Ship pipeline path optimization method based on improved artificial chimpanzee army algorithm

Abstract

A ship pipeline path optimization method based on an improved artificial chimpanzee army comprises the steps that firstly, chaotic mapping is used for initializing a population position, after an objective function is determined, Logistic chaos is used for creating an initialized population, chimpanzee is randomly divided into three types, namely, the chimpanzee is migrated to an unknown position, a known position and other chimpanzee positions, and the chimpanzee is migrated to the unknown position, the known position and the other chimpanzee positions; the method is used for quickly searching the position of the silver back ompanzee. And then, in combination with a position updating strategy in an eagle optimization algorithm, optimizing a silver-back-type chimpanzee position in a chimpanzee army through a silver-back-type competition mechanism. And finally, fusing the Levy flight strategy in a later stage to update the individual position, introducing external impanzee to compete with silver-back impanzee, avoiding too fast convergence, reaching a preset number of iterations, and outputting an optimal solution. The method guarantees the diversity of population in the later period of the algorithm, jumps out of a local optimal solution, expands the searching capability, and is applied to ship pipeline path planning.

Description

Technical Field

[0001] The present invention belongs to the field of ship pipeline design and construction, and particularly relates to a ship pipeline path optimization method based on an improved artificial gorilla troops algorithm. Background Art

[0002] The essence of ship pipeline layout design is to find a connected path between equipment interfaces in three-dimensional space and meet the path legality, effectiveness, and optimization. Current mainstream CAD software, such as CATIA, FORAN, TRIBON, and SolidWorks, although provides an automatic routing module, can only generate simple pipe paths in a limited mode and does not consider obstacle avoidance, and requires manual adjustment by designers. In addition, ship production has the characteristics of customization, and pipeline design often needs to be reworked with the adjustment of equipment layout, resulting in a large amount of manual design work and low efficiency. Therefore, exploring an automatic pipeline layout design algorithm is of great significance for improving ship design efficiency, and can also provide reference for pipeline or cable layout in the fields of aeroengines, building spaces, and mechatronic equipment, as well as robot path planning.

[0003] Research on pipeline layout has received extensive attention from scholars at home and abroad. Representative research includes: Park et al. were the first to propose using the unit generation method to automatically layout pipelines in a ship engine room; Asmara et al. compared the differences of various ship pipeline design algorithms and proposed a solution strategy that combines a heuristic algorithm (PSO) with a deterministic algorithm (A*); Fan et al. were the first to apply the ant colony algorithm to ship pipeline layout design. Jiang et al. and Wang et al. respectively improved the ant colony algorithm for ship pipe laying using co-evolution and human-machine combination strategies, but the algorithm efficiency and layout diversity still need to be improved; Liu et al. proposed an ant colony path-finding algorithm based on an adaptive dynamic adjustment strategy and applied it to two-dimensional plane pipeline design; Niu et al. proposed to layout ship pipelines based on the multi-objective optimization algorithm NSGA-II, reflecting the value of providing multiple pipe laying schemes at the same time. The above research methods have developed from solving small-scale simple problems to solving complex practical problems, but there are still deficiencies. The main reason is that ship pipeline layout is a multi-objective optimization problem under complex constraints, with great difficulty in solving. A single method has limitations, and it is more feasible to select a suitable algorithm according to user requirements.

[0004] The Gorilla Troop Optimization (GTO) algorithm is a global optimization meta-heuristic algorithm proposed by Abdollahzadeh et al. that simulates the social behavior and movement patterns of gorillas in the wild. The general process of the GTO algorithm is mainly divided into two major stages: exploration and exploitation. The GTO algorithm has the advantages of strong optimization ability and fast convergence speed, and is applied to research in engineering fields such as multi-objective optimization, annual runoff time series prediction, and distribution network reconfiguration. However, the GTO algorithm also has problems such as weak global optimization ability in the early stage, weak convergence ability in the later stage, and being prone to falling into local optima. In view of the above deficiencies, the present invention proposes a ship pipeline path optimization method based on an improved Gorilla Troop algorithm with a fusion strategy. Summary of the Invention

[0005] To solve the above problems, the present invention provides a ship pipeline path optimization method based on an improved Gorilla Troop algorithm, and the technical solution adopted is as follows:

[0006] The ship pipeline path optimization method based on the improved Gorilla Troop algorithm has the following specific operation steps:

[0007] Step 1: Determine the objective function and preset the maximum number of iterations.

