Wiring harness section layout optimization design and simulation verification method

Abstract

The invention discloses a wire harness section layout optimization design and simulation verification method. The method comprises the following steps: S1, model establishment and legal judgment of a wire harness section layout pattern; s2, calculating an adjacent attribute value Lij of any two circles, and obtaining a non-adjacent circle judgment matrix of the whole layout; s3, calculating the real porosity of the wire harness section layout X; s4, according to the data index of each circle in the cross section layout X, on the basis of the non-adjacent circle judgment matrix, calculating to obtain extrusion elastic potential energy E under the cross section layout X; s5, according to the potential energy function U (X) of the pattern X, converging the initial pattern to a corresponding local optimal pattern U (X *) based on a gradient descent continuous optimization method; and S6, based on multiple iterations and searches of a multi-neighborhood operator combination strategy and a simulated annealing algorithm, outputting an optimal solution with a certain probability according to an annealing function, obtaining a final layout after global optimization, and outputting a layout result of each circle in the optimal layout. According to the invention, an optimal solution for wiring harness layout can be sought.

Description

Technical Field

[0001] The invention belongs to the field of secondary cable harness design of power systems, and in particular relates to a harness cross-section layout optimization design and simulation verification method based on a simulated annealing algorithm and a multi-neighborhood operator combination strategy. Background Art

[0002] The flexible wiring harness in the electrical system refers to the wiring harness (also called wire harness) formed by bundling and fixing cables, network cables, optical fibers, etc. The wiring harness often has multiple branches and a complex topological structure. The structural characteristics are long, thin and soft. Its ends are connected to relay plug-ins, terminal blocks, and aviation plugs to form a wiring harness assembly. The secondary cable harness, as the neural network of the secondary system of relay protection, transmits various types of photoelectric signals and is an important signal transmission medium.

[0003] Traditional cable layout design is to wait for the geometric mechanical structure of the product to be determined, and then install and debug it based on the actual product. It relies on the experience of wiring personnel, and has problems such as low installation efficiency, difficult to control the quality of wire harness laying, and unreasonable layout. It is difficult to adapt to complex spaces, high laying density, and wire harness laying layout scenarios with a large number of equipment. At present, the optimization, automatic layout, and path optimization of cable harnesses during movement are a research hotspot and difficulty. There is little research on the optimization of secondary cable harnesses in actual field applications in the power field, and there is a large room for optimization and improvement of cable harnesses.

[0004] During the on-site operation and maintenance, it was found that the unreasonable design of the wire harness cross section led to frequent wire harness failures, which in turn affected the stability and reliability of the entire system. The wire harness cross section problem involves the layout of the wire harness, the selection of wiring channels, the protection of the wire harness, and the size design of the wire harness.

[0005] Therefore, in the process of wiring harness arrangement, it is necessary to optimize the wiring channels and select appropriate wiring channels, such as wire troughs, wire troughs, pipes, etc., so that the wiring harness can be arranged in an orderly manner and easy to repair and replace. According to the shortest path principle: the wiring harness arrangement should try to use the shortest path, reduce the length of the wiring harness, and reduce the impact of resistance and voltage reduction. Reasonable protection of the wiring harness: Arrange the wiring harness in protective facilities, such as sheaths, protective tubes, etc. to prevent mechanical damage and the influence of the external environment. Wire harness size design: According to the power, current and voltage requirements of the system, reasonably select the cross-sectional area of ​​the wiring harness to ensure the stability and reliability of power transmission. Isolate high and low voltage harnesses: The high-voltage harness should be isolated from the low-voltage harness to prevent mutual interference and safety accidents.

[0006] The wiring harness layout should be designed according to the specific application scenario and system requirements. Reasonable wiring harness layout can ensure the stability of signal transmission and avoid problems such as electromagnetic interference and voltage reduction. In the actual wiring harness layout design, it is also necessary to consider the specific application scenario, equipment layout, and relevant standards and specifications. At the same time, using tools such as CAD software for three-dimensional design and simulation analysis of wiring harnesses can help optimize wiring harness layout and improve system performance and reliability.

[0007] Reasonable wiring harness layout can reduce the length and volume of the wiring harness, reduce the crossover and interference of the wiring harness, and improve the heat dissipation performance of the wiring harness, thereby improving the performance and reliability of the entire electrical system. Optimization of wiring harness material selection: Selecting the right wiring harness material can improve the heat resistance, corrosion resistance and insulation performance of the wiring harness, thereby increasing the service life and reliability of the wiring harness. At the same time, the right material selection can also reduce the weight and cost of the wiring harness.

