based on the relationship between the target train's current position and the partition boundary position; a corresponding partition state variable is constructed based on the train's operating state; when the target train crosses the partition boundary, a discrete state switching event is triggered, and the partition state variable is updated based on the switching event; and communication state information after the switch is generated based on the updated partition state variable.

[0047] In this embodiment, in the vacuum tube high-speed transportation system, the line is divided into multiple continuous zones, each corresponding to an independent wireless base station and ground control equipment. Therefore, the train corresponds to different communication service objects at different locations. To describe the train's operating status in different zones, a state description structure composed of continuous and discrete states is introduced. Continuous states represent information such as train position, speed, and communication quality parameters that change continuously over time, while discrete states represent the current zone number to which the train belongs. When the target train is within the coverage area of ​​the current zone, the discrete state remains unchanged; when the train approaches the zone boundary, the discrete state enters a waiting-to-switch state, thus achieving state representation of the cross-zone operation process. After determining the train's operating status, corresponding zone state variables are constructed based on the train's operating status. Zone state variables are used to uniformly describe the relationship between the train's operating status and communication status, and can be represented as a set of states composed of continuous and discrete state variables. Continuous state variables reflect the changes in the train's position and communication quality within the current zone, while discrete state variables reflect the zone affiliation of the train.

[0048] When a target train crosses a zone boundary, a discrete state switching event is triggered, and the zone state variables are updated based on the switching event. The discrete state switching event characterizes the process of a train moving from the current zone to the next. A switching event is considered triggered when the train reaches a preset boundary position or meets the switching conditions after entering the communication relay area. After the switching event occurs, the discrete state variables are updated from the current zone number to the next zone number. Simultaneously, the set of communication parameters corresponding to the current zone is switched to the set of communication parameters corresponding to the next zone. Communication status information after the switch is generated based on the updated zone state variables. This communication status information characterizes the train's current corresponding radio base station, communication resource allocation, and communication link availability. Because the updated zone state variables already contain the new zone affiliation and corresponding continuous operating status, the communication service object after the train's switch can be accurately determined. For example, when a target train moves from zone three to zone four, the updated discrete state variables indicate that the train has completed the zone affiliation change, and the communication status information is updated to the communication status information corresponding to the radio base station in zone four. This allows subsequent communication quality assessment, communication delay prediction, and communication relay control to be processed based on the latest zone status. By using the co-evolution of continuous and discrete states, it is possible not only to describe the continuous motion characteristics during train operation, but also to describe the discrete changes in communication service objects during cross-regional operation, thereby achieving unified modeling and state management of the region switching process.

[0049] It should be noted that the zone switching process involves the interaction between continuous states (train position, speed) and discrete states (zone operating mode). Each zone... Define a finite number of discrete states: , in, This indicates that there are no trains in the zone, and the gate valve can be closed for maintenance or vacuum initialization. This indicates that the zone has been evacuated to a vacuum, the gate valve is open, the trackside radio is on standby, and the train is waiting to enter. This indicates that a train is running within this section, the traction control system is activated, and real-time communication is maintained. This indicates that a leak, fire, or train malfunction has occurred, and the zone enters isolation mode, closing the gate valves, opening the escape doors, and cutting off the traction power.

[0050] Train from zone When the train approaches section M, the following sequence of events will occur (assuming the train is traveling from left to right): Triggering condition: Train position ,in The coordinates of the partition boundary, To trigger the distance in advance (e.g., 200 m).

[0051] Event 1, Partition The operation control system sends a "train approaching" message to section M.

[0052] Event 2: Section M switches from STANDBY to OCCUPIED and activates the traction control system to prepare for train reception.

[0053] Event 3: The vehicle-mounted radio unit began simultaneously monitoring different zones. and partitions The timing of switching is determined based on the signal strength or signal-to-noise ratio (SNR) of the wireless signal.

[0054] Event 4: When When there is hysteresis, communication relay is completed, and vehicle-mounted operation control is switched to zone control. Ground instructions.

[0055] Incident 5: The train completely leaves the zone. After that, partition Switch back from OCCUPIED to STANDBY (if there are no other trains behind) or IDLE.

