Rail transit braking energy recycling method and device

Abstract

The invention relates to a rail transit braking energy recycling method and device, and relates to the technical field of rail transit system energy conservation, and the method comprises the steps that a braking parameter set of a target rail transit vehicle is obtained; based on weather correction parameters, credibility verification is conducted on the braking parameter set of the target rail transit vehicle, braking parameter sets with the credibility smaller than a credibility threshold value are removed, and a credible braking parameter set is obtained; constructing an improved particle swarm optimization algorithm, and optimizing the braking energy recovery parameters of the target rail transit vehicle to obtain optimized energy recovery parameters; and the optimized energy recovery parameters are adopted, and rail transit braking energy is recycled. The method solves the problem that the traditional rail transit braking energy recovery scheme does not fully consider environmental factors, so that the reliability of the used data is insufficient, and the vehicle driving safety and the energy recovery efficiency are difficult to effectively balance.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit system energy saving, in particular to a rail transit braking energy recovery and utilization method and device. BACKGROUND

[0002] With the growing demand for energy saving and driving safety in the rail transit industry, the comprehensive performance and application stability of braking energy recovery technology as a key energy saving means have become the focus of the industry.

[0003] At present, the traditional rail transit braking energy recovery scheme does not fully consider the influence of environmental factors, resulting in insufficient reliability of braking-related data collected and used, which not only makes it difficult to accurately support parameter setting in the energy recovery link, but also cannot effectively balance vehicle driving safety and energy recovery efficiency, ultimately restricting the practical application value of braking energy recovery technology. SUMMARY

[0004] The present application provides a rail transit braking energy recovery and utilization method and device, which improves the status of insufficient data reliability caused by not fully considering environmental factors in the traditional rail transit braking energy recovery scheme, and the difficulty in effectively balancing vehicle driving safety and energy recovery efficiency.

[0005] The embodiments of the present application disclose the following technical solutions: In a first aspect, the embodiments of the present application provide a rail transit braking energy recovery and utilization method, which comprises: obtaining a braking parameter set of a target rail transit vehicle, wherein the braking parameter set comprises a plurality of braking parameter groups, each braking parameter group comprising braking intensity, braking duration and braking distance; based on a weather correction parameter, verifying the credibility of the braking parameter set of the target rail transit vehicle, eliminating braking parameter groups with credibility less than a credibility threshold, and obtaining a credible braking parameter set; constructing an improved particle swarm optimization algorithm to optimize braking energy recovery parameters of the target rail transit vehicle, and obtaining optimized energy recovery parameters, wherein the braking energy recovery parameters comprise charge and discharge thresholds, energy conversion module operating frequency and recovery power distribution ratio, and the improved particle swarm optimization algorithm comprises optimization stagnation judgment and dynamic adjustment; using the optimized energy recovery parameters to perform rail transit braking energy recovery and utilization.

[0006] In a second aspect, the embodiments of the present application provide a rail transit braking energy recovery and utilization device, which comprises: The brake parameter set acquisition module is configured to acquire a brake parameter set of the target rail transit vehicle, wherein the brake parameter set comprises a plurality of brake parameter groups, and each brake parameter group comprises brake intensity, brake duration and brake distance; The brake parameter set screening module is configured to perform credibility verification on the brake parameter set of the target rail transit vehicle based on the weather correction parameter, eliminate brake parameter groups with credibility less than a credibility threshold, and acquire a credible brake parameter set. The recovery parameter optimization module is configured to construct an improved particle swarm optimization algorithm, optimize brake energy recovery parameters of the target rail transit vehicle, and acquire optimized energy recovery parameters, wherein the brake energy recovery parameters comprise a charge-discharge threshold, an energy conversion module operating frequency and a recovery power distribution ratio, and the improved particle swarm optimization algorithm comprises optimization stagnation determination and dynamic adjustment. The energy recovery execution module is configured to perform rail transit brake energy recovery and utilization by using the optimized energy recovery parameters.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The present application provides a rail transit brake energy recovery and utilization method and device. By constructing an improved particle swarm optimization algorithm, acquiring an initial particle swarm population, calculating particle fitness, updating the optimal solution and dynamically adjusting the optimization process, and finally applying optimized energy recovery parameters, the rail transit vehicle brake safety and energy recovery efficiency are synergistically improved. First, based on the credible brake parameter set, the historical brake energy recovery parameters and the initial constraints of the particles are obtained, the real particle set is integrated and the random particle set is generated, and after merging and deduplication, the initial particle swarm population is obtained. Then, the influence of brake intensity, duration and distance corresponding to the particles is obtained, three types of fitness functions of safety, efficiency and matching are constructed, the comprehensive fitness calculation function is obtained by combining the road section and the driving demand weighting, and the fitness of each particle is calculated. Then, the particle swarm algorithm is used to iteratively update the individual and population optimal solution of the particles. If optimization stagnation occurs for N consecutive rounds, the particle generation space is adjusted by calculating the brake parameter comprehensive fluctuation degree, and the adjustment particles are generated to supplement the population. Finally, the optimized energy recovery parameters are applied to the vehicle brake energy recovery system, and the parameter execution strategy is dynamically adapted in combination with the real-time working condition, so that the brake energy is efficiently recovered and utilized.

[0008] The technical solutions of the present application solve the problems of poor adaptability, optimization easily falling into local optimum, and difficulty in balancing brake safety and recovery efficiency caused by parameter fixation in traditional rail transit brake energy recovery. The brake risks or energy waste caused by weather and road section changes are avoided. At the same time, the initialization method combining historical data and random particles ensures the diversity and reliability of the population, improves the accuracy of brake energy recovery parameter optimization and the stability in actual application, and provides implementable technical support for energy saving and consumption reduction of rail transit vehicles. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic flowchart illustrating a method for recovering and utilizing braking energy in rail transit, provided as an embodiment of this application; Figure 2 This is a schematic diagram of a rail transit braking energy recovery and utilization device provided in an embodiment of this application.

[0011] The components represented by each number in the attached diagram are explained below: Braking parameter set acquisition module 01, braking parameter set filtering module 02, recovery parameter optimization module 03, energy recovery execution module 04. Detailed Implementation

[0012] This application provides a method and apparatus for recovering and utilizing braking energy in rail transit, which addresses the technical problem that existing rail transit braking energy recovery schemes do not fully consider environmental factors, resulting in insufficient reliability of the data used, and thus making it difficult to effectively balance vehicle driving safety and energy recovery efficiency.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0015] In the description of the present application, the term "for example" is used to indicate "as an example, instance, or illustration." Any embodiment described as "for example" in this application is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, for the purpose of explanation, details are set forth. It is apparent to those skilled in the art that the present application can be practiced without the use of these specific details. In other instances, well-known structures and processes are not elaborated in order not to obscure the description of the present application with unnecessary details. Therefore, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0016] Embodiment one, as shown in the accompanying drawings Figure 1 The present application provides a rail transit braking energy recovery method, which comprises the following steps: S110: Obtain a braking parameter set of a target rail transit vehicle, wherein the braking parameter set comprises a plurality of braking parameter groups, and each braking parameter group comprises braking intensity, braking duration and braking distance; In the present application, in the scenario of performing braking energy recovery operation of the rail transit vehicle, in order to provide effective basic data support for braking parameter credibility verification and energy recovery parameter optimization, the braking parameter set of the target rail transit vehicle is obtained first, so as to ensure the reliability of the subsequent braking energy recovery process.

[0017] Specifically, the braking pressure of the target rail transit vehicle is obtained first, and the braking pressure is mapped by a preset mapping rule to obtain the braking intensity which can reflect the braking force of the vehicle.

[0018] Further, the trigger time when the braking instruction of the target rail transit vehicle is issued is captured, and the release time when the braking instruction stops acting is recorded, and the time difference between the two times is calculated, which is determined as the braking duration.

