Track parameter design method for tracked vehicles collecting manganese nodules on the seabed

Abstract

This invention belongs to the field of deep-sea mining, specifically relating to a method for designing track parameters for a tracked vehicle used for collecting manganese nodules on the seabed. The method for designing track parameters for a tracked vehicle used for collecting manganese nodules on the seabed comprises the following steps: (1) Constructing a deep-sea seabed sediment creep model under the driving action of the deep-sea mining vehicle; (2) Deriving the formula for calculating the driving resistance of the deep-sea mining vehicle; (3) Performing secondary development and modeling based on the seabed creep model and driving resistance using multibody dynamics software; (4) Data interaction between the improved PSO algorithm based on Python and the multibody dynamics software; (5) Dynamically changing the search interval based on the current iteration number of the improved PSO algorithm; (6) Reaching the maximum iteration number and outputting the parameters. This invention comprehensively considers the rheological characteristics of the soft seabed sediment and the driving performance of the mining vehicle, and combines the improved multi-objective PSO algorithm to provide optimal values ​​for track parameters based on the subsidence amount and traction force of the mining tracked vehicle. It is expected to achieve good results in the design of track parameters for mining tracked vehicles in different sea areas.

Description

Technical Field

[0001] This invention belongs to the field of deep-sea mining, specifically relating to a method for designing track parameters for a tracked vehicle used for collecting manganese nodules from the seabed. Background Technology

[0002] Deep-sea manganese nodule mining primarily employs the fluid lift mining method, which involves using tracked mining vehicles to cut and crush the manganese nodules, collect them, and then transport them via pipeline to surface mining vessels. As the core component of the mining system, the tracked mining vehicle's mobility on the soft seabed significantly impacts the efficiency of manganese nodule mining. The close relationship between track parameters—track length, track width, and track tooth height—and the tracked vehicle's mobility performance—subsidence and traction—has long been proven by numerous engineering examples and experimental studies.

[0003] Previous studies have typically employed models such as the Bekker model, Janosi-Hanamoto model, Wong model, and rheological constitutive models when designing track parameters based on subsidence and traction. However, these models do not simultaneously consider the combined effects of the rheological properties of the soft seabed and the driving resistance, and usually only consider one factor affecting the tracked vehicle's performance when designing parameters.

[0004] To address this, this invention provides a RecurDyn secondary development model based on the rheological properties of the soft seabed and its driving resistance. Based on this model and an improved multi-objective PSO algorithm, optimal track parameter values ​​are obtained. This method fully considers the rheological properties of the soft seabed and the walking performance of the tracked vehicle. Combined with the improved multi-objective PSO algorithm, it provides optimal track parameter values ​​based on the subsidence amount and traction force of the mining tracked vehicle, and is expected to achieve good results in the design of track parameters for mining tracked vehicles in different sea areas. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a method for designing track parameters for a tracked vehicle used for collecting manganese nodules from the seabed.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] (1) Construct a deep-sea seabed creep model under the action of a deep-sea tracked vehicle;

[0008] (2) Derive the formula for calculating the driving resistance of deep-sea tracked vehicles;

[0009] (3) Based on multibody dynamics software, secondary development and modeling based on the bottom creep model and driving resistance are carried out, with the track parameters to be inverted as input values ​​and the calculated sinking-time curve and vehicle speed response curve as output values.

[0010] (4) The improved PSO algorithm based on Python language interacts with the multibody dynamics software. The calculated sinking-time curve and vehicle speed response curve are transmitted to the improved PSO algorithm for track parameter updates. The updated track parameters are transmitted to the software for the next calculation.

[0011] (5) During the iteration of the improved PSO algorithm, the search interval is dynamically changed according to the current iteration number;

[0012] (6) Reach the maximum number of iterations and output the optimal track parameters.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] (1) The compression and shearing effects of the tracked vehicle on the soft seabed and the rheological properties of the seabed during operation are comprehensively considered, and the driving resistance is also taken into account, which improves the accuracy of the analysis of the deep-sea tracked vehicle's walking performance.

[0015] (2) This invention can obtain the optimal values ​​of track parameters for mining tracked vehicles by using an improved PSO algorithm while simultaneously obtaining the maximum traction force and minimum subsidence of the tracked vehicle. Combined with the seabed parameters obtained from the actual sea area, this provides some guidance for the track design of mining tracked vehicles. Attached Figure Description

[0016] Figure 1 A schematic diagram illustrating the design method for track parameters of a tracked vehicle for collecting manganese nodules from the seabed.

