Design method for working device of hydraulic backhoe excavator

Abstract

The invention discloses a method for designing a working device of a hydraulic backhoe excavator, which adopts a staged design mode, takes minimization of design target loss as a target function, and performs automatic search through an improved myxomycete algorithm to achieve the purpose of designing the working device. The method adopts the non-feedback process, the optimal leader strategy and the sparrow leading search strategy to improve the standard myxomycete algorithm, improves the search precision of the algorithm, and reduces the overall design error. And constructing a backhoe hydraulic excavator working device design model by taking an improved myxomycete algorithm as an optimization search algorithm and taking design target loss minimization as a target function. According to the design model, the size parameters can be automatically searched without using the size parameters as known parameters, so that the design of determining the size parameters of the excavator working device according to the working performance is realized, the problem that the working performance evaluation needs to be performed after the size parameters are determined in the design process of the excavator working device is solved, and the design period is greatly shortened.

Description

Technical Field

[0001] The invention belongs to the technical field of hydraulic backhoe excavator design, and in particular relates to a design method for a hydraulic backhoe excavator working device. Background Art

[0002] With the construction of modern infrastructure around the world, hydraulic excavators have been widely used in many fields such as construction, farmland, water conservancy and transportation construction. Due to its wide application and high value creation, hydraulic excavators have accounted for about 65% to 70% of the earthwork volume in earthwork projects, and have a pivotal position in the field of engineering machinery. The working device of an excavator usually refers to the general term for mechanical devices including booms, dipper arms, buckets, rocker arms, and connecting rods. It is a device that directly completes the excavation task of the excavator and is one of the most important parts of the excavator. The main evaluation indicators of the excavator's operating performance are the excavation range and excavation force, and working devices with different size parameters have different excavation ranges. Therefore, the design of the excavator's working device has become an important task. With the development of science and technology, researchers have launched research on the design of excavator working devices, and most of them have conducted research based on virtual prototype technology and computer technology. For example, Chen Zhi et al. combined Pro / E and ADAMS for kinematic simulation and verified the correctness of the kinematic model. Zhang Guiju et al. combined Pro / E and ADAMS to simulate the kinematics and dynamics of a 23t excavator working device. The simulation results were consistent with the actual situation, providing a basis and method for the structural design of the excavator working device. Deng Fang et al. conducted three-dimensional modeling with SolidWorks and used finite element software to perform strength analysis, verifying that the designed bucket rod can meet the strength requirements, thus completing the design of the excavator bucket rod. Cao Yongli et al. used SolidWorks and RecurDyn to perform dynamic simulation analysis. The results showed that the simulation results were basically consistent with the actual situation, providing a basis for the design of the excavator working device. Li Xianjie et al., based on the analysis and research of the working device's motion trajectory and force law, used Matlab to develop a visualization software for the excavation performance of the working device. The software can quickly and accurately draw the excavation range of the working device and the excavation force cloud map, providing a theoretical basis and practical means for the design of the excavator working device.

[0003] In summary, most of the existing work device design research is based on the excavator's dimensional parameters to assess its working performance. It is difficult to design a work device that meets expectations. There are two main reasons: first, the work device involves many dimensional parameters and working performance, and the influence of dimensional parameters on working performance is highly nonlinear; second, there is a complex coupling relationship between the dimensional parameters. Therefore, the design of the work device often cannot meet all the working performance requirements at one time, and repeated experiments are required, and the design cycle is long. In fact, the design process of the work device is to determine the dimensional parameters of the work device according to the design goals, and a few researchers have also studied this in order to determine the dimensional parameters of the excavator according to the working performance. For example, Dong Hanlin used the method of solving the system of equations to design the excavator working device based on the known design goals, and developed a practical and effective working device design software. On this basis, kinematic and dynamic simulations were performed to optimize the position of the hinge point of the excavator working device. However, because the solution of the system of equations requires the same number of equations and unknown quantities, this method still requires certain dimensional parameters to be known parameters, and the excavator dimensional parameters cannot be determined only by working performance, and the design cycle is relatively long.