[0019] Further, the distance traveled by the target rail transit vehicle from the start of deceleration after receiving the braking instruction to the termination of the braking action and the stop of the running state is monitored, and the distance is determined as the braking distance.

[0020] Finally, the braking intensity, the braking duration and the braking distance are obtained by the above-mentioned manner respectively, each group of corresponding three parameters is integrated to form a braking parameter group, and a plurality of braking parameter groups jointly constitute the braking parameter set of the target rail transit vehicle, which provides a complete and accurate data basis for subsequent screening of credible parameters based on weather correction parameters and optimization of energy recovery parameters.

[0021] The step S110 in the method provided in the embodiment of the application comprises: obtaining the braking pressure of the target rail transit vehicle, performing mapping processing, and obtaining the braking intensity; obtaining the triggering moment and the release moment of the braking instruction of the target rail transit vehicle, calculating the time difference between the two moments as the braking duration; obtaining the distance from when the target rail transit vehicle receives the braking instruction to when the braking ends as the braking distance.

[0022] In the embodiment of the application, in order to avoid the efficiency of the subsequent energy recovery link being reduced and the safety risk being increased due to the missing or deviation of the basic braking data, the key parameters reflecting the real braking state of the vehicle are obtained through the step-by-step acquisition and calculation, so as to ensure the adaptability of the entire braking energy recovery process.

[0023] Specifically, the braking pressure of the target rail transit vehicle is first obtained. In actual operation, the pressure data in the braking pipeline during the braking process can be collected in real time through the pressure sensor of the vehicle braking system, and the collection frequency needs to match the vehicle braking response speed, so as to ensure that the complete change process of the braking pressure from starting to peak and then to release can be captured.

[0024] Further, the obtained braking pressure is subjected to mapping processing to obtain the braking intensity. Specifically, the mapping processing needs to be based on the fixed rules preset based on the characteristics of the vehicle braking system, for example, the maximum braking pressure and the minimum braking pressure of the target rail transit vehicle are first determined, and the actual collected braking pressure value is proportionally converted with the difference between the maximum and minimum braking pressures.

[0025] For example, if the maximum braking pressure of a vehicle is 10 bar and the minimum braking pressure is 1 bar, when the actually collected braking pressure is 6.4 bar, the mapping processing can obtain the braking intensity of 0.6 ((6.4-1) / (10-1)=0.6), which converts the abstract pressure data into a quantitative index directly reflecting the braking intensity, and ensures that the braking intensity under different working conditions is comparable.

[0026] Further, the triggering moment and the release moment of the braking instruction of the target rail transit vehicle are obtained. The triggering moment is the moment when the vehicle control system issues the braking instruction, which can be accurately extracted from the system instruction log. In addition, the release moment is the moment when the vehicle control system stops the braking instruction and the braking action begins to subside, which is also obtained from the instruction log.

[0027] After obtaining the two time points, the time difference between the two time points is calculated, and the time difference is the duration of the braking. For example, the braking instruction triggering time is 10:05:23.120, and the release time is 10:05:25.460. The calculated time difference of 0.34 seconds is the duration of this braking, and the time difference can directly reflect the length of the braking process, and provide a basis for subsequent energy recovery time.

[0028] Further, the distance of the target rail transit vehicle from receiving the braking instruction to the end of braking is obtained, that is, the braking distance. Specifically, data collection can be achieved through the vehicle's own GPS positioning system. The GPS positioning system can calculate the straight-line distance (or the actual distance combined with the travel trajectory) between the coordinates at the time of the braking instruction and the coordinates at the end of braking by continuously recording the vehicle position coordinates.

[0029] For example, the position corresponding to the GPS coordinates when the vehicle receives the braking instruction is point A, and the position when the braking is terminated is point B. The actual travel distance between points A and B is 15 meters, which is obtained by coordinate calculation. The 15 meters is the braking distance of this braking.

[0030] At the same time, during the entire step execution process, it is necessary to ensure that the collection and calculation of the braking intensity, the braking duration, and the braking distance are time-synchronized, that is, each set of parameters corresponds to the same braking process, to avoid mixing of parameters of different braking events and to ensure that the subsequently formed braking parameter set can truly reflect the complete state of a single braking.

[0031] S120: Based on the weather correction parameter, the braking parameter set of the target rail transit vehicle is verified for credibility, braking parameter groups with credibility less than a credibility threshold are removed, and a credible braking parameter set is obtained; In the embodiment of the application, in order to exclude invalid braking parameters affected by environmental factors and avoid unreliable data from causing deviation of the subsequent energy recovery parameter optimization direction, the braking parameter set needs to be verified for credibility and screened in combination with the weather correction parameter, so as to obtain a credible parameter set that can truly reflect the braking state of the vehicle and ensure the accuracy of the subsequent energy recovery optimization process.

[0032] Specifically, a multivariate verification model is first constructed, the inputs of the model are the braking intensity, the braking duration, and the weather correction parameter obtained based on the current weather in the braking parameter set, and the output is a verification braking distance matched with the input parameters, and the correlation between the braking parameters and the weather factors is established through the model.

[0033] Further, the obtained target rail transit vehicle braking parameter set is input into the multivariate verification model in groups, and the model outputs a corresponding verification braking distance for each group of braking parameters. Then, the deviation degree of the actual braking distance and the verification braking distance in each group of parameters is calculated, and the deviation degree directly reflects the deviation degree of the actual collected data from the theoretical reasonable range.

[0034] Further, the deviation degree is converted into a reliability of measuring the reliability of the parameter by using 1 minus the deviation degree. The higher the reliability is, the more the group of braking parameters is consistent with the reasonable braking state under the weather condition. Meanwhile, the braking parameter groups with a reliability lower than a reliability threshold are deleted from the braking parameter set, and the remaining parameter groups are integrated to form a reliable braking parameter set.

[0035] This step can effectively exclude the parameters that do not meet the reasonable braking rule under the weather condition by combining the weather correction parameter and the multivariate verification model, and provide high-quality data support for subsequent construction of an improved particle swarm optimization algorithm and optimization of the braking energy recovery parameter.

[0036] The step S120 in the method provided in the embodiments of the present application includes: A multivariate verification model is constructed, wherein the input of the multivariate verification model is the braking intensity, the braking duration and the weather correction parameter, and the output result is a verification braking distance, wherein the weather correction parameter is obtained based on the current weather; The braking parameter set is input into the multivariate verification model, a plurality of verification braking distances are obtained, and the deviation degree of the plurality of verification braking distances and the plurality of braking distances is calculated; The reliability is obtained by using 1 minus the deviation degree; The braking parameter groups with a reliability less than a reliability threshold are deleted from the braking parameter set to obtain a reliable braking parameter set.

[0037] In the embodiments of the present application, in order to avoid that unreliable data deviates the subsequent energy recovery parameter optimization direction, and further affects the energy recovery efficiency and the vehicle driving safety, the weather correction parameter needs to be combined to carry out systematic reliability verification and screening on the braking parameter set, so as to obtain a reliable parameter set that can truly reflect the vehicle braking state under different weather conditions.

[0038] Specifically, a multivariate verification model is first constructed. The core function of the multivariate verification model is to realize the braking data rationality judgment through the correlation of multiple parameters, and the input items are set as the braking intensity, the braking duration in the braking parameter set, and the weather correction parameter obtained based on the current weather.

[0039] The weather correction parameter needs to be set in advance through the mapping relationship between the weather type and the braking influence law, for example, the correction value corresponding to the dry road braking resistance characteristic in sunny weather, the correction value corresponding to the wet road braking resistance reduction in rainy weather, and the correction value corresponding to the significant reduction of the road friction coefficient in snowy weather, so as to ensure that the parameter can accurately quantify the influence of weather on the braking process.

