Multi-vehicle cooperative intelligent marshalling speed regulation method and system for railway transportation of power batteries

Abstract

The invention discloses a multi-vehicle cooperative intelligent marshalling speed regulation method for railway transportation of power batteries. The method comprises the following steps: establishing state characteristics of each transportation truck, and constructing a position sensitive damage cost function for evaluating the risk cost of each truck at different marshalling positions according to the state characteristics; in the railway vehicle transportation process, an event-driven time pane dynamic optimization method is adopted, a vehicle marshalling and train speed control curve is dynamically updated according to state parameters collected in real time, and in the optimization process, a variable damping distribution strategy of a coupler and a cooperative vehicle speed curve generation algorithm are adopted, so that the train speed control curve is dynamically updated. Meanwhile, time dispersion is carried out on coupling impact events through an impact time peak shifting strategy; and when the comprehensive risk index of any transport truck exceeds a preset threshold value, an emergency re-marshalling strategy based on the energy absorption remaining amount of the battery packaging unit and the coupler health degree is executed. The method can comprehensively consider the thermal characteristics of the power lithium battery, the dynamic response of the vehicle and the risk evolution.

Description

Technical Field

[0001] This application relates to the field of power lithium battery railway transportation technology, specifically to a method, device, and system for intelligent train formation and speed regulation of power battery railway transportation with multi-car coordination. Background Technology

[0002] As the core energy unit for new energy vehicles, energy storage power stations, and special equipment, the transportation demand for power lithium batteries has increased significantly in recent years. Railway transportation offers advantages such as large capacity, low cost, high safety, and wide route coverage, providing a better option for medium- and long-distance transportation of power lithium batteries. Currently, power lithium batteries typically have high energy density and volume, making them susceptible to physical damage during railway transportation, potentially leading to thermal runaway. In particular, collisions during shunting at level crossings, deceleration and stopping during hump shunting, and long downhill curves can generate lateral centripetal forces that cause cargo shifting.

[0003] In existing railway transportation trial operation schemes, power lithium batteries are usually loaded in containers. Train formation and shunting processes follow the general cargo operation rules. Due to the high energy density and heat-sensitive characteristics of power lithium batteries, transient impacts generated during train coupling, shunting, or braking may cause deformation of the internal structure of the battery module, misalignment of the electrode plates, or damage to the insulation layer. Under high temperature environments or continuous vibration, thermal runaway or even combustion and explosion accidents may also occur. Furthermore, railway transportation scheduling mostly adopts fixed speed curves and static formation methods, making it difficult to respond in real time to environmental changes and differences in vehicle status.

[0004] Therefore, there is an urgent need for a multi-vehicle collaborative intelligent train formation and speed control method that can comprehensively consider the thermal characteristics of power lithium batteries, vehicle dynamic response, and risk evolution. Summary of the Invention

[0005] Purpose of the invention: In order to overcome the above shortcomings, the purpose of this application is to provide a method, device and system for intelligent train formation and speed regulation of multi-vehicle collaborative power battery railway transportation, which can sense the operating status of each vehicle in real time and quantify the risks, and then realize the dynamic joint control of vehicle position and speed through optimization algorithm, and establish a multi-layer energy regulation mechanism of variable damping coupler and cooperative speed, and have the ability to quickly re-form train when the risk exceeds the limit.

[0006] To address the aforementioned technical problems, this application provides a multi-vehicle collaborative intelligent train formation speed control method for power battery railway transportation, comprising: S1: collecting real-time state parameters of each transport vehicle and establishing state characteristics for each transport vehicle, wherein the state characteristics include at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type, and packaging protection parameters; S2: constructing a position-sensitive damage cost function based on the state characteristics to evaluate the risk cost of each freight car at different formation positions, thereby quantifying the comprehensive risk cost of each freight car at a specific formation position; S3: during the rail vehicle transportation process, employing event-driven... The time-pane dynamic optimization method dynamically updates the vehicle formation and train speed control curves based on real-time collected state parameters. During the optimization process, a variable damping distribution strategy for the coupler and a cooperative speed curve generation algorithm are adopted. At the same time, the impact time staggering strategy is used to disperse the time of coupled impact events to reduce the superposition effect of local impact peaks. S4: When the comprehensive risk index of any transport vehicle exceeds the preset threshold, an emergency re-formation strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler is executed to adjust high-risk vehicles to low-risk positions or safety buffer zones, thereby maintaining the overall risk balance of transportation.

[0007] As a preferred embodiment of this application, in step S2, the method includes: S21: establishing a longitudinal impact transfer matrix using a railway vehicle dynamics model and calculating the failure probability of the power battery under different temperature gradients using a thermodynamic simulation model, thereby mapping the results of both to a standardized risk probability; S22: constructing a location-sensitive damage cost function. ,in To quantify the risk cost of the i-th freight car at different train formation positions pos, Let be the probability of battery thermal runaway at the formation position pos of the i-th truck. Let be the probability of mechanical impact damage to the i-th freight car at different train formation positions pos. Let be the expected economic loss of the i-th freight car at different train formation positions pos; These are weighting coefficients, and S23: Calculate the total cost of each group arrangement using the location-sensitive damage cost function, and select the grouping order using the minimum cost criterion; S24: If the total risk is less than the preset safety threshold, then determine the current arrangement as a safe solution; otherwise, proceed to the re-optimization stage.

[0008] As a preferred embodiment of this application, in step S3, the event-driven time pane dynamic optimization method includes: S301: Defining the scheduling time axis as a sliding pane: ,in The time window length, S302: If the preset monitoring parameter is found to exceed the threshold, an event is triggered; S303: Local optimal adjustment is performed on the set of transport vehicles in the time pane to which the triggered event belongs. , in Map the new location of the i-th truck. Let be the local speed adjustment value for the i-th truck. For speed smoothing weights, S304: Use a greedy search algorithm combined with a local exchange algorithm to optimally replace adjacent vehicles in the sliding pane; S305: Restore the global scheduling cycle after the risk event is resolved or the window times out.

[0009] As a preferred embodiment of this application, step S3 employs a variable damping distribution strategy for the couplers, including: S306: obtaining the current load, temperature, and wear level parameter set for each coupler of the rail vehicle; S307: calculating the upper limit of coupler energy absorption. ,in The upper limit of energy absorption for the j-th coupler. For material coefficients, For the wear level of the j-th coupler, S308: Solve the energy distribution optimization problem for the entire trainset, where the temperature correction function is for the j-th coupler. ,in For allocation coefficients, S309: The energy absorption value of the j-th coupler; S309: Adjusts the damping value of the magnetorheological damper controlling the coupler. ,in This is the initial damping value. Sensitivity factor; S310: When the energy absorption of the coupler exceeds the energy safety threshold, a deceleration command is triggered. ,in For energy safety threshold, This is the speed reduction coefficient.

[0010] As a preferred embodiment of this application, in step S3, the cooperative speed curve generation algorithm includes: S311: defining the speed curve for each train and setting the objective as minimizing the overall risk energy function: ,in Let be the heat accumulation amount corresponding to the i-th truck. The impact energy function is calculated based on the speed difference and dynamic model, and it is proportional to the square of the speed difference between adjacent vehicles. Weighting coefficients; S312: Establish velocity constraints: And introduce collaborative constraints between rail vehicles: ,in S313: Use model predictive control to iteratively optimize the speed profile and share the current speed of each transport vehicle to form a multi-vehicle coordinated speed queue and output the speed reference trajectory to realize the synchronous acceleration and deceleration of each transport vehicle.

[0011] As a preferred embodiment of this application, in step S3, reducing the superposition of local impact peaks through an impact time staggering strategy includes: S314: Establishing a time series for the expected n consecutive impact events: S315: Calculate the energy of each impact: And define the cumulative impact function: ,in S316: When the cumulative impact function exceeds the energy absorption limit of the battery pack cell, the insertion time delay staggers the impact events. , ,in This represents the upper limit of energy absorption for a battery packaging unit. S317: Adjust train speed trajectory to meet the new impact schedule, and when all impact intervals... and At that time, the final speed regulation scheme is determined and output.