[0004] In recent years, with the vigorous development of artificial intelligence and computer technology, a large number of swarm intelligence optimization algorithms inspired by natural phenomena have emerged, such as: Particle Swarm Optimization Algorithm (PSO), Ant Colony Optimization Algorithm (ACO), Artificial Bee Colony Algorithm (ABC), Glowworm Swarm Optimization Algorithm (GSO), Cuckoo Search Algorithm (CS), etc. These intelligent algorithms are simple in principle, easy to implement, and do not require gradient information. They can solve various complex high-dimensional nonlinear problems and are often used in various engineering optimization problems. However, most of the current intelligent optimization algorithms are prone to fall into local optimality. The Slime Mould Algorithm (SMA) is an intelligent optimization algorithm proposed by Li et al. in 2020 to simulate the foraging of slime mold. Compared with other intelligent optimization algorithms, it has the characteristics of simple structure, few adjustment parameters, and strong optimization ability. It meets the needs of rapid design of multi-parameter excavator working devices. However, the standard slime mold algorithm still has shortcomings such as weak local search ability and low search accuracy. Summary of the invention

[0005] In view of the problems of the prior art, the present invention provides a design method for a hydraulic backhoe excavator working device. The method adopts a non-feedback process, an optimal leader strategy, and a sparrow-leading search strategy to improve the standard slime mold algorithm, improve the search accuracy of the algorithm, and reduce the overall design error. With the Improved Slime Mold Algorithm (ISMA) as the optimization search algorithm and the minimization of the design target loss as the objective function, a design model for a backhoe hydraulic excavator working device is constructed. The design model does not require the size parameters to be known parameters, and can automatically search for the size parameters, realizing the design of the excavator working device size parameters determined by the working performance, solving the problem that the excavator working device design process needs to determine the size parameters before the working performance evaluation, greatly shortening the design cycle.

[0006] A method for designing a working device of a hydraulic backhoe excavator comprises the following steps:

[0007] Step 1: Overall size design stage

[0008] (1) According to the design requirements, the design objectives are obtained, including the maximum digging depth h1, the maximum digging height h2, the maximum digging radius r1, the maximum unloading height h3, the ratio of the boom to the arm length K, the boom angle range Δθ1, the arm angle range Δθ2, and the bucket angle range Δθ3. h1, h2, r1, and h3 are used as the target values ​​of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0009]

[0010] Where: Y i is the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are h1, h2, r1, and h3 respectively; y i is the predicted value of the corresponding parameter calculated using the mathematical model in the overall size design stage;

[0011] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set the horizontal distance B from the hinge point between the variable boom and the turntable to the turntable rotation center x , vertical distance B from the hinge point of the boom and turntable to the horizontal ground y , boom length L1, arm length L2, bucket length L3, minimum boom angle θ 1min 、Maximum boom angle θ 1max 、Minimum arm angle θ 2min 、Maximum arm rotation angle θ 2max 、Minimum bucket rotation angle θ 3min 、Maximum bucket rotation angle θ3max The boundary conditions are as follows:

[0012] lb i ≤x i ≤ub i 1≤i≤11

[0013] Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of x ,B y ,L1,L2,L3,θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max ];

[0014] (3) Take the design variable x = [B x ,B y ,L1,L3,θ 1min ,θ 2max ,θ 3max ], update the design variables L1, θ according to the following formula 1min ,θ 2max ,θ 3max The upper and lower bounds of :

[0015]

[0016] Where: ub, lb are the new upper and lower bounds of the corresponding design variables; lb0, lb1, lb2, lb3 are L1, θ 1min ,θ 2max ,θ 3max The original lower bound of ub0, ub1, ub2, and ub3 are L1, θ 1min ,θ 2max ,θ 3max The original upper bound; LB0, LB1, LB2, LB3 are L2, θ 1max ,θ 2min ,θ 3min The lower bound of UB0, UB1, UB2, and UB3 are L2, θ 1max ,θ 2min ,θ 3min The upper bound of

[0017] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables, and then obtain the variables L2 and θ according to the following formula 1max ,θ 2min ,θ3min Finally, all design values ​​B of this design stage are output x , B y , L1, L2, L3, θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max , and then proceed to the next stage of design;

[0018]

[0019] Step 2: Boom and arm angle design stage

[0020] (1) Taking the design value L1 in the overall dimension design stage as a known parameter, taking the design value θ in the overall dimension design stage as a known parameter 1min ,θ 1max ,θ 2min ,θ 2max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0021]

[0022] Where: Y i is the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are θ 1min ,θ 1max ,θ 2min ,θ 2max ;y i It is the predicted value of the corresponding parameter calculated by the mathematical model of the boom and arm angle design stage;

[0023] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set variables: length AB of the connecting line between hinge points A and B, length DE of the connecting line between hinge points D and E, length EF of the connecting line between hinge points E and F, angle ∠OAB between hinge point connecting line OA and AB, angle ∠CBE between hinge point connecting line BE and BC, angle ∠BEC between hinge point connecting line BE and CE, angle ∠CED between hinge point connecting line CE and DE, angle ∠FEI between hinge point connecting line EF and EI, shortest installation distance AC of boom hydraulic cylinder min , The longest installation distance of boom hydraulic cylinder AC max , the shortest installation distance DF of the bucket hydraulic cylinder min , The longest installation distance DF of the bucket hydraulic cylinder max The boundary conditions are as follows:

[0024] lb i≤x i ≤ub i 1≤i≤12

[0025] Where: ub i lb i is the i-th variable x i The user-defined upper and lower bounds of , where x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ].

