Rail point position regulation and control suppression system based on multi-objective optimization

Abstract

This invention relates to the field of electromagnetic compatibility technology in power systems, and more particularly to a rail position control and suppression system based on multi-objective optimization. The invention generates a target potential distribution prediction scheme through a potential distribution prediction generation module, which then corrects the scheme using a spatial calibration unit based on real-time train position offset to generate spatial dynamic correction coefficients. The spatial optimization unit then optimizes the scheme based on historical soil resistivity drift data. Simultaneously, a target drainage action timing scheme is generated through a timing scheme generation module, corrected by a timing calibration unit based on real-time rail potential deviation to generate timing dynamic correction coefficients, and then optimized by a timing optimization unit based on historical equipment performance data. Finally, a comprehensive point-based control module achieves coordinated spatial and temporal control.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic compatibility technology for power systems, and in particular to a rail position control and suppression system based on multi-objective optimization. Background Technology

[0002] With the rapid development of urban rail transit, when subway trains operate using DC traction power supply systems, poor insulation between the running rails and the ground causes some traction current to leak into the ground, forming stray currents. These stray currents are conducted through the soil, causing changes in the ground potential gradient, which leads to a shift in the neutral point potential of transformers in nearby substations. This triggers DC bias in the transformers, resulting in half-wave saturation of the core, distortion of the excitation current, increased vibration and noise, and localized overheating, seriously threatening the safety of power grid equipment.

[0003] For example, Chinese patent CN111324966B discloses a simulation calculation method for stray current in urban rail transit based on CDEGS. Specifically, it involves building a simulation model of urban rail transit leakage current using CDEGS software, based on the urban rail transit line topology. Relevant parameters are set in the simulation model of the urban rail transit leakage circuit based on parameters such as the contact wire resistance per unit length, track system parameters, drainage system parameters, locomotive system parameters, and traction substation parameters. The model is then simulated using CDEGS software to obtain the rail leakage current, and the stray current in the urban rail transit system is obtained through calculation. However, this method still suffers from problems such as static rail point distribution prediction, delayed drainage timing control, and low point suppression efficiency and rapid equipment lifespan degradation due to the fragmentation of spatial and temporal control. Summary of the Invention

[0004] To address this, the present invention provides a rail position control and suppression system based on multi-objective optimization, which overcomes the problems of low position suppression efficiency and rapid equipment lifespan decay caused by static prediction of rail position distribution, delayed drainage timing control, and fragmented spatial and temporal control in the prior art.

[0005] To achieve the above objectives, the present invention provides a rail position control and suppression system based on multi-objective optimization, comprising: The data acquisition module is used to collect basic track topology data, real-time train location data, historical soil resistivity drift data, basic train operation diagram data, real-time track potential monitoring data, and historical equipment performance data. The potential distribution prediction generation module is used to generate a potential distribution prediction scheme based on basic track topology data, real-time train position data and historical soil resistivity drift data, and obtain the target potential distribution prediction scheme. The timing scheme generation module is used to generate the drainage action timing scheme based on the basic train operation diagram data, real-time rail potential monitoring data and equipment operation history performance data, so as to obtain the target drainage action timing scheme. The point-based integrated control module is used to generate the final point-based control scheme based on the target potential distribution prediction scheme and the target drainage action timing scheme.

[0006] Furthermore, the potential distribution prediction generation module includes: The basic scheme generation unit is used to generate a basic potential distribution prediction scheme based on the basic track topology data, thereby obtaining the basic potential distribution prediction scheme. A spatial calibration unit is used to calibrate the generation process of the basic potential distribution prediction scheme based on real-time train position data to obtain a first potential distribution prediction scheme. The spatial optimization unit is used to optimize the spatial calibration process based on historical soil resistivity drift data to obtain a target potential distribution prediction scheme. The basic scheme generation unit generates a basic potential distribution prediction scheme based on the basic track topology data. When obtaining the basic potential distribution prediction scheme, based on the track segment length TL, track longitudinal resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS in the basic track topology data, a rail-ground return current network model is established to calculate the predicted potential value VP and the predicted stray current density value CD at each monitoring point along the rail line, where: The predicted potential value VP = (PL × RR × TL) / (2 × TG × TS + TL), where PL is the traction load power; The predicted stray current density is CD = (PL × TG) / (2 × π × RL × TS), where RL is the soil resistivity. The predicted potential value VP and the predicted stray current density value CD are output as the basic potential distribution prediction scheme.

[0007] Further, when the spatial calibration unit calibrates the generation process of the basic potential distribution prediction scheme based on real-time train position data, it calculates the position offset PO based on the real-time train mileage marker RM in the real-time train position data and the preset train mileage marker PM in the basic potential distribution prediction scheme, setting PO = |RM-PM| / TS, where TS is the traction substation spacing. The position offset PO is compared with a first preset offset threshold P1 and a second preset offset threshold P2, and the spatial dynamic correction coefficient SC is assigned a value based on the comparison result, wherein: When PO≤P1, set SC=1.0; When P1 < PO ≤ P2, set SC = 1.0 + 0.2 × (PO - P1) / (P2 - P1); When PO > P2, set SC = 1.25; The potential prediction value VP is corrected according to the spatial dynamic correction coefficient SC. The corrected potential prediction value VC = VP × SC is set. The value of the potential prediction value VP output by the basic potential distribution prediction scheme is replaced with the value of the corrected potential prediction value to obtain the first potential distribution prediction scheme.

[0008] Furthermore, when the spatial optimization unit optimizes the spatial calibration process based on historical soil resistivity drift data, it calculates the spatial calibration convergence index CI based on the historical mean soil resistivity RH, historical soil moisture variation coefficient VH, and seasonal drift amplitude SA in the historical soil resistivity drift data. CI is set as (SA / RH) × (1 + VH). The spatial calibration convergence index CI is compared with a preset convergence threshold CT, and the optimization parameter OP of the spatial calibration process is assigned a value based on the comparison result. When CI ≤ CT, set OP = 0.1; When CI > CT, set OP = 0.2; The calculation logic of the spatial dynamic correction coefficient SC is optimized according to the optimization parameter OP of the spatial calibration process. The optimized spatial correction coefficient SO is set as SC×[1+OP×(SA / RH)], and the value of the spatial dynamic correction coefficient SC is replaced with the value of the optimized spatial correction coefficient SO. The predicted value of the corrected potential is recalculated to obtain the target potential distribution prediction scheme.

[0009] Furthermore, the timing scheme generation module includes: The timing sequence basic scheme generation unit is used to generate the basic drainage action timing scheme based on the basic driving diagram data, so as to obtain the basic drainage action timing scheme; The timing calibration unit is used to calibrate the generation process of the drainage action timing scheme based on real-time track potential monitoring data to obtain the first drainage action timing scheme. The timing optimization unit is used to optimize the timing calibration process based on the historical performance data of the equipment operation to obtain the target drainage operation timing scheme. In the basic timing scheme generation unit, the basic drainage operation timing scheme is generated based on the basic train schedule data. When obtaining the basic drainage operation timing scheme, the train departure timetable TD, train interval running time RT, and train stop time ST in the basic train schedule data are used to calculate the drainage cabinet switching time TC, rail potential limiter action threshold timing TH, and stray current collection network switch status timing SS through the drainage demand prediction algorithm, wherein: The start-up time of the drain cabinet is TC=TD+RT+0.5×ST-ΔT, where ΔT is the drain advance amount; The timing sequence of the rail potential limiter's operating threshold TH is dynamically set according to the distance between the train position and the traction substation. The basic operating threshold TH1 for the near-end rail potential limiter is 60V, and the basic operating threshold TH2 for the far-end rail potential limiter is 80V. The stray current collection network switch state timing SS is set according to the peak time of the predicted stray current density value CD, closing 5 minutes before the peak of CD and opening 5 minutes after the peak. The switching time TC of the drain cabinet, the action threshold timing TH of the rail potential limiter, and the switch status timing SS of the stray current collection network are output as the basic drain action timing scheme.