[0056] According to the speed control method for a vacuum tube high-speed train provided in this application, the real-time operating speed and actual position of the target train are determined by a positioning speed measuring unit, and a speed reference curve sent by a ground control center is received. Based on the speed reference curve, real-time operating speed, and actual position, a speed tracking controller predicts the speed tracking error between the train's operating speed and the speed reference curve within a preset time period. The traction force of the target train is optimized based on the speed tracking error to determine a target traction force that minimizes the speed tracking error. The traction force includes traction and braking force. The operating speed of the target train is controlled based on the target traction force. This application avoids the problem of insufficient safety distance caused by delays in traditional feedback control by predicting the speed tracking error within a preset time period and proactively optimizing the traction force.

[0057] Figure 2 This application provides a speed control system for a vacuum tube high-speed train, which can be used to implement the speed control method for the vacuum tube high-speed train described in the foregoing embodiments. Figure 2 As shown, the speed control system of this vacuum tube high-speed train mainly includes: The determination module 10 is used to determine the real-time running speed and actual position of the target train through the positioning and speed measurement unit, and to receive the speed reference curve sent by the ground control center. Prediction module 20 is used to predict the speed tracking error between the train's running speed and the speed reference curve within a preset time period, based on the speed reference curve, real-time running speed, and actual position, through the speed tracking controller. The optimization module 30 is used to optimize the traction force of the target train based on the speed tracking error, and determine the target traction force that minimizes the speed tracking error; wherein, the traction force includes traction force and braking force; The control module 40 is used to control the running speed of the target train according to the target restraining force.

[0058] In one optional implementation, the prediction module includes: a generation unit, used to generate a position deviation based on the difference between the actual position and the corresponding reference position in the speed reference curve; and to generate a speed deviation based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve; a processing unit, used to combine the position deviation and the speed deviation into a first predicted state variable, and to process the first predicted state variable according to a preset state transition matrix to generate a predicted state variable for the next sampling period; wherein, the state transition matrix is ​​jointly determined by the sampling period, the train mass, and the linearized damping coefficient; the predicted state variable for the next sampling period is used as a second predicted state variable and combined with the preset state transition matrix and recursively generated to generate predicted state variables corresponding to the next N consecutive sampling periods; and an extraction unit, used to extract the speed deviation component from the predicted state variables corresponding to the next N consecutive sampling periods to obtain the vehicle speed tracking error corresponding to the next N consecutive sampling periods.

[0059] In one optional implementation, the optimization module is specifically used to: generate a speed deviation evaluation value based on the vehicle speed tracking error over the next N consecutive sampling periods and a preset speed tracking error weight; generate a restraint force fluctuation evaluation value based on the candidate restraint forces over the next N consecutive sampling periods and a preset control input penalty weight; generate an accumulated evaluation value for the next N consecutive sampling periods based on the speed deviation evaluation value and the restraint force fluctuation evaluation value; generate a terminal evaluation based on the predicted state variables over the next N consecutive sampling periods and a preset terminal cost matrix; and generate a comprehensive evaluation value by combining the accumulated evaluation value; and, under preset constraints, determine the candidate restraint force corresponding to the minimum comprehensive evaluation value as the target restraint force.

[0060] In an optional implementation, the optimization module is further configured to: predict the target operating speed and target predicted position of the target train within a preset time period based on the target restraining force; generate a spacing constraint violation instruction when the distance between the target predicted position and the predicted position of the preceding train is less than a safe distance at any time within the preset time period; obtain a candidate restraining force sequence generated in the current optimization cycle based on the spacing constraint violation instruction, and identify the number of times the candidate restraining force sequence exceeds the upper or lower limit threshold of the restraining force; when the number of exceedances exceeds a preset threshold, generate a downgraded speed reference value by successively decreasing the reference speed in the speed reference curve according to a preset deceleration step size, and perform secondary optimization of the restraining force of the target train based on the downgraded speed reference value.

[0061] According to the speed control system of a vacuum tube high-speed train provided in this application, the real-time running speed and actual position of the target train are determined by a positioning and speed measuring unit, and a speed reference curve sent by the ground control center is received. Based on the speed reference curve, real-time running speed, and actual position, the speed tracking controller predicts the speed tracking error between the train's running speed and the speed reference curve within a preset time period. The traction force of the target train is optimized based on the speed tracking error to determine the target traction force that minimizes the speed tracking error. The traction force includes traction force and braking force. The running speed of the target train is controlled based on the target traction force. This application avoids the problem of insufficient safety distance caused by delay in traditional feedback control by predicting the speed tracking error within a preset time period and proactively optimizing the traction force.