[0026] (3) Take the design variables x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints

[0027]

[0028] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min , AC max DF min DF max , and then proceed to the next stage of design;

[0029] Step 3: Bucket Angle Design Stage

[0030] (1) Taking the design value L2 of the overall dimension design stage and the design values ​​EF and ∠FEI of the boom and arm angle design stage as known parameters, the overall dimension design stage θ 3min ,θ 3max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0031]

[0032] Where: Y i is the design target value of the i-th design target parameter, that is, Y1 and Y2 are θ 3min ,θ 3max ;y i is the predicted value of the corresponding parameter calculated using the mathematical model of the bucket angle design stage;

[0033] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the SSA search threshold T S , adjustment coefficient m; set variables: length of the line between hinge points H and I HI, length of the line between hinge points G and I GI, length of the line between hinge points H and K HK, length of the line between hinge points K and J KJ, length of the line between hinge points I and J IJ, angle ∠EIH between hinge point line EI and HI, angle ∠FIG between hinge point line FI and GI, angle ∠JIL between hinge point line IJ and IL, shortest installation distance GK for bucket hydraulic cylinder min , Maximum installation distance of bucket hydraulic cylinder GK max The boundary conditions are as follows:

[0034] lb i ≤x i ≤ub i 1≤i≤10

[0035] Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of , where x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GK max ];

[0036] (3) Take the design variables x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GK max ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints. In addition, no interference should be ensured:

[0037]

[0038] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min , GK max , and then finish the design.

[0039] As a further explanation of the present invention, during the iteration of the improved slime mold algorithm, the feedback process of the slime mold quality is not used for feedback adjustment, so as to reduce the randomness of the algorithm iteration and improve the algorithm convergence performance.

[0040] Further explanation of the present invention, the improved slime mold algorithm adopts the optimal leader strategy and the sparrow-led search strategy to improve the standard slime mold algorithm; the optimal leader strategy indicates that the best individual in the population acts as a leader to lead the population for optimization, and other individuals act as its followers, which can effectively improve the convergence speed and accuracy of the algorithm, balance the global and local search capabilities of the algorithm, and define the best individual as the leader of the population. The formula for the position of the slime mold individual is expressed as:

[0041] X(t+1)=X b (t)+v b ·(W·X b (t)-X A (t))

[0042] Where: t is the current iteration number; X b (t) is the individual position with the best current fitness; X A (t) is the random individual position; v b is a random number between [-b, b], which determines the moving direction of the slime mold individual; W is the weight of the slime mold, and its calculation formula is shown in the following formula;

[0043]

[0044] Where: SI = sort (S) represents the fitness value sequence of slime molds sorted from the best to the worst fitness value; condition represents the individuals whose fitness values ​​are ranked in the first half of the group; others represents the individuals whose fitness values ​​are ranked in the second half of the group; r represents a random number between [0, 1]; bF and wF represent the best fitness value and the worst fitness value in the current iteration (in this paper, the objective function is taken as fitness, and the minimum fitness value is the best);

[0045] The sparrow-led search strategy uses the sparrow search algorithm (SSA) to perform a global search in the early stage of algorithm iteration, and at the same time adjusts the parameter v b The boundary b is improved to perform local search in the later stage of algorithm iteration, improve the global search ability and local search ability of the improved slime mold algorithm, and thus improve the convergence performance of the algorithm. The definition of b is as follows:

[0046]

[0047] Where, t is the tth iteration; T is the maximum number of iterations; m is the adjustment coefficient, which adjusts the decreasing speed of the b value, that is, the larger the m, the faster the decrease;

[0048] The improved mathematical model of the improved slime mold algorithm is:

[0049]

[0050] Where, TS The threshold of the iterative search times of the sparrow search strategy led by sparrows through the sparrow search algorithm (SSA); v b is a random number in [-b, b], t is the current iteration number; r1, r2 are random numbers between [0, 1]; ub, lb are the upper and lower bounds of the search area; z is the conversion probability, used to switch between the global search stage and the local search stage of the algorithm, take z = 0.03; X b (t) is the position of the individual with the best fitness currently; X A (t) is a random individual position.