[0008] The essence of ship pipeline layout is a multi-objective optimization problem under certain constraints, and the objectives and constraints can be described in the following form:

[0009]

[0010] s.t. g(path) = 0, h((path) = 0 (1).

[0011] Among them, min f length (path) calculates the path length, min f bend (path) calculates the number of path bends, minf power (path) calculates the energy value of the positions passed by the path, min f pocket (path) calculates the number of "concave pocket" structures formed by the path in the vertical direction, min f times (path) calculates the number of times the path bend spacing is less than the limit length. g(path) = 0 means that the path does not collide with the layout space, and h(path) = 0 means that the path is orthogonally arranged with the floor or wall (the bend is a right angle).

[0012] After determining the objective function, perform function transformation on it according to the actual optimization objective to obtain the fitness function, and convert the objective function value into a relative fitness value for subsequent algorithm calculations.

[0013] Set the maximum number of iterations to end the iteration.

[0014] Step 2: Use Logistic chaos to create an initial population.

[0015] The chaos mapping is used to generate a chaotic sequence, which is a random sequence generated by a simple deterministic system and has characteristics such as nonlinearity, ergodicity, and randomness. Use the Logistic mapping to generate a chaotic sequence to initialize the population positions of the gorilla troops. The expression of the Logistic mapping is:

[0016] X n+1 = aX n (1 - X n ) (2).

[0017] In the formula: Xn+1 ∈ (0, 1), n = 1, 2, …, representing the position of an individual in the population; X0 {0, 0.25, 0.5, 0.75, 1.0}; a ∈ [0, 4].

[0018] Step 3: Randomly divide the gorillas into three types, namely migrating to unknown positions, known positions, and the positions of other gorillas, for quickly retrieving the positions of silverback gorillas.

[0019] The GTO algorithm uses three mechanisms for position update, namely migrating to unknown positions, migrating to known positions, and migrating to the positions of other gorillas. The mathematical descriptions of the three position update mechanisms are:

[0020]

[0021] In the formula: GX(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the current position vector of the gorilla; r1, r2, r3, and rand are random numbers generated in the range of [0, 1] for each iteration; p represents a given parameter that determines the probability of choosing to migrate to an unknown position; UB and LB represent the upper and lower bounds of the search space respectively; Xr(t) represents the position vector of a randomly selected gorilla in the population; GXr(t) represents the position of the randomly selected gorilla at the t-th iteration in the population; the control parameters C, L, and H are calculated through formulas (3), (5), and (6) respectively.

[0022]

[0023] G = cos(2 × r4) + 1 (5).

[0024] L = C × l (6).

[0025] H = Z × X(t) (7).

[0026] Z = [-C, C] (8).

[0027] Where: It represents the current iteration value; MaxIt represents the total number of iterations for performing the optimization operation; G is the cosine function coefficient; r4 is a random number generated between [0, 1] in each iteration update; l is a random number between [-1, 1]; Z is a random value within the range of [-C, C] in the problem dimension.

[0028] Enter the exploration phase. When rand < p or rand ≥ p and rand < 0.5, update the position of the gorilla individual using Equation (3) in the previous step, otherwise update the position of the gorilla individual using Equation (9).

[0029] In the GTO algorithm, the communication between the silverback gorilla and its members is an important part of making decisions. In the exploration phase, when rand ≥ 0.5, there are more parameters in the formula for updating the gorilla position in Equation (3), which will affect the optimization effect of the algorithm. Therefore, combine the position update strategy formula in the GTO algorithm with the position update strategy formula in the OOA algorithm.