[0012] As a preferred embodiment of this application, in step S4, the emergency re-grouping strategy for the remaining energy absorption capacity of the battery pack unit and the health of the coupler includes: S41: Calculating the remaining energy absorption capacity of the battery pack unit: ,in Let be the remaining energy absorbed by the battery pack unit of the i-th transport vehicle at time t. The maximum absorbable energy of the battery pack unit of the i-th transport vehicle (which is set by the designer based on the design and materials of the battery pack unit). The input impact energy rate applied to the battery packaging unit per unit time. This refers to the energy dissipated per unit time by the battery packaging unit during processes such as phase change, plastic deformation, and viscoelastic energy dissipation. S42: Establish the priority of transport vehicles for the battery pack unit to begin service or for the most recent reset / replacement time. ,in This represents the maximum energy absorption value for the battery packaging unit corresponding to the packaging type. For the health of the coupler, S43: Arrange transport vehicles in descending order of priority, and select the subset with the highest risk; S44: Use the minimum number of vehicle position adjustments as the objective function. Solve for the new parking space mapping; where, S45: Output and execute the transposition sequence to move high-risk vehicles to the end of the column or the buffer and lock the cooperative speed curve during regrouping to maintain energy balance.

[0013] As a preferred embodiment of this application, during the execution of the emergency regrouping in step S4, the method includes: S401: Constructing a vehicle arrangement state diagram. , where the vertex For vehicle position, edge To establish commutative paths, each edge is then assigned a weight. , S402: For the subset with the highest risk, establish a set of target transfer locations and define the objective function: (The original text contains some formatting errors and inconsistencies. A more accurate translation would require the full context.) ,in The subset with the highest risk Set of target locations; S403: Use A Heuristic search is used to find the optimal match, and after obtaining the match, a sequence of transposition instructions is generated. Each instruction includes adjusting speed, sequence, and target time; S404: Execute the repositioning action step by step and verify the impact energy change in real time. If the impact energy change exceeds the preset energy threshold, the execution is paused and the path is replanned.

[0014] This application also provides a multi-vehicle cooperative intelligent train formation speed control device for power battery railway transportation using the aforementioned method, comprising: a feature construction module for establishing state features of each transport vehicle in the rail vehicle, the state features including at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type, and packaging protection parameters; a risk assessment module for constructing a position-sensitive damage cost function based on the state features to assess the risk cost of each freight car at different formation positions, thereby quantifying the comprehensive risk cost of each freight car at a specific formation position; and a speed coordination module for employing event-driven time windows during rail vehicle transportation. The system employs a dynamic optimization method to dynamically update vehicle formation and train speed control curves based on real-time collected state parameters. During the optimization process, it utilizes a variable damping allocation strategy for couplers and a collaborative speed curve generation algorithm. Simultaneously, it employs an impact time staggering strategy to distribute the impact events of coupled trains over time, thereby reducing the superposition effect of local impact peaks. The re-formation control module is used to execute an emergency re-formation strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler when the comprehensive risk index of any transport vehicle exceeds a preset threshold. This strategy adjusts high-risk vehicles to low-risk positions or safety buffer zones, maintaining the overall risk balance of transportation.

[0015] The technical solution described in this application has the following advantages over the prior art: 1. This application introduces a real-time perception and collective intelligence collaboration mechanism at the train formation level, which can continuously monitor the operating parameters of each transport vehicle, such as temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type, and packaging protection parameters, and dynamically assess the overall transportation safety status. On this basis, optimization algorithms are used to realize dynamic joint decision-making for vehicle position adjustment and speed control, so that the formation structure and running speed can be adaptively adjusted according to the risk distribution.

[0016] 2. This application constructs a multi-layer energy regulation mechanism of variable damping coupler and coordinated vehicle speed control. Through coupler damping adaptation and multi-vehicle speed coordination, the peak impact of coupling is reduced and the energy transfer path is balanced.

[0017] 3. When the risk index of any vehicle exceeds the preset safety threshold, this application can automatically trigger an emergency regrouping strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler, so as to achieve rapid isolation and redistribution of high-risk vehicles without affecting the overall stability of train operation.

[0018] 4. The battery packaging unit designed in this application can prevent the power lithium battery from being damaged by collisions and drops during transportation, and can also prevent the power lithium battery from thermal runaway under high temperature conditions, thereby causing safety accidents such as fires and explosions.

[0019] 5. This application combines PLC control with the structure of the battery transport container to achieve dual prevention and control of thermal abuse and mechanical abuse during the railway transportation of power lithium batteries, and to monitor the environmental status inside the container and provide early warning and fire handling. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the intelligent train formation and speed control method for power battery railway transportation provided in this application embodiment.

[0022] Figure 2 This is a flowchart illustrating the method for quantifying the comprehensive risk cost of each truck provided in the embodiments of this application.

[0023] Figure 3 This is a flowchart illustrating the event-driven time pane dynamic optimization method provided in an embodiment of this application.

[0024] Figure 4 This is a flowchart illustrating the variable damping distribution strategy method for car couplers provided in the embodiments of this application.

[0025] Figure 5 This is a flowchart illustrating the collaborative vehicle speed curve generation method provided in an embodiment of this application.

[0026] Figure 6 This is a flowchart illustrating the method for reducing the superposition of local impact peaks through an impact time staggering strategy provided in this application embodiment.

[0027] Figure 7 This is a flowchart illustrating the emergency re-grouping strategy method for battery packaging unit energy absorption capacity and coupler health provided in the embodiments of this application.

[0028] Figure 8 This is a flowchart illustrating the emergency regrouping execution method provided in the embodiments of this application.

[0029] Figure 9 This is a schematic diagram of the module connection of the intelligent train formation and speed control device for power battery railway transportation provided in the embodiments of this application.

[0030] Figure 10 This is a schematic diagram of the 1P7S model power battery provided in the embodiments of this application.

[0031] Figure 11 This is a schematic diagram of the battery packaging unit provided in the embodiments of this application.

[0032] Figure 12 These are three-dimensional and three-view schematic diagrams of the battery packaging unit provided in the embodiments of this application.

[0033] Explanation of reference numerals in the instruction manual: 201. Feature Construction Module, 202. Risk Assessment Module, 203. Speed ​​Coordination Module, 204. Regrouping Control Module, 3. Power Battery, 41. First Pad, 42. Second Pad, 43. Hole and Slot, 411. Upper Base, 412. Lower Base. Detailed Implementation

[0034] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0035] refer to Figures 1 to 10 As shown, in some embodiments, the intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation includes: S1: Collect real-time status parameters of each transport vehicle on the rail vehicle and establish the status characteristics of each transport vehicle. The status characteristics include at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type, and packaging protection parameters.

[0036] Specifically, in step S1, to achieve comprehensive monitoring of the transportation status of power lithium batteries, each transport vehicle is equipped with a status monitoring unit. This monitoring unit includes at least a temperature sensor, strain gauge, acceleration sensor, BeiDou positioning module, wireless communication module, and tension sensor. The temperature sensor is used to detect the ambient temperature inside the transport vehicle, the surface temperature of the battery packaging, and the hook temperature in real time. The strain gauge is used to measure longitudinal stress changes, thereby reflecting the stress on the vehicle body. The acceleration sensor is deployed on the vehicle's bottom beam and the battery packaging unit to detect longitudinal impact and vibration acceleration. The BeiDou positioning module is used to obtain the vehicle's position index, which is used to identify the vehicle's position in the train formation sequence. The wireless communication module uploads real-time data to the rail train control system via the train's intranet (such as LoRa, ZigBee, or 5G-R communication). The tension sensor collects hook force parameters for subsequent damping distribution strategy calculations.

[0037] The load type information for each transport vehicle includes the type of power battery, the quantity loaded, the mass of the modules, and the stacking method; the packaging protection parameters are used to characterize the protective performance of the battery outer packaging, including the packaging stiffness coefficient, energy absorption capacity, phase change material heat capacity, and packaging integrity index. The packaging protection parameters can be entered before loading or read by the RFID chip embedded in the packaging.