[0017] Figure 2 A schematic diagram of the generalized Kelvin model and the Burgers model;

[0018] Figure 3 Flowchart of the improved PSO algorithm. Detailed Implementation

[0019] like Figure 1 As shown, the design method for track parameters of a tracked vehicle for collecting manganese nodules on the seabed is as follows:

[0020] 1. Construct a deep-sea seabed sediment creep model under the action of a deep-sea tracked vehicle. The specific method is as follows:

[0021] Compression creep experiments and compression-shear coupled creep experiments were conducted using deep-sea seabed sediment to obtain compression creep curves under different compressive stresses, compression-shear creep curves under different compressive stresses, and compression-shear creep curves under different shear stresses.

[0022] Figure 2 (a) is a schematic diagram of the generalized Kelvin model; to describe the compression creep curve obtained from the experiment, based on the generalized Kelvin model, the equation for compression amount versus compression time can be written as:

[0023]

[0024] In equation (1): σ and z are compressive stress and compression amount, respectively; E1 and E2 are viscoelastic modulus; η is viscosity coefficient; and t is time. The values ​​of E1, E2, and η are obtained by fitting the compression creep curve.

[0025] Figure 2 (b) is a schematic diagram of the Burgers model; to describe the experimentally obtained compression-shear creep curve, based on the Burgers model, the equations for shear displacement, shear stress, and compressive stress can be written as follows:

[0026]

[0027] In equation (2): τ and s are shear stress and shear displacement, respectively; K1(σ) and K2(σ) are viscoelastic shear moduli; β1(σ) and β2(σ) are viscosity coefficients; K1(σ), K2(σ), β1(σ), and β2(σ) are functions of σ, determined by fitting experimental data.

[0028] 2. The formula for calculating the driving resistance of deep-sea tracked vehicles is derived as follows:

[0029] The running resistance of a tracked vehicle consists of two parts: track compaction resistance R. c The raised ground at the front of the track of the tracked vehicle generates bulldozing resistance R on the vehicle body. b ;

[0030] The track shoe compaction resistance is calculated as follows:

[0031]

[0032] In equation (3): E is the work done by the track plate with a compaction depth of z, and L is half the length of a single track plate; the formula for calculating E is as follows:

[0033]

[0034] In equation (4): B is the track width, and p is the ground pressure of the track plate; the formula for calculating p is as follows:

[0035]

[0036] In equation (5): G is the pressure borne by the track plate; substituting equations (1), (4), and (5) into equation (3) yields:

[0037]

[0038] The bulldozing resistance generated by the raised subgrade at the front of the track of a tracked vehicle on the vehicle body is calculated as follows:

[0039] Rb =B(0.67czK) c +0.5z 2 γK γ (7)

[0040] In equation (7), γ is the specific gravity of the substrate, and K c , K γ is the passive earth pressure coefficient, and c is the soil cohesion; where N c N γ The bearing capacity coefficient is uniquely determined by the internal friction angle. Let be the internal friction angle of the soil; rearranging equation (7) gives:

[0041]

[0042] 3. Based on multibody dynamics software, a secondary development and modeling based on the seabed creep model and driving resistance is performed. The track parameters to be inverted are used as input values, and the calculated sinking-time curve and vehicle speed response curve are used as output values. The specific method is as follows:

[0043] Taking the multibody dynamics software RecurDyn as an example, based on the deep-sea tracked vehicle's creep model under driving action and the calculation formula for driving resistance, the software is written in Fortran. The completed .for file is placed in the RecurDyn folder and the FORTRAN folder to generate a dynamic link library for use during calculation. Based on the existing tracked vehicle parameters and deep-sea seabed physical parameters, a simplified tracked vehicle model and a soil model are established, and the drive wheel speed, simulation time, and step size are set. Based on the established simplified tracked vehicle model and soil model, the track parameters are input, and the software calculates and outputs the settlement-time curve and vehicle speed response curve data under the current track parameters.

[0044] 4. Based on the improved PSO algorithm using Python, data interaction is performed between the calculated sinking-time curve and the vehicle speed response curve. The calculated data is then transmitted to the improved PSO algorithm for track parameter updates. The updated track parameters are then transmitted to the software for the next calculation. The specific method is as follows:

[0045] Taking the multibody dynamics software RecurDyn as an example, open the ProcessNet module in RecurDyn and select Python in the ProcessNet type; write an improved PSO algorithm based on Python; based on the PWLCM chaotic mapping, input the initialized track vehicle track parameters into the established track vehicle model. After calculation, input the generated sink-time curve and vehicle speed response curve data into the improved PSO algorithm to update the track parameters. The updated track parameters are then input back into RecurDyn to calculate the sink-time curve and vehicle speed response curve data under the current parameters, and input back into the improved PSO algorithm for the next track parameter update; the PWLCM chaotic mapping can make the initial parameters uniformly distributed in the search space. The definition of the PWLCM chaotic mapping is as follows:

[0046]

[0047] In equation (9), X represents the iteration value, n is the current iteration number, and p is the control parameter.