[0051] The specific steps of the improved slime mold algorithm are as follows:

[0052] Step 1: Set parameters: the maximum number of iterations T, the population number N, the threshold of the iterative search times T of the sparrow search algorithm SSA S and the adjustment coefficient m, and then initialize the population and calculate the fitness value; set the parameters of the algorithm and generate the initial values of the size parameters to be designed, and calculate the fitness value to understand its distance from the design goal, providing initial conditions for the design;

[0053] Step 2: Perform SSA search until the iteration number t is not less than T S , and then proceed to Step 3;

[0054] Step 3: Calculate the weight W;

[0055] Step 4: Generate a random number r1, and judge the size of the random number r1 and the parameter z. If r1 < z, update the individual position; otherwise, update the parameter b and calculate v b , and then update the individual position; update the size parameters according to the position update formula, in order to gradually reduce the distance between the size parameter combination and the design goal;

[0056] Step 5: Boundary condition constraint, calculate the fitness value, and update the global optimal solution; perform boundary constraint on the updated size parameters, and replace the values exceeding the boundary with a random value within the design range to prevent the designed size parameters from exceeding the design range; then calculate the fitness value and update the global optimal solution according to the fitness value, that is, the current optimal size parameter combination;

[0057] Step 6: Judge whether the end condition is satisfied, that is, the iteration number reaches the maximum number of iterations or other end conditions. If satisfied, output the global optimal solution and the fitness and end the algorithm; otherwise, execute Steps 3 to 6 in sequence. Check whether the algorithm operation meets the end requirements. If the end requirements are not met, repeatedly execute Steps 3 to 6; otherwise, output the optimal solution corresponding to the current optimal fitness, that is, the current optimal size parameter combination. This step ensures the iterative operation of the algorithm.

[0058] The beneficial effects of the present invention:

[0059] This method uses an improved slime mold algorithm for the optimization search process of the excavator working device design, automatically searches for the optimal size parameter combination, does not rely on the traditional trial and error method, reduces the need for manual intervention and repeated experiments, and thus greatly shortens the design cycle. Due to the complex coupling relationship between the size parameters of the working device, the improved slime mold algorithm can effectively handle these relationships and find a size parameter combination that meets all design requirements. This method uses the improved slime mold algorithm, so the design process does not need to preset size parameters, and the size parameters are automatically searched based on the design goal (working performance), which can directly convert the working performance requirements into size parameters. The application of the improved slime mold algorithm makes it possible to customize the excavator working device according to specific working performance requirements (design goals), rather than relying solely on standardized designs.

[0060] Instruction Manual

[0061] Figure 1 It is a schematic diagram of the process of the present invention;

[0062] Figure 2 A schematic diagram for reference of variables of the design method of the present invention;

[0063] Figure 3 To obtain the maximum number of iterations of 1000, the b parameters and v of the improved slime mold algorithm and the standard slime mold algorithm are b See the figure for comparison of oscillation vectors. DETAILED DESCRIPTION

[0064] The present invention will be further described below in conjunction with specific embodiments.

[0065] Embodiment 1:

[0066] A method for designing a working device of a hydraulic backhoe excavator comprises the following steps:

[0067] Step 1: Overall size design stage

[0068] (1) According to the design requirements, the design objectives are obtained, including the maximum digging depth h1, the maximum digging height h2, the maximum digging radius r1, the maximum unloading height h3, the ratio of the boom to the arm length K, the boom angle range Δθ1, the arm angle range Δθ2, and the bucket angle range Δθ3. h1, h2, r1, and h3 are used as the target values ​​of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0069]

[0070] Where: Y iis the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are h1, h2, r1, and h3 respectively; y i is the predicted value of the corresponding parameter calculated using the mathematical model in the overall size design stage;

[0071] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set the horizontal distance B from the hinge point between the variable boom and the turntable to the turntable rotation center x , vertical distance B from the hinge point of the boom and turntable to the horizontal ground y , boom length L1, arm length L2, bucket length L3, minimum boom angle θ 1min 、Maximum boom angle θ 1max 、Minimum arm angle θ 2min 、Maximum arm rotation angle θ 2max 、Minimum bucket rotation angle θ 3min 、Maximum bucket rotation angle θ 3max The boundary conditions are as follows:

[0072] lb i ≤x i ≤ub i 1≤i≤11

[0073] Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of x ,B y ,L1,L2,L3,θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max ];

[0074] (3) Take the design variable x = [B x ,B y ,L1,L3,θ 1min ,θ 2max ,θ 3max ], update the design variables L1, θ according to the following formula 1min ,θ 2max ,θ 3max The upper and lower bounds of :

[0075]