[0010] Further, when the timing calibration unit calibrates the generation process of the foundation drainage action timing scheme based on real-time rail potential monitoring data, it calculates the potential deviation PE based on the real-time rail potential RP in the real-time rail potential monitoring data and the predicted rail potential PP in the foundation drainage action timing scheme, setting PE=|RP-PP| / PP, comparing the potential deviation PE with the first preset potential deviation threshold E1 and the second preset potential deviation threshold E2, and assigning a value to the timing dynamic correction coefficient TE based on the comparison result, wherein: When PE≤E1, set TE=1.0; When E1 < PE ≤ E2, set TE = 1.0 + 0.15 × (PE - E1) / (E2 - E1); When PE > E2, set TE = 1.2; The timing of the drain cabinet switching TC is corrected according to the timing dynamic correction coefficient TE. The corrected drain time TM is set as TC-TE×ΔT, where ΔT is the drain advance amount. The value of the drain cabinet switching time TC in the basic drain action timing scheme is replaced with the value of the corrected drain time TM to obtain the first drain action timing scheme.

[0011] Further, when the timing optimization unit optimizes the timing calibration process based on the historical performance data of the equipment operation, it calculates the timing calibration performance index EI based on the average monthly number of drainage cabinet operations (MA), the operation success rate (SR), and the operation response delay (RD) in the historical performance data of the equipment operation. EI is set as EI = (MA / 30) × (1 - SR) × (RD / 100), where MA is the average monthly number of operations, SR is the operation success rate, and RD is the response delay in milliseconds. The timing calibration performance index EI is compared with a preset performance threshold ET, and the timing calibration process optimization parameter TO is assigned a value based on the comparison result, where: When EI≤ET, set TO=0.05; When EI > ET, set TO = 0.12; The calculation logic of the timing dynamic correction coefficient TE is optimized according to the timing calibration process optimization parameter TO. The optimized timing correction coefficient TI is set as TE×[1+TO×(1-SR)], and the value of the timing dynamic correction coefficient TE is replaced with the value of the optimized timing correction coefficient TI. The corrected drainage time is recalculated to obtain the target drainage action timing scheme.

[0012] Furthermore, the point-based integrated control module generates a basic coordinated control scheme based on the corrected potential prediction value VC in the target potential distribution prediction scheme and the corrected drainage time TM in the target drainage action timing scheme, thereby obtaining a basic control threshold matrix. The basic control threshold matrix is ​​set as TH=[TH1, TH2, TC], where TH1 is the basic threshold for the near-end rail potential limiter action, set as TH1=VC×k1, where k1 is the near-end safety factor; TH2 is the basic threshold for the far-end rail potential limiter action, set as TH2=VC×k2, where k2 is the far-end safety factor; and TC is the basic switching time of the drainage cabinet, set as TC=TM-ΔT, where ΔT is the drainage advance.

[0013] Further, the point-based integrated control module calculates the potential prediction deviation PV based on the real-time instantaneous rail potential value RT from the real-time rail potential monitoring data and the corrected potential prediction value VC, setting PV=|RT-VC| / VC. The potential prediction deviation PV is then compared with a first preset deviation threshold V1 and a second preset deviation threshold V2. Based on the comparison result, a value is assigned to the dynamic control correction coefficient DC, wherein: When PV≤V1, set DC=1.0; When V1 < PV ≤ V2, set DC = 1.0 + 0.1 × (PV - V1) / (V2 - V1); When PV > V2, set DC = 1.15; The basic threshold values ​​for the rail potential limiter in the basic control threshold matrix TH are corrected according to the dynamic control correction coefficient DC. The corrected basic threshold value for the near-end rail potential limiter is set to TH1' = TH1 × DC, and the corrected basic threshold value for the far-end rail potential limiter is set to TH2' = TH2 × DC. Then, TH1 and TH2 in the basic control threshold matrix TH are replaced with the corrected basic threshold values ​​for the near-end rail potential limiter TH1' and TH2' for the far-end rail potential limiter, thus obtaining the first control threshold matrix TH'.

[0014] Furthermore, the point-based integrated control module calculates the time deviation degree DV based on the corrected discharge time TC and the preset safe time threshold TS, setting DV=|TC-TS| / TS, comparing the time deviation degree DV with the first preset deviation threshold D1 and the second preset deviation threshold D2, determining the control risk priority PR based on the comparison result, and optimizing the first control threshold matrix TH' based on the determination result, wherein: When DV≤D1, the control risk priority PR is determined to be low risk, the first control threshold matrix TH' is not optimized for risk, and the first control threshold matrix TH' is output as the final point control scheme. When D1 < DV ≤ D2, the risk priority PR of the regulation is determined to be medium risk. The first regulation threshold matrix TH' is optimized for risk. The modified near-end rail potential limiter action base threshold TH1' is lowered by the risk compensation coefficient RC. RC is set to 0.9-0.1×(DV-D1) / (D2-D1), and the optimized near-end rail potential limiter action base threshold TH1'' = TH1'×RC is obtained. The basic switching time TC of the drain cabinet is adjusted to TC-2min. TH1' in the first regulation threshold matrix TH' is replaced with the optimized near-end rail potential limiter action base threshold TH1'', resulting in the second regulation threshold matrix TH''. The second regulation threshold matrix TH'' is output as the final point regulation scheme. When DV > D2, the control risk priority PR is determined to be high risk. Risk optimization is performed on the first control threshold matrix TH', the rail potential limiter forced grounding mode is immediately triggered, the forced grounding threshold TH0 = 0.5 × VC is set, and the switching time TC of the drain cabinet foundation is adjusted to immediate action. An emergency control command is generated as the final point control scheme output.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The system uniformly collects basic track topology data, real-time train position data, historical soil resistivity drift data, basic train operation diagram data, real-time track potential monitoring data, and historical equipment performance data through the data acquisition module, providing a data foundation for subsequent spatial and temporal dual-layer optimization; the system generates a target potential distribution prediction scheme through a potential distribution prediction generation module, solving the prediction inaccuracy problem caused by relying solely on static topology in existing technologies, and realizing a baseline potential distribution prediction based on track topology; the system calibrates the generation process of the basic potential distribution prediction scheme through a spatial calibration unit, solving the potential prediction inaccuracy problem caused by the deviation between the actual and planned train positions, and realizing dynamic spatial correction based on real-time train positions; the system optimizes the spatial calibration process through a spatial optimization unit, solving the problem of slow convergence of spatial calibration coefficients caused by seasonal drift of soil resistivity. The system addresses several key issues: First, it addresses the problem of large overshoot by implementing adaptive optimization based on historical soil parameter drift. Second, it generates target drainage action timing schemes through a timing scheme generation module, resolving the lack of correlation between drainage actions and train operation diagrams in existing technologies, thus achieving baseline drainage timing planning based on train operation diagrams. Third, it uses a timing calibration unit to calibrate the generation process of the basic drainage action timing scheme, addressing the issue that fixed timing schemes cannot adapt to real-time potential fluctuations, and achieving dynamic timing correction based on real-time track potential monitoring. Fourth, it optimizes the timing calibration process through a timing optimization unit, addressing the issue of equipment lifespan degradation caused by frequent drainage cabinet operations, and achieving timing optimization based on historical equipment performance. Fifth, it uses a point-based integrated control module to coordinate the target potential distribution prediction scheme and the target drainage action timing scheme, resolving the problem of the separation between spatial potential control and timing drainage control, and achieving multi-objective collaborative optimization of point suppression effect and equipment action performance. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the rail position control and suppression system based on multi-objective optimization in this embodiment; Figure 2 A schematic diagram of the potential distribution prediction generation module; Figure 3 A schematic diagram of the timing scheme generation module. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 The diagram shown is a structural schematic of the rail position control and suppression system based on multi-objective optimization in this embodiment. The system includes: The data acquisition module is used to collect basic track topology data, real-time train location data, historical soil resistivity drift data, basic train operation diagram data, real-time track potential monitoring data, and historical equipment performance data. The potential distribution prediction generation module is used to generate a potential distribution prediction scheme based on basic track topology data, real-time train position data and historical soil resistivity drift data, and obtain a target potential distribution prediction scheme. The potential distribution prediction generation module is connected to the data acquisition module. The timing scheme generation module is used to generate the drainage action timing scheme based on the basic train operation diagram data, real-time track potential monitoring data and equipment operation history performance data, so as to obtain the target drainage action timing scheme. The timing scheme generation module is connected to the potential distribution prediction generation module. The point-based integrated control module is used to generate the final point-based control scheme based on the target potential distribution prediction scheme and the target drainage action timing scheme. The point-based integrated control module is connected to the timing scheme generation module.