[0062] According to the scheme provided in this application Figure 3 An electronic device is provided as an embodiment of this application. This electronic device can be used to implement the speed control method for a vacuum tube high-speed train in the foregoing embodiments, mainly including: The system includes a memory 301, a processor 302, and a computer program 303 stored on the memory 301 and executable on the processor 302. The memory 301 and the processor 302 are connected via communication. When the processor 302 executes the computer program 303, it implements the speed control method for the vacuum tube high-speed train described in the foregoing embodiments. The number of processors can be one or more.

[0063] The memory 301 can be a high-speed random access memory (RAM) or a non-volatile memory, such as a disk storage device. The memory 301 is used to store executable program code, and the processor 302 is coupled to the memory 301.

[0064] Furthermore, embodiments of this application also provide a computer-readable storage medium, which may be disposed in the electronic device described in the above embodiments, and the computer-readable storage medium may be as described above. Figure 3 The memory in the illustrated embodiment.

[0065] The computer-readable storage medium stores a computer program that, when executed by a processor, implements the speed control method for the vacuum tube high-speed train described in the foregoing embodiments. Furthermore, the computer-readable storage medium can also be a USB flash drive, a portable hard drive, a read-only memory (ROM), RAM, a magnetic disk, or an optical disk, or any other medium capable of storing program code.

[0066] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0067] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0068] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method of controlling the speed of a vacuum tube train, characterized by, include: The real-time operating speed and actual position of the target train are determined by the positioning and speed measurement unit, and the speed reference curve sent by the ground control center is received. Based on the speed reference curve, the real-time operating speed, and the actual position, the speed tracking controller predicts the speed tracking error between the train's operating speed and the speed reference curve within a preset time period in the future. The traction force of the target train is optimized based on the speed tracking error to determine the target traction force that minimizes the speed tracking error; wherein, the traction force includes traction force and braking force. The speed of the target train is controlled according to the target restraining force. The step of predicting the speed tracking error between the train's operating speed and the speed reference curve within a preset time period based on the speed reference curve, the real-time operating speed, and the actual position, using a speed tracking controller, includes: A position deviation is generated based on the difference between the actual position and the corresponding reference position in the speed reference curve, and a speed deviation is generated based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve. The position deviation and the speed deviation are combined into a first predicted state variable, and the first predicted state variable is processed according to a preset state transition matrix to generate the predicted state variable for the next sampling period; wherein, the state transition matrix is ​​determined by the sampling period, the train mass and the linearized damping coefficient. The predicted state variable of the next sampling period is used as the second predicted state variable and combined with the preset state transition matrix and recursively to generate the predicted state variables corresponding to the next N consecutive sampling periods. The speed deviation component is extracted from the predicted state variables corresponding to N consecutive sampling periods in the future to obtain the vehicle speed tracking error corresponding to N consecutive sampling periods in the future. After the step of optimizing the traction force of the target train based on the speed tracking error and determining the target traction force that minimizes the speed tracking error, the method further includes: Based on the target restraint force, predict the target speed and target predicted position of the target train within a future preset time period; When the distance between the predicted target position and the predicted position of the preceding vehicle is less than the safe distance at any time within the preset future time period, a spacing constraint violation instruction is generated. Based on the spacing constraint violation instruction, obtain the candidate restraint force sequence generated in the current optimization cycle, and identify the number of times the upper or lower restraint force threshold is reached in the candidate restraint force sequence. When the number of times the limit is exceeded exceeds a preset threshold, the target train undergoes active degradation processing. The reference speed in the speed reference curve is gradually reduced according to the preset deceleration step size to generate a degraded speed reference value. The traction force of the target train is then optimized a second time based on the degraded speed reference value.