[0030] When rand ≥ 0.5, the improved formula for updating the gorilla position is:

[0031] GX(t + 1) = X(t) + r × (X best - I × X(t)) (9).

[0032] Where: G(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the current position vector of the gorilla; r represents a random number within the range of [0, 1], I is a random number in the set {1, 2}; Xbest is the position of the gorilla with the best current objective function value in the population.

[0033] Calculate the fitness value of the gorilla according to the objective function. If the new fitness value is better than the previous one, replace it, and set the optimal solution (the best position) as the position of the silverback gorilla Xsilverback.

[0034] Step 4: Optimize the position of the silverback gorilla within the gorilla troop through the silverback competition mechanism.

[0035] The GTO algorithm uses two behaviors, following the silverback gorilla and competing with adult female gorillas, to update the position. The silverback gorilla leads the population, makes decisions, determines the actions of the group, and guides the gorillas to find food, and is responsible for the safety and well-being of the team. At the same time, the silverback gorilla may become weak, old and eventually die, and the blackback gorilla in the group may become the new leader, and other male gorillas may also fight with the silverback gorilla and rule the population. The value of C in Equation (3) can be used to select whether to follow the silverback or compete with adult females. If C ≥ W, then choose to follow the silverback mechanism; if C < W, then adopt the mechanism of competing with adult females. W is a parameter that needs to be set before the optimization operation.

[0036] When the population is just initialized, the silverback gorilla is young and healthy, and other gorillas in the population follow the silverback gorilla to go to various regions to look for food. In addition, members can influence other members during the action. When C ≥ W, its position update formula is:

[0037] GX(t + 1) = L × M × (X(t) - X silverback ) + X(t) (10).

[0038]

[0039] g = 2 L (12).

[0040] In the formula: GXi(t) represents the position vector of each candidate gorilla at the t-th iteration; Xsilverback represents the position of the silverback gorilla, that is, the global optimal solution; N represents the population size.

[0041] When male gorillas enter maturity, they will fight with other male gorillas to expand the group of adult female gorillas they can choose from, and this behavior may last for several days. When C < W, this behavior can be described by the formula:

[0042] GX(i) = X silverback - (X silverback × Q - X(t) × Q) × A (13).

[0043] Q = 2r5 + 1(14).

[0044] A = β × E (15).

[0045]

[0046] In the formula: GX(i) represents the i-th male gorilla; Q represents the impact force factor; r5 represents a random number between [0, 1]; A represents the coefficient of violent conflict degree; β is a parameter that needs to be given a value before the optimization operation. E is used to simulate the impact of violent conflict on the dimension. If rand ≥ 0.5, the value of E is equal to a random value in the normal distribution and the problem dimension; if rand < 0.5, the value of E is equal to a random value in the normal distribution; the meanings of other parameters are the same as above.

[0047] Obtain the value of W through formula (17), compare its relationship with C, if C ≥ W, then use formula (10) to update the position of the gorilla individual, otherwise use formula (13) to update the position of the gorilla individual.

[0048] Generally speaking, in the early stage of the algorithm, W should be set to a smaller value; in the later stage, W should be set to a larger value. In the traditional GTO algorithm, W takes a fixed value, which is prone to problems such as weak global optimization ability in the early stage and weak convergence ability in the later stage. Therefore, a new calculation formula is introduced to make the value of W increase linearly within the range of [0.011, 1.1] as the number of iterations increases. The specific calculation formula is:

[0049] W = w max - (w max - w min) × It ÷ Maxit (17).

[0050] In the formula: w max and w min represent the maximum value and the minimum value respectively, It represents the current iteration value; MaxIt represents the total number of iterations for performing the optimization operation.

[0051] Step 5: Incorporate the Levy flight strategy to increase randomness, introduce foreign gorillas to compete with the silverback gorilla, and avoid premature convergence.