[0048] 5. During the iterative process of the improved PSO algorithm, the search interval is dynamically changed according to the current iteration number. The specific method is as follows:

[0049] according to Figure 3 The flowchart of the improved PSO algorithm shown illustrates that if the current iteration number is not a multiple of 50, within the initially set velocity and position boundaries, the inertia weight w, learning factors c1 and c2, and the particle's velocity and position are updated. The update rules for w, c1, and c2 are as follows:

[0050]

[0051]

[0052]

[0053] In the formula, N is the total number of iterations, λ is a random variable greater than 0, and Γ(λ, 1-n / N) is the incomplete gamma function; the subscripts start and end represent the initial and final values ​​of the value; the particle velocity and position update mechanism is as follows:

[0054]

[0055]

[0056] In the formula, P best With G best These represent the individual optimal position and the group optimal position, respectively. i 、x iLet r1 and r2 be the velocity and position of the particle, respectively, and r1 and r2 be random variables between 0 and 1.

[0057] After the parameters are updated, calculate the fitness f of each particle. i The average fitness f of all particles a If f i <f a If f i >f a If the fitness of the perturbed or mutated particle is greater than that of the original particle, it is replaced; if it is less than that of the original particle, the original particle is retained; then P is performed. best With G best Update; the rules for chaotic perturbation and Gaussian mutation are as follows:

[0058]

[0059] X n =X n ×[1+σN(0,1)] (16)

[0060] Where rand(0,1) is a random number in the range [0,1], and σN(0,1) is a random number with a mean of 0 and a variance of 1 that follows a Gaussian distribution;

[0061] If the current iteration number is a multiple of 50, then use the current G... best Generate a new search interval; perform iterative calculations within the new search interval, following the same iteration method as described above; when the number of iterations reaches 25, exit the iteration and return to the original iteration loop to continue the calculation; with the current G... best The rule for generating a new search interval is:

[0062] X ub =λ1G best (17)

[0063] X lb =λ2G best (18)

[0064] Among them, X ub X lb λ1 and λ2 are the upper and lower bounds of the search space, respectively, and scaling factors.

[0065] 6. Reach the maximum number of iterations and output the optimal track parameters. The specific method is as follows:

[0066] When the current iteration count reaches the maximum iteration count, the algorithm stops iterating. The track parameters obtained at this time are the optimal track parameters, and these parameters are output.

Claims

1. A method for designing track parameters for a tracked vehicle used for collecting manganese nodules from the seabed, characterized in that, Here are the steps: (1) Construct a deep-sea seabed creep model under the action of a deep-sea tracked vehicle; The specific method is as follows: Compression creep experiment and compression-shear coupled creep experiment were carried out using deep seabed sediment to obtain compression creep curves under different compressive stresses, compression-shear creep curves under different compressive stresses and compression-shear creep curves under different shear stresses; The compression creep curve and the compression-shear creep curve are described using the generalized Kelvin model and the Burgers model, respectively, with the following equations: In the formula: σ and z are compressive stress and compression, respectively; E1 and E2 are viscoelastic moduli, η is viscosity coefficient, and t is time; the values ​​of E1, E2, and η are obtained by fitting the compression-creep curve; τ and s are shear stress and shear displacement, respectively; K1(σ) and K2(σ) are viscoelastic shear moduli, and β1(σ) and β2(σ) are viscosity coefficients; K1(σ), K2(σ), β1(σ), and β2(σ) are functions of σ, determined by fitting experimental data. (2) Derive the formula for calculating the driving resistance of deep-sea tracked vehicles; The specific method is as follows: The running resistance of a tracked vehicle consists of two parts: the track plate compaction resistance R. c The raised ground at the front of the track of the tracked vehicle generates bulldozing resistance R on the vehicle body. b ; The track shoe compaction resistance is calculated as follows: In equation (3): E is the work done by the track plate with a compaction depth of z, and L is half the length of a single track plate; the formula for calculating E is as follows: In equation (4): B is the track width, and p is the ground pressure of the track plate; the formula for calculating p is as follows: In equation (5): G is the pressure borne by the track plate; substituting equations (1), (4), and (5) into equation (3) yields: The bulldozing resistance generated by the raised subgrade at the front of the track of a tracked vehicle on the vehicle body is calculated as follows: R b =B(0.67czK c +0.5z 2 γK γ ) (7) In equation (7), γ is the specific gravity of the substrate, and K c K γ is the passive earth pressure coefficient, and c is the soil cohesion; where N c N γ The bearing capacity coefficient is uniquely determined by the internal friction angle. Let be the internal friction angle of the soil; rearranging equation (7) yields: (3) Based on multibody dynamics software, secondary development and modeling based on the bottom creep model and driving resistance are carried out, with the track parameters to be inverted as input values ​​and the calculated sinking-time curve and vehicle speed response curve as output values. (4) The improved PSO algorithm based on Python language interacts with the multibody dynamics software. The calculated sinking-time curve and vehicle speed response curve are transmitted to the improved PSO algorithm for track parameter updates. The updated track parameters are transmitted to the software for the next calculation. (5) During the iteration of the improved PSO algorithm, the search interval is dynamically changed according to the current iteration number; (6) Reach the maximum number of iterations and output the optimal track parameters.