[0038] To avoid state deviations caused by sensor errors or transient interference, this application employs an existing sliding time window filtering algorithm to correct the collected data. If the rate of change of a certain parameter exceeds an empirical threshold (e.g., temperature change rate > 5℃ / min), a verification request is triggered, and corresponding data from adjacent transport vehicles is used for interpolation and repair. This application includes a local clock module in each transport vehicle, using the IEEE 1588 Precision Time Protocol (PTP) for vehicle-wide synchronization. This ensures that data uploaded from different vehicles is recorded under the same time base, and automatically synchronizes the time when the time drift exceeds 5ms to guarantee the consistency of timing in subsequent risk calculations.

[0039] Therefore, all the state features constructed in this application are stored in the ring cache of the on-board control system of the rail vehicle in time sequence, and periodically uploaded to the rail vehicle central control platform via wireless link.

[0040] S2: Construct a location-sensitive damage cost function based on the state characteristics to evaluate the risk cost of each freight car at different formation positions, so as to quantify the comprehensive risk cost of each freight car at a specific formation position.

[0041] Specifically, in step S2, the probability of thermal runaway and the probability of mechanical impact damage are calculated based on the state characteristics of each transport vehicle. The expected economic loss is then combined to form a position-sensitive damage cost function through weighting. Finally, the total cost is obtained by summing all possible grouping arrangements and minimizing it to select the grouping order.

[0042] Therefore, for reference Figure 2 As shown, the specific methods include: S21: The longitudinal impact transmission matrix is ​​established using the railway vehicle dynamics model and combined with the thermodynamic simulation model to calculate the failure probability of the power battery under different temperature gradients. The results of the two are then mapped to the standardized risk probability.

[0043] Among them, a longitudinal multibody dynamics model is established using multibody dynamics simulation software (such as SIMPACK / Adams): ,in Let the mass of the i-th transport vehicle be... For vehicle body damping, For vehicle body rigidity, External impact input; Several impact inputs (e.g., pulse velocity difference) are applied to the coupler at the grouping position pos, the response (time-domain acceleration or energy absorption) is recorded, and the received energy of the k-th car is calculated. Define the transfer coefficient: Furthermore, longitudinal impact transmission matrices are established for different train positions (pos) and different coupling modes (multiple trains impacting multiple trains, one train impacting multiple trains, multiple trains impacting one train) and stored in a lookup table / function form.

[0044] The calculation process for the probability of mechanical impact damage is as follows: Let the impact energy received by the i-th transport vehicle in a certain event be: , The input kinetic energy at impact location j (or the measured impact energy of the coupler); and the breakdown / damage threshold energy of the i-th vehicle is defined as follows: (Determined by the battery packaging unit), energy is mapped to probability using a logarithmic / logic mapping: ,in This represents the probability of mechanical impact damage. The sensitivity coefficient, This represents the 50% probability point of occurrence corresponding to the energy (which can be taken as 0.8). Furthermore, if cumulative damage needs to be considered, cumulative energy can be used. Replace the instantaneous energy and add a decay factor to the formula.

[0045] The thermal failure probability calculation uses a short-term probability model: ,in The measured maximum temperature (including instantaneous temperature rise due to impact). This refers to the critical temperature (e.g., 80°C is the threshold for thermal runaway). Temperature sensitivity.

[0046] because and Essentially, these are probability values ​​(0–1), but their origins differ from uncertainty. Therefore, it is necessary to introduce sensor reliability / model confidence to correct the probability and normalize it using soft maximum or weighted average.

[0047] S22: Constructing a location-sensitive damage cost function: ,in To quantify the risk cost of the i-th freight car at different train formation positions pos, Let be the probability of battery thermal runaway at the formation position pos of the i-th truck. Let be the probability of mechanical impact damage to the i-th freight car at different train formation positions pos. Let be the expected economic loss of the i-th freight car at different train formation positions pos; These are weighting coefficients, and .

[0048] in, Defined as: ,in For repair costs (such as replacement / outage / compensation, etc.). The total losses resulting from thermal runaway (such as fire extinguishing, train shutdown, and related losses).

[0049] in, for The weighting coefficients, for The weighting coefficients, for The weighting coefficients are calculated by taking the standardized variance of each indicator from historical transportation records to determine its contribution to the overall risk; for example, using a 2000-time shunting scenario, the weighting coefficients are obtained. The relative contributions are 0.44:0.29:0.27, then the corresponding settings are... =0.44, =0.29, =0.27.

[0050] S23: Calculate the total cost of each group arrangement using the location-sensitive damage cost function: Therefore, the minimum cost criterion is used to solve the problem: ,in For the number of transport vehicles. It is the set of all possible groupings and permutations.

[0051] S24: If the total risk is less than the preset safety threshold, the current arrangement is determined to be a safe solution; otherwise, proceed to the re-optimization stage.

[0052] The re-optimization phase may involve adjusting weights (e.g., temporarily increasing the thermal risk weight to prioritize fire prevention), changing train scheduling windows or speed constraints within permissible limits, reducing impact by increasing time staggering (as in step S314 below), or triggering emergency regrouping (as in step S4 below) to move high-risk vehicles to the end of the train or a buffer zone.

[0053] The preset safety threshold is based on historical accidents and operational experience, plus a safety margin setting, and the safety margin ranges from [0.05, 0.2].

[0054] S3: During the rail vehicle transportation process, an event-driven time pane dynamic optimization method is adopted, and the vehicle formation and train speed control curves are dynamically updated based on the real-time collected state parameters. In addition, during the optimization process, a variable damping allocation strategy for couplers and a cooperative speed curve generation algorithm are adopted. At the same time, the impact time staggering strategy is used to disperse the time of coupled impact events to reduce the superposition effect of local impact peaks.

[0055] During the operation of rail vehicles, the state characteristics of each transport vehicle are constructed in real time. Then, an event-driven control architecture combined with a sliding time window is adopted. When any key parameter exceeds the threshold or an external trigger (such as a shunting command) is received, local joint optimization (car position adjustment and local speed correction) is performed in the corresponding time window. The optimization also considers the coupler damping distribution, the coordinated speed curve and the impact time staggering strategy, and then outputs an executable speed reference trajectory and grouping / transposition instructions.

[0056] Specifically, refer to Figure 3 As shown, the event-driven time pane dynamic optimization method includes: S301: Define the scheduling timeline as a sliding pane: ,in The length of the time window used to limit the time range of local optimization is within the range of [value missing]. s, where the short window can be used for high-frequency event response, and the long window can be used for complex regrouping. The actual value is set by the designer according to the grouping size, communication delay and the response speed of the actuator; For scheduling time.

[0057] In actual implementation, only locally feasible parking space swaps and speed fine-tuning are allowed within the defined sliding pane to avoid affecting long-term remote operation plans, and all sliding pane operations must be performed within the specified time frame. Complete within (e.g., ≤5s) to ensure real-time performance.

[0058] S302: If it is determined that the preset monitoring parameter exceeds the threshold, an event is triggered.

[0059] Specifically, the conditions for triggering the event may include, but are not limited to, temperature greater than or equal to a temperature threshold, temperature rise rate greater than or equal to a rate threshold, acceleration greater than or equal to a velocity threshold, longitudinal stress greater than or equal to a stress threshold, remaining energy absorption of the battery packaging unit less than or equal to an energy absorption threshold, and coupler health less than or equal to a health threshold. For example, a temperature threshold of 60°C, a rate threshold of 2°C / min, and a velocity threshold of 5 m / s are set. The stress threshold and energy absorption threshold are set by the designer based on the actual battery packaging unit, and the health threshold is set by the designer based on the actual rail vehicle coupler.

[0060] Therefore, when any of the above triggering conditions are met, a triggering event is generated, and the relevant transport vehicles are grouped together with the time in the sliding pane. As an optimized input.

[0061] S303: Perform locally optimal adjustments on the set of transport vehicles within the time pane to which the triggering event belongs. ,in Map the new location of the i-th truck. Let be the local speed adjustment value for the i-th truck. For speed smoothing weights ( ), This is the reference speed for the time pane.