[0022] Specifically, the rail position control and suppression system based on multi-objective optimization is applied to the safety monitoring scenarios of subway traction power supply systems and grounding grids, such as urban rail transit lines with DC750V or DC1500V power supply systems. The system uses a data acquisition module to uniformly collect basic track topology data, real-time train position data, historical soil resistivity drift data, basic train operation diagram data, real-time rail potential monitoring data, and historical equipment performance data, providing a data foundation for subsequent spatial and temporal dual-layer optimization. The system generates a target potential distribution prediction scheme through a potential distribution prediction generation module, solving the prediction inaccuracy problem caused by relying solely on static topology in existing technologies, and achieving a baseline potential distribution prediction based on track topology. The system uses a spatial calibration unit to calibrate the generation process of the basic potential distribution prediction scheme, solving the potential prediction inaccuracy problem caused by the deviation between the actual and planned train positions, and achieving dynamic spatial correction based on real-time train positions. The system uses a spatial optimization unit to optimize the spatial calibration process. The system is optimized to address the issues of slow convergence and large overshoot of spatial calibration coefficients caused by seasonal drift in soil resistivity, achieving adaptive optimization based on historical drift of soil parameters. The system generates target drainage action timing schemes through a timing scheme generation module, addressing the lack of correlation between drainage actions and train operation diagrams in existing technologies, and achieving baseline drainage timing planning based on train operation diagrams. The system also verifies the generation process of the basic drainage action timing scheme through a timing calibration unit, addressing the inability of fixed timing schemes to adapt to real-time potential fluctuations, and achieving dynamic timing correction based on real-time track potential monitoring. Furthermore, the system optimizes the timing calibration process through a timing optimization unit, addressing the equipment lifespan degradation caused by frequent operation of the drainage cabinet, and achieving timing optimization based on historical equipment performance. Finally, the system uses a point-based integrated control module to coordinate the target potential distribution prediction scheme and the target drainage action timing scheme, resolving the disconnect between spatial potential control and timing drainage control, and achieving multi-objective collaborative optimization of point suppression effects and equipment action performance.

[0023] Specifically, when the data acquisition module collects the basic track topology data, the real-time train position data, the historical soil resistivity drift data, the basic train timetable data, the real-time track potential monitoring data, and the historical equipment operation performance data, it collects the track segment length TL, track longitudinal resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS from the basic track topology data through the track parameter database; it collects the real-time train mileage marker RM from the real-time train position data through the Automatic Train Monitoring System (ATS); and it collects the data from the soil resistivity monitoring terminal... The system collects the historical average soil resistivity (RH), historical coefficient of variation of soil moisture (VH), and seasonal drift amplitude (SA) from the historical soil resistivity drift data. It also collects the train departure timetable (TD), train interval running time (RT), and train stop time (ST) from the basic train schedule data through the train dispatching system interface. Furthermore, it collects the real-time rail potential (RP) from the real-time rail potential monitoring data through the trackside potential sensor. Finally, it collects the average monthly number of drainage cabinet actions (MA), action success rate (SR), and action response delay (RD) from the equipment action history performance data through the drainage cabinet monitoring terminal.

[0024] Specifically, the track parameter database refers to a database system that stores design parameters and as-built electrical parameters of subway lines, including track section length TL, longitudinal track resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS. The track section length TL refers to the length of a single rail between two adjacent traction substations, measured in km. The longitudinal track resistance RR refers to the DC resistance per unit length of rail, measured in Ω / km, determined by the resistivity of the rail material and the quality of rail joint welding. The track-to-ground transition conductance TG refers to the leakage conductance per unit length between the rail and the ground, measured in S / km, determined by the insulation resistance of the rail fasteners and... The track bed resistance is comprehensively measured; the traction substation spacing TS refers to the difference in center mileage between two adjacent traction substations, in km; the Automatic Train Monitoring System (ATS) is a signal system that monitors the train's position and running status in real time, detecting the real-time mileage marker RM of the train through axle counters or transponders, in m; the soil resistivity monitoring terminal refers to a network of soil resistivity sensors buried along the track, collecting the historical average soil resistivity RH, the historical coefficient of variation of soil moisture VH, and the seasonal drift amplitude SA; the historical average soil resistivity RH refers to the arithmetic mean of soil resistivity in the same season over the past 3 to 5 years, in Ω·m. The soil moisture historical variation coefficient VH refers to the ratio of the standard deviation of soil resistivity to the mean, which is dimensionless; the seasonal drift amplitude SA refers to the difference between the peak and trough values ​​in the annual cycle of soil resistivity, with the unit being Ω·m; the train dispatching system interface refers to the data communication interface with the subway operation control center ATS system, which collects the train departure timetable TD, the train interval running time RT, and the train stop time ST; the train departure timetable TD refers to the planned departure time of the train from the originating station; the train interval running time RT refers to the planned running time of the train between two stations, with the unit being min; the train stop time ST ... running time of the train between two stations, with the unit being min; the train interval running time RT refers to the planned running time of the train between two stations, with the unit being min; the train stop time ST refers to the planned running time of the train between two stations, with the unit being min; the train departure timetable TD refers to the planned running time of the train between two stations, with the unit being min; the train departure timetable TD refers to The planned stopping time at the station is in minutes; the trackside potential sensor refers to a voltage monitoring device installed beside the rail, which collects the real-time rail potential RP, in volts; the drain cabinet monitoring terminal refers to an intelligent terminal that monitors the operating status of the drain cabinet, collecting the average monthly number of drain cabinet actions (MA), action success rate (SR), and action response delay (RD); the average monthly number of drain cabinet actions (MA) refers to the total number of switching actions performed by the drain cabinet in the past 30 days; the action success rate (SR) refers to the proportion of successfully completed switching actions to the total number of actions; the action response delay (RD) refers to the time difference from the issuance of the control command to the closure of the main contact, in milliseconds.