[0052] In the GTO algorithm, the position of the silverback gorilla represents the optimal solution, and all gorillas follow the silverback gorilla and act under its decision. This action trend often causes the population to converge prematurely and fall into a local optimum during the iteration process. Therefore, in the later stage of the GTO algorithm development, the Levy flight strategy is applied to the position update of the gorillas, that is, the Levy flight strategy is applied once again to update the individual position after the original position update. The improved position update formula is:

[0053] X(t + 1) = X silverback (t) + (X silverback - X(t)) × α × Levy(λ) (18).

[0054] In the formula: X(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the position vector of the gorilla at the t-th iteration; Xsilverback(t) represents the position vector of the current silverback gorilla; Levy(λ) represents the random search path adopted by the Levy flight strategy, and λ usually takes values within [0, 2]; α is the step size scaling factor.

[0055] Step 6: When the preset number of iterations is reached, output the optimal solution.

[0056] Judge whether the algorithm iteration ends. If the maximum number of iterations is reached, return the position of the optimal gorilla, that is, the global optimal solution; otherwise, return to Step 3 to continue the loop experiment.

[0057] For the above ship pipeline path optimization method based on the improved artificial gorilla troops algorithm, further, in Step 1, the maximum number of iterations is set to 800 times.

[0058] The above ship pipeline path optimization method based on the improved artificial gorilla troops algorithm, further, the positions of the population individuals in step 2 are arbitrary.

[0059] The above ship pipeline path optimization method based on the improved artificial gorilla troops algorithm, further, the population size N in step 4 is 50.

[0060] The above ship pipeline path optimization method based on the improved artificial gorilla troops algorithm, further, in step 4, W is set to 0.011.

[0061] For this ship pipeline path optimization method of the present invention, first, the population positions are initialized using chaotic mapping; then, the position update strategy in the osprey optimization algorithm is combined; finally, the Levy flight strategy is integrated in the later stage to update the individual positions, ensuring the diversity of the population in the later stage of the algorithm, jumping out of the local optimal solution, expanding the search ability, and applying it to the ship pipeline path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is the operation flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0063] The cooling system of a certain LNG ship needs to arrange a ring pipeline network with a total length of about 200 meters, and the following constraints are required to be met: avoiding the hull structure obstacles (12 machinery compartments + 4 load-bearing beams); the number of elbows ≤ 10; the total path length is the shortest.

[0064] Step 1: Determine the objective function and the number of iterations, initialize the population positions, and generate the predator matrix. The decomposition steps are as follows:

[0065] a) Set the following constraint conditions and their weights, and then determine the objective function through formula (1).

[0066]

[0067] Step 2: Use Logistic chaos to create the initial population.

[0068] Set the following parameters and then create 50 initial populations through formula (2).

[0069] Parameter Name Set Value Initial Value n Population Size Positive Integer 50 <![CDATA[X0]]> Individual Position Arbitrary Coordinates {0,0.25,0.5,0.75,1.0}

[0070] Taking three as an example, assume the initial values are X0 = {0, 0.25, 0.5, 0.75, 1.0}

[0071] Mapped to three-dimensional space coordinates:

[0072] X = 55.0, Y = 45.0, Z = -28.75 (safe path);

[0073] X = 170.0, Y = 250.0, Z = -22.5 (Collision with obstacle, eliminated);

[0074] X = 110.0, Y = 675.0, Z = 16.25 (Safe path);

[0075] X = 5.0, Y = 250.0, Z = -28.75 (Safe path);

[0076] X = 215.0, Y = 5.0, Z = -22.5 (Collision with obstacle, eliminated);

[0077] ……

[0078] The final population obtained is as follows:

[0079]

[0080] Initial population after screening (32 safe individuals retained).