[0062] When performing local optimal adjustments, constraints are set, including parking space feasibility constraints (e.g., specific vehicles cannot be moved to dangerous locations) and upper and lower limits of physical speed. ), the number of transposition operands must not exceed the set number to ensure that the execution can be completed, and the coupler / buffer capacity is constrained (i.e., steps S307 to S309 below).

[0063] In practice, the local optimal adjustment adopts a heuristic and hierarchical strategy, that is, first fixes... And on Solve the continuum subproblem, and then... A greedy search algorithm combined with a local exchange algorithm is used to find the near-optimal permutation, and then the optimal permutation is fine-tuned and its feasibility is checked.

[0064] S304: A greedy search algorithm combined with a local exchange algorithm is used to optimally replace adjacent vehicles in the sliding pane.

[0065] The actual steps of the greedy search combined with the local exchange algorithm are as follows: Read the actual train formation sequence at the current moment and record it as the initial arrangement. And set a reference speed based on the previous scheduling cycle or the default running plan. At the same time, determine the set S of transport vehicles that need to be optimized (usually the transport vehicles within the influence range of the triggering event). Calculate the total cost under the current grouping based on the location-sensitive damage cost function defined in step S2. Within the set of transport vehicles S, determine the set of interchangeable neighborhoods (i.e., pairs of vehicles that are physically adjacent or logically interchangeable; the neighborhood is usually limited to a range of no more than three transport vehicles to ensure that the swapping action can be completed within the time pane). For each pair of neighboring transport vehicles, temporarily swap their positions to obtain a new arrangement. While keeping the positions of other transport vehicles unchanged, recalculate the cost changes for the affected transport vehicles and solve for the optimal adjustment value of the continuous speed variable in this local area. , making the cost item Minimum; If the total cost of the new plan Less than the total cost of the current solution If this exchange improves the overall risk level and speed balance, the current plan should be updated immediately. And update the corresponding speed adjustment amount simultaneously. ; Repeat the above process of neighborhood swapping and cost comparison until any of the following conditions are met: the swapping has failed to further reduce the total cost for several consecutive rounds (e.g., 3–5 rounds), the maximum number of iterations has been reached, or the computation time has reached the set upper limit. When any of these conditions are met, the current arrangement is considered to have reached a local optimum. The final grouping scheme and the corresponding local speed adjustment results are output. If the scheme can be implemented within the set time window and meets the energy and safety constraints, it is issued as the optimal execution scheme for the current time period.

[0066] S305: Restore the global scheduling cycle after the risk event is resolved or the window times out.

[0067] Specifically, when the event in the sliding pane is cleared (the sensor returns to normal) or reaches... At the upper limit, the adjustments implemented in the sliding pane are processed as follows: if the adjustment is successful and meets the energy balance and safety constraints, it is written into the global plan and the global cycle continues; if a new anomaly occurs during the execution process, a new event is immediately triggered and the process enters the next sliding pane for optimization or triggers the emergency process of step S4.

[0068] Specifically, refer to Figure 4 As shown, the variable damping distribution strategy using the coupler includes: S306: Obtain the current load, temperature, and wear level parameter set for each coupler of the rail vehicle.

[0069] S307: Calculate the upper limit of energy absorption by the coupler: ,in The upper limit of energy absorption for the j-th coupler. For material coefficients, For the wear level of the j-th coupler, This is the temperature correction function for the j-th coupler.

[0070] in, , To set the operating temperature, This is the temperature decay coefficient.

[0071] S308: The optimization objective is to reduce the actual impact energy. With allocation ability Matching, minimizing the difference of squares: ,in For allocation coefficients, Let be the energy absorption value of the j-th coupler; where the constraints are set as follows: , , This is a safety margin factor (e.g., 1.05, to ensure that the absorption capacity exceeds the demand).

[0072] If the problem has a small dimension, the existing constrained least squares (QP) solver is used for real-time solution; if faster speed is required, analytical normalized assignment is used. And normalize to [0,1].

[0073] S309: Adjusting the damping value of the magnetorheological damper controlling the coupler, that is, mapping the allocation result to a damping control signal (this application uses a magnetorheological damper): ,in This is the initial damping value. For sensitivity factor, The target damping is issued.

[0074] The damping value is issued by updating the magnetorheological valve current via the vehicle controller through the local bus (CAN / LVDS) or a dedicated drive interface. The valve current is nearly linearly mapped to the damping.

[0075] S310: If the energy absorption of any coupler exceeds the energy safety threshold, a speed correction command is triggered. ,in For energy safety threshold, This is the speed reduction coefficient.

[0076] Among them, when hour When the speed is exceeded, the speed will be reduced proportionally.

[0077] Specifically, refer to Figure 5 As shown, the cooperative vehicle speed curve generation algorithm includes: S311: Define the speed curve for each train and set the objective as minimizing the overall risk energy function: ,in Let be the heat accumulation amount corresponding to the i-th truck. The impact energy function is calculated based on the speed difference and dynamic model, and it is proportional to the square of the speed difference between adjacent vehicles. These are the weighting coefficients.

[0078] in, , , The energy transfer coefficient of the coupler. Thermosensitive coefficient.

[0079] S312: Establish velocity constraints: And introduce collaborative constraints between rail vehicles: This is to prevent large impacts caused by the high speed difference between adjacent trains or vehicles.

[0080] in, For speed coordination tolerance between adjacent transport vehicles; acceleration constraint is , For passenger comfort restrictions.

[0081] S313: The speed profile is iteratively optimized using model predictive control methods and all transport vehicles share the current speed to form a multi-vehicle coordinated speed queue and output the speed reference trajectory, thereby realizing the synchronous acceleration and deceleration of all transport vehicles.

[0082] Specifically, the process of step S313 is as follows: Based on the longitudinal dynamics model of railway vehicles, a discrete-time prediction model for each transport vehicle is established. For the i-th transport vehicle, its prediction model is described as follows: , ,in Sampling time (e.g., 0.5-1s). For vehicle acceleration or deceleration commands, For the location of railway vehicles, For railway vehicle speed; The above prediction model is used to predict the future position and speed changes of each transport vehicle within a limited prediction time domain (e.g., 10–30 seconds). In each prediction period, a function is established to minimize the overall risk energy, thereby simultaneously minimizing the impact energy and the thermal accumulation energy, and constraining the smoothness of acceleration changes. In terms of setting constraints, multiple constraints are imposed on the speed, acceleration, and relative speed difference of each transport vehicle to ensure safe and stable operation. The objective function and constraints are then combined to form an optimization problem, which is then solved iteratively using the Model Predictive Control (MPC) algorithm. This involves predicting the velocity and acceleration trajectory within a set time step based on the current state at the current moment, and finding the optimal control sequence that minimizes the objective function. To ensure real-time computation, a quadratic programming (QP) solution method is used, and the previous optimization result is used for a warm start to reduce the number of iterations and improve the solution speed. Specifically, the state variables of MPC include the position, speed, acceleration, and cumulative impact energy of the transport vehicle, and the control variables include the desired acceleration control value of the transport vehicle and the coupler damping. The formula for calculating the MPC objective function is as follows: in To predict the step size, For speed tracking error, The center trajectory for velocity planning, For the speed consistency item among transport vehicles; The cumulative impact energy of the i-th truck The weight of the control is determined by the degree of importance attached to the suppression of impact energy, and is taken as 0.3–0.5; Let i be the probability of thermal runaway of the i-th truck. The weight of the indicator is set to 0.2-0.4, which emphasizes the urgency of thermal safety indicators in the overall optimization. The weight for speed tracking error, balancing risk control and operational efficiency, is set between 0.05 and 0.25. When the risk is high, the weight is adjusted and increased. and reduce ; The weight for the speed consistency term among transport vehicles, which forces the entire train to maintain a consistent speed trajectory, is 0.3–0.6. in, ,specific The specific values ​​of each weight can be determined by organizing an expert group (scheduling, safety, materials, vehicle dynamics) to compare the importance of each of the four items pairwise, forming a judgment matrix, and then calculating the largest eigenvector and normalizing it to obtain the weight value. Alternatively, existing weighting methods, such as economic methods, can be used. Before weighting, each item needs to be mapped to a unified dimension, that is, the items are normalized.