[0025] Specifically, the basic scheme generation unit generates a basic potential distribution prediction scheme based on the basic track topology data. When obtaining the basic potential distribution prediction scheme, based on the track segment length TL, track longitudinal resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS in the basic track topology data, a rail-ground return current network model is established to calculate the predicted potential value VP and the predicted stray current density value CD at each monitoring point along the rail line, where: The predicted potential value VP = (PL × RR × TL) / (2 × TG × TS + TL), where PL is the traction load power; The predicted stray current density is CD = (PL × TG) / (2 × π × RL × TS), where RL is the soil resistivity. The predicted potential value VP and the predicted stray current density value CD are output as the basic potential distribution prediction scheme.

[0026] Specifically, the basic track topology data refers to a set of static parameters describing the physical structure and electrical connection attributes of a subway line, including track section length TL, track longitudinal resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS. The track section length TL refers to the length of a single rail between two adjacent traction substations, measured in km, obtained from line design drawings or as-built survey data. The track longitudinal resistance RR refers to the DC resistance per unit length of rail, measured in Ω / km, determined through inspection of rail material and rail joint welding quality. The track-to-ground transition conductance TG refers to the leakage conductance per unit length between the rail and the ground, measured in S / km, determined by a combination of rail fastener insulation resistance, track bed resistance, and tunnel structure leakage. The traction substation spacing TS... The spacing TS refers to the difference in mileage between the centers of two adjacent traction substations, in km, typically 2 km to 4 km; the predicted potential value VP refers to the theoretical rail-to-ground potential calculated based on transmission line theory, in V, where the numerator PL×RR×TL represents the voltage drop of the traction current across the rail resistance, and the denominator 2×TG×TS+TL represents the current shunting coefficient of the rail-to-ground return network; the predicted stray current density CD refers to the current density leaking from the rail into the ground, in mA / m², where PL×TG represents the total leakage current, and 2×π×RL×TS represents the equivalent hemispherical area of ​​the current diffusion; the basic potential distribution prediction scheme refers to a benchmark dataset containing VP and CD values ​​for each monitoring point, used as an initial reference for subsequent spatial calibration.

[0027] Specifically, when the spatial calibration unit calibrates the generation process of the basic potential distribution prediction scheme based on real-time train position data, it calculates the position offset PO based on the real-time train mileage marker RM in the real-time train position data and the preset train mileage marker PM in the basic potential distribution prediction scheme, setting PO = |RM - PM| / TS, where TS is the traction substation spacing. The position offset PO is compared with a first preset offset threshold P1 and a second preset offset threshold P2, and a spatial dynamic correction coefficient SC is assigned a value based on the comparison result, wherein: When PO≤P1, set SC=1.0; When P1 < PO ≤ P2, set SC = 1.0 + 0.2 × (PO - P1) / (P2 - P1); When PO > P2, set SC = 1.25; The potential prediction value VP is corrected according to the spatial dynamic correction coefficient SC. The corrected potential prediction value VC = VP × SC is set. The value of the potential prediction value VP output by the basic potential distribution prediction scheme is replaced with the value of the corrected potential prediction value to obtain the first potential distribution prediction scheme.

[0028] Specifically, the real-time train position data refers to the real-time running position information of the train obtained through the Automatic Train Control (ATS) system, including the real-time train mileage marker RM. The real-time train mileage marker RM refers to the track mileage coordinates corresponding to the current position of the train's head, in meters (m), detected by a trackside axle counter or transponder and transmitted via a communication network. The preset train mileage marker PM refers to the mileage coordinates of the train at a planned position as specified in the basic train timetable, in meters (m). The position offset PO refers to the relative deviation between the actual and planned positions of the train, dimensionless, and in the calculation formula, |RM-PM| represents the absolute deviation, with the denominator TS used for normalization to reflect the offset. The proportion of the power supply interval length; the first preset offset threshold P1 is 0.05, indicating that when the offset is less than 5% of the power supply interval length, the train is considered on time and no correction is needed; the second preset offset threshold P2 is 0.15, indicating that linear interpolation correction is initiated when the offset is in the range of 5% to 15%; the spatial dynamic correction coefficient SC refers to the amplification factor of the basic potential prediction value, with a value of 1.0 to 1.25. The coefficient 1.0 indicates that the original prediction value is maintained when there is no offset, and the coefficient 1.25 indicates that the potential peak value may increase by 25% when the offset exceeds 15%; the corrected potential prediction value VC refers to the rail potential prediction value after position offset correction, with the unit being V.

[0029] Specifically, when the spatial optimization unit optimizes the spatial calibration process based on historical soil resistivity drift data, it calculates the spatial calibration convergence index CI based on the historical mean soil resistivity RH, historical soil moisture variation coefficient VH, and seasonal drift amplitude SA in the historical soil resistivity drift data. CI is set as (SA / RH) × (1 + VH). The spatial calibration convergence index CI is compared with a preset convergence threshold CT, and the optimization parameter OP of the spatial calibration process is assigned a value based on the comparison result. When CI ≤ CT, set OP = 0.1; When CI > CT, set OP = 0.2; The calculation logic of the spatial dynamic correction coefficient SC is optimized according to the optimization parameter OP of the spatial calibration process. The optimized spatial correction coefficient SO is set as SC×[1+OP×(SA / RH)], and the value of the spatial dynamic correction coefficient SC is replaced with the value of the optimized spatial correction coefficient SO. The predicted value of the corrected potential is recalculated to obtain the target potential distribution prediction scheme.

[0030] Specifically, the historical soil resistivity drift data refers to historical statistical data reflecting the seasonal variation of soil resistivity, including the historical mean soil resistivity (RH), the historical coefficient of variation of soil moisture (VH), and the seasonal drift amplitude (SA). The historical mean soil resistivity (RH) is the arithmetic mean of soil resistivity in the same season over the past 3 to 5 years, measured in Ω·m. The historical coefficient of variation of soil moisture (VH) is the ratio of the standard deviation of soil resistivity to the mean; it is dimensionless and characterizes the dispersion of soil resistivity, with a larger VH indicating more drastic fluctuations in soil moisture. The seasonal drift amplitude (SA) is the difference between the peak and trough values ​​in the annual cycle of soil resistivity, measured in Ω·m; typically, SA can reach 50 Ω·m to 100 Ω·m during the rainy season. The spatial correction convergence index (CI) is the index that characterizes the spatial correction process. The quantitative index of soil parameter drift is dimensionless. In the calculation formula, SA / RH represents the relative drift intensity, and (1+VH) represents the humidity fluctuation amplification factor. The preset convergence threshold CT is set to 0.2. When CI≤0.2, the soil parameters are considered stable, and when CI>0.2, the soil parameter drift is considered significant. The spatial calibration process optimization parameter OP refers to the step size factor that adjusts the calculation logic of the spatial dynamic correction coefficient. OP=0.1 represents the normal step size, and OP=0.2 represents the accelerated step size. The basis for the value is that when the soil drift is significant, the correction coefficient update speed needs to be accelerated to track parameter changes. The optimized spatial correction coefficient SO refers to the final spatial correction coefficient after soil drift compensation. In the calculation formula, OP×(SA / RH) represents the additional correction amount added according to the drift intensity.