[0081] Step 3: Randomly divide the gorillas into three types, namely migrating to unknown locations, known locations, and the locations of other gorillas, for quickly retrieving the positions of silverback gorillas. The following parameters need to be set

[0082]

[0083] The following parameters need to be set in formulas (4)(5)(6)(7)(8)

[0084]

[0085]

[0086] The following parameters need to be set in formula (9)

[0087] Parameter Name Set Value Initial Value I Random Number Set {1, 2} 0.5 r Random Number [0,1] 0.2

[0088] According to the above set values, the updated population status:

[0089]

[0090] Step 4: Optimize the positions of silverback gorillas within the gorilla troop through the silverback competition mechanism. The following parameters need to be set in formulas (10)(12)(13)(14)(15)(16)

[0091]

[0092] The following parameters need to be set in formula (17)

[0093]

[0094]

[0095] According to the parameters defined above, the result is obtained:

[0096] (C = 0.798 > W = 0.007): Gx(i) = X silverback -(X silverback ×Q - X(t)×Q)×A

[0097] Assume the current silverback position: X silverback = [55.0, 45.0, 28.75]

[0098] Original position: X(t) = [66.5, 45.0, 28.75]

[0099]

[0100] The updated population matrix P competed :

[0101]

[0102] Step 5: Incorporate the Levy flight strategy to increase randomness and introduce foreign gorillas to compete with the silverback gorilla to avoid premature convergence. The following parameters need to be set

[0103]

[0104] Update example:

[0105]

[0106] Step 6: When the preset number of iterations is reached, output the optimal solution.

[0107] The population matrix P after 800 iterations final (Partial):

[0108]

[0109] Optimal solution X silverback :

[0110]

[0111] The results show that the global optimal solution (length 76.4m, number of elbows 5) was found at 650 iterations.