[0083] After obtaining the optimal control sequence, only the first control action in the prediction result is executed, that is, the first acceleration or speed correction value, and this control quantity is sent to each transport vehicle. Then, the time sliding pane is advanced according to the rolling time domain principle (RecedingHorizon) to enter the next sampling period, re-collect the state, update the prediction and repeat the solution. Through this iterative optimization method, the running trajectory can be continuously corrected to adapt to real-time environmental changes and sudden risks.

[0084] Meanwhile, at the end of each control cycle, the current speed of all transport vehicles is uploaded and transmitted to each node of the rail vehicle, thus forming a coordinated speed queue. This speed queue records the current target speed sequence of all transport vehicles on the rail vehicle and serves as the reference input for the next cycle. Each transport vehicle performs synchronous acceleration and deceleration control according to the shared speed queue to ensure smooth longitudinal movement of the entire train. If any transport vehicle experiences communication delay or data anomaly, it can temporarily use the reference speed value of the previous cycle to maintain synchronization, and then return to the global queue after communication is restored.

[0085] During the optimization process, if the impact energy, temperature rise rate, or coupler stress exceeds a preset threshold, the weighting coefficients are dynamically adjusted. and speed coordination tolerance This is to strengthen the priority of safety control and appropriately reduce operational efficiency, thereby achieving a balance between safety and efficiency.

[0086] Specifically, refer to Figure 6 As shown, reducing the superposition of local impact peaks through impact time staggering strategy includes: S314: Establish a time series for the expected n consecutive shock events: ; S315: Calculate the energy of each impact: And define the cumulative impact function: ,in is the energy decay constant, representing the timescale of recovery in time (e.g., 30–300 s). The smaller the value, the faster the energy decays or recovers.

[0087] S316: When the cumulative impact function exceeds the energy absorption limit of the battery pack cell, the time delay is inserted to adjust the subsequent impact timing. , To stagger the impact events; among which This represents the upper limit of energy absorption for a battery packaging unit. The scale factor is used to control the peak shifting amplitude.

[0088] S317: Target Timeline Convert to speed adjustment, that is: If the impact event occurs during a physical coupling operation, its time is related to the relative position and speed of the trains. Assuming the coupling is completed within a distance *s* of relative displacement between the two trains, then the solution is as follows: ,in Then, an approximately linear speed change strategy is used to calculate the required speed difference: and adjust under constraints , To achieve time offset; if it cannot be achieved within the allowable constraints, it degenerates into increasing. Or choose other grouping actions; when all impact intervals meet and If the conditions cannot be met, the final speed regulation scheme is determined and output; if the conditions cannot be met, the emergency regrouping in step S4 is triggered.

[0089] Therefore, to quantify the impact of the shock time staggering strategy on the probability of thermal runaway, this application uses a cumulative shock function to describe the energy load of the local package on a short timescale, and then applies this energy according to... Converting the temperature to a local maximum temperature and mapping the temperature to the probability of thermal runaway using a short-term probability model, this is used for the third thermal runaway under the baseline condition of n=3. =2811J, if we take the example parameter: energy conversion coefficient Local equivalent mass involved in heating (This is typically the effective mass near the hot spot, much smaller than the total mass of the package), the specific heat capacity of this local equivalent mass. initial temperature Critical temperature Temperature sensitivity Then the maximum temperature The thermal runaway probability was calculated by decreasing the temperature from approximately 53.1°C to approximately 42.8°C. from dropped to about ,Right now This represents a relative reduction of approximately 92%. Therefore, by adopting the aforementioned multi-vehicle collaborative dynamic optimization scheme for rail vehicles throughout the entire transportation cycle, the train speed curve and formation control strategy can be continuously updated based on real-time status data. This reduces local impact peaks, minimizes the risk of thermal runaway of the power battery, and maintains overall train operational stability under complex railway conditions.

[0090] S4: When the comprehensive risk index of any transport vehicle exceeds the preset threshold, an emergency regrouping strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler is implemented to adjust high-risk vehicles to low-risk positions or safety buffer zones, thereby maintaining the overall risk balance of transportation.

[0091] Specifically, when the comprehensive risk index of any transport vehicle exceeds a preset threshold (or a higher alarm / critical threshold), an emergency regrouping strategy is triggered. The emergency regrouping strategy calculates the remaining energy absorption of the battery pack unit and the health of the coupler of each transport vehicle in real time and selects a high-risk subset based on priority ranking. Then, it generates a repositioning mapping and execution sequence with the goal of minimizing the number of vehicle position adjustments. Finally, the high-risk vehicles are moved to the end of the line or a safety buffer zone. During the repositioning process, the cooperative speed curve is locked to maintain energy balance and longitudinal stability.

[0092] Among them, reference Figure 7 As shown, the emergency re-grouping strategies for the remaining energy absorption capacity of the battery pack unit and the health of the coupler include: S41: Calculate the remaining energy absorption capacity of the battery pack unit: ,in Let be the remaining energy absorbed by the battery pack unit of the i-th transport vehicle at time t. The maximum absorbable energy of the battery pack unit of the i-th transport vehicle (which is set by the designer based on the design and materials of the battery pack unit). The input impact energy rate applied to the battery packaging unit per unit time. This refers to the energy dissipated per unit time by the battery packaging unit during processes such as phase change, plastic deformation, and viscoelastic energy dissipation. This refers to the time when the battery pack unit began service or the most recent reset / replacement.

[0093] in, ,in To achieve equivalent participation in quality, This represents a sudden change in relative velocity. ,in The elastomer energy dissipation coefficient of the battery packaging unit. For strain amplitude, and These are the equivalent heat capacity and temperature rise of the phase change material, respectively.

[0094] when Less than the set packaging safety threshold (which can be set by the maximum absorbable energy of the battery packaging unit, such as 10%-20%). When the battery pack is nearing depletion, its priority is increased.

[0095] S42: Establish priority for transport vehicles: ,in This represents the maximum energy absorption value for the battery packaging unit corresponding to the packaging type. For the health of the coupler, The coupler's health score is perfect.

[0096] The health status of the coupler can be constructed by combining static inspection (such as wear level), real-time peak load statistics, and temperature rise: ,in For normalization function, Wear level This represents the real-time peak load. This is temperature rise data.

[0097] In practice, priority Calculate immediately upon triggering an event and use it for subsequent sorting and subset selection; if time sensitivity needs to be considered, add time decay or weighting factors to reflect the rate of risk escalation.

[0098] S43: Arrange the transport vehicles in descending order of the aforementioned priority, and select the subset with the highest risk.

[0099] Specifically, calculating all vehicles and according to Sort the data in descending order to obtain a sequence and select the highest-risk subset using either the Top-k strategy or the threshold strategy. The Top-k strategy selects the top k transport vehicles, while the threshold strategy selects all transport vehicles that meet the set conditions. If the subset is empty (no vehicles meet the rules), no further grouping is required; otherwise, proceed to the transposition mapping solution.

[0100] In actual implementation, when selecting a subset, the time cost of transposition and train operation plan should be considered to avoid selecting a scheme that causes excessive interference due to transposition.

[0101] S44: Objective function is to minimize the number of parking space adjustments. Solve for the new parking space mapping; where, This is the original location index of the i-th transport vehicle.

[0102] Specifically, the goal of this step is to minimize the total displacement (total transposition distance) of high-risk vehicles, that is, to minimize the number of transpositions and the difficulty of transposition. The transposition cost can also be calculated using weighted distance, taking into account the actual transposition cost (time, path, obstacles).

[0103] S45: Outputs and executes a transposition sequence to move high-risk vehicles to the end of the column or a buffer zone and locks the cooperative speed curve during regrouping to maintain energy balance.