[0031] Specifically, when the basic timing scheme generation unit generates the basic drainage operation timing scheme based on the basic train schedule data, and obtains the basic drainage operation timing scheme, it calculates the drainage cabinet switching time TC, the rail potential limiter operation threshold timing TH, and the stray current collection network switch status timing SS based on the train departure timetable TD, train interval running time RT, and train stop time ST in the basic train schedule data using the drainage demand prediction algorithm. The start-up time of the drain cabinet is TC=TD+RT+0.5×ST-ΔT, where ΔT is the drain advance amount; The timing sequence of the rail potential limiter's operating threshold TH is dynamically set according to the distance between the train position and the traction substation. The basic operating threshold TH1 for the near-end rail potential limiter is 60V, and the basic operating threshold TH2 for the far-end rail potential limiter is 80V. The stray current collection network switch state timing SS is set according to the peak time of the predicted stray current density value CD, closing 5 minutes before the peak of CD and opening 5 minutes after the peak. The switching time TC of the drain cabinet, the action threshold timing TH of the rail potential limiter, and the switch status timing SS of the stray current collection network are output as the basic drain action timing scheme.

[0032] Specifically, the basic train schedule data refers to the train operation plan data specified in the metro operating timetable, including the train departure timetable TD, train interval running time RT, and train stop time ST. The train departure timetable TD refers to the planned departure time of the train from the originating station, in the format of hours:minutes:seconds. The train interval running time RT refers to the planned running time of the train between two adjacent stations, in minutes. The train stop time ST refers to the planned stopping time of the train at the station, in minutes. The drainage advance ΔT refers to the time margin for initiating drainage in advance to prevent potential over-limit, typically taken as 30s to 60s. The drainage cabinet switching time TC refers to the planned action sequence of the drainage cabinet negative terminal cabinet from opening to closing. The rail potential limiter action threshold timing TH refers to the potential at which the rail potential limiter triggers grounding. The threshold value sequence changes over time. The basic threshold value TH1=60V for the near-end rail potential limiter refers to the protection threshold within 500m of the traction substation. The basic threshold value TH2=80V for the far-end rail potential limiter refers to the protection threshold within 500m to 2000m of the traction substation. The difference in value is based on the following: the stray current is large at the near end, but the absolute value of the rail potential is affected by the substation clamping. The stray current is small at the far end, but the potential may increase due to long-distance accumulation. The stray current collection network switch state sequence SS refers to the on / off state sequence of the collection network isolating switch. Closing 5 minutes before the peak means closing the switch 5 minutes before the predicted stray current density CD reaches the peak value to provide a low-resistance path for the current. Disconnecting 5 minutes after the peak means disconnecting 5 minutes after the peak value to ensure sufficient collection of stray current.

[0033] Specifically, when the timing calibration unit calibrates the generation process of the foundation drainage action timing scheme based on real-time rail potential monitoring data, it calculates the potential deviation PE based on the real-time rail potential RP in the real-time rail potential monitoring data and the predicted rail potential PP in the foundation drainage action timing scheme, setting PE=|RP-PP| / PP. The potential deviation PE is then compared with a first preset potential deviation threshold E1 and a second preset potential deviation threshold E2. Based on the comparison result, the timing dynamic correction coefficient TE is assigned a value, where: When PE≤E1, set TE=1.0; When E1 < PE ≤ E2, set TE = 1.0 + 0.15 × (PE - E1) / (E2 - E1); When PE > E2, set TE = 1.2; The timing of the drain cabinet switching TC is corrected according to the timing dynamic correction coefficient TE. The corrected drain time TM is set as TC-TE×ΔT, where ΔT is the drain advance amount. The value of the drain cabinet switching time TC in the basic drain action timing scheme is replaced with the value of the corrected drain time TM to obtain the first drain action timing scheme.

[0034] Specifically, the real-time rail potential monitoring data refers to the rail-to-ground potential data collected in real time by the railside potential sensor, including the real-time rail potential RP; the real-time rail potential RP refers to the measured rail-to-ground potential value at the current moment, in V, with a sampling frequency of once every 1 second; the predicted rail potential PP refers to the theoretical predicted potential value at the corresponding moment in the drainage action timing scheme, in V; the potential deviation PE refers to the relative error between the real-time value and the predicted value, dimensionless; the first preset potential deviation threshold E1 is set to 0.1, indicating that a deviation of less than 10% is considered an accurate prediction; The second preset potential deviation threshold E2 is set to 0.3, indicating that linear interpolation correction is initiated when the deviation is between 10% and 30%. The timing dynamic correction coefficient TE is the adjustment coefficient for the advance of drainage, with a value between 1.0 and 1.2. A coefficient of 1.0 indicates that the original drainage time is maintained when there is no deviation, and a coefficient of 1.2 indicates that drainage needs to be initiated 20% ΔT time in advance when the deviation exceeds 30%. Here, TE = 1.2 corresponds to TM = TC - 1.2 ΔT, which is 0.2 ΔT earlier than the original plan. The corrected drainage time TM refers to the final drainage start time after real-time potential deviation correction.

[0035] Specifically, when the timing optimization unit optimizes the timing calibration process based on the historical performance data of the equipment operation, it calculates the timing calibration performance index EI based on the average monthly number of drainage cabinet operations (MA), the operation success rate (SR), and the operation response delay (RD) in the historical performance data of the equipment operation. EI is set as (MA / 30) × (1-SR) × (RD / 100), where MA is the average monthly number of operations, SR is the operation success rate, and RD is the response delay in milliseconds. The timing calibration performance index EI is compared with a preset performance threshold ET, and the timing calibration process optimization parameter TO is assigned a value based on the comparison result, where: When EI≤ET, set TO=0.05; When EI > ET, set TO = 0.12; The calculation logic of the timing dynamic correction coefficient TE is optimized according to the timing calibration process optimization parameter TO. The optimized timing correction coefficient TI is set as TE×[1+TO×(1-SR)], and the value of the timing dynamic correction coefficient TE is replaced with the value of the optimized timing correction coefficient TI. The corrected drainage time is recalculated to obtain the target drainage action timing scheme.

[0036] Specifically, the historical performance data of the equipment refers to statistical data reflecting the long-term operating performance of the drain cabinet and rail potential limiter, including the average monthly number of drain cabinet operations (MA), the success rate (SR), and the response delay (RD). The average monthly number of drain cabinet operations (MA) refers to the total number of switching operations performed by the drain cabinet in the past 30 days, measured in times. The success rate (SR) is the proportion of successful switching operations to the total number of operations, dimensionless, and successful switching operations refer to no mechanical jamming or electrical faults. The response delay (RD) is the time difference from the issuance of the control command to the actual closure of the main contacts of the drain cabinet, measured in milliseconds (ms), typically required to be less than 100 ms. The timing calibration performance index (EI) is a dimensionless quantitative indicator characterizing the impact of equipment performance degradation on the timing calibration process. In the calculation formula, (MA / 30) represents the average daily operation frequency, (1-SR) represents the failure rate, and (RD / 100) represents the response delay factor. The preset performance threshold ET is set to 0.5. When EI≤0.5, the equipment performance is considered good, and when EI>0.5, the equipment performance is considered to be degraded and needs to be accelerated for optimization. The timing calibration process optimization parameter TO refers to the acceleration factor that adjusts the timing dynamic correction coefficient calculation logic. TO=0.05 represents the normal convergence speed, and TO=0.12 represents accelerated convergence. The basis for the value is: when the equipment failure rate is high, the timing calibration speed needs to be accelerated to compensate for the equipment response delay. The optimized timing calibration coefficient TI refers to the final timing calibration coefficient after equipment performance feedback optimization. In the calculation formula, TO×(1-SR) represents the amount of calibration acceleration added based on the failure rate.