[0040] In addition, the output item of the multivariate verification model is the verification braking distance, and the correlation between the braking intensity, the braking duration, the weather correction parameter and the reasonable braking distance is established through the multiple linear regression algorithm, so as to realize the rationality verification of the actually collected braking distance and play the role of mutual verification of multiple parameters.

[0041] The multiple linear regression algorithm determines the weight coefficients and constant terms corresponding to the braking intensity, the braking duration and the weather correction parameter by fitting the historical braking data of the target rail transit vehicle under different weather and different braking conditions, forms a calculation model of "verification braking distance = α × braking intensity + β × braking duration + γ × weather correction parameter + δ" (α, β, γ are weight coefficients of each input parameter, and δ is a constant term), and quantitatively correlates multiple parameters with the reasonable braking distance. By comparing the verification braking distance output by the model with the actually collected braking distance, it is determined whether the actual data conforms to the reasonable law under the current weather and braking condition.

[0042] Further, after the multivariate verification model is constructed, the obtained braking parameter set of the target rail transit vehicle is input into the multivariate verification model in groups. Specifically, for each group of input braking intensity, braking duration and corresponding weather correction parameter, the multivariate verification model will output a verification braking distance conforming to the current condition according to the preset algorithm logic.

[0043] Further, the deviation degree between the actually collected braking distance and the verification braking distance output by the model in each group of parameters is calculated. The deviation degree is determined based on the ratio of the numerical difference between the two and the verification braking distance, for example, the actual braking distance in a certain group of parameters is 18 meters, the verification braking distance output by the model is 20 meters, the absolute value of the difference between the two is 2 meters, and the deviation degree is 2 meters / 20 meters = 0.1. The smaller the deviation degree, the higher the degree of fit between the actually collected data and the theoretical reasonable range.

[0044] Further, the calculation method of 1 minus the deviation degree is adopted to convert the deviation degree into the reliability for measuring the reliability of the parameter. Continuing the above example, when the deviation degree is 0.1, the reliability is 1-0.1 = 0.9. The reliability value range is between 0-1, and the closer the value is to 1, the more matched the group of braking parameters is with the reasonable braking law under the current weather condition, and the stronger the data reliability is.

[0045] Further, according to the braking safety requirement of the rail transit vehicle and the accuracy requirement of the energy recovery parameter optimization, a credibility threshold is set. For example, combined with long-term operation data and industry standards, the credibility threshold is set to 0.7, and the parameter groups with a credibility lower than 0.7 are deleted from the braking parameter set. Among them, such parameter groups mostly belong to unreasonable situations, such as when the weather is sunny (dry road surface), the braking intensity has reached a high level but the braking duration is still far beyond the reasonable range under the same working condition, or when it is rainy, the braking intensity is low but the braking distance is abnormally short. If such data is used for subsequent optimization, it will seriously interfere with the accuracy of parameter setting.

[0046] At the same time, during the entire verification and screening process, it is necessary to ensure that the acquisition of weather correction parameters and the collection of braking parameters are strictly time-synchronized, that is, the weather correction parameters corresponding to each set of braking parameters are the values mapped by the real-time weather at the time of the braking behavior, so as to avoid distortion of the verification result due to time mismatch between weather and parameters.

[0047] Finally, through the above screening process combining weather correction parameters and multivariate verification models, the remaining braking parameter groups with required credibility are integrated to form a complete set of credible braking parameters. The credible braking parameter set not only excludes invalid interference data, but also reflects the braking characteristic differences under different weather conditions, providing a data basis that fits actual working conditions for subsequent optimization of charging and discharging thresholds, energy conversion module working frequencies and other recovery parameters.

[0048] S130: An improved particle swarm optimization algorithm is constructed to optimize the braking energy recovery parameters of the target rail transit vehicle, and an optimized energy recovery parameter is obtained, wherein the braking energy recovery parameters include a charging and discharging threshold, an energy conversion module working frequency and a recovery power distribution ratio, and the improved particle swarm optimization algorithm includes optimization stagnation judgment and dynamic adjustment. In the embodiments of the present application, in order to enable the three types of recovery parameters, i.e., the charging and discharging threshold, the energy conversion module working frequency and the recovery power distribution ratio, to adapt to the braking characteristics reflected by the credible braking parameters, and to avoid the problem that the traditional optimization algorithm is prone to local optimization or optimization stagnation, an improved particle swarm optimization algorithm incorporating optimization stagnation judgment and dynamic adjustment is constructed to optimize the recovery parameters, so as to maximize the braking energy recovery utilization rate while ensuring the braking safety of the vehicle.

[0049] In the method provided by the embodiments of the present application, the improved particle swarm optimization algorithm is constructed, including: An initial particle swarm population is obtained, wherein the initial particle swarm population includes a plurality of particles, and each particle represents a set of braking energy recovery parameters; A comprehensive fitness calculation function is obtained to calculate the fitness of the plurality of particles; The individual optimal solution and the population optimal solution of the particles are updated, particle swarm optimization is performed, and optimized energy recovery parameters are obtained, wherein the particle swarm optimization process further includes optimization stagnation determination and dynamic adjustment.

[0050] Specifically, first, a real particle set is formed by extracting historical recovery parameters, a random particle set is generated according to constraints, and the initialization particle swarm population is obtained by merging and removing duplicates, so as to balance the reliability and diversity of the population. The initialization particle swarm population includes a plurality of particles, and each particle represents a set of brake energy recovery parameters.

[0051] In the method provided in the embodiments of the present application, the initialization particle swarm population is obtained, including: Based on the set of trusted braking parameters, a set of historical brake energy recovery parameters is obtained, and a particle initial constraint is obtained; The historical brake energy recovery parameters in the set of historical brake energy recovery parameters are integrated into real particles to obtain a set of real particles; Based on the particle initial constraint, a set of random particles is obtained, wherein the set of random particles includes a plurality of random particles, and the number of random particles is obtained based on the number of real particles; The set of random particles and the set of real particles are merged, and duplicate particles are removed to obtain an initialization particle population.

[0052] Specifically, first, a set of historical brake energy recovery parameters is obtained based on a set of trusted braking parameters, and a particle initial constraint is determined. The set of trusted braking parameters is effective data screened by weather correction parameters, and includes a plurality of sets of braking intensity, braking duration and braking distance that can reflect the real braking state. The historical brake energy recovery parameters corresponding to these braking parameters need to be extracted from the database to form the set of historical brake energy recovery parameters.

[0053] At the same time, the particle initial constraint is determined according to the maximum and minimum values of each parameter in the set of historical brake energy recovery parameters. For example, the maximum value of the historical charge and discharge threshold is 3.2V, and the minimum value is 2.5V. The maximum value of the energy conversion module operating frequency is 80Hz, and the minimum value is 50Hz. The maximum value of the recovery power distribution ratio is 0.8, and the minimum value is 0.3. The constraint ranges of the particles in the three parameter dimensions are set to 2.5V-3.2V, 50Hz-80Hz and 0.3-0.8, respectively, to ensure that the particles generated subsequently meet the safety and performance boundaries of the actual operation of the vehicle.

[0054] Further, the parameters in the set of historical brake energy recovery parameters are integrated into real particles to form a set of real particles. Each set of complete historical brake energy recovery parameters corresponds to a real particle, and each real particle can match a set of braking conditions reflected by the trusted braking parameters.

[0055] Exemplarily, a certain real particle corresponds to a historical energy recovery parameter that has caused the vehicle to produce a braking effect of a braking intensity of 0.6, a braking duration of 1.2 seconds, and a braking distance of 18 meters, and the braking effect is verified as effective after weather correction, so that the real particle set can provide basis data support for subsequent optimization that conforms to actual working conditions, and avoid deviation of the optimization direction from actual demand.

[0056] Further, a set of random particles is obtained based on the initial constraints of the particles, and the number of the random particles needs to be determined according to the number of the real particles. For example, if there are 180 real particles, 40 random particles need to be generated to ensure that the population size is sufficient to cover more potential optimization directions and to reserve space for subsequent de-duplication screening.