[0104] Specifically, the process of outputting and executing the transposition sequence is as follows: The mapping π is transformed into atomic transposition actions, and the transposition order is optimized to minimize intermediate moves and total execution time, thus avoiding concurrent conflicts. Before issuing each transposition command, energy verification, longitudinal stability verification, and time feasibility verification are performed.

[0105] During the transposition execution, the coordinated speed curve is locked, prohibiting large-scale global adjustments to the train speed; only minor corrections are permitted to ensure transposition feasibility and limit impact energy. During the locking period, MPC operates with tighter safety weights (e.g., ...). Increase, decrease ); For each atomic action, synchronous control is issued, including deceleration / acceleration commands, turnout switching commands, braking and holding commands, and target damping adjustment of the magnetorheological damper, to ensure that each transition is completed under permissible dynamic constraints.

[0106] Short-term observation and online verification are performed between the repositioning actions. If an uncontrollable impact or a sudden decrease in energy absorption of the battery pack unit is detected, the subsequent actions are immediately stopped and a safe stop or rollback plan is initiated. The rollback plan restores the repositioned vehicle to its original position or transitions it to a safe stop position.

[0107] In some embodiments of this application, reference is made to Figure 8 As shown, the method includes the following during the execution of emergency regrouping: S401: Construct a vehicle arrangement diagram , where the vertex For vehicle position, edge To establish commutative paths, each edge is then assigned a weight. , Let x be the cost of adjusting from position x to position y.

[0108] Specifically, to characterize the transposition operations that can be performed within a train formation and their costs, a weighted directed graph is constructed. , Corresponding to the physical location slots of transport vehicles, each vertex represents a specific location number (e.g., the sequential number from the front to the rear of the vehicle). This represents a commutative path from position x to position y.

[0109] Each vertex is labeled with: location index, track number (if multiple tracks), available turnout information, currently occupied vehicle ID (if any), whether it is a special position such as buffer / tail / head of a train, and whether there are time window constraints (such as through signals or passing times).

[0110] If a rail vehicle at position x can be moved to position y by shunting under the current station / track constraints, then an edge is established in the graph. If a transfer from x to y requires a transfer to another intermediate position, it can be represented as a multi-segment path (edge ​​sequence). For ease of calculation, common multi-segment composite operations can be pre-converted into a composite edge and assigned a corresponding total cost. For concurrent swapping (parallel movement), a set of mutually exclusive edges can be marked on the graph to represent edges that cannot be executed simultaneously.

[0111] Among them, each edge Assign a non-negative weight , to represent the operation cost of adjusting from position x to position y.

[0112] S402: For the subset with the highest risk, establish a set of transfer target locations and define the objective function: ,in The high-risk transport vehicles identified in step S43 above that require priority scheduling or isolation. For the set of target locations.

[0113] S403: Using A Heuristic search is used to find the optimal match, and after obtaining the match, a sequence of transposition instructions is generated. Each instruction includes a starting position, a target position, an estimated start time and end time, a speed adjustment instruction, and a damping adjustment instruction.

[0114] Among them, A A-Star is a heuristic search algorithm used to find the path with the minimum cost in the state space. It guides the search by combining a heuristic function (the estimated cost to the goal) with known costs (the cost from the starting point to the current state), significantly reducing the number of search nodes while ensuring the optimal solution is found. In the regrouping problem, A-Star... It can be used to search for transposition sequences in a transposition state space (containing multiple vehicle position states), thereby handling path conflicts and multi-step transposition problems.

[0115] Specifically, to ensure the lowest possible cost for emergency regrouping and repositioning during multi-vehicle collaborative transportation, this application adopts A A heuristic search algorithm finds the optimal matching path between the vehicle's current location set S and the target location set Q. The cost function for the search is defined as: ,in This represents the cumulative actual transposition cost from the initial state to the current node. For a heuristic estimate of the cost from the current node to the target state, ,in Let i be the position of the risk transport vehicle in the current arrangement. For high-risk transport vehicles that require priority dispatch or isolation, This represents the average displacement length of a single allowed exchange action.

[0116] After calculating the optimal sequence, a transposition instruction sequence is generated. And it will be carried out step by step.

[0117] Specifically, the heuristic function uses the sum of the shortest path costs from the current position to the nearest available target position for each high-risk vehicle that has not yet been placed on a target as the lower bound.

[0118] S404: Execute the repositioning action step by step and verify the impact energy change in real time. If the impact energy change exceeds the preset energy threshold, pause the execution and replan the path.

[0119] The transposition sequence is divided into several atomic action batches. A pre-check is performed before each batch is executed and an effect verification is performed after execution. The batch division is based on mutual exclusion (actions that can be executed in parallel are placed in the same batch) and time sequence requirements. Before executing a batch, speed trajectory locking, related coupler damping presets, turnout action plans, and execution time sliding panes are issued.

[0120] For each action to be executed, the peak impact energy that may occur during the execution process is estimated before the action actually begins; during the execution of the action, the actual impact energy is sampled and calculated in real time; if the actual impact energy exceeds the acceptable deviation of the peak impact energy, a fault response (deceleration / braking, stopping subsequent actions) is immediately executed and immediate safety isolation is carried out (e.g., parking the at-risk vehicle nearby); or the manual takeover process can be initiated directly.

[0121] In some embodiments of this application, three security thresholds are set and corresponding to different response strategies: The first threshold is for triggering only a warning and a slight speed correction; The second threshold is used to trigger local grouping adjustments and activate the high-damping mode. The third threshold is to initiate emergency regrouping and lock the maximum buffer mode.

[0122] Therefore, the thresholds for each layer are determined using a statistical learning method; when in a high-risk zone, all transport vehicles containing power batteries are automatically moved to the rear buffer zone.

[0123] Therefore, for reference Figure 9 As shown, in some embodiments of this application, this application also provides a multi-vehicle cooperative intelligent trainset speed control device for power battery railway transportation using the method described in any of the above embodiments, comprising: The feature construction module 201 is used to establish the state features of each transport vehicle in the rail vehicle, and the state features include at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type and packaging protection parameters; The risk assessment module 202 is used to construct a location-sensitive damage cost function based on the state characteristics to assess the risk cost of each truck at different formation positions, so as to quantify the comprehensive risk cost of each truck at a specific formation position. The speed coordination module 203 is used to dynamically update the vehicle formation and train speed control curves by adopting an event-driven time pane dynamic optimization method during the transportation of rail vehicles and based on the real-time collected state parameters. In the optimization process, a variable damping distribution strategy for couplers and a coordinated speed curve generation algorithm are adopted. At the same time, the impact time staggering strategy is used to disperse the time of coupled impact events to reduce the superposition effect of local impact peaks. The regrouping control module 204 is used to execute an emergency regrouping strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler when the comprehensive risk index of any transport vehicle exceeds a preset threshold. This strategy is used to adjust high-risk vehicles to low-risk positions or safety buffer zones, thereby maintaining the overall risk balance of transportation.

[0124] For example, experimental data are given below to help understand the advantages of this application; assuming the energy input for each impact is E=1000J, and considering 3 consecutive impacts (i.e., n=3), the energy decay constant... Take 30 seconds (that is, the packaging partially recovers within tens of seconds).

[0125] The calculation results are as follows: In the above data, under the example parameter conditions, increasing the interval between adjacent impacts from 2s to 20s can reduce the cumulative impact peak at the third event from approximately 2,811J to approximately 1,777J, a decrease of approximately 36.7%. If the interval is extended to 30s, the decrease can reach approximately 46%. By changing Δt and τ, different degrees of peak value reduction can be obtained within the range of 10%–60%, thereby significantly reducing the probability of mechanical damage and the risk of thermal runaway caused by impact superposition.