[0008] In order to establish an accurate model of the wire harness in the design stage, the wire harness fault location and evaluation can be achieved through simulation motion analysis of the wire harness. It is necessary to determine the wire harness cross-section and accurate and reliable parameters such as wire harness stiffness. Summary of the invention

[0009] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a wiring harness cross-section layout optimization design and simulation verification method.

[0010] A technical solution to achieve the above purpose is: a wiring harness cross-section layout optimization design and simulation verification method, comprising the following steps:

[0011] S1, model establishment and legality judgment of wiring harness cross-section arrangement pattern;

[0012] S2, calculate the adjacency attribute value Lij of any two circles, and obtain the non-adjacent circle judgment matrix of the entire layout;

[0013] S3, calculate the real porosity of the wire bundle cross-section layout X;

[0014] S4, according to the data index of each circle in the cross-sectional layout X, on the basis of the non-adjacent circle judgment matrix, calculate and obtain the extrusion elastic potential energy E under the cross-sectional layout X;

[0015] S5, according to the potential energy function U(X) of the pattern X, the initial pattern is converged to the corresponding local optimal pattern U(X*) based on the continuous optimization method of gradient descent;

[0016] S6, based on multiple iterations and searches of multi-neighborhood operator combination strategies and simulated annealing algorithms, outputs the optimal solution with a certain probability according to the annealing function, obtains the final layout after global optimization, and outputs the layout results of each circle in the optimal layout.

[0017] Furthermore, the step S1 specifically includes:

[0018] S11, establishing a physical model X=(x1, y1, x2, y2, ..., xn, yn) of the cross-sectional arrangement of the wire harness;

[0019] S12, introducing illegal and legal judgment criteria for the wiring harness cross-section arrangement pattern, and establishing constraint conditions for the legal pattern of the wiring harness cross-section arrangement.

[0020] Furthermore, the step S2 specifically includes:

[0021] S21, define and calculate the intrusion depth Dij between any two circles, and use this as a basis to define the non-adjacent attribute value Lij of the two circles;

[0022] S22, calculating and obtaining the non-adjacent attribute values ​​of any two circles, and establishing a non-adjacent circle determination matrix A.

[0023] Furthermore, the step S3 specifically includes:

[0024] S31, calculating the ratio of the sum of the areas of each circle to the area of ​​the minimum enclosing circle as the porosity, and using the porosity as a basis for evaluating the compactness and space utilization of each circle in a certain cross-sectional layout, and as an evaluation and acceptance criterion for each solution in the iterative optimization of the harness layout;

[0025] S32, considering the influence of the discrete degree of the area data of each circle in the wiring harness cross-section layout on the evaluation of space utilization, a porosity correction coefficient δ is introduced, and the calculated true porosity is used as the acceptance criterion of the optimal solution of the wiring harness cross-section layout.

[0026] Furthermore, the step S4 specifically includes:

[0027] S41, calculate the elastic potential energy W1 between each small circle in the cross-sectional layout of the harness and the outer enveloping large circle: the outer sheath material of the cable harness is a nonlinear material, and its elastic behavior is described by a viscoelastic model. According to its stress-strain curve, the area enclosed by the curve is obtained by integration, and the extrusion elastic potential energy W1 is obtained by solution;

[0028] S42, calculate the elastic potential energy W2 between the small circles in the wiring harness cross-sectional layout: the material hardness of the cable insulation layer is relatively large, and its deformation is small after being squeezed to a certain extent. There is a linear relationship between the stress and strain of elastic solids within a small deformation range. Therefore, the extrusion elastic potential energy W2 between the cables is calculated using Hooke's law.

[0029] Furthermore, in step S4, the data indicators of each circle specifically include the material, diameter, spring constant, and penetration depth of each circle.

[0030] Furthermore, the step S5 specifically includes:

[0031] S51, defining a potential energy function U(X) of the pattern X, and calculating the gradient functions h(x) and h(y) of U(X) in the X direction and the Y direction;

[0032] S52, set the initial gradient values ​​x0 and y0, the learning rate α, the energy value convergence accuracy δ, the step size convergence accuracy σ, and the upper limit of the number of iterations m; and continuously iterate and search for the minimum value of the elastic potential energy of U(X);

[0033] S53, output the minimum value W(X)min of the elastic potential energy of the layout U(X), the specific coordinates of the layout X*(x1*, y1*, x2*, y2*, ..., xn*, yn*), and the number of iterations m.

[0034] Furthermore, the step S6 specifically includes:

[0035] S61, for the local optimal layout X*, call multiple neighborhood operators according to the combination strategy;

[0036] S62, use simulated annealing algorithm to perform global optimization of the pattern based on probability: by accepting a certain probability of poor solution during the search process to avoid falling into the local optimum, and accepting better solutions with a higher probability, a better solution is found in the global search space.

[0037] Furthermore, the multi-neighborhood operator includes a tossing operator, an exchange operator and an oscillation operator.

[0038] Furthermore, the layout results of each circle in the optimal layout are specifically output as the center position Oij,ri, the equivalent outer diameter R of the harness and the gap ratio φ.

[0039] The beneficial effects of the wiring harness cross-section layout optimization design and simulation verification method of the present invention are:

[0040] First, the present invention provides a feasible idea for optimizing the design of wire harness cross-section arrangement by converting the problem of wire harness cross-section arrangement optimization into an unequal circle arrangement optimization model. It has good performance and effect in solving the problem of wire harness cross-section arrangement, is applicable to circles of different numbers and shapes, and can obtain near-optimal layout solutions. It is efficient in solving and has good flexibility and robustness. In addition, the computational complexity of the algorithm is relatively low, and it can be efficiently solved in practical applications. It has practical guiding significance in the design stage and on-site construction of secondary cable flexible harnesses. Secondly, the adjacency attribute values ​​proposed in the present invention and the non-adjacent circle judgment matrix introduced can improve the calculation speed of the extrusion elastic potential energy between circles, reducing the computational complexity of the underlying algorithm from O(N). 2) is reduced to O(N), which greatly shortens the calculation time and meets the practical needs. Secondly, the present invention incorporates factors such as the material of the wire harness, the size of the wire harness, the linear elastic model and the viscoelastic model into the consideration of the total elastic potential energy of the entire layout, so that the overall energy calculation of the layout X is more in line with the actual situation and the measurement results are more accurate and reliable. Finally, the layout optimization design is realized based on the multi-neighborhood operator combination strategy and the simulated annealing algorithm. During the search process, a certain probability of differential solution is accepted to avoid falling into the local optimum, and a better solution is accepted with a higher probability. Finally, the center position O of each circle in the optimal layout is given. i (x i ,y i ), equivalent radius, porosity, energy value and other specific data, which are very practical and operational in subsequent wiring harness design and on-site installation scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a wiring harness cross-section layout optimization design and simulation verification method of the present invention;

[0042] Figure 2 The problem of optimizing the arrangement of wire harness cross sections is transformed into a schematic diagram of the unequal circle Packing model;

[0043] Figure 3 Schematic diagram of the penetration depth between any two circles in the wiring harness cross-section layout;

[0044] Figure 4 Stress-strain curve of PVC (polyvinyl chloride);

[0045] Figure 5 Flowchart of local optimal solution of wire harness cross section based on gradient descent;

[0046] Figure 6 Cross-sectional layout of wire harness sample 1. DETAILED DESCRIPTION

[0047] In order to better understand the technical solution of the present invention, the following is a detailed description through specific embodiments:

[0048] The optimization design and simulation verification method of wiring harness cross-section layout based on multi-neighborhood operator combination strategy and simulated annealing algorithm is as follows: Figure 1 As shown, the following steps are included:

[0049] S1, Model establishment and legality judgment of wire harness cross-section arrangement pattern.

[0050] a) Convert the wire harness cross-section arrangement pattern into an unequal circle Packing model. The wire harness cross-section arrangement model is as follows Figure 2As shown, constraints are formulated on this basis to determine whether it is a legal pattern. The specific process is as follows:

[0051] The center of the large disk is (x0, y0), its radius is denoted as R0, and the radius of the other circles is denoted as r i (i=1,2,…,n), the corresponding coordinates of the circle center are recorded as (x i ,y i ), the problem of wiring harness cross-section arrangement is transformed into finding a pattern X = (x1, y1, x2, y2, ..., x n ,y n ), so that

[0052] Minimize R0(1)

[0053]

[0054] Constraint (2) requires that each circle is within the range of the large circle; Constraint (3) requires that any two circles do not interfere with each other. A pattern that strictly satisfies all constraints is called a legal pattern, otherwise it is called an illegal pattern.

[0055]

[0056] L ij is the adjacency attribute value. From the above formula, we can see that: if and only if E(X) is 0, it is a legal layout.

[0057] b) Determine the cable model, quantity, cross-sectional area, coating material and other data that make up the wiring harness. Provide data support for establishing an accurate wiring harness model.

[0058] S2, non-adjacent circle judgment matrix and algorithm acceleration.

[0059] a) Define and calculate the penetration depth D between any two circles ij , and based on this, define the non-adjacent attribute value L of the two circles ij .

[0060] For any two circles in the layout X, r i (x i ,y i ) and r j (x j ,y j ), define the penetration depth D of the two circles ij for

[0061]

[0062] A schematic diagram of the penetration depth is shown in Figure 3 shown.

[0063] L ij =0,D ij >0, define circle r i (x i ,y i ) and r j (x j ,y j ) is a non-adjacent circle

[0064] L ij =1,D ij ≤0, define circle r i (x i ,y i ) and r j (x j ,y j ) is an adjacent circle

[0065] Non-adjacent circle judgment matrix A:

[0066]

[0067] L ij is the adjacency attribute value, when L ij = 0, it is determined that the two circles have no adjacent relationship. When calculating the total elastic potential energy of the entire pattern X, L ij = 0 is not included in the subsequent calculation and is defaulted to 0 (when i = j, L ij =0).

[0068] b) Calculate the non-adjacent attribute values ​​of any two circles in the wire harness cross-section layout and establish the non-adjacent circle judgment matrix A.

[0069] The introduction of the non-adjacent circle judgment matrix can speed up the calculation of the sum of the extrusion elastic potential energy between each circle, and can reduce the computational complexity of the underlying algorithm from O(N 2 ) is reduced to O(N), greatly shortening the calculation time T.

[0070] S3, calculate the true porosity of the wire harness cross-sectional layout X.

[0071] Considering that the proportion of large circles and small circles in different examples is different, it affects the optimization target of the solution. If the difference between the maximum and minimum values ​​in the example is large, and the numerical dispersion is higher, the data distribution is uneven, and the porosity requirement in the optimization target is lower.

[0072] The number of large circles is relatively large, and the number of small circles is relatively small, so the porosity requirement is relatively low. The number of large circles is relatively large, and the number of small circles is relatively small, so the porosity requirement is relatively high.

[0073] a) Calculation of the porosity of the wiring harness cross-sectional layout X

[0074] The porosity is calculated as the ratio of the sum of the areas of each circle to the area of ​​the minimum enclosing circle. The porosity is used as the basis for evaluating the compactness of the arrangement of each circle and the space utilization rate in a certain cross-sectional layout, and as one of the evaluation and acceptance criteria for each solution in the iterative optimization of the harness layout. Reducing the porosity is the goal of cross-sectional layout optimization.

[0075] The center of the large disk is (x0, y0), its radius is R0, its area is S0, and the radius of the other circles is r i (i=1,2,…,n), the area is S i (i=1,2,…,n), the corresponding coordinates of the circle center are recorded as (x i ,y i ), for the pattern X=(x1,y1,x2,y2,…,x n ,y n )

[0076] Porosity (α) = (cross-sectional area of ​​the smallest enclosing circle - sum of cross-sectional areas of each circle) / cross-sectional area of ​​the smallest enclosing circle × 100%

[0077]

[0078] b) Calculation of porosity correction factor δ and true porosity

[0079] Considering the influence of the discrete degree of area data of each circle on the evaluation of space utilization in the wiring harness cross-section layout, the porosity correction coefficient δ is introduced, and the calculated true porosity is used as the acceptance criterion for the optimal solution of the wiring harness cross-section layout.

[0080] Porosity correction factor δ:

[0081] Maximize m

[0082]

[0083] Acceptance criteria for corrected true porosity

[0084] S4, calculation of the extrusion elastic potential energy W of the cross-sectional arrangement.

[0085] In order to more accurately reflect the severity of extrusion under a specific layout, it is proposed to use the corresponding elastic model to calculate the total elastic potential energy of the wire harness cross-section layout according to the material properties of different parts of the wire harness. The specific process is as follows:

[0086] (a) First, calculate the elastic potential energy W1 between each small circle in the wiring harness cross-section layout and the outer envelope large circle

[0087] PVC (polyvinyl chloride) is a common wire harness outer wrapping material. As a typical nonlinear material, its elastic behavior can be described by the viscoelastic model. The viscoelastic model of PVC (polyvinyl chloride) outer wrapping material is shown in the figure below. Figure 4 As shown. Ignoring the influencing factors such as temperature and deformation rate, its elastic potential energy can be expressed by the relationship between stress and strain.

[0088] W1(ε)=∫σ·dε (9)

[0089] Among them, σ is stress and dε is small strain.

[0090] For common cable outer envelope coating materials, the extrusion elastic potential energy can be accurately solved by obtaining its stress-strain curve and integrating the area enclosed by the curve.

[0091] (b) Next, calculate the elastic potential energy W2 between the small circles in the wiring harness cross-section layout

[0092] The material hardness of the cable insulation layer is relatively high, and its deformation is small after being squeezed to a certain extent. There is a linear relationship between the stress and strain of elastic solids within a small deformation range, and the extrusion elastic potential energy between cables can be calculated using Hooke's law.

[0093] For any two circular cross-sections in the harness cross-section layout, whose diameters are D1 and D2 respectively, and the elastic modulus of the cable insulation material is E, is the pressure applied between the two cables.

[0094] According to Hooke's law, the elastic potential energy can be calculated by the following formula:

[0095] U=(1 / 2)*k*(ΔL) 2

[0096] Where U is the elastic potential energy, k is the spring constant, and ΔL is the deformation (compression distance) between the spheres. The total elastic potential energy can be expressed as:

[0097] U_total=U1+U2=(1 / 2)*k1*(ΔL) 2 +(1 / 2)*k2*(ΔL) 2 =(1 / 2)

[0098] *(ΔL)^2*(k1+k2)

[0099] Where U1 and U2 are the elastic potential energies of the first and second spheres, respectively. k1 and k2 are the spring constants of the two cables, respectively.

[0100] It should be noted that the above calculation is based on a simplified model, assuming that there is no sliding and rolling friction between the cables, and the deformation is small, so the spring constant can be approximated to be constant.

[0101] S5, according to the potential energy function U(X) of the pattern X, based on the gradient descent algorithm, the initial pattern is converged to the corresponding local optimal pattern U(X*) through continuous optimization.

[0102] a) Define the potential energy function U(X) of the pattern X

[0103] Define the potential energy function U(X) of the configuration X.

[0104]

[0105] Where: W 2 oi is the elastic potential energy between the ith disk and the large disk, which is defined as:

[0106] W 2 oi =W1 + W2 = (1 / 2) * k0 * (ΔL oi ) 2 + (1 / 2) * k i * (ΔL oi ) 2 (11)

[0107] k0 and k i are the spring constants of the outer enclosing circle and the i-th sphere respectively

[0108] in:

[0109] ΔL oi =0, when

[0110]

[0111] Where: W 2 ij is the elastic potential energy between the i-th and j-th disks, which is defined as W 2 ij =W i +W j =(1 / 2)*k i *(ΔL ij ) 2 +(1 / 2)*k j *(ΔL ij ) 2

[0112] k i and k jare the spring constants of the i-th sphere and the j-th sphere respectively

[0113] in,

[0114] ΔL ij =0, when

[0115]

[0116] From the above formula, we can know that: U(X)≥0, and it is a legal layout when and only when U(X) is 0. The process of optimizing the cross-section arrangement of the wire harness can be transformed into the process of reducing U(X).

[0117] Define and calculate the gradient functions h(x) and h(y) of U(X) in the X and Y directions.

[0118] (b) Based on the gradient descent algorithm, the initial pattern is converged to the corresponding local optimal pattern U(X*) through continuous optimization;

[0119] Set the initial gradient values ​​x0 and y0, the learning rate α, the energy value convergence accuracy δ, the step size convergence accuracy σ, and the upper limit of the number of iterations m; and continuously iterate to search for the minimum value of the elastic potential energy of U(X).

[0120] The algorithm flow is:

[0121] 1) Import pattern X =(x1,y1,x2,y2,…,x n ,y n ), including the center coordinates of each circle (x i ,y i ), the radius of each circle is r i (i=1, 2,…, n).

[0122] 2) For the pattern X, calculate the gradient of the objective function U(X) in the current pattern. The gradient functions in the x and y directions are: and Find the values ​​of h(x) and h*y) respectively.

[0123] 3) Set the initial gradient values ​​x0 and y0, the iteration step size α, the energy value convergence accuracy δ, the step size convergence accuracy σ, and set the upper limit of the number of iterations m.

[0124] 4) Update the parameter value according to the direction of the gradient. The parameter update formula is: Where learning_rate is the learning rate, which controls the update step size of each step.

[0125] 5) After updating the parameters, calculate U(X) again k+1 ) energy value W(X k+1 ),if

[0126] |W(X k+1 )-W(X k )|<δ,|x k+1 -x k |<σ, or |y k+1 -y k |<σ, or the number of iterations reaches the upper limit m, then the iterative search stops and the current layout X*(x1*, y1*, x2*, y2*, …, xn*, yn*) is output.

[0127] 6) Otherwise, k=k+1, go to 4) and continue iterative search;

[0128] 7) Output the minimum value W(X) of the elastic potential energy of the layout U(X) min , layout the specific coordinates of X*(x1*, y1*, x2*, y2*,…, xn*, yn*), the number of iterations m, etc.

[0129] The algorithm flow chart of the local optimal solution of the wire section based on gradient descent is as follows Figure 5 shown.

[0130] S6, a global layout optimization method based on multi-neighborhood operator combination strategy and simulated annealing algorithm.

[0131] a) Use the throwing operator, the swap operator, and the oscillation operator once each to perform a multi-neighborhood operator combination call on the layout X, make small changes to the layout, and generate a new layout X*.

[0132] Toss operator: randomly picks a circle and "tosses" it somewhere randomly within the layout;

[0133] Exchange operator: Exchange the positions of two nodes to change the layout, and exchange the positions of any two unequal circles in the current layout;

[0134] Oscillation operator: Changes the layout by slightly oscillating the nodes in the current layout. You can apply a small offset or rotation angle to each node, and then evaluate the quality of the new layout by shifting all circles slightly in their own random directions.

[0135] b) Use simulated annealing algorithm to perform global optimization of the pattern based on probability.

[0136] The simulated annealing algorithm is a probability-based global optimization algorithm that avoids falling into the local optimum by accepting a certain probability of poor solution during the search process, and accepts a better solution with a higher probability.

[0137] The simulated annealing algorithm is selected as the acceptance criterion for a certain layout to find the optimal solution in the spatial layout. After each iteration, a new layout is produced through a combination strategy of multiple neighborhood operators. The algorithm calculates the probability of accepting a new solution based on the difference between the current solution and the new solution and the current temperature. A common formula for calculating the probability is: P = exp(-ΔE / T), which is the annealing function.

[0138] P=1,ΔU>0

[0139]

[0140] Among them, ΔU=U(X)-U(X * )

[0141] Among them, P represents the probability of accepting a new solution, exp represents the natural exponential function, ΔU represents the energy difference between the current solution and the new solution, and T* represents the current temperature.

[0142] When ΔU is positive, it means that the new solution is better than the current solution, so the probability P will be 1, that is, the new solution will definitely be accepted. When ΔU is negative, it means that the new solution is worse than the current solution, and the probability P will change according to the size of the energy difference and the influence of the current temperature. When the temperature is high, the probability of accepting a worse solution is higher, and as the temperature decreases, the probability of accepting a worse solution will also decrease.

[0143] By continuously changing the temperature, the simulated annealing algorithm will gradually reduce the probability of accepting a bad solution, so that the algorithm tends to accept a better solution during the search process. In this way, the algorithm can find a better solution in the global search space.

[0144] Embodiment 2:

[0145] The following is a detailed description of the embodiment of the optimized layout of the cross-section of the wire harness in conjunction with the diagram. It should be emphasized that the following description is only exemplary and is not intended to limit the scope and application of the present invention.

[0146] The present invention is further explained by taking the cable harness sample in the AC and DC panel cabinet of the 110kV Xutang substation in Songjiang area power grid as an example.

[0147] A wire harness cross-section layout optimization design and simulation verification method based on simulated annealing algorithm and multi-neighborhood operator combination strategy, the steps of this method are:

[0148] The specific process of step 1) is:

[0149] For wire harness sample 1, determine the cable model, quantity, cross-sectional area, coating material and other data that make up the wire harness based on the parameters of common cable types listed in the table below.

[0150] Common cable models and parameter table

[0151]

[0152]

[0153] According to statistics, the wire harness sample 1 is composed of 4 cables with a diameter of 2.96mm, 2 cables with a diameter of 3.64mm, and 1 cable with a diameter of 4.72mm. The initial pattern of the cross-sectional arrangement of the wire harness is measured as X = (x1, y1, x2, y2, ..., xn, yn). As shown in the following table

[0154] Components of wiring harness sample 1

[0155]

[0156] The center of the large disk is (x0, y0), its radius is denoted as R0, and the radius of the other circles is denoted as r. i (i=1,2,…,n), the corresponding coordinates of the circle center are recorded as (x i ,y i ). The cross-sectional arrangement pattern of the wire harness sample 1 is converted into an unequal circle packing model. The wire harness cross-sectional arrangement model is as follows: Figure 6 As shown, and on this basis, it is determined whether the initial pattern is a legal pattern.

[0157] After calculation and verification, the pattern X = (x1, y1, x2, y2, ..., xn, yn) cannot simultaneously satisfy the constraints:

[0158] Minimize R0

[0159]

[0160] Constraint (2) requires that each circle is within the range of the large circle; Constraint (3) requires that any two circles do not interfere with each other, and X is judged to be an illegal layout.

[0161] The specific process of step 2) is:

[0162] Calculate any two circles in the layout X, r i (x i ,y i ) and r j (x j ,y j The penetration depth of ) is:

[0163]

[0164] in:

[0165] in:

[0166] When Dij >0, define circle r i (x i ,y i ) and r j (x j ,y j ) are non-adjacent circles, there is no extrusion elastic potential energy between them, and the adjacent attribute L ij =0.

[0167] When D ij ≤0, define circle r i (x i ,y i ) and r j (x j ,y j ) are adjacent circles, there is no extrusion elastic potential energy between them, and the adjacent attribute L ij =1.

[0168] The non-adjacent attribute values ​​of any two circles in the wire harness section layout can effectively eliminate the calculation of energy between non-adjacent circles when subsequently calculating the energy value in the wire harness layout by establishing the non-adjacent circle judgment matrix A.

[0169] Non-adjacent circle judgment matrix A:

[0170]

[0171] The amount of calculation in the iterative process of wire harness section layout optimization based on the simulated annealing algorithm acceptance criterion and the multi-neighborhood operator combination strategy is reduced from C(n,k)=n! / (k!*(nk)!) to the cumulative number of times Lij=0.

[0172] In this example, the number of operations is reduced from n(n-1) / 2=28 times to 10 times when Lij=1, and the number of operations is reduced by 64.28%, which significantly improves the operation speed.

[0173] The specific process of step 3) is:

[0174] In the pattern X = (x1, y1, x2, y2, ..., xn, yn), the area of ​​each circle is Si (i = 1, 2, ..., n), and the corresponding coordinates of the center of the circle are recorded as (xi, yi)

[0175] Porosity (α) = (cross-sectional area of ​​the smallest enclosing circle - sum of cross-sectional areas of each circle) / cross-sectional area of ​​the smallest enclosing circle × 100%

[0176]

[0177] The diameter of the smallest enclosing circle in the wire harness cross-section arrangement pattern example 1 is 11.32 mm. Substituting the above data into the above formula, we can get α = 34.59%

[0178] Porosity correction factor δ:

[0179]

[0180] The maximum value of m in the above formula is 4, and the porosity correction factor

[0181]

[0182] The corrected true porosity φ = δ·α = 0.3459 × 0.5714 = 0.1977

[0183] In this example, the true porosity will be used as one of the acceptance criteria for the optimal solution in the subsequent iterative process of wire harness cross-section layout optimization.

[0184] The specific process of step 4) is:

[0185] The elastic behavior of PVC (polyvinyl chloride) conforms to the viscoelastic model, and its stress-strain curve is as follows: Figure 4 As shown, by integrating the area enclosed by the curve, the extrusion elastic potential energy W1 between the internal cables of the wiring harness and the external covering material can be accurately solved.

[0186] W1(ε)=∫(σ·dε)

[0187] According to the total elastic potential energy formula between the internal cables:

[0188] U ij _total=U i +U j =(1 / 2)*(ΔL)^2*(k i +k j ), the total elastic potential energy W2 can be obtained:

[0189]

[0190] Among them, U i and U j are the elastic potential energies of the first sphere and the second sphere respectively.

[0191] Among them, k i and k j are the spring constants of the two cables (generally between 1.0-4GPa).

[0192] The elastic potential energy of the entire pattern is W total :

[0193]

[0194] The specific process of step 5) is:

[0195] According to the potential energy function U(X) of the pattern X, based on the gradient descent algorithm, the initial pattern is converged to the corresponding local optimal pattern U(X*) through continuous optimization;

[0196] In order to verify that the continuous optimization method based on the gradient descent method converges to the local optimal pattern, two examples were randomly called, and the gradient descent method, a classic continuous optimization method, was used for continuous optimization, which can converge from any initial pattern to the corresponding local optimal pattern.

[0197] The specific process of step 6) is:

[0198] For the initial layout example X1, use the throwing operator, the exchange operator, and the oscillation operator to make a small perturbation to the layout once, and generate a new layout X*

[0199] Toss operator: randomly picks a circle and "tosses" it somewhere randomly within the layout;

[0200] Exchange operator: Exchange the positions of two nodes to change the layout, and exchange the positions of any two unequal circles in the current layout;

[0201] Oscillation operator: Changes the layout by slightly oscillating the nodes in the current layout. You can apply a small offset or rotation angle to each node, and then evaluate the quality of the new layout by shifting all circles slightly in their own random directions.

[0202] The simulated annealing algorithm is selected as the acceptance criterion for a certain layout to find the optimal solution in the spatial layout. After each iteration, a new layout is produced through a multi-neighborhood operator combination strategy. The algorithm calculates the probability of accepting the new solution based on the difference between the current solution and the new solution and the current temperature. The algorithm flow is as follows Figure 1 shown.

[0203] Four wire harness samples were selected to verify the algorithm operation. The results showed that the algorithm operation speed and the error of the operation result were within the design range and met the practical requirements. For any given wire harness cross section, the optimized cross-section arrangement pattern can be given. The algorithm can find the optimal solution in the global search space.

[0204] Those skilled in the art should recognize that the above embodiments are only used to illustrate the present invention, and are not intended to limit the present invention. As long as they are within the spirit of the present invention, any changes or modifications to the above embodiments will fall within the scope of the claims of the present invention.

Claims

1. A wire harness cross-section layout optimization design and simulation verification method, characterized in that: The steps include: S1, model establishment and legality judgment of wiring harness cross-section arrangement pattern; S2, calculate the adjacency attribute value Lij of any two circles, and obtain the non-adjacent circle judgment matrix of the entire layout; S3, calculate the real porosity of the wire bundle cross-section layout X; S4, according to the data index of each circle in the cross-sectional layout X, on the basis of the non-adjacent circle judgment matrix, calculate and obtain the extrusion elastic potential energy E under the cross-sectional layout X; S5, according to the potential energy function U(X) of the pattern X, the initial pattern is converged to the corresponding local optimal pattern U(X*) based on the continuous optimization method of gradient descent; S6, based on multiple iterations and searches of multi-neighborhood operator combination strategies and simulated annealing algorithms, outputs the optimal solution with a certain probability according to the annealing function, obtains the final layout after global optimization, and outputs the layout results of each circle in the optimal layout.

2. A wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S1 specifically includes: S11, establishing a physical model X=(x1, y1, x2, y2, ..., xn, yn) of the cross-sectional arrangement of the wire harness; S12, introducing illegal and legal judgment criteria for the wiring harness cross-section arrangement pattern, and establishing constraint conditions for the legal pattern of the wiring harness cross-section arrangement.

3. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S2 specifically includes: S21, define and calculate the intrusion depth Dij between any two circles, and use this as a basis to define the non-adjacent attribute value Lij of the two circles; S22, calculating and obtaining the non-adjacent attribute values ​​of any two circles, and establishing a non-adjacent circle determination matrix A.

4. The wiring harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S3 specifically includes: S31, calculating the ratio of the sum of the areas of each circle to the area of ​​the minimum enclosing circle as the porosity, and using the porosity as a basis for evaluating the compactness and space utilization of each circle in a certain cross-sectional layout, and as an evaluation and acceptance criterion for each solution in the iterative optimization of the harness layout; S32, considering the influence of the discrete degree of the area data of each circle in the wiring harness cross-section layout on the evaluation of space utilization, a porosity correction coefficient δ is introduced, and the calculated true porosity is used as the acceptance criterion of the optimal solution of the wiring harness cross-section layout.

5. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S4 specifically includes: S41, calculate the elastic potential energy W1 between each small circle in the cross-sectional layout of the harness and the outer enveloping large circle: the outer sheath material of the cable harness is a nonlinear material, and its elastic behavior is described by a viscoelastic model. According to its stress-strain curve, the area enclosed by the curve is obtained by integration, and the extrusion elastic potential energy W1 is obtained by solution; S42, calculate the elastic potential energy W2 between the small circles in the wiring harness cross-sectional layout: the material hardness of the cable insulation layer is relatively large, and its deformation is small after being squeezed to a certain extent. There is a linear relationship between the stress and strain of elastic solids within a small deformation range. Therefore, the extrusion elastic potential energy W2 between the cables is calculated using Hooke's law.

6. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: In step S4, the data indicators of each circle specifically include the material, diameter, spring constant, and penetration depth of each circle.

7. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S5 specifically includes: S51, defining a potential energy function U(X) of the pattern X, and calculating the gradient functions h(x) and h(y) of U(X) in the X direction and the Y direction; S52, set the initial gradient values ​​x0 and y0, the learning rate α, the energy value convergence accuracy δ, the step size convergence accuracy σ, and the upper limit of the number of iterations m; and continuously iterate and search for the minimum value of the elastic potential energy of U(X); S53, output the minimum value W(X)min of the elastic potential energy of the layout U(X), the specific coordinates of the layout X*(x1*, y1*, x2*, y2*, ..., xn*, yn*), and the number of iterations m.

8. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The step S6 specifically includes: S61, for the local optimal layout X*, call multiple neighborhood operators according to the combination strategy; S62, use simulated annealing algorithm to perform global optimization of the pattern based on probability: by accepting a certain probability of poor solution during the search process to avoid falling into the local optimum, and accepting better solutions with a higher probability, a better solution is found in the global search space.

9. A wire harness cross-section layout optimization design and simulation verification method according to claim 8, characterized in that: Multi-neighborhood operators include throwing operators, exchange operators, and oscillation operators.

10. The wire harness cross-section layout optimization design and simulation verification method according to claim 1, characterized in that: The layout results of each circle in the output optimal layout are specifically the center position Oij,ri, the equivalent outer diameter R of the harness and the gap ratio φ.

Images