[0037] Specifically, the point-based integrated control module generates a basic coordinated control scheme based on the corrected potential prediction value VC in the target potential distribution prediction scheme and the corrected drainage time TM in the target drainage action timing scheme, thereby obtaining a basic control threshold matrix. The basic control threshold matrix is ​​set as TH=[TH1, TH2, TC], where TH1 is the basic threshold for the near-end rail potential limiter action, set as TH1=VC×k1, where k1 is the near-end safety factor; TH2 is the basic threshold for the far-end rail potential limiter action, set as TH2=VC×k2, where k2 is the far-end safety factor; and TC is the basic switching time of the drainage cabinet, set as TC=TM-ΔT, where ΔT is the drainage advance.

[0038] Specifically, the near-end safety factor k1 refers to a safety margin factor set based on the rail potential attenuation characteristics near the traction substation, with a value of 0.85. The reason is that the near-end area is close to the traction substation and is significantly affected by traction return current, resulting in a higher absolute value of the rail potential, typically 1.2 to 1.4 times that of the far-end potential. To prevent malfunction of the potential limiter and ensure personal safety, a 15% safety margin needs to be reserved; therefore, k1 is set to 0.85, ensuring that the near-end action threshold TH1 is lower than the theoretically predicted value, triggering protection earlier. The far-end safety factor k2 refers to a safety margin factor based on the cumulative attenuation of track potential over long distances. The risk compensation coefficient for the effect setting is set to 1.15. The reason is that although the remote area is far from the substation, considering that the seasonal drift of soil resistivity may cause abnormal potential rise, the drift range can reach 10% to 20%, and the sparse remote monitoring points may lead to missed detections, a 15% risk compensation margin needs to be reserved. Therefore, k2 is set to 1.15 so that the remote action threshold TH2 is slightly higher than the theoretical prediction value, avoiding missed protection. The drainage advance amount ΔT refers to the time window in which the drainage cabinet acts in advance relative to the train arrival time, and is set to 3 to 5 minutes. The reason is that, according to the "Metro Design Code" GB Based on 50157 and stray current diffusion dynamics, it takes 2 to 3 minutes for the rail potential to stabilize to a safe level after the drain cabinet is activated. Considering the relay protection device's response delay of about 30 to 60 seconds and the control signal transmission delay of about 1 to 2 seconds, in order to ensure that the potential has dropped to a safe range when the obstructing train passes, the drain needs to be activated 3 to 5 minutes in advance. Therefore, ΔT is set to 3 to 5 minutes, which can be dynamically adjusted according to the train's running speed. 5 minutes is used for high speed and 3 minutes is used for low speed.

[0039] Specifically, the point-based integrated control module calculates the potential prediction deviation PV based on the real-time instantaneous rail potential value RT from the real-time rail potential monitoring data and the corrected potential prediction value VC, setting PV=|RT-VC| / VC. The potential prediction deviation PV is then compared with a first preset deviation threshold V1 and a second preset deviation threshold V2. Based on the comparison result, a dynamic control correction coefficient DC is assigned a value, wherein: When PV≤V1, set DC=1.0; When V1 < PV ≤ V2, set DC = 1.0 + 0.1 × (PV - V1) / (V2 - V1); When PV > V2, set DC = 1.15; The basic threshold values ​​for the rail potential limiter in the basic control threshold matrix TH are corrected according to the dynamic control correction coefficient DC. The corrected basic threshold value for the near-end rail potential limiter is set to TH1' = TH1 × DC, and the corrected basic threshold value for the far-end rail potential limiter is set to TH2' = TH2 × DC. Then, TH1 and TH2 in the basic control threshold matrix TH are replaced with the corrected basic threshold values ​​for the near-end rail potential limiter TH1' and TH2' for the far-end rail potential limiter, thus obtaining the first control threshold matrix TH'.

[0040] Specifically, the real-time rail potential instantaneous value RT refers to the instantaneous value of the rail-to-ground potential obtained in real time by rail potential sensors deployed at monitoring points along the rail line. In this embodiment, it is obtained through a distributed rail potential monitoring device, which includes gapless zinc oxide surge arrester-type rail potential sensors installed at the traction substation outlet, the middle of the section, and the end. The sampling frequency is set to 100 times per second to capture the transient peak value of the potential when the train passes. The first preset deviation threshold V1 refers to the upper limit of the allowable normal deviation range between the potential prediction model and the actual monitoring value, and is set to 0.10. The reason is that, considering that there is a measurement error of about 5% to 8% in the soil resistivity parameter in the rail-earth return network model, and the train... The vehicle load current fluctuates in real time. A deviation of less than 10% between the predicted and actual values ​​is within the normal system error range and does not require threshold correction. Therefore, V1 is set to 0.10, meaning that the prediction is considered accurate when the deviation is less than 10%. The second preset deviation threshold V2 is the critical value for determining that the potential prediction is significantly inaccurate, and it is set to 0.25. The reason is that when the deviation exceeds 25%, it indicates that the soil resistivity may change abruptly or the train position may deviate significantly. At this time, the original prediction model is unreliable and a large correction mechanism must be activated. By increasing the dynamic adjustment correction coefficient DC to 1.15, the protection threshold is corrected downward by 15% to ensure system safety in advance. Therefore, V2 is set to 0.25 as the boundary condition for activating strong correction measures.

[0041] Specifically, the point-based integrated control module calculates the time deviation degree DV based on the corrected discharge time TC and the preset safe time threshold TS, setting DV=|TC-TS| / TS. It then compares the time deviation degree DV with a first preset deviation threshold D1 and a second preset deviation threshold D2. Based on the comparison result, it determines the control risk priority PR, and optimizes the first control threshold matrix TH' based on the determination result. When DV≤D1, the control risk priority PR is determined to be low risk, the first control threshold matrix TH' is not optimized for risk, and the first control threshold matrix TH' is output as the final point control scheme. When D1 < DV ≤ D2, the risk priority PR of the regulation is determined to be medium risk. The first regulation threshold matrix TH' is optimized for risk. The modified near-end rail potential limiter action base threshold TH1' is lowered by the risk compensation coefficient RC. RC is set to 0.9-0.1×(DV-D1) / (D2-D1), and the optimized near-end rail potential limiter action base threshold TH1'' = TH1'×RC is obtained. The basic switching time TC of the drain cabinet is adjusted to TC-2min. TH1' in the first regulation threshold matrix TH' is replaced with the optimized near-end rail potential limiter action base threshold TH1'', resulting in the second regulation threshold matrix TH''. The second regulation threshold matrix TH'' is output as the final point regulation scheme. When DV > D2, the control risk priority PR is determined to be high risk. Risk optimization is performed on the first control threshold matrix TH', the rail potential limiter forced grounding mode is immediately triggered, the forced grounding threshold TH0 = 0.5 × VC is set, and the switching time TC of the drain cabinet foundation is adjusted to immediate action. An emergency control command is generated as the final point control scheme output.

