Loading optimization method under power grid material target intelligent algorithm

Abstract

The invention discloses a loading optimization method under a power grid material target intelligent algorithm. The method comprises the following steps: constructing a multi-constraint mathematical model aiming at maximizing the loading weight and the priority material quantity and minimizing the gravity center deviation; irregular materials to be loaded are rotated, combined and subjected to external rectangular pretreatment, and the space utilization rate of a cargo hold is increased; dynamically matching an optimal transport vehicle according to the transport vehicle load, the cargo hold size and the material priority; a multi-target genetic algorithm is adopted to carry out material distribution on the front section of the cargo hold, a Pareto leading-edge solution set is generated through binary coding and non-dominated sorting, and convergence efficiency is ensured in combination with an elitist strategy; filling residual materials at the rear section of the cargo hold in a descending order according to the unloading priority by utilizing a greedy algorithm; material layout is optimized based on a single-target genetic algorithm of hybrid coding, coordinates are adjusted to minimize center-of-gravity deviation, and pressure bearing and collision constraints are verified; and carrying out transport vehicle selection and distribution iteratively until all loading is completed.

Description

Technical Field

[0001] The present invention belongs to the field of power grid material optimization, and more specifically relates to a loading optimization method under a power grid material target intelligent algorithm. Background Art

[0002] With the continuous growth of global energy demand and the rapid expansion of power infrastructure, the transportation and management of power grid materials play a vital role in the construction and maintenance of power systems. Power grid materials primarily include cables, transformers, switchgear, and other related heavy and irregularly shaped cargo. The transportation of these materials places high demands on loading methods, transportation vehicle selection, and transportation route planning. However, traditional material loading and transportation methods often exhibit the following deficiencies when faced with the complexity and diversity of power grid materials:

[0003] First, power grid materials are heavy, irregular in size, and come in a wide variety of types. Traditional loading methods make it difficult to fully utilize the space on transport vehicles, resulting in low transportation efficiency. Maximizing space utilization while ensuring loading safety and transport economy, especially under multiple objectives and constraints, is a pressing issue.

[0004] Secondly, power grid material transportation involves numerous constraints, including vehicle size restrictions, cargo hold pressure limits, and material loading and unloading priorities. Traditional optimization methods often consume significant computing resources to address these complex constraints and struggle to quickly arrive at effective solutions in practical applications, resulting in limited responsiveness and flexibility in transportation planning.

[0005] Furthermore, with the increasing complexity of power grid construction and maintenance, the scale and frequency of transportation tasks have significantly increased. Traditional manual planning and empirical decision-making methods are no longer able to meet the needs of efficient and accurate transportation. Human interference can also easily lead to irrational loading, increased transportation costs, and even safety hazards. Therefore, there is an urgent need for an optimization method based on intelligent algorithms that can intelligently optimize power grid material loading plans under multi-objective and multi-constraint conditions, improve transportation efficiency, reduce costs, and ensure transportation safety. Summary of the Invention

[0006] The present invention constructs a multi-objective and multi-constrained mathematical model and combines it with advanced intelligent algorithms to achieve intelligent optimization of power grid material loading plans, solve many bottleneck problems in existing technologies, and significantly improve the overall efficiency and management level of power grid material transportation.

[0007] In order to achieve the above object, the present invention is implemented by adopting the following technical solution: the method comprises the following steps:

[0008] The method comprises the following steps:

[0009] A multi-objective and multi-constraint mathematical model is constructed, the objective function is to maximize the loading weight, maximize the number of priority loading materials and minimize the distance between the actual center of gravity and the ideal center of gravity, and the constraint conditions include the loading capacity of the transport vehicle, the size of the cargo compartment, the center of gravity envelope and the pressure limit;

[0010] The irregular materials to be loaded, such as cables and coils, are rotated and combined for pretreatment to improve the utilization rate of the vehicle width;

[0011] According to the weight, size and priority of the pretreated materials, a set of materials to be loaded is determined;

[0012] Based on the state of the transport vehicle, the remaining loading capacity and the size of the cargo compartment, a transport vehicle matching the materials to be loaded is selected;

[0013] A multi-objective genetic algorithm is used to allocate materials to the front section of the cargo compartment of the transport vehicle, including initializing a binary coded population, generating a Pareto front by non-dominated sorting, crossover and mutation operations and elite reservation;

[0014] The greedy algorithm is used to allocate the remaining materials to the rear section of the cargo compartment, fill in descending order of unloading priority and verify the constraints;

[0015] The single-objective genetic algorithm is used to optimize the layout of the materials in the cargo compartment, hybrid coding is used to adjust the order and position of the materials, the deviation of the center of gravity is minimized, and the collision and pressure constraints are verified;

[0016] Iteratively remove the loaded materials and repeat steps 4-7 until all loading is completed, and output the final solution.

[0017] In one solution, the pretreatment includes: rotating the materials to place them along the longitudinal or transverse direction of the vehicle, combining multiple small-sized materials into regular combinations to reduce gaps; geometric simplification is performed on irregular materials using the minimum circumscribed rectangle, and the merging strategy is dynamically adjusted based on the width of the vehicle cargo compartment to ensure that the size of the pretreated materials fits the pressure limit of the cargo compartment partition.

[0018] In one solution, the transport vehicle selection criterion includes: the remaining loading capacity of the transport vehicle must be greater than the total weight of the materials to be loaded, the remaining space of the front and rear sections of the cargo compartment must be greater than the combined maximum length and width of the pretreated materials, and the state parameters of the transport vehicle preferentially select the transport vehicle that can accommodate high-priority materials.

[0019] In one solution, the multi-objective genetic algorithm includes: individual coding is a binary sequence consistent with the number of materials, indicating whether the material is allocated to the front section; the crossover operation is single-point or uniform crossover, and the mutation operation is random bit flipping; the constraint processing adopts feasibility screening or penalty function to eliminate individuals with overload or oversize.

[0020] In one scheme, the greedy algorithm comprises: sorting the remaining materials in descending order of unloading priority, sequentially attempting to load into the rear section, skipping if the current material causes the total weight, size or pressure to exceed the limit, until it cannot continue to load or all materials are traversed, and finally outputting the rear section allocation result.

[0021] In one scheme, the hybrid encoding comprises: the order of the front section and the rear section materials adopts permutation encoding to represent the loading position, and the position offset adopts real number encoding to represent the coordinates of the materials in the cargo hold; the fitness function is the Euclidean distance between the actual barycenter and the ideal barycenter, and the collision detection algorithm is used to verify that the materials do not overlap.

[0022] In one scheme, the genetic algorithm adopts an elitist strategy, directly retains the individuals ranked high in the non-dominated set in each generation to the next generation, and accelerates the population evaluation through parallel computing