2. The speed control method for a high-speed train in a vacuum tube according to claim 1, characterized in that, The predicted state variables for the next sampling period are obtained through the following state update model: ， in, , For train quality, The linearized damping coefficient, The sampling period is k At the current sampling time, For model perturbation terms, As the first predicted state variable, To control the amount of restraint, This is the second predicted state variable.

3. The speed control method for a vacuum tube high-speed train according to claim 1, characterized in that, The step of optimizing the train's traction force based on the speed tracking error and determining the target traction force that minimizes the speed tracking error includes: Based on the vehicle speed tracking error in the next N consecutive sampling periods and the preset speed tracking error weight, a speed deviation evaluation value is generated, and based on the candidate restraining force in the next N consecutive sampling periods and the preset control input penalty weight, a restraining force fluctuation evaluation value is generated. The cumulative evaluation value for the next N consecutive sampling periods is generated based on the speed deviation evaluation value and the restraining force fluctuation evaluation value. Based on the predicted state variables for the next N consecutive sampling periods and the preset terminal cost matrix, a terminal evaluation is generated, and a comprehensive evaluation value is generated by combining the accumulated evaluation value. Under preset constraints, the candidate restraining force corresponding to the minimum comprehensive evaluation value is determined as the target restraining force.

4. The speed control method for a vacuum tube high-speed train according to claim 3, characterized in that, The expression for the target restraining force that minimizes the vehicle speed tracking error is: ， in, To minimize vehicle speed tracking error, This represents the step offset at the current sampling time. For speed tracking error weights, To control the input penalty weights, For the terminal cost matrix, This is the reference speed for the speed baseline curve.

5. A speed control system for a vacuum tube high-speed train, characterized in that, The speed control system of the vacuum tube high-speed train is used to implement the speed control method of the vacuum tube high-speed train according to claim 1. The speed control system of the vacuum tube high-speed train includes: The determination module is used to determine the real-time running speed and actual position of the target train through the positioning and speed measurement unit, and to receive the speed reference curve sent by the ground control center; The prediction module is used to predict the speed tracking error between the train's running speed and the speed reference curve within a future preset time period, based on the speed reference curve, the real-time running speed, and the actual position, through the vehicle speed tracking controller. An optimization module is used to optimize the traction force of the target train based on the speed tracking error, and determine the target traction force that minimizes the speed tracking error; wherein the traction force includes traction force and braking force. The control module is used to control the running speed of the target train according to the target restraining force; The prediction module includes: a generation unit, used to generate a position deviation based on the difference between the actual position and the corresponding reference position in the speed reference curve, and to generate a speed deviation based on the difference between the real-time running speed and the corresponding reference speed in the speed reference curve; a processing unit, used to combine the position deviation and the speed deviation into a first predicted state variable, and to process the first predicted state variable according to a preset state transition matrix to generate a predicted state variable for the next sampling period; wherein, the state transition matrix is ​​jointly determined by the sampling period, train mass, and linearized damping coefficient; the predicted state variable for the next sampling period is used as a second predicted state variable and combined with the preset state transition matrix and recursively generated to generate predicted state variables corresponding to the next N consecutive sampling periods; and an extraction unit, used to extract the speed deviation component from the predicted state variables corresponding to the next N consecutive sampling periods to obtain the vehicle speed tracking error corresponding to the next N consecutive sampling periods. The control module is further configured to predict the target running speed and target predicted position of the target train within a future preset time period based on the target restraining force; generate a spacing constraint violation instruction when the distance between the target predicted position and the predicted position of the preceding vehicle is less than a safe distance at any time within the future preset time period; obtain a candidate restraining force sequence generated in the current optimization cycle based on the spacing constraint violation instruction, and identify the number of times the candidate restraining force sequence exceeds the upper or lower limit threshold of the restraining force; when the number of exceedances exceeds a preset threshold, the running target undergoes active degradation processing, successively reducing the reference speed in the speed reference curve according to a preset speed reduction step size to generate a degraded speed reference value, and perform secondary optimization of the restraining force of the target train based on the degraded speed reference value.

6. An electronic device, characterized in that, Includes memory and processor, of which: The processor is used to execute computer programs stored in the memory; When the processor executes the computer program, it implements the steps in the speed control method for the vacuum tube high-speed train according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps in the speed control method for the vacuum tube high-speed train according to any one of claims 1 to 4.

Images