[0126] Furthermore, power lithium batteries typically have high energy density and volume, making them susceptible to physical damage during rail transport, potentially leading to thermal runaway. This is particularly true during shunting operations, such as collisions during flatbed shunting, deceleration and stopping during hump shunting, and long downhill curves where lateral centripetal forces can cause cargo shifting. Additionally, the transport of power lithium batteries is subject to international and domestic regulations and safety standards. Therefore, proper transport packaging can provide insulation and protection for power lithium batteries, reducing the impact of temperature fluctuations, preventing hotspot propagation and thermal runaway, and minimizing the effects of impact, vibration, and temperature, thereby reducing the risk of battery damage and thermal runaway. Considering the potential hazards during rail transport, such as electrolyte leakage, damage to conductive separators, and battery casing rupture, power lithium batteries require special design and packaging to prevent contact with the external environment.

[0127] refer to Figure 10 As shown, Figure 10 The diagram illustrates the structure of the 1P7S power lithium battery. This application designs the packaging based on the specifications of this lithium battery. The dimensions of the power lithium battery are: length × width × height: 1419.2 × 1019 × 211.7 mm.

[0128] In the design of transport packaging, attention should be paid to the design of the external protective layer. Battery packs usually require an external protective layer to provide physical protection and heat insulation. These protective layers are usually made of insulating or fire-resistant materials to reduce the impact of the external environment on the battery pack. If the protective layer is damaged or broken, the battery pack may be exposed to the external environment, which may be easily damaged or cause a fire.

[0129] Phase change materials (PCMs), also known as latent heat storage materials, absorb heat during melting and release it again during crystallization / freezing. These materials can maintain the required storage temperature during transportation. During the phase change from solid to liquid, the heat or cold entering the transport container is absorbed by the PCM, and vice versa. Organic PCMs, in particular, possess advantages such as high heat storage density, low cost, resistance to supercooling and phase separation, suitable phase change temperature, and chemical stability. They are widely used in solar energy, building energy conservation, electronic equipment, and the textile industry. When PCMs are adsorbed into the cushioning liner of transport packaging, they can effectively achieve cushioning and insulation. This application uses PCMs with higher thermal conductivity as the cushioning liner material to achieve cushioning and insulation effects during transportation.

[0130] Therefore, for reference Figure 11As shown in some embodiments of this application, this application provides a battery packaging unit for a power lithium battery, which includes a first liner and a second liner. The first liner wraps at least two power lithium batteries and separates the power lithium batteries, and the first liner is provided with a groove to buffer the force on the corners. The second liner is placed between the power lithium batteries and connected to the first liner, and the second liner is provided with a plurality of holes and slots to increase the contact surface area with air convection.

[0131] Regarding the cushioning material, the outer periphery of the first pad can effectively fit the power lithium battery and the outer corrugated cardboard box, allowing two 1P7S model power lithium batteries to be stacked tightly together, improving transportation efficiency and reducing the transportation risk of mutual reaction between the positive and negative electrodes of the battery; optionally, the corner portion of the first pad adopts a rounded arc structure with a radius of 40mm to cushion the force at the corner and fit the size specifications of the material.

[0132] In terms of phase change materials, the second gasket between the batteries is a flame-retardant composite phase change material. Pure paraffin (OP44) is adsorbed by expanded graphite (EG), with the mass ratio of expanded graphite to paraffin controlled at 1:9. The resulting OP44 / EG composite material is then mixed with ethylene propylene diene monomer (EPDM, model 3745P), decabromodiphenyl ethane (DBDPE), and antimony trioxide (ATO) to prepare the OP44 / EG composite material. This application fuses the prepared composite material with a porous extruded polystyrene thermal insulation board to prepare a flame-retardant composite phase change material to prevent damage such as overheating caused by friction during transportation of power lithium batteries.

[0133] The second liner has small holes and grooves cut into it (radius 10mm, height 200mm (400mm for the upper and lower bases, groove spacing 40mm, 27 in total) to increase the surface area of ​​the composite phase change material in convective contact with air, enhance the thermal conductivity of the material, and increase the safety between the two battery packs.

[0134] Therefore, due to the excellent heat absorption and energy storage properties of phase change materials, two batteries can be kept within a suitable temperature range even when they are tightly stacked, maintaining an average temperature during transportation and preventing thermal runaway of power lithium batteries. In addition, in actual implementation, the size of the first liner can be increased as needed to accommodate more power lithium batteries, thereby reducing transportation costs.

[0135] During the impact of rail vehicles, in addition to the impact forces from the interaction between the rail vehicles, there is also friction between the wheels and rails. Considering only the longitudinal force, each transport vehicle is regarded as a single-degree-of-freedom rigid body, and the rigid bodies are connected by a buffer dynamic model. The entire train is regarded as a nonlinear elastic damping system with multiple rigid bodies in series. By constructing the dynamic equations of this system, the operating characteristics such as coupler force, displacement, velocity, and acceleration are further solved. The dynamic equations of the vehicles during the shunting impact are shown in equation (1): (1) in, This represents the total number of moving vehicles and stationary vehicles. For the quality of each transport vehicle, This represents the absolute displacement of each transport vehicle. and The first The instantaneous inertial force and running resistance of the transport vehicle, For the first Force on the coupling.

[0136] refer to Figure 12 As shown, this is for fitting. Figure 12 The schematic diagram of the first pad structure designed for the power lithium battery shown is divided into two parts: an upper base and a lower base. The dimensions of the upper base are length × width × height: 1119 × 563.4 × 250 × 1119 mm. The internal dimensions are divided into two grooves on the left and right sides. Each groove is designed to fit the size of the power lithium battery tightly. The internal dimensions of the grooves on the left and right sides are length × width × height: 1019 × 211.7 × 200 mm. Both the top and bottom bases have rectangular grooves to buffer the force on the corners.

[0137] Furthermore, a second pad is arranged between the two power lithium batteries for transporting the lithium batteries in pairs. Its specifications are: length × width × height: 1019 × 550 × 40 mm. To further enhance the safety of transportation, cylindrical grooves with a radius of 10 mm and a height of 200 mm (400 mm for the upper and lower bases, with 50 mm reserved on each base) are made in the second pad. The grooves are spaced 40 mm apart and there are a total of 27 grooves to increase the convective heat exchange area between the material and the air, and to ensure thermal insulation between the two batteries.

[0138] The relevant parameters of transport packaging, such as thermal conductivity, density, tensile strength, impact strength, and flexural strength, are shown in the table below: Table: Material Parameters of Various Components of Transport Packaging Given the basic attribute parameters, the thermo-mechanical coupling of the above parameters during transportation can be determined by referring to relevant formulas: Solar radiation is also a type of thermal radiation. The heat flow that produces thermal radiation can be calculated using the Stefan-Boltzmann law formula, as shown in equation (2): (2) Where Ф represents the total radiative heat transfer; Emissivity is the surface emissivity of an object, also known as blackness, and is a number between 0 and 1 (blackbody). The area of ​​the object exposed to radiation. This is the Stefan-Boltzmann constant, with a value of 5.6704 × 10⁻⁸. These are the fourth powers of the temperatures of the two objects, respectively. For phase change materials, enthalpy-porosity techniques are used for numerical simulation calculations of the solidification-melting process, employing a general heat conduction equation in the phase change material layer: (3) in, The density of the composite phase change material; H It is the enthalpy value (J·kg) -1 The enthalpy of the material shown in the formula is the sensible enthalpy ( ). h ) and latent heat ( ); Thermal conductivity, T It's temperature. x, y, z The three heating directions represent the micro-element of the material.

[0139] in , , It is the baseline enthalpy value. It is the reference temperature. It is the specific heat capacity at normal pressure. It is the liquid phase ratio. It is the latent heat of the phase change material; the liquidus fraction of the phase change material. It can be defined as follows: (4) in, It is the temperature at which a phase change material begins to transform from a solid phase to a liquid phase. It is the temperature at which a phase change material completely transforms into a liquid state.