[0076] Where: ub, lb are the new upper and lower bounds of the corresponding design variables; lb0, lb1, lb2, lb3 are L1, θ 1min ,θ 2max ,θ 3max The original lower bound of ub0, ub1, ub2, and ub3 are L1, θ 1min ,θ 2max ,θ 3max The original upper bound; LB0, LB1, LB2, LB3 are L2, θ 1max ,θ 2min ,θ 3min The lower bound of UB0, UB1, UB2, and UB3 are L2, θ 1max ,θ 2min ,θ 3min The upper bound of

[0077] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables, and then obtain the variables L2 and θ according to the following formula 1max ,θ 2min ,θ 3min Finally, all design values ​​B of this design stage are output x , B y , L1, L2, L3, θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max , and then proceed to the next stage of design;

[0078]

[0079] Step 2: Boom and arm angle design stage

[0080] (1) Taking the design value L1 in the overall dimension design stage as a known parameter, taking the design value θ in the overall dimension design stage as a known parameter 1min ,θ 1max ,θ 2min ,θ 2max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0081]

[0082] Where: Y i is the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are θ 1min ,θ 1max ,θ 2min ,θ 2max ;yi It is the predicted value of the corresponding parameter calculated by the mathematical model of the boom and arm angle design stage;

[0083] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set variables: length AB of the connecting line between hinge points A and B, length DE of the connecting line between hinge points D and E, length EF of the connecting line between hinge points E and F, angle ∠OAB between hinge point connecting line OA and AB, angle ∠CBE between hinge point connecting line BE and BC, angle ∠BEC between hinge point connecting line BE and CE, angle ∠CED between hinge point connecting line CE and DE, angle ∠FEI between hinge point connecting line EF and EI, shortest installation distance AC of boom hydraulic cylinder min , The longest installation distance of boom hydraulic cylinder AC max , the shortest installation distance DF of the bucket hydraulic cylinder min , The longest installation distance DF of the bucket hydraulic cylinder max The boundary conditions are as follows:

[0084] lb i ≤x i ≤ub i 1≤i≤12

[0085] Where: ub i lb i is the i-th variable x i The user-defined upper and lower bounds of , where x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ].

[0086] (3) Take the design variables x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints:

[0087]

[0088] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min , AC max DFmin DF max , and then proceed to the next stage of design;

[0089] Step 3: Bucket Angle Design Stage

[0090] (1) Taking the design value L2 of the overall dimension design stage and the design values ​​EF and ∠FEI of the boom and arm angle design stage as known parameters, the overall dimension design stage θ 3min ,θ 3max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value:

[0091]

[0092] Where: Y i is the design target value of the i-th design target parameter, that is, Y1 and Y2 are θ 3min ,θ 3max ;y i is the predicted value of the corresponding parameter calculated using the mathematical model of the bucket angle design stage;

[0093] (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the SSA search threshold T S , adjustment coefficient m; set variables: length of the line between hinge points H and I HI, length of the line between hinge points G and I GI, length of the line between hinge points H and K HK, length of the line between hinge points K and J KJ, length of the line between hinge points I and J IJ, angle ∠EIH between hinge point line EI and HI, angle ∠FIG between hinge point line FI and GI, angle ∠JIL between hinge point line IJ and IL, shortest installation distance GK for bucket hydraulic cylinder min , Maximum installation distance of bucket hydraulic cylinder GK max The boundary conditions are as follows:

[0094] lb i ≤x i ≤ub i 1≤i≤10

[0095] Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of , where x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GK max ];

[0096] (3) Take the design variables x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GKmax ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints. In addition, no interference should be ensured:

[0097]

[0098] (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min , GK max , and then finish the design.

[0099] As a further explanation of the present invention, during the iteration of the improved slime mold algorithm, the feedback process of the slime mold quality is not used for feedback adjustment, so as to reduce the randomness of the algorithm iteration and improve the algorithm convergence performance.

[0100] Further explanation of the present invention, the improved slime mold algorithm adopts the optimal leader strategy and the sparrow-led search strategy to improve the standard slime mold algorithm; the optimal leader strategy indicates that the best individual in the population acts as a leader to lead the population for optimization, and other individuals act as its followers, which can effectively improve the convergence speed and accuracy of the algorithm, balance the global and local search capabilities of the algorithm, and define the best individual as the leader of the population. The formula for the position of the slime mold individual is expressed as:

[0101] X(t+1)=X b (t)+v b ·(W·X b (t)-X A (t))

[0102] Where: t is the current iteration number; X b (t) is the individual position with the best current fitness; X A (t) is the random individual position; v b is a random number between [-b, b], which determines the moving direction of the slime mold individual; W is the weight of the slime mold, and its calculation formula is shown in the following formula;

[0103]