[0057] In the formula, the charge and discharge threshold of each random particle is randomly selected in the range of 2.5V-3.2V, the working frequency is randomly selected in the range of 50Hz-80Hz, and the power allocation ratio is randomly selected in the range of 0.3-0.8, and it is necessary to ensure that the random selection process is uniformly distributed to avoid the concentration of particles in a certain parameter interval, so as to improve the diversity of the population.

[0058] Finally, the set of random particles is combined with the set of real particles, and the duplicate particles are removed. Specifically, during the combination process, the three parameters (charge and discharge threshold, working frequency, and power allocation ratio) of each particle in the two sets of particle groups are compared one by one, and if the three parameters of a certain random particle are completely consistent with those of a certain real particle, one of the particles is removed to avoid the occupation of computing resources or interference with the optimization process by duplicate data.

[0059] Exemplarily, after the combination, a total of 220 particles are obtained, and after de-duplication, 200 particles remain, which together constitute the initial particle group population.

[0060] Further, after obtaining the initial particle group population, the braking parameters influenced by the particles are obtained, three types of fitness calculation functions are constructed and weighted, a comprehensive fitness calculation function is obtained, and the fitness of the particles is calculated to determine the advantages and disadvantages of the parameters and provide a basis for subsequent optimization.

[0061] In the method provided by the embodiment of the application, the comprehensive fitness calculation function is obtained, and the fitness of the plurality of particles is calculated, including: obtaining the influenced braking intensity, the influenced braking duration, and the influenced braking distance after the braking energy recovery parameter of the plurality of particles is influenced; constructing a plurality of fitness calculation functions, wherein the plurality of fitness calculation functions include a safety fitness calculation function, an efficiency fitness calculation function, and a matching fitness calculation function; performing weighted calculation on the safety fitness, the efficiency fitness, and the matching fitness to obtain a comprehensive fitness calculation function. Adopt the comprehensive fitness calculation function, the particle in the initialization particle population is calculated, and a plurality of particle fitnesses are obtained.

[0062] Specifically, first, the influence braking intensity, influence braking duration and influence braking distance of a plurality of particles after the influence of the braking energy recovery parameters are obtained. Each particle represents a set of braking energy recovery parameters, and the braking effect data generated after the set of parameters acts on the target rail transit vehicle is obtained through actual working condition test.

[0063] Illustratively, the braking energy recovery parameters corresponding to a particle are the charging and discharging threshold 2.9V, the energy conversion module working frequency 70Hz, and the recovery power distribution ratio 0.65. Through actual working condition test, it is obtained that the influence braking intensity of the vehicle under the braking energy recovery parameters is 0.55, the influence braking duration is 1.1 seconds, and the influence braking distance is 17.5 meters.

[0064] In addition, another particle braking energy recovery parameter is the charging and discharging threshold 2.6V, the working frequency 55Hz, and the power distribution ratio 0.4, corresponding to the influence braking intensity 0.7, the influence duration 0.9 seconds, and the influence distance 19 meters.

[0065] Further, a plurality of fitness calculation functions are constructed, including a safety fitness calculation function, an efficiency fitness calculation function and a matching fitness calculation function.

[0066] Among them, the safety fitness calculation function takes the influence braking intensity as the core basis, adopts the calculation method of "1-influence braking intensity", and the greater the influence braking intensity, the more extreme the braking force of the vehicle during braking, and the higher the safety risk. For example, when the influence braking intensity is 0.7, the safety fitness is 0.3; when the influence braking intensity is 0.5, the safety fitness is 0.5, so as to intuitively reflect the safety level of the parameters through the numerical size.

[0067] In addition, the efficiency fitness calculation function is constructed based on the influence braking duration, and adopts the formula "benchmark braking duration / influence braking duration". Among them, the benchmark braking duration is the reasonable braking time of the target vehicle under standard working condition (for example, 1.0 seconds), if the influence braking duration of a particle is 0.9 seconds, the efficiency fitness is 1.0 / 0.9≈1.11; if the influence duration is 1.2 seconds, the efficiency fitness is 1.0 / 1.2≈0.83, the shorter the duration, the higher the efficiency fitness, representing the lower time cost and higher efficiency of the energy recovery process.

[0068] In addition, the matching fitness calculation function needs to combine the relationship between the influence braking distance and the target braking distance, and the target braking distance is a reasonable braking distance (for example, 18 meters) set according to the driving section of the vehicle. If the influence braking distance is less than or equal to the target braking distance, it is calculated according to "1-(1-influence braking distance / target braking distance) * 0.5". For example, when the influence distance is 17.5 meters, the matching fitness is 1-(1-17.5 / 18) * 0.5 ≈ 0.97.

[0069] On the contrary, if the influence distance exceeds the target distance, such as 19 meters, it is calculated according to "target braking distance / influence braking distance", and the matching fitness is 18 / 19 ≈ 0.95. In this way, the adaptability of the braking distance to the road section requirement is ensured, and the safety hidden danger caused by too short or too long is avoided.

[0070] Further, the safety fitness, the efficiency fitness and the matching fitness are weighted and calculated to obtain a comprehensive fitness calculation function, so as to integrate the multi-dimensional fitness indexes into a unified evaluation standard.

[0071] In the method provided by the embodiments of the present application, the safety fitness, the efficiency fitness and the matching fitness are weighted and calculated to obtain a comprehensive fitness calculation function, which includes: The safety fitness is obtained based on the influence braking intensity of the particle, the efficiency fitness is obtained based on the influence braking duration of the particle, and the matching fitness is obtained based on the influence braking distance of the particle. Based on the driving section and driving requirement of the target rail transit vehicle, a safety weight, an efficiency weight and a matching weight are obtained, and the safety fitness, the efficiency fitness and the matching fitness are weighted and calculated to obtain a comprehensive fitness calculation function.

[0072] Firstly, the correspondence between the three types of fitness and the influence braking parameters of the particle is determined. Specifically, the safety fitness is directly determined by the influence braking intensity corresponding to the particle. The greater the influence braking intensity, the closer the braking force of the vehicle during braking to the limit, the higher the safety risk, and the lower the safety fitness value. For example, the influence braking intensity of a certain particle is 0.7, and the corresponding safety fitness is 0.3. The safety fitness value directly reflects the parameter's ability to ensure braking safety.

[0073] In addition, the efficiency fitness is determined by the influence braking duration corresponding to the particle. The shorter the influence braking duration, the lower the time cost of the energy recovery process, the higher the energy recovery efficiency, and the higher the efficiency fitness value. For example, the influence braking duration of a certain particle is 0.9 seconds, and the reference braking duration is 1.0 second. The corresponding efficiency fitness is about 1.11, which quantifies the energy recovery efficiency level of the parameter.

[0074] In addition, the matching fitness is determined by the influence braking distance corresponding to the particle. The difference between the influence braking distance and the target braking distance is compared. If the influence distance is within a reasonable range (≤ target braking distance), the matching fitness value is higher, and if it exceeds the reasonable range, the value decreases. For example, the target braking distance is 18 meters, and the influence braking distance of a particle is 17 meters. The corresponding matching fitness is about 0.97. If the influence distance is 20 meters, the corresponding matching fitness is 0.9. The matching fitness value reflects the adaptability of the parameters to the braking distance requirement.

[0075] Further, based on the driving section and driving demand of the target rail transit vehicle, the safety weight, efficiency weight and matching weight are determined. Specifically, the weight distribution needs to be combined with the section characteristics and operation priority. If the vehicle mainly drives on a mountain railway section, such section has large slope and dense curves, and braking safety is the primary requirement. The safety weight proportion needs to be increased, for example, the safety weight is set to 0.4, the efficiency weight is set to 0.3, and the matching weight is set to 0.3, to ensure that the optimized parameters prioritize braking safety.