[0140] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0141] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for intelligent train formation and speed control of multi-vehicle collaborative power battery railway transportation, characterized in that, Includes the following steps: S1: Collect real-time status parameters of each transport vehicle on the rail vehicle and establish the status characteristics of each transport vehicle. The status characteristics include at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type and packaging protection parameters. S2: Construct a location-sensitive damage cost function based on the state characteristics to evaluate the risk cost of each truck at different formation positions, so as to quantify the comprehensive risk cost of each truck at a specific formation position; S3: In the process of rail vehicle transportation, an event-driven time pane dynamic optimization method is adopted and the vehicle formation and train speed control curves are dynamically updated according to the real-time collected state parameters. In the optimization process, a variable damping distribution strategy for couplers and a cooperative speed curve generation algorithm are adopted. At the same time, the impact time staggering strategy is used to disperse the time of coupled impact events to reduce the superposition effect of local impact peaks. S4: When the comprehensive risk index of any transport vehicle exceeds the preset threshold, an emergency regrouping strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler is implemented to adjust high-risk vehicles to low-risk positions or safety buffer zones, thereby maintaining the overall risk balance of transportation.

2. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 1, characterized in that, In step S2, the method includes: S21: Use the railway vehicle dynamics model to establish the longitudinal impact transmission matrix and combine it with the thermodynamic simulation model to calculate the failure probability of the power lithium battery under different temperature gradients, and then map the results of the two into a standardized risk probability. S22: Constructing a location-sensitive damage cost function: ,in To quantify the risk cost of the i-th freight car at different train formation positions pos, Let be the probability of battery thermal runaway at the formation position pos of the i-th truck. Let be the probability of mechanical impact damage to the i-th freight car at different train formation positions pos. Let be the expected economic loss of the i-th freight car at different train formation positions pos; These are weighting coefficients, and ; S23: Calculate the total cost of each group arrangement using a location-sensitive damage cost function, and select the grouping order using the minimum cost criterion; S24: If the total risk is less than the preset safety threshold, the current arrangement is determined to be a safe solution; otherwise, proceed to the re-optimization stage.

3. A method for intelligent train formation and speed regulation of multi-vehicle cooperative power battery railway transportation according to claim 1 or 2, characterized in that, In step S3, the event-driven time pane dynamic optimization method includes: S301: Define the scheduling timeline as a sliding pane: ,in The time window length, For time; S302: If it is determined that the preset monitoring parameter exceeds the threshold, an event is triggered; S303: Perform locally optimal adjustments on the set of transport vehicles within the time pane to which the triggering event belongs. ,in Map the new location of the i-th truck. Let be the local speed adjustment value for the i-th truck. For speed smoothing weights, The reference speed for the time pane; S304: Use a greedy search algorithm combined with a local exchange algorithm to optimally replace adjacent vehicles in the sliding pane; S305: Restore the global scheduling cycle after the risk event is resolved or the window times out.

4. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 3, characterized in that, In step S3, the variable damping distribution strategy of the coupler includes: S306: Obtain the current load, temperature, and wear level parameter set for each coupler of the rail vehicle; S307: Calculate the upper limit of energy absorption by the coupler: ,in The upper limit of energy absorption for the j-th coupler. For material coefficients, For the wear level of the j-th coupler, Let j be the temperature correction function for the j-th coupler; S308: Solve the energy allocation optimization problem for the entire formation: ,in For allocation coefficients, Let j be the energy absorption value of the j-th coupler; S309: Adjusting the damping value of the magnetorheological damper controlling the coupler: ,in This is the initial damping value. Sensitivity factor; S310: When the energy absorption of the coupler exceeds the energy safety threshold, a deceleration command is triggered. ,in For energy safety threshold, This is the speed reduction coefficient.

5. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 4, characterized in that, In step S3, the cooperative vehicle speed curve generation algorithm includes: S311: Define the speed curve for each train and set the objective as minimizing the overall risk energy function: ,in Let be the heat accumulation amount corresponding to the i-th truck. The impact energy function is calculated based on the speed difference and dynamic model, and it is proportional to the square of the speed difference between adjacent vehicles. These are the weighting coefficients; S312: Establish velocity constraints: And introduce collaborative constraints between rail vehicles: ,in Tolerance for speed coordination between adjacent transport vehicles; S313: The speed profile is iteratively optimized using model predictive control methods and all transport vehicles share the current speed to form a multi-vehicle coordinated speed queue and output the speed reference trajectory, thereby realizing the synchronous acceleration and deceleration of all transport vehicles.

6. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 5, characterized in that, In step S3, reducing the superposition of local impact peaks through the impact time staggering strategy includes: S314: Establish a time series for the expected n consecutive shock events: ; S315: Calculate the energy of each impact: And define the cumulative impact function: ,in It is the energy decay constant; S316: When the cumulative impact function exceeds the energy absorption limit of the battery pack cell, the insertion time delay staggers the impact events. , ,in This represents the upper limit of energy absorption for a battery packaging unit. The scale factor is used to control the magnitude of peak shifting; S317: Adjust train speed trajectory to meet the new impact schedule, and when all impact intervals... and At that time, the final speed regulation scheme is determined and output.

7. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 6, characterized in that, In step S4, the emergency re-grouping strategy based on the remaining energy absorption capacity of the battery pack unit and the health of the coupler includes: S41: Calculate the remaining energy absorption capacity of the battery pack unit: ,in Let be the remaining energy absorbed by the battery pack unit of the i-th transport vehicle at time t. The maximum absorbable energy of the battery pack unit of the i-th transport vehicle (which is set by the designer based on the design and materials of the battery pack unit). The input impact energy rate applied to the battery packaging unit per unit time. This refers to the energy dissipated per unit time by the battery packaging unit during processes such as phase change, plastic deformation, and viscoelastic energy dissipation. The date when the battery pack unit began service or was last reset / replaced; S42: Establish priority for transport vehicles: ,in This represents the maximum energy absorption value for the battery packaging unit corresponding to the packaging type. For the health of the coupler, The coupler's health score is perfect; S43: Arrange the transport vehicles in descending order of the aforementioned priority, and select the subset with the highest risk; S44: Objective function is to minimize the number of parking space adjustments. Solve for the new parking space mapping; where, Let be the original location index of the i-th transport vehicle; S45: Outputs and executes a transposition sequence to move high-risk vehicles to the end of the column or a buffer zone and locks the cooperative speed curve during regrouping to maintain energy balance.

8. The intelligent train formation and speed control method for multi-vehicle cooperative power battery railway transportation according to claim 7, characterized in that, During the execution of the emergency regrouping in step S4, the method includes: S401: Construct a vehicle arrangement diagram , where the vertex For vehicle position, edge To establish commutative paths, each edge is then assigned a weight. , The cost of adjusting from position x to position y; S402: For the subset with the highest risk, establish a set of transfer target locations and define the objective function: ,in The subset with the highest risk For the set of target locations; S403: Using A Heuristic search is used to find the optimal match, and after obtaining the match, a sequence of transposition instructions is generated. Each instruction includes adjustments to speed, sequence, and target time; S404: Execute the repositioning action step by step and verify the impact energy change in real time. If the impact energy change exceeds the preset energy threshold, pause the execution and replan the path.

9. A multi-vehicle cooperative intelligent trainsetting and speed control system for power battery railway transportation using the method described in any one of claims 1-8, characterized in that, include: The feature construction module is used to establish the state features of each transport vehicle in the rail vehicle. The state features include at least temperature, longitudinal stress, longitudinal acceleration, vehicle position index, load type, and packaging protection parameters. The risk assessment module is used to construct a location-sensitive damage cost function based on the state characteristics to assess the risk cost of each truck at different formation positions, so as to quantify the comprehensive risk cost of each truck at a specific formation position. The speed coordination module is used to dynamically update the vehicle formation and train speed control curves in the process of rail vehicle transportation by adopting an event-driven time pane dynamic optimization method and based on real-time collected state parameters. In the optimization process, a variable damping distribution strategy for couplers and a coordinated speed curve generation algorithm are adopted. At the same time, the impact time staggering strategy is used to disperse the time of coupled impact events to reduce the superposition effect of local impact peaks. The regrouping control module is used to execute an emergency regrouping strategy based on the remaining energy absorption of the battery pack unit and the health of the coupler when the comprehensive risk index of any transport vehicle exceeds a preset threshold. This strategy aims to adjust high-risk vehicles to low-risk positions or safety buffer zones, thereby maintaining the overall risk balance of transportation.

Images