[0042] Specifically, the preset safety time threshold TS refers to the ideal switching reference time of the drain cabinet determined based on the train timetable. It is typically set to the time ΔT before the train arrives at the monitoring section, and is typically 3 to 5 minutes before the train's arrival. The reason is that, according to the protection coordination requirements of the traction power supply system, the drain cabinet needs to fully establish a drain channel before the train enters the section, ensuring that the rail potential has dropped to a safe level the instant the train's pantograph is raised. This time window ensures both the stray current suppression effect and avoids premature switching leading to energy waste. The first preset deviation threshold D1 refers to the critical proportion for determining that the deviation of the drain time is within the normal tolerance range, and is typically 0.20. The reason is that, considering the ±5% to ±10% variation during train operation... The fluctuation is % and the rail potential prediction model has an error accumulation of about 10%. The total deviation after the two are combined is within 20%, which is within the controllable range. At this time, there is no need to adjust the protection threshold. Therefore, D1 is set to 0.20, that is, when the deviation is less than 20%, it is judged as low risk. The second preset deviation threshold D2 is the critical proportion for judging that the deviation of the drainage time has constituted a significant safety risk. The value is set to 0.50. The reason is that when the deviation exceeds 50%, it indicates that the actual switching time has seriously lagged behind the train arrival time. It may be that the drainage cabinet has not yet been activated or has just been put into operation and is not stable when the train passes. At this time, the rail potential may exceed the safety limit and the emergency protection mode must be activated. Therefore, D2 is set to 0.50 as the dividing line between activating forced grounding and immediate action.

[0043] Please see Figure 2As shown, this is a schematic diagram of the potential distribution prediction and generation module in this embodiment. The potential distribution prediction and generation module includes: The basic scheme generation unit is used to generate a basic potential distribution prediction scheme based on the basic track topology data, thereby obtaining the basic potential distribution prediction scheme. A spatial calibration unit is used to calibrate the generation process of the basic potential distribution prediction scheme based on real-time train position data to obtain a first potential distribution prediction scheme. The spatial calibration unit is connected to the basic scheme generation unit. A spatial optimization unit is used to optimize the spatial calibration process based on historical soil resistivity drift data to obtain a target potential distribution prediction scheme. The spatial optimization unit is connected to the spatial calibration unit.

[0044] Please see Figure 3 As shown, this is a structural diagram of the timing scheme generation module in this embodiment. The timing scheme generation module includes: The timing sequence basic scheme generation unit is used to generate the basic drainage action timing scheme based on the basic driving diagram data, so as to obtain the basic drainage action timing scheme; The timing calibration unit is used to calibrate the generation process of the drainage action timing scheme based on real-time track potential monitoring data to obtain the first drainage action timing scheme. The timing calibration unit is connected to the timing basic scheme generation unit. The timing optimization unit is used to optimize the timing calibration process based on the historical performance data of the equipment operation to obtain the target drainage operation timing scheme. The timing optimization unit is connected to the timing calibration unit.

[0045] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A rail point position regulation and suppression system based on multi-objective optimization, characterized in that, include: The data acquisition module is used to collect basic track topology data, real-time train location data, historical soil resistivity drift data, basic train operation diagram data, real-time track potential monitoring data, and historical equipment performance data. The potential distribution prediction generation module is used to generate a potential distribution prediction scheme based on basic track topology data, real-time train position data and historical soil resistivity drift data, and obtain the target potential distribution prediction scheme. The timing scheme generation module is used to generate the drainage action timing scheme based on the basic train operation diagram data, real-time rail potential monitoring data and equipment operation history performance data, so as to obtain the target drainage action timing scheme. The point-based integrated control module is used to coordinate the action threshold of the rail potential limiter and the switching time of the drainage cabinet according to the target potential distribution prediction scheme and the target drainage action timing scheme, so as to obtain the final point-based control scheme.

2. The rail point position regulation and suppression system based on multi-objective optimization according to claim 1, characterized in that, The potential distribution prediction generation module includes: The basic scheme generation unit is used to generate a basic potential distribution prediction scheme based on the basic track topology data, thereby obtaining the basic potential distribution prediction scheme. A spatial calibration unit is used to calibrate the generation process of the basic potential distribution prediction scheme based on real-time train position data to obtain a first potential distribution prediction scheme. The spatial optimization unit is used to optimize the spatial calibration process based on historical soil resistivity drift data to obtain a target potential distribution prediction scheme. The basic scheme generation unit generates a basic potential distribution prediction scheme based on the basic track topology data. When obtaining the basic potential distribution prediction scheme, based on the track segment length TL, track longitudinal resistance RR, track-to-ground transition conductance TG, and traction substation spacing TS in the basic track topology data, a rail-ground return current network model is established to calculate the predicted potential value VP and the predicted stray current density value CD at each monitoring point along the rail line, where: The predicted potential value VP = (PL × RR × TL) / (2 × TG × TS + TL), where PL is the traction load power; The predicted stray current density is CD = (PL × TG) / (2 × π × RL × TS), where RL is the soil resistivity. The predicted potential value VP and the predicted stray current density value CD are output as the basic potential distribution prediction scheme.

3. The rail position control and suppression system based on multi-objective optimization according to claim 2, characterized in that, When the spatial calibration unit calibrates the generation process of the basic potential distribution prediction scheme based on real-time train position data, it calculates the position offset PO based on the real-time train mileage marker RM in the real-time train position data and the preset train mileage marker PM in the basic potential distribution prediction scheme. PO is set to |RM-PM| / TS, where TS is the traction substation spacing. The position offset PO is compared with a first preset offset threshold P1 and a second preset offset threshold P2. Based on the comparison result, the spatial dynamic correction coefficient SC is assigned a value, where: When PO≤P1, set SC=1.0; When P1 < PO ≤ P2, set SC = 1.0 + 0.2 × (PO - P1) / (P2 - P1); When PO > P2, set SC = 1.25; The potential prediction value VP is corrected according to the spatial dynamic correction coefficient SC. The corrected potential prediction value VC = VP × SC is set. The value of the potential prediction value VP output by the basic potential distribution prediction scheme is replaced with the value of the corrected potential prediction value to obtain the first potential distribution prediction scheme.

4. The rail position control and suppression system based on multi-objective optimization according to claim 3, characterized in that, When the spatial optimization unit optimizes the spatial calibration process based on historical soil resistivity drift data, it calculates the spatial calibration convergence index CI based on the historical mean soil resistivity RH, historical soil moisture variation coefficient VH, and seasonal drift amplitude SA in the historical soil resistivity drift data. CI is set as (SA / RH) × (1 + VH). The spatial calibration convergence index CI is compared with a preset convergence threshold CT, and the optimization parameter OP of the spatial calibration process is assigned a value based on the comparison result. When CI≤CT, set OP=0.1; When CI > CT, set OP = 0.2; The calculation logic of the spatial dynamic correction coefficient SC is optimized according to the optimization parameter OP of the spatial calibration process. The optimized spatial correction coefficient SO is set as SC×[1+OP×(SA / RH)], and the value of the spatial dynamic correction coefficient SC is replaced with the value of the optimized spatial correction coefficient SO. The predicted value of the corrected potential is recalculated to obtain the target potential distribution prediction scheme.