[0076] In addition, if the vehicle drives on a flat urban subway line, such section has smooth road surface and stable braking conditions, and the energy recovery efficiency can be appropriately emphasized. At this time, the efficiency weight is increased to 0.4, and the safety weight and the matching weight are each set to 0.3, so that the optimization direction is more in line with the energy saving demand.

[0077] In addition, if the vehicle drives on a suburban mixed section (both flat section and small slope section), balanced weight distribution is adopted, such as setting the safety, efficiency and matching weights to 0.333, to balance the multi-dimensional requirements.

[0078] At the same time, in the weight determination process, the historical braking accident statistical data of the section, the vehicle energy consumption report and other actual operation data need to be referred to, to avoid the disconnection between the weight and the actual demand caused by subjective setting.

[0079] Further, the safety fitness, efficiency fitness and matching fitness are weighted and calculated to construct a comprehensive fitness calculation function. Taking the mountain section weight (safety weight 0.4, efficiency weight 0.3, matching weight 0.3) as an example, if the safety fitness of a particle is 0.4, the efficiency fitness is 1.05, and the matching fitness is 0.96, the corresponding comprehensive fitness = 0.4 x 0.4 + 1.05 x 0.3 + 0.96 x 0.3 = 0.763. If another particle has a comprehensive fitness of 0.82 under the weight of the section, it means that the latter corresponds to the recovery parameters that are more in line with the safety and efficiency balance requirements of the mountain section.

[0080] Through the above weighting mode, the three types of scattered fitness indicators are integrated into a single comprehensive fitness value, which not only retains the evaluation logic of each dimension, but also realizes the unified quantification of multiple needs. Subsequently, the pros and cons of different particle recovery parameters can be quickly judged by comparing the comprehensive fitness values.

[0081] Further, the constructed comprehensive fitness calculation function is used to calculate the fitness of the particles in the initialized particle population to obtain multiple particle fitnesses.

[0082] Specifically, first, all particles in the initialized particle population are traversed. For each particle, its corresponding safety fitness, efficiency fitness, and matching fitness are extracted.

[0083] For example, a particle corresponds to a charge and discharge threshold of 2.8V, a working frequency of 65Hz, and a power distribution ratio of 0.6, with a safety fitness of 0.42, an efficiency fitness of 1.08, and a matching fitness of 0.95. Another particle corresponds to a charge and discharge threshold of 2.7V, a working frequency of 60Hz, and a power distribution ratio of 0.55, with a corresponding safety fitness of 0.38, an efficiency fitness of 0.98, and a matching fitness of 0.93.

[0084] Further, the driving section weight of the particle adaptation is confirmed. If both particles adapt to urban flat subway lines (efficiency weight 0.4, safety weight 0.3, and matching weight 0.3), the fitness is substituted into the function to calculate: the first particle comprehensive fitness = 0.42 x 0.3 + 1.08 x 0.4 + 0.95 x 0.3 = 0.843; the second particle comprehensive fitness = 0.38 x 0.3 + 0.98 x 0.4 + 0.93 x 0.3 = 0.785.

[0085] In addition, after all particles are calculated, the particle recovery parameter and the corresponding relationship of the comprehensive fitness can be sorted and archived, and the adaptation section type of each particle is labeled to form a particle fitness list, which provides data support for subsequent updating of individual optimal solution and screening of population optimal solution.

[0086] Further, according to the obtained multiple particle fitnesses, the individual optimal solution and the population optimal solution of the particle are updated, and the particle swarm optimization is performed to obtain better brake energy recovery parameters. The particle swarm optimization process further includes optimization stagnation determination and dynamic adjustment.

[0087] In the method provided by the embodiments of the present application, the individual optimal solution and the population optimal solution of the particle are updated, and the particle swarm optimization is performed, wherein the particle swarm optimization process further includes optimization stagnation determination and dynamic adjustment, which includes: The particle swarm algorithm is used to iteratively update the multiple particles to obtain the individual optimal solution and the population optimal solution of the particles; If the fitness improvement amplitude of the global optimal solution is less than the fitness improvement threshold, or the change amplitude of any one of the brake intensity improvement amplitude, brake duration, and brake distance is less than the brake change threshold in the continuous N iterations, it is determined that the optimization is stagnant, and the dynamic adjustment scheme is used to adjust the particle swarm, wherein N is obtained based on the weather correction parameter.

[0088] Specifically, first, the particle swarm algorithm is used to iteratively update a plurality of particles to obtain a particle individual optimal solution and a global optimal solution. In each iteration, each particle determines the individual optimal solution according to the parameter combination corresponding to the optimal fitness (i.e., the individual optimal fitness) in the historical iteration process.

[0089] At the same time, the population determines the global optimal solution according to the parameter combination corresponding to the current highest fitness (i.e., the population optimal fitness) of all particles. For example, the comprehensive fitness of a particle in the first iteration is 0.75, the fitness is improved to 0.82 after adjusting the parameters in the second iteration, and the fitness does not exceed 0.82 in the third iteration. The individual optimal solution of the particle is the recovery parameter corresponding to the second iteration. If the comprehensive fitness of a particle in the population reaches 0.88, which is the current highest, in the fifth iteration, the parameter corresponding to the particle is the global optimal solution.

[0090] During the iterative update process, the particle adjusts its parameter value according to the preset speed formula. The speed update needs to refer to the deviation of the individual optimal solution and the global optimal solution. For example, if the current parameter of a particle deviates less from the individual optimal solution and deviates more from the global optimal solution, the speed weight of approaching the global optimal solution will be increased, and the parameter difference will be gradually reduced, to ensure that each iteration can move in a better direction.

[0091] Further, the number of continuous iterations N is determined based on the weather correction parameter, which is used to determine whether the optimization is stagnant. The influence of different weather conditions on vehicle braking characteristics is significantly different, and N needs to be adjusted to adapt to weather changes. Specifically, if the current weather is rainy, the road friction coefficient is low, and the braking parameters (brake intensity, duration, and distance) fluctuate greatly, the optimization state needs to be monitored more frequently to avoid optimization deviation caused by unstable parameters. At this time, N is set to 5; if it is sunny, the road is dry, and the braking characteristics are stable, the monitoring period can be appropriately extended, and N is set to 8; if it is snowy, the road friction coefficient is extremely low, and the braking risk is high, N needs to be further reduced to 3 to ensure that the stagnation problem can be found in time.

[0092] Further, after each iteration, the fitness improvement amplitude of the global optimal solution is compared with the preset fitness improvement threshold, and whether the change amplitudes of the brake intensity, brake duration, and brake distance are less than the brake change threshold is checked.

[0093] Exemplarily, if the fitness improvement threshold is 0.02, the braking change threshold is 0.01, and the fitness of the global optimal solution is only improved from 0.88 to 0.89 in 5 consecutive iterations, the fitness improvement amplitude is 0.01 < 0.02; or the braking intensity is only changed from 0.52 to 0.525, the braking change amplitude is 0.005 < 0.01, and any condition is satisfied, which is determined as optimization stagnation.

[0094] Further, after determining the optimization stagnation, a dynamic adjustment scheme needs to be used to adjust the particle swarm, so as to break the solidification state of the current parameter iteration, introduce new parameter diversity, and ensure that the subsequent iteration can continue to approach the more optimal braking energy recovery parameter, and adapt to the braking demand under different weather and road sections.

[0095] In the method provided by the embodiments of the present application, the dynamic adjustment scheme is used to adjust the particle swarm, including: Based on the set of braking parameters, a comprehensive fluctuation degree of braking parameters is obtained, wherein the comprehensive fluctuation degree of braking parameters is obtained based on the variance and mean value of the braking intensity, the braking duration and the braking distance; Based on the comprehensive fluctuation degree of braking parameters and the initial constraint of the particle, an adjustment space is obtained; Based on the adjustment space, a plurality of adjustment particles are generated and added to the initial particle population for iterative optimization.