[0104] Where: SI = sort (S) represents the fitness value sequence of slime molds sorted from the best to the worst fitness value; condition represents the individuals whose fitness values ​​are ranked in the first half of the group; others represents the individuals whose fitness values ​​are ranked in the second half of the group; r represents a random number between [0, 1]; bF and wF represent the best fitness value and the worst fitness value in the current iteration (in this paper, the objective function is taken as fitness, and the minimum fitness value is the best);

[0105] The sparrow-leading search strategy conducts global search in the early stage of algorithm iteration through the sparrow search algorithm (SSA), and at the same time improves the boundary b of the parameter v b to conduct local search in the later stage of algorithm iteration, improve the global search ability and local search ability of the improved slime mold algorithm, thereby improving the algorithm convergence performance. The definition of b is as follows:

[0106]

[0107] In the formula, t is the t-th iteration; T is the maximum number of iterations; m is a regulation coefficient that regulates the descent speed of the b value, that is, the larger m is, the faster the descent is;

[0108] The improved mathematical model of the slime mold algorithm is as follows:

[0109]

[0110] In the formula, T S is the threshold of the number of iteration searches of the sparrow-leading search strategy through the sparrow search algorithm (SSA); v b is a random number in [-b, b], t is the current iteration number; r1, r2 are random numbers between [0, 1]; ub, lb are the upper and lower bounds of the search area; z is the conversion probability used to switch between the global search stage and the local search stage of the algorithm, and z = 0.03 is taken; X b (t) is the position of the individual with the best fitness currently; X A (t) is the position of a random individual.

[0111] The specific steps of the improved slime mold algorithm are as follows:

[0112] Step 1: Set parameters: the maximum number of iterations T, the population size N, the threshold T of the number of iteration searches of the sparrow search algorithm SSA S and the regulation coefficient m, and then initialize the population and calculate the fitness value; set the parameters of the algorithm and generate the initial values of the size parameters to be designed, and calculate the fitness value to understand its distance from the design goal, providing initial conditions for the design;

[0113] Step 2: Conduct SSA search until the iteration number t is not less than T S , and then proceed to Step 3;

[0114] Step 3: Calculate the weight W;

[0115] Step 4: Generate a random number r1, and judge the size of the random number r1 and the parameter z. If r1 < z, update the individual position; otherwise, update the parameter b and calculate v b, and then update the individual position; update the size parameters according to the position update formula, in order to gradually reduce the distance between the size parameter combination and the design target;

[0116] Step 5: Boundary condition constraints, calculate fitness values, and update the global optimal solution; impose boundary constraints on the updated size parameters, and replace the values ​​beyond the boundary with a random value within the design range to prevent the designed size parameters from exceeding the design range; then calculate the fitness value and update the global optimal solution, i.e., the current optimal size parameter combination, according to the fitness value;

[0117] Step 6: Determine whether the termination condition is met, that is, the number of iterations reaches the maximum number of iterations or other termination conditions. If it is met, the global optimal solution and fitness are output and the algorithm ends; otherwise, steps 3 to 6 are executed in sequence. Check whether the algorithm operation meets the termination requirements. If it does not meet the termination requirements, steps 3 to 6 are executed repeatedly; otherwise, the optimal solution corresponding to the current optimal fitness, that is, the current optimal size parameter combination, is output. This step ensures the iterative operation of the algorithm.

[0118] Example 2

[0119] Take the maximum number of iterations as 1000, and compare the b parameter and v of ISMA with the standard slime mold algorithm. b The oscillation vectors are compared, such as Figure 3 As shown in the figure, during the algorithm iteration, ISMA has a smaller b value than the standard slime mold algorithm, thus limiting v b The size of helps ISMA to perform local search and improve the algorithm's search accuracy.