5. The rail position control and suppression system based on multi-objective optimization according to claim 4, characterized in that, The timing scheme generation module includes: The timing sequence basic scheme generation unit is used to generate the basic drainage action timing scheme based on the basic driving diagram data, so as to obtain the basic drainage action timing scheme; The timing calibration unit is used to calibrate the generation process of the drainage action timing scheme based on real-time track potential monitoring data to obtain the first drainage action timing scheme. The timing optimization unit is used to optimize the timing calibration process based on the historical performance data of the equipment operation to obtain the target drainage operation timing scheme. In the basic timing scheme generation unit, the basic drainage operation timing scheme is generated based on the basic train schedule data. When obtaining the basic drainage operation timing scheme, the train departure timetable TD, train interval running time RT, and train stop time ST in the basic train schedule data are used to calculate the drainage cabinet switching time TC, rail potential limiter action threshold timing TH, and stray current collection network switch status timing SS through the drainage demand prediction algorithm, wherein: The start-up time of the drain cabinet is TC=TD+RT+0.5×ST-ΔT, where ΔT is the drain advance amount; The timing sequence of the rail potential limiter's operating threshold TH is dynamically set according to the distance between the train position and the traction substation. The basic operating threshold TH1 for the near-end rail potential limiter is 60V, and the basic operating threshold TH2 for the far-end rail potential limiter is 80V. The stray current collection network switch state timing SS is set according to the peak time of the predicted stray current density value CD, closing 5 minutes before the peak of CD and opening 5 minutes after the peak. The switching time TC of the drain cabinet, the action threshold timing TH of the rail potential limiter, and the switch status timing SS of the stray current collection network are output as the basic drain action timing scheme.

6. The rail position control and suppression system based on multi-objective optimization according to claim 5, characterized in that, When the timing calibration unit calibrates the generation process of the foundation drainage action timing scheme based on real-time rail potential monitoring data, it calculates the potential deviation PE based on the real-time rail potential RP in the real-time rail potential monitoring data and the predicted rail potential PP in the foundation drainage action timing scheme, setting PE=|RP-PP| / PP. The potential deviation PE is then compared with a first preset potential deviation threshold E1 and a second preset potential deviation threshold E2. Based on the comparison result, the timing dynamic correction coefficient TE is assigned a value, where: When PE≤E1, set TE=1.0; When E1 < PE ≤ E2, set TE = 1.0 + 0.15 × (PE - E1) / (E2 - E1); When PE > E2, set TE = 1.2; The timing of the drain cabinet switching TC is corrected according to the timing dynamic correction coefficient TE. The corrected drain time TM is set as TC-TE×ΔT, where ΔT is the drain advance amount. The value of the drain cabinet switching time TC in the basic drain action timing scheme is replaced with the value of the corrected drain time TM to obtain the first drain action timing scheme.

7. The rail position control and suppression system based on multi-objective optimization according to claim 6, characterized in that, When the timing optimization unit optimizes the timing calibration process based on the historical performance data of the equipment operation, it calculates the timing calibration performance index EI based on the average monthly number of drainage cabinet operations (MA), the operation success rate (SR), and the operation response delay (RD) in the historical performance data of the equipment operation. EI is set as (MA / 30) × (1-SR) × (RD / 100), where MA is the average monthly number of operations, SR is the operation success rate, and RD is the response delay in milliseconds. The timing calibration performance index EI is compared with a preset performance threshold ET, and the timing calibration process optimization parameter TO is assigned a value based on the comparison result, where: When EI≤ET, set TO=0.05; When EI > ET, set TO = 0.12; The calculation logic of the timing dynamic correction coefficient TE is optimized according to the timing calibration process optimization parameter TO. The optimized timing correction coefficient TI is set as TE×[1+TO×(1-SR)], and the value of the timing dynamic correction coefficient TE is replaced with the value of the optimized timing correction coefficient TI. The corrected drainage time is recalculated to obtain the target drainage action timing scheme.

8. The rail position control and suppression system based on multi-objective optimization according to claim 7, characterized in that, The point-based integrated control module generates a basic coordinated control scheme based on the corrected potential prediction value VC in the target potential distribution prediction scheme and the corrected drainage time TM in the target drainage action timing scheme, thereby obtaining a basic control threshold matrix. The basic control threshold matrix is ​​set as TH=[TH1, TH2, TC], where TH1 is the basic threshold for the near-end rail potential limiter action, set as TH1=VC×k1, k1 is the near-end safety factor; TH2 is the basic threshold for the far-end rail potential limiter action, set as TH2=VC×k2, k2 is the far-end safety factor; and TC is the basic switching time of the drainage cabinet, set as TC=TM-ΔT, ΔT is the drainage advance.

9. The rail position control and suppression system based on multi-objective optimization according to claim 8, characterized in that, The point-based integrated control module calculates the potential prediction deviation PV based on the real-time rail potential instantaneous value RT from the real-time rail potential monitoring data and the corrected potential prediction value VC. PV is set to |RT-VC| / VC. The potential prediction deviation PV is compared with a first preset deviation threshold V1 and a second preset deviation threshold V2. Based on the comparison result, a dynamic control correction coefficient DC is assigned a value, wherein: When PV≤V1, set DC=1.0; When V1 < PV ≤ V2, set DC = 1.0 + 0.1 × (PV - V1) / (V2 - V1); When PV > V2, set DC = 1.15; The basic threshold values ​​for the rail potential limiter in the basic control threshold matrix TH are corrected according to the dynamic control correction coefficient DC. The corrected basic threshold value for the near-end rail potential limiter is set to TH1' = TH1 × DC, and the corrected basic threshold value for the far-end rail potential limiter is set to TH2' = TH2 × DC. Then, TH1 and TH2 in the basic control threshold matrix TH are replaced with the corrected basic threshold values ​​for the near-end rail potential limiter TH1' and TH2' for the far-end rail potential limiter, thus obtaining the first control threshold matrix TH'.

10. The rail position control and suppression system based on multi-objective optimization according to claim 9, characterized in that, The point-based integrated control module calculates the time deviation degree DV based on the corrected discharge time TC and the preset safe time threshold TS, setting DV=|TC-TS| / TS. It compares the time deviation degree DV with the first preset deviation threshold D1 and the second preset deviation threshold D2, determines the control risk priority PR based on the comparison result, and optimizes the first control threshold matrix TH' based on the determination result. When DV≤D1, the control risk priority PR is determined to be low risk, the first control threshold matrix TH' is not optimized for risk, and the first control threshold matrix TH' is output as the final point control scheme. When D1 < DV ≤ D2, the risk priority PR of the regulation is determined to be medium risk. The first regulation threshold matrix TH' is optimized for risk. The modified near-end rail potential limiter action base threshold TH1' is lowered by the risk compensation coefficient RC. RC is set to 0.9-0.1×(DV-D1) / (D2-D1), and the optimized near-end rail potential limiter action base threshold TH1'' = TH1'×RC is obtained. The basic switching time TC of the drain cabinet is adjusted to TC-2min. TH1' in the first regulation threshold matrix TH' is replaced with the optimized near-end rail potential limiter action base threshold TH1'', resulting in the second regulation threshold matrix TH''. The second regulation threshold matrix TH'' is output as the final point regulation scheme. When DV > D2, the control risk priority PR is determined to be high risk. Risk optimization is performed on the first control threshold matrix TH', the rail potential limiter forced grounding mode is immediately triggered, the forced grounding threshold TH0 = 0.5 × VC is set, and the switching time TC of the drain cabinet foundation is adjusted to immediate action. An emergency control command is generated as the final point control scheme output.

Images