[0096] Specifically, first, the comprehensive fluctuation degree of braking parameters is obtained based on the set of braking parameters. The set of braking parameters includes a plurality of groups of braking intensity, braking duration and braking distance verified by credibility, and the variance / mean value of the three types of parameters needs to be calculated respectively, and then the average value of the three is taken as the comprehensive fluctuation degree of braking parameters.

[0097] The comprehensive fluctuation degree of braking parameters can directly reflect the stability of the braking parameters. The greater the fluctuation degree, the more significant the influence of external factors on the braking working condition, and the greater the range of adjustment particle parameters needs to be adjusted to adapt to the fluctuation. The smaller the fluctuation degree, the more stable the braking working condition, and the adjustment range can be appropriately reduced.

[0098] Exemplarily, in a certain set of braking parameters, the variance of the braking intensity is 0.09, the mean value is 0.6, the corresponding fluctuation degree is 0.09 / 0.6=0.15; the variance of the braking duration is 0.04, the mean value is 1.0, the fluctuation degree is 0.04 / 1.0=0.04; the variance of the braking distance is 2.25, the mean value is 15, the fluctuation degree is 2.25 / 15=0.15, and the average of the three is obtained. The comprehensive fluctuation degree is (0.15+0.04+0.15) / 3≈0.113, indicating that there is a certain fluctuation in the current braking parameters, and the diversity needs to be introduced by adjusting the particle range.

[0099] Further, based on the comprehensive fluctuation degree of the braking parameter and the initial constraint of the particle, an adjustment space is obtained to adapt the particle generation range to the actual fluctuation of the braking parameter. The initial constraint of the particle is a range determined based on the maximum and minimum values of the historical recovery parameter, and the adjustment space needs to expand or reduce the initial constraint according to the size of the comprehensive fluctuation degree.

[0100] Specifically, if the comprehensive fluctuation degree is large, it means that the braking parameter stability is poor, and the constraint range needs to be widened to cover more potential adaptive parameters. If the comprehensive fluctuation degree is small, it means that the braking parameter is stable, and only a small amount of fine-tuning of the initial constraint is needed, such as keeping the charge and discharge threshold at 2.5V-3.2V and adjusting the working frequency to 48Hz-82Hz, to avoid excessive expansion and cause the parameter to deviate from the reasonable operation interval.

[0101] For example, based on the above-mentioned comprehensive fluctuation degree of 0.113, the initial constraint of the charge and discharge threshold 2.5V-3.2V can be adjusted to 2.4V-3.3V, the working frequency 50Hz-80Hz can be adjusted to 47Hz-83Hz, and the recovery power distribution ratio 0.3-0.8 can be adjusted to 0.28-0.82, to form a new adjustment space.

[0102] Further, based on the new adjustment space, a plurality of adjustment particles are generated and added to the initial particle population. The number of adjustment particles needs to be determined in combination with the current population size, for example, the original initial particle population has 200 particles, in order to ensure that the new particles can effectively enrich the population diversity, and at the same time avoid increasing the calculation burden due to excessive number, 80 adjustment particles can be generated.

[0103] Similarly, the parameters of each adjustment particle need to be randomly selected within the adjustment space, and the selection needs to be uniformly distributed to avoid excessive concentration of particles in a certain parameter interval. For example, in the charge and discharge threshold adjustment space of 2.4V-3.3V, 2.45V, 2.8V, 3.25V, etc. are randomly generated; in the working frequency adjustment space of 47Hz-83Hz, 49Hz, 65Hz, 81Hz, etc. are randomly generated; in the power distribution ratio adjustment space of 0.28-0.82, 0.3, 0.55, 0.8, etc. are randomly generated. Each adjustment particle contains a complete set of "charge and discharge threshold-working frequency-power distribution ratio" parameters.

[0104] Further, after generating the adjustment particles, they are all added to the current initial particle population to form a new population and restart the iterative optimization.

[0105] Specifically, the new population needs to be repeated particle verification during the adding process. If the three parameters of an adjustment particle are completely consistent with those of a particle in the original population, the repeated particle is removed to avoid occupying redundant computing resources. For example, after adding 80 adjustment particles, 60 effective particles are left after deduplication, and the size of the new population becomes 200+60=260 particles. These particles not only retain the parameters in the original population that fit the historical working conditions, but also add new parameters that adapt to the fluctuation of the braking parameters, providing a wider exploration space for subsequent iterations.

[0106] Finally, through multiple rounds of iteration optimization of the new population, in each round of iteration, the particles adjust the parameters according to the individual optimal solution and the population optimal solution, while continuously monitoring whether optimization stagnation occurs again and timely dynamic adjustment, until the comprehensive fitness of the global optimal solution is stable at a high level for multiple rounds in a row, and the corresponding influence braking intensity, influence braking duration, and influence braking distance all meet the demand of the driving section of the target rail transit vehicle and weather adaptability.

[0107] At this time, the "charge and discharge threshold-energy conversion module operating frequency-recovery power distribution ratio" parameter combination corresponding to the global optimal solution is the final optimized energy recovery parameter.

[0108] For example, for the rain condition in the mountain area, the finally determined optimized energy recovery parameter may be "charge and discharge threshold 2.8V, energy conversion module operating frequency 81Hz, and recovery power distribution ratio 0.8". This parameter can not only guarantee braking safety, but also achieve high energy recovery efficiency, and accurately adapt to the braking and energy saving demand in actual operation.

[0109] S140: Adopting the optimized energy recovery parameter, the rail transit braking energy recovery utilization is carried out.

[0110] In the embodiments of the present application, in order to apply the charge and discharge threshold, energy conversion module operating frequency, and recovery power distribution ratio obtained through the previous optimization to the actual braking process, these optimized parameters need to be applied to the braking energy recovery system of the vehicle, so that the parameters and the vehicle braking action cooperate to ensure that the vehicle can stably recover energy while meeting the safety braking requirements when braking.

[0111] Specifically, before applying the optimized energy recovery parameter, it is necessary to confirm the adaptability of the parameter to the vehicle braking energy recovery related equipment to avoid the parameter exceeding the operating range of the equipment. For example, the voltage range of the energy storage device (power battery) of the urban subway vehicle is 2.5V-3.3V, and the optimized charge and discharge threshold needs to be within this range. The working frequency adaptation range of the energy conversion device (inverter) is 50Hz-80Hz, and the optimized working frequency also needs to meet this requirement.

[0112] Exemplarily, for a vehicle of a certain urban subway line, the optimized energy recovery parameters are a charge-discharge threshold of 2.8V-3.1V, an energy conversion module working frequency of 65Hz, and a recovery power distribution ratio of 0.7. When applying, these parameters need to be input into the brake control unit of the vehicle, and the control unit converts the parameters into executable instructions for the device.

[0113] When the vehicle brakes into the station, the brake control unit controls the energy storage device to start charging when the voltage is lower than 2.8V and stop charging when the voltage is higher than 3.1V according to the optimized parameters. At the same time, the energy conversion module is controlled to work at a frequency of 65Hz to efficiently convert the electric energy generated by braking, and the recovered electric energy is distributed to the power battery at a ratio of 0.7, and the remaining part is supplied to the facilities along the line or other trains through the train catenary.

[0114] In addition, during the application of the optimized energy recovery parameters, the braking state of the vehicle and the energy recovery effect need to be monitored in real time. For example, when the vehicle brakes in the rain, if it is monitored that the braking distance has a trend of becoming longer, the recovery power distribution ratio needs to be adjusted to 0.5 for a short time to preferentially guarantee the braking safety; when braking on a flat road in sunny weather, the original optimized parameters are maintained to maximize the recovery of energy.