Claims

1. A method for designing a working device of a hydraulic backhoe excavator, characterized in that: The following steps are involved: Step 1: Overall size design stage (1) According to the design requirements, the design objectives are obtained, including the maximum digging depth h1, the maximum digging height h2, the maximum digging radius r1, the maximum unloading height h3, the ratio of the boom to the arm length K, the boom angle range Δθ1, the arm angle range Δθ2, and the bucket angle range Δθ3. h1, h2, r1, and h3 are used as the target values ​​of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value: Where: Y i is the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are h1, h2, r1, and h3 respectively; y i is the predicted value of the corresponding parameter calculated using the mathematical model in the overall size design stage; (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set the horizontal distance B from the hinge point between the variable boom and the turntable to the turntable rotation center x , vertical distance B from the hinge point of the boom and turntable to the horizontal ground y , boom length L1, arm length L2, bucket length L3, minimum boom angle θ 1min 、Maximum boom angle θ 1max 、Minimum arm angle θ 2min 、Maximum arm angle θ 2max 、Minimum bucket rotation angle θ 3min 、Maximum bucket rotation angle θ 3max The boundary conditions are as follows: lb i ≤x i ≤ub i 1≤i≤11 Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of x ,B y ,L1,L2,L3,θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max ]; (3) Take the design variable x = [B x ,B y ,L1,L3,θ 1min ,θ 2max ,θ 3max ], update the design variables L1, θ according to the following formula 1min ,θ 2max ,θ 3max The upper and lower bounds of : Where: ub, lb are the new upper and lower bounds of the corresponding design variables; lb0, lb1, lb2, lb3 are L1, θ 1min ,θ 2max ,θ 3max The original lower bound of ub0, ub1, ub2, and ub3 are L1, θ 1min ,θ 2max ,θ 3max The original upper bound; LB0, LB1, LB2, LB3 are L2, θ 1max ,θ 2min ,θ 3min The lower bound of UB0, UB1, UB2, and UB3 are L2, θ 1max ,θ 2min ,θ 3min The upper bound of (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables, and then obtain the variables L2 and θ according to the following formula 1max ,θ 2min ,θ 3min Finally, all design values ​​B of this design stage are output x , B y , L1, L2, L3, θ 1min ,θ 1max ,θ 2min ,θ 2max ,θ 3min ,θ 3max , and then proceed to the next stage of design; Step 2: Boom and arm angle design stage (1) Taking the design value L1 in the overall dimension design stage as a known parameter, taking the design value θ in the overall dimension design stage as a known parameter 1min ,θ 1max ,θ 2min ,θ 2max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value: Where: Y i is the design target value of the i-th design target parameter, that is, Y1, Y2, Y3, and Y4 are θ 1min ,θ 1max ,θ 2min ,θ 2max ;y i It is the predicted value of the corresponding parameter calculated by the mathematical model of the boom and arm angle design stage; (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the sparrow search algorithm SSA search threshold T S , adjustment coefficient m, set variables: length AB of the connecting line between hinge points A and B, length DE of the connecting line between hinge points D and E, length EF of the connecting line between hinge points E and F, angle ∠OAB between hinge point connecting line OA and AB, angle ∠CBE between hinge point connecting line BE and BC, angle ∠BEC between hinge point connecting line BE and CE, angle ∠CED between hinge point connecting line CE and DE, angle ∠FEI between hinge point connecting line EF and EI, shortest installation distance AC of boom hydraulic cylinder min , The longest installation distance of the boom hydraulic cylinder AC max , the shortest installation distance DF of the bucket hydraulic cylinder min , The longest installation distance DF of the bucket hydraulic cylinder max The boundary conditions are as follows: lb i ≤x i ≤ub i 1≤i≤12 Where: ub i lb i is the i-th variable x i The user-defined upper and lower bounds of , where x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ]. (3) Take the design variables x = [AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min ,AC max ,DF min ,DF max ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints: (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage AB, DE, EF, ∠OAB, ∠CBE, ∠BEC, ∠CED, ∠FEI, AC min , AC max DF min DF max , and then proceed to the next stage of design; Step 3: Bucket Angle Design Stage (1) Taking the design value L2 of the overall dimension design stage and the design values ​​EF and ∠FEI of the boom and arm angle design stage as known parameters, the overall dimension design stage θ 3min ,θ 3max The design value is the target value of the objective function. The subsequent improved slime mold algorithm uses the following formula to calculate the error value, i.e., the fitness value: Where: Y i is the design target value of the i-th design target parameter, that is, Y1 and Y2 are θ 3min ,θ 3max ;y i is the predicted value of the corresponding parameter calculated using the mathematical model of the bucket angle design stage; (2) Set the slime mold algorithm parameters, including the maximum number of iterations T, the population number N, and the SSA search threshold T S , adjustment coefficient m; set variables: length of the line between hinge points H and I HI, length of the line between hinge points G and I GI, length of the line between hinge points H and K HK, length of the line between hinge points K and J KJ, length of the line between hinge points I and J IJ, angle ∠EIH between hinge point line EI and HI, angle ∠FIG between hinge point line FI and GI, angle ∠JIL between hinge point line IJ and IL, shortest installation distance GK for bucket hydraulic cylinder min , Maximum installation distance of bucket hydraulic cylinder GK max The boundary conditions are as follows: lb i ≤x i ≤ub i 1≤i≤10 Where: ub i lb i is the i-th variable x i The custom upper and lower bounds of , where x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GK max ]; (3) Take the design variables x = [HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min ,GK max ], during the calculation process, the design variables should be guaranteed to satisfy the following constraints. In addition, no interference should be ensured: (4) Run the improved slime mold algorithm and use it to automatically search for design variables; after the algorithm runs, obtain the design values ​​of the design variables and output all the design values ​​of this design stage HI, GI, HK, KJ, IJ, ∠EIH, ∠FIG, ∠JIL, GK min , GK max , and then finish the design.