2. The track parameter design method for the tracked vehicle for collecting manganese nodules on the seabed according to claim 1, characterized in that, This paper presents a secondary development and modeling based on a multibody dynamics software, using a seabed creep model and driving resistance calculation. The track parameters to be inverted are used as input values, and the calculated subsidence-time curve and vehicle speed response curve are used as output values. The specific method is as follows: Taking the multibody dynamics software RecurDyn as an example, based on the constructed deep-sea tracked vehicle driving action deep-sea seabed creep model and the deep-sea tracked vehicle driving resistance calculation formula, the software is written in Fortran. The completed .for file is placed in the RecurDyn folder and the FORTRAN folder to generate a dynamic link library for use during calculation. Based on the existing tracked vehicle parameters and deep-sea seabed physical parameters, a simplified tracked vehicle model and a soil model are established, and the drive wheel speed, simulation time, and step size are set. Based on the established simplified tracked vehicle model and soil model, the track parameters are input, and the software calculates and outputs the subsidence-time curve and vehicle speed response curve data under the current track parameters.

3. The method for designing track parameters for a tracked vehicle used for collecting manganese nodules on the seabed according to claim 1, characterized in that, An improved PSO algorithm based on Python interacts with multibody dynamics software. The calculated sinking-time curve and vehicle speed response curve are transmitted to the improved PSO algorithm for track parameter updates. The updated track parameters are then transmitted to the software for the next calculation. The specific method is as follows: Taking the multibody dynamics software RecurDyn as an example, open the ProcessNet module in RecurDyn and select Python in the ProcessNet type; write the improved PSO algorithm based on Python. Based on the PWLCM chaotic mapping, the initially generated track parameters of the tracked vehicle are input into the established tracked vehicle model. After calculation, the generated sink-time curve and vehicle speed response curve data are input into the improved PSO algorithm to update the track parameters. The updated track parameters are then input back into RecurDyn to calculate the sink-time curve and vehicle speed response curve data under the current parameters, and input back into the improved PSO algorithm for the next track parameter update. The PWLCM chaotic mapping enables the initial parameters to be evenly distributed in the search space. The definition of the PWLCM chaotic mapping is as follows: In the formula, X represents the iteration value, n is the current iteration number, and p is the control parameter.

4. The method for designing track parameters for a tracked vehicle used for collecting manganese nodules from the seabed according to claim 1, characterized in that, In the iterative process of the improved PSO algorithm, the search interval is dynamically changed according to the current iteration number. The specific method is as follows: If the current iteration number is not an integer multiple of 50, within the initially set velocity and position boundaries, the inertia weight w, learning factors c1 and c2, and the particle's velocity and position are updated; the update rules for w, c1, and c2 are as follows: In the formula, N is the total number of iterations, λ is a random variable greater than 0, and Γ(λ, 1-n / N) is the incomplete gamma function; the subscripts start and end represent the initial and final values ​​of the value; the particle velocity and position update mechanism is as follows: In the formula, P best With G best These represent the individual optimal position and the group optimal position, respectively. i x i Let r1 and r2 be the velocity and position of the particle, respectively, and r1 and r2 be random variables between 0 and 1. After the parameters are updated, calculate the fitness f of each particle. i The average fitness f of all particles a If f i <f a If f i >f a If the fitness of the perturbed or mutated particle is greater than that of the original particle, it is replaced; if it is less than that of the original particle, the original particle is retained; then P is performed. best With G best Update; the rules for chaotic perturbation and Gaussian mutation are as follows: X n =X n ×[1+σN(0,1)] (16) Where rand(0,1) is a random number in the range [0,1], and σN(0,1) is a random number with a mean of 0 and a variance of 1 that follows a Gaussian distribution; If the current iteration number is a multiple of 50, then use the current G... best Generate a new search interval; perform iterative calculations within the new search interval, following the same iteration method as described above; when the number of iterations reaches 25, exit the iteration and return to the original iteration loop to continue the calculation; with the current G... best The rule for generating a new search interval is: X ub =λ1G best (17) X lb =λ2G best (18) Among them, X ub X lb λ1 and λ2 are the upper and lower bounds of the search space, respectively, and scaling factors.

5. The method for designing track parameters for a tracked vehicle used for collecting manganese nodules from the seabed according to claim 1, characterized in that, The algorithm stops iterating when the current iteration count reaches the maximum number of iterations. The track parameters obtained at this time are the optimal track parameters, and these parameters are output.

Images