[0115] Finally, by combining the optimized energy recovery parameters with the vehicle braking process, the problem of poor adaptability of traditional parameters can be solved, and the utilization rate of braking energy recovery can be improved, which provides practical support for energy saving and consumption reduction of rail transit vehicles.

[0116] Through the specific implementation manner described above, the embodiments of the present application achieve the following technical effects: This application proposes a method for recovering and utilizing braking energy in rail transit. First, by collecting the braking pressure, braking command time, and travel distance of the target rail transit vehicle, braking intensity, braking duration, and braking distance are obtained, and integrated to form a braking parameter set containing multiple braking parameter groups. Then, a multivariate verification model is constructed based on weather correction parameters. The braking parameter set is input into the model group by group to calculate the verification braking distance. The deviation between the actual and verified distances is converted into a confidence level. Parameter groups with confidence levels below a threshold are eliminated to obtain a reliable braking parameter set that reflects the true braking characteristics. Subsequently, historical braking energy recovery parameters are extracted from the reliable braking parameter set to form a true particle set. Combined with initial particle constraints, a random particle set is generated. After deduplication, an initial particle swarm population is obtained. Then, three fitness functions—safety, efficiency, and matching—are constructed. Weights are allocated according to road segment requirements to obtain a comprehensive fitness function, and particle fitness is calculated. Next, the particle swarm algorithm is used to iteratively update the optimal solutions for individuals and the population. Combined with weather correction parameters, the stagnation determination round N is determined. If the fitness increases for N consecutive rounds or the parameter change is less than a threshold, stagnation is determined. The particle generation space is adjusted by calculating the comprehensive fluctuation of braking parameters. After adding adjusted particles, the iteration continues until the optimized energy recovery parameters are obtained. Finally, the optimized parameters are input into the vehicle braking control unit to control the charging and discharging of the energy storage device, the operation of the energy conversion module, and power distribution. At the same time, dynamic adaptation is performed according to real-time operating conditions to realize the recovery and utilization of braking energy.

[0117] The method provided in this application, through a technical solution of "data acquisition - reliable screening - algorithm optimization - parameter application," solves the problems of poor adaptability to operating conditions, unreliable data due to weather interference, and easy optimization getting stuck in local optima or stagnation caused by fixed parameters in traditional rail transit braking energy recovery. It avoids optimization deviations caused by invalid data or braking safety risks caused by incompatible parameters, improves the balance between braking energy recovery utilization rate and braking safety, and can be adapted to different road sections such as mountainous areas, cities, and suburbs, as well as different weather conditions such as sunny, rainy, and snowy weather, providing reliable technical support for energy saving and consumption reduction of rail transit vehicles.

[0118] Example 2, as shown in the appendix Figure 2 As shown, based on the inventive concept of a method for recovering and utilizing braking energy in rail transit provided in Embodiment 1, this application also provides a device for recovering and utilizing braking energy in rail transit, specifically including: Braking parameter set acquisition module 01 is used to acquire the braking parameter set of the target rail transit vehicle, wherein the braking parameter set includes multiple braking parameter groups, and each braking parameter group includes braking intensity, braking duration and braking distance; Braking parameter set filtering module 02 is used to verify the credibility of the braking parameter set of the target rail transit vehicle based on weather correction parameters, remove braking parameter groups with credibility less than the credibility threshold, and obtain a credible braking parameter set. The recovery parameter optimization module 03 is configured to construct an improved particle swarm optimization algorithm to optimize the braking energy recovery parameters of the target rail transit vehicle, and obtain the optimized energy recovery parameters, wherein the braking energy recovery parameters include the charge-discharge threshold, the energy conversion module operating frequency, and the recovery power distribution ratio, and the improved particle swarm optimization algorithm includes optimization stagnation determination and dynamic adjustment. The energy recovery execution module 04 is configured to perform rail transit braking energy recovery and utilization by using the optimized energy recovery parameters.

[0119] In one embodiment, the braking parameter set acquisition module 01 is further configured to: acquire the braking pressure of the target rail transit vehicle, perform mapping processing, and acquire the braking intensity; acquire the triggering time and the release time of the braking instruction of the target rail transit vehicle, calculate the time difference between the two times as the braking duration; and acquire the distance from when the target rail transit vehicle receives the braking instruction to when the braking ends as the braking distance.

[0120] In one embodiment, the braking parameter set screening module 02 is further configured to: construct a multivariate verification model, wherein the input of the multivariate verification model is the braking intensity, the braking duration, and the weather correction parameter, and the output result is the verification braking distance, wherein the weather correction parameter is obtained based on the current weather; input the braking parameter set into the multivariate verification model, obtain a plurality of verification braking distances, calculate the deviation degree of a plurality of verification braking distances and a plurality of braking distances; obtain a plurality of credibility by subtracting the deviation degree from 1; delete the braking parameter group with a credibility less than a credibility threshold from the braking parameter set to obtain a credible braking parameter set.

[0121] In one embodiment, the recovery parameter optimization module 03 is further configured to: obtain an initial particle swarm population, wherein the initial particle swarm population includes a plurality of particles, and each particle represents a set of braking energy recovery parameters; obtain a comprehensive fitness calculation function to calculate the fitness of a plurality of particles; update the individual optimal solution and the population optimal solution of the particles, perform particle swarm optimization, and obtain the optimized energy recovery parameters, wherein the particle swarm optimization process further includes optimization stagnation determination and dynamic adjustment.

[0122] Further, the recovery parameter optimization module 03 further includes: Based on the set of trusted braking parameters, a set of historical braking energy recovery parameters is obtained, and a particle initial constraint is obtained; the historical braking energy recovery parameters in the set of historical braking energy recovery parameters are integrated into real particles to obtain a set of real particles; based on the particle initial constraint, a set of random particles is obtained, wherein the set of random particles includes a plurality of random particles, and the number of random particles is obtained based on the number of real particles; the set of random particles and the set of real particles are merged, and duplicate particles are removed to obtain an initial particle population.

[0123] Further, the recovery parameter optimization module 03 further includes: The influence braking intensity, influence braking duration and influence braking distance of the plurality of particles after the influence of the braking energy recovery parameters are obtained; a plurality of fitness calculation functions are constructed, wherein the plurality of fitness calculation functions include a safety fitness calculation function, an efficiency fitness calculation function and a matching fitness calculation function; the safety fitness, the efficiency fitness and the matching fitness are weighted calculated to obtain a comprehensive fitness calculation function; the particles in the initial particle population are calculated for fitness using the comprehensive fitness calculation function to obtain a plurality of particle fitnesses.

[0124] Further, the recovery parameter optimization module 03 further includes: The safety fitness is obtained based on the influence braking intensity of the particles, the efficiency fitness is obtained based on the influence braking duration of the particles, and the matching fitness is obtained based on the influence braking distance of the particles; based on the driving route and driving demand of the target rail transit vehicle, a safety weight, an efficiency weight and a matching weight are obtained, the safety fitness, the efficiency fitness and the matching fitness are weighted calculated to obtain a comprehensive fitness calculation function.

[0125] Further, the recovery parameter optimization module 03 further includes: The plurality of particles are iteratively updated using a particle swarm algorithm to obtain a particle individual optimal solution and a population optimal solution; if the fitness of the global optimal solution is improved by less than a fitness improvement threshold, or any one of the braking intensity improvement amplitude, the braking duration and the braking distance is changed by less than a braking change threshold, within N consecutive iterations, it is determined that the optimization is stagnant, and a dynamic adjustment scheme is used to adjust the particle swarm, wherein N is obtained based on the weather correction parameter.

[0126] Further, the recovery parameter optimization module 03 further includes: Based on the brake parameter set, a brake parameter comprehensive fluctuation degree is obtained, wherein the brake parameter comprehensive fluctuation degree is obtained based on the variance and mean value of the brake intensity, the brake duration and the brake distance; based on the brake parameter comprehensive fluctuation degree and the particle initial constraint, an adjustment space is obtained; based on the adjustment space, a plurality of adjustment particles are generated and added into the initial particle population for iterative optimization.

[0127] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. Moreover, the above-mentioned embodiments of the present application are described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0128] The above only describes the preferred embodiments of the present application and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. The specification and drawings are only exemplary descriptions of the present application, and any and all modifications, changes, combinations or equivalents within the scope of the present application are considered to be covered. Obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope of the present application. Thus, if these modifications and changes of the present application belong to the scope of the present application and its equivalent technology, the present application intends to include these modifications and changes.

Claims

1. A method for recovering and utilizing braking energy in rail transit, characterized in that, include: Obtain the braking parameter set of the target rail transit vehicle, wherein the braking parameter set includes multiple braking parameter groups, and each braking parameter group includes braking intensity, braking duration and braking distance; Based on weather correction parameters, the reliability of the braking parameter set of the target rail transit vehicle is verified, and braking parameter sets with reliability less than the reliability threshold are eliminated to obtain a reliable braking parameter set. An improved particle swarm optimization algorithm is constructed to optimize the braking energy recovery parameters of the target rail transit vehicle and obtain the optimized energy recovery parameters. The braking energy recovery parameters include the charging and discharging threshold, the operating frequency of the energy conversion module, and the recovery power allocation ratio. The improved particle swarm optimization algorithm includes optimized stagnation judgment and dynamic adjustment. The optimized energy recovery parameters are used to recover and utilize braking energy in rail transit.

2. The method for recovering and utilizing braking energy in rail transit according to claim 1, characterized in that, Obtain the braking parameter set of the target rail transit vehicle, wherein the braking parameter set includes multiple braking parameter groups, each braking parameter group including braking intensity, braking duration, and braking distance, including: Obtain the braking pressure of the target rail transit vehicle, perform mapping processing, and obtain the braking intensity; Obtain the trigger time and release time of the braking command for the target rail transit vehicle, and calculate the time difference between the two times as the braking duration; The distance from when the train receiving the braking command to when the train stops braking is used as the braking distance.

3. The method for recovering and utilizing braking energy in rail transit according to claim 1, characterized in that, Based on weather correction parameters, the reliability of the braking parameter set of the target rail transit vehicle is verified. Braking parameter sets with reliability below the reliability threshold are removed to obtain a reliable braking parameter set, including: A multivariate validation model is constructed, wherein the input of the multivariate validation model is the braking intensity, the braking duration and the weather correction parameters, and the output is the validation braking distance, wherein the weather correction parameters are obtained based on the current weather. The braking parameter set is input into the multivariate verification model to obtain multiple verification braking distances, and the deviation between the multiple verification braking distances and the multiple braking distances is calculated. Multiple confidence levels are obtained by subtracting the deviation from 1. Remove braking parameter groups with a confidence level lower than the confidence level threshold from the braking parameter set to obtain a reliable braking parameter set.

4. The method for recovering and utilizing braking energy in rail transit according to claim 1, characterized in that, Constructing an improved particle swarm optimization algorithm, including: Obtain an initial particle swarm population, wherein the initial particle swarm population includes multiple particles, each particle representing a set of braking energy recovery parameters; Obtain the comprehensive fitness calculation function and calculate the fitness of multiple particles; The individual optimal solution and the population optimal solution of the particles are updated to perform particle swarm optimization and obtain optimized energy recovery parameters. The particle swarm optimization process also includes optimization stagnation determination and dynamic adjustment.

5. A method for recovering and utilizing braking energy in rail transit according to claim 4, characterized in that, Obtain the initial particle swarm population, including: Based on the reliable braking parameter set, the historical braking energy recovery parameter set is obtained, and the initial particle constraints are obtained; The historical braking energy recovery parameters in the historical braking energy recovery parameter set are integrated into real particles to obtain a real particle set. Based on the initial particle constraints, a random particle set is obtained, wherein the random particle set includes multiple random particles, and the number of random particles is obtained based on the number of real particles; The random particle set and the real particle set are merged, and duplicate particles are removed to obtain an initial particle population.

6. The method for recovering and utilizing braking energy in rail transit according to claim 4, characterized in that, Obtain the comprehensive fitness calculation function and calculate the fitness of multiple particles, including: The effects of the braking energy recovery parameters of multiple particles on braking intensity, braking duration, and braking distance are obtained. Construct multiple fitness calculation functions, wherein the multiple fitness calculation functions include a safety fitness calculation function, an efficiency fitness calculation function, and a matching fitness calculation function; The safety fitness, efficiency fitness, and matching fitness are weighted and calculated to obtain a comprehensive fitness calculation function; The fitness calculation function is used to calculate the fitness of particles in the initial particle population and obtain the fitness of multiple particles.

7. A method for recovering and utilizing braking energy in rail transit according to claim 6, characterized in that, The safety fitness, efficiency fitness, and matching fitness are weighted and calculated to obtain a comprehensive fitness calculation function, including: The safety fitness is obtained based on the braking intensity of the particle's influence, the efficiency fitness is obtained based on the braking duration of the particle's influence, and the matching fitness is obtained based on the braking distance of the particle's influence. Based on the travel segments and travel demands of the target rail transit vehicles, safety weights, efficiency weights, and matching weights are obtained. The safety fitness, efficiency fitness, and matching fitness are then weighted and calculated to obtain a comprehensive fitness calculation function.

8. A method for recovering and utilizing braking energy in rail transit according to claim 4, characterized in that, The individual optimal solution and the population optimal solution of the particles are updated to perform particle swarm optimization. The particle swarm optimization process also includes optimization stagnation determination and dynamic adjustment, including: The particle swarm optimization algorithm is used to iteratively update multiple particles to obtain the optimal solution for each individual particle and the optimal solution for the swarm. If, within N consecutive iterations, the fitness improvement of the global optimal solution is less than the fitness improvement threshold, or the change in any of the parameters—braking intensity improvement, braking duration, and braking distance—is less than the braking change threshold, then the optimization is considered stalled, and a dynamic adjustment scheme is adopted to adjust the particle swarm, wherein N is obtained based on the weather correction parameters.

9. A method for recovering and utilizing braking energy in rail transit according to claim 5, characterized in that, The particle swarm is adjusted using a dynamic adjustment scheme, including: Based on the braking parameter set, the comprehensive fluctuation of braking parameters is obtained, wherein the comprehensive fluctuation of braking parameters is obtained based on the variance and mean of the braking intensity, the braking duration, and the braking distance; Based on the comprehensive fluctuation of the braking parameters and the initial constraints of the particles, the adjustment space is obtained; Based on the adjustment space, multiple adjustment particles are generated and added to the initial particle population for iterative optimization.

10. A device for recovering and utilizing braking energy in rail transit, characterized in that, The apparatus is used to perform a method for recovering and utilizing braking energy in rail transit as described in any one of claims 1-9, and the apparatus comprises: A braking parameter set acquisition module is used to acquire the braking parameter set of the target rail transit vehicle, wherein the braking parameter set includes multiple braking parameter groups, and each braking parameter group includes braking intensity, braking duration and braking distance; The braking parameter set filtering module is used to verify the credibility of the braking parameter set of the target rail transit vehicle based on weather correction parameters, remove braking parameter sets with credibility less than the credibility threshold, and obtain a credible braking parameter set. The recovery parameter optimization module is used to construct an improved particle swarm optimization algorithm to optimize the braking energy recovery parameters of the target rail transit vehicle and obtain optimized energy recovery parameters. The braking energy recovery parameters include charging and discharging thresholds, energy conversion module operating frequency, and recovery power allocation ratio. The improved particle swarm optimization algorithm includes optimized stagnation determination and dynamic adjustment. An energy recovery execution module is used to recover and utilize braking energy in rail transit using the optimized energy recovery parameters.

Images