2. A method for designing a working device of a hydraulic backhoe excavator according to claim 1, characterized in that: During the iteration of the improved slime mold algorithm, the feedback process of the slime mold quality is not used for feedback adjustment, which reduces the randomness of the algorithm iteration and improves the algorithm convergence performance.

3. The method for designing a working device of a hydraulic backhoe excavator according to claim 1, characterized in that: The improved slime mold algorithm adopts the optimal leader strategy and the sparrow-led search strategy to improve the standard slime mold algorithm; the optimal leader strategy means that the best individual in the population acts as a leader to lead the population for optimization, and other individuals act as its followers, which can effectively improve the convergence speed and accuracy of the algorithm, balance the global and local search capabilities of the algorithm, and define the best individual as the leader of the population. The formula for the position of the slime mold individual is expressed as: X(t+1)=X b (t)+v b ·(W·X b (t)-X A (t)) Where: t is the current iteration number; X b (t) is the individual position with the best current fitness; X A (t) is the random individual position; v b is a random number between [-b, b], which determines the moving direction of the slime mold individual; W is the weight of the slime mold, and its calculation formula is shown in the following formula; Where: SI = sort (S) represents the fitness value sequence of slime molds sorted from the best to the worst fitness value; condition represents the individuals whose fitness values ​​are ranked in the first half of the group; others represents the individuals whose fitness values ​​are ranked in the second half of the group; r represents a random number between [0, 1]; bF and wF represent the best fitness value and the worst fitness value in the current iteration (in this paper, the objective function is taken as fitness, and the minimum fitness value is the best); The sparrow-led search strategy uses the sparrow search algorithm (SSA) to perform a global search in the early stage of algorithm iteration, and at the same time adjusts the parameter v b The boundary b is improved to perform local search in the later stage of algorithm iteration, improve the global search ability and local search ability of the improved slime mold algorithm, and thus improve the convergence performance of the algorithm. The definition of b is as follows: Where t is the tth iteration; T is the maximum number of iterations; m is the adjustment coefficient, which adjusts the decreasing speed of the b value, that is, the larger the m, the faster the decrease; The improved mathematical model of the improved slime mold algorithm is: Where, T S The threshold value of the number of iterative searches for the sparrow-led search strategy through the sparrow search algorithm (SSA); v b is a random number between [-b, b], t is the current iteration number; r1 and r2 are random numbers between [0, 1]; ub and lb are the upper and lower bounds of the search area; z is the conversion probability, which is used for the algorithm to switch between the global search stage and the local search stage, and z = 0.03; X b (t) is the individual position with the best current fitness; X A (t) is a random individual position. The specific steps of the improved slime mold algorithm are as follows: Step 1: Set parameters: maximum number of iterations T, population number N, and SSA iterative search threshold T S , adjust the coefficient m, and then initialize the population and calculate the fitness value; set the algorithm parameters and generate the initial values ​​of the size parameters to be designed, and calculate the fitness value to understand its distance from the design target, providing initial conditions for the design; Step 2: Perform SSA search until the number of iterations t is not less than T S , then proceed to step 3; Step 3: Calculate the weight W; Step: 4: Generate a random number r1, and determine the magnitude relationship between the random number r1 and the parameter z. If r1 < z, update the individual position; otherwise, update the parameter b and calculate v b , and then update the individual position; Update the size parameters according to the position update formula, so as to gradually reduce the distance between the size parameter combination and the design target; Step 5: Boundary condition constraints, calculate fitness values, and update the global optimal solution; boundary constraints are imposed on the updated size parameters, and values ​​beyond the boundary are replaced with a random value within the design range to prevent the designed size parameters from exceeding the design range; Then the fitness value is calculated and the global optimal solution, i.e. the current optimal size parameter combination, is updated according to the fitness value; Step 6: Determine whether the termination condition is met, that is, the number of iterations reaches the maximum number of iterations or other termination conditions. If it is met, the global optimal solution and fitness are output and the algorithm ends; otherwise, steps 3 to 6 are executed in sequence. Check whether the algorithm operation meets the termination requirements. If it does not meet the termination requirements, steps 3 to 6 are executed repeatedly; otherwise, the optimal solution corresponding to the current optimal fitness, that is, the current optimal size parameter combination, is output. This step ensures the iterative operation of the algorithm.

Images