Claims

1. A method for optimizing the path of ship pipelines based on an improved artificial gorilla troops algorithm, characterized in that The specific operations are as follows: Step 1: Determine the objective function and preset the maximum number of iterations; The layout of ship pipelines is essentially a multi-objective optimization problem under certain constraints. The objectives and constraints can be described in the following form: s.t.g(path)) = 0, h(path) = 0 (1); where, min f length (path) calculates the path length, min f bend (path) calculates the number of path bends, minf power (path) calculates the energy value of the positions passed by the path, min f pocket (path) calculates the number of "concave pocket" structures formed by the path in the vertical direction, min f times (path) calculates the number of times the path bend spacing is less than the limit length; g(path) = 0 indicates that the path does not collide with the layout space, and h(path) = 0 indicates that the path is orthogonally arranged with the floor or wall, and the bend is a right angle; After determining the objective function, perform a function transformation on it according to the actual optimization objective to obtain the fitness function, and convert the objective function value into a relative fitness value; Step 2: Use Logistic chaos to create the initial population; Use the Logistic mapping in formula (2) to generate a chaotic sequence to initialize the population positions of the gorilla troops; X n+1 = aX n (1 - X n )(2); In the formula: Xn+1 ∈ (0, 1), n = 1, 2, …, represents the position of an individual in the population; X0 {0, 0.25, 0.5, 0.75, 1.0}; a ∈ [0, 4]; Step 3: Randomly divide the gorillas into three types, namely migrating to unknown positions, known positions, and the positions of other gorillas, for quickly retrieving the positions of silverback gorillas; The GTO algorithm uses 3 mechanisms for position update, namely migrating to unknown positions, migrating to known positions, and migrating to the positions of other gorillas. The mathematical descriptions of the 3 position update mechanisms are as follows: Where: GX(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the current position vector of the gorilla; r1, r2, r3, and rand are random numbers generated in the range of [0, 1] for each iteration; p represents a given parameter that determines the probability of migrating to an unknown location; UB and LB represent the upper and lower bounds of the search space, respectively; X r (t) represents the position vector of a randomly selected gorilla in the population; GX r (t) represents the position vector of the gorilla at the t-th iteration randomly selected in the population; the control parameters C, L, and H are obtained through equations (3), (5), and (6), respectively; G = cos(2 × r4) + 1 (5); L = C × l (6); H = Z × X(t) (7); Z = [-C, C] (8); In the formula: It represents the current iteration value; MaxIt represents the total number of iterations for performing the optimization operation; G is the coefficient of the cosine function; r4 is a random number between [0, 1] generated in each iteration update; l is a random number between [-1, 1]; Z is a random value in the problem dimension with a range of [-C, C]; Enter the exploration stage. When rand < p or rand ≥ p and rand < 0.5, update the positions of gorilla individuals using formula (3) in the previous step, otherwise update the positions of gorilla individuals using formula (9); When rand ≥ 0.5, obtain the position update of the gorillas through formula (9); GX(t + 1)=X(t)+r×(X best -I×X(t)) (9); Where: G(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the position vector of the current gorilla; r represents a random number in the range [0, 1], and I is a random number in the set {1, 2}; X best is the position of the gorilla with the optimal current objective function value in the population; Calculate the fitness value of the gorilla according to the objective function. If the new fitness value is better than the previous one, replace it and set the optimal solution (the best position) as the position of the silverback gorilla X silverback ; Step 4: Optimize the positions of silverback gorillas within the gorilla troops through the silverback competition mechanism; Use the C value in formula (3) to select whether to follow the silverback or compete with adult females. If C ≥ W, select formulas (10), (11), (12) to follow the silverback mechanism; if C < W, adopt formulas (13), (14), (15), (16) for the adult female competition mechanism, where W is a parameter that needs to be set before the optimization operation; GX(t + 1) = L × M × (X(t) - X silverback ) + X(t) (10); g＝2 L (12)； Where: GX i (t) represents the position vector of the t-th iteration of each candidate gorilla; X silverback represents the position of the silverback gorilla, i.e., the global optimal solution; N represents the population size; GX(i) = X silverback -(X silverback ×Q - X(t)×Q)×A (13); Q＝2r5+1 (14)； A = β × E (15); In the formula: GX(i) represents the i-th male gorilla; Q represents the impact force factor; r5 represents a random number between [0, 1]; A represents the coefficient of the violent conflict degree; β is a parameter that needs to be given a value before the optimization operation, and E is used to simulate the impact of violent conflict on the dimension. If rand ≥ 0.5, the value of E is equal to the random value in the normal distribution and the problem dimension; if rand < 0.5, the value of E is equal to the random value in the normal distribution; the meanings of other parameters are the same as above; The value of W is obtained through formula (17), and its relationship with C is compared. If C ≥ W, the position of the gorilla individual is updated using formula (10); otherwise, the position of the gorilla individual is updated using formula (13). W = w max - (w max - w min) × It ÷ Maxit (17); In the formula: wmax and wmin represent the maximum value and the minimum value respectively, It represents the current iteration value; MaxIt represents the total number of iterations for performing the optimization operation; Step 5: Incorporate the Levy flight strategy to increase randomness, and introduce foreign gorillas to compete with the silverback gorilla through formula (18) to avoid premature convergence; X(t + 1) = X silverback (t) + (X silverback - X(t)) × ɑ × Levy(λ) (18); Where: X(t + 1) represents the position vector of the gorilla at the (t + 1)-th iteration; X(t) represents the position vector of the gorilla at the t-th iteration; X silverback (t) represents the position vector of the current silverback gorilla; Levy(λ) represents the path randomly searched using the Levy flight strategy, and λ usually takes values within [0, 2]; α is the step size scaling factor; Step 6: When the preset number of iterations is reached, output the optimal solution; Judge whether the algorithm iteration ends. If the maximum number of iterations is reached, return the optimal gorilla position, that is, the global optimal solution; otherwise, return to Step 3 to continue the loop experiment.

2. The ship pipeline path optimization method based on the improved artificial gorilla troops algorithm according to claim 1, characterized in that, In Step 1, the maximum number of iterations is set to 800 times.

3. A ship pipeline path optimization method based on an improved artificial gorilla troops algorithm according to claim 1, characterized in that, In Step 2, the positions of the population individuals are arbitrary.

4. A method for optimizing the path of a ship pipeline based on an improved artificial gorilla troops algorithm according to claim 1, characterized in that In Step 4, the population size N is 50.

5. The ship pipeline path optimization method based on the improved artificial gorilla troops algorithm according to claim 1, characterized in that In Step 4, W is set to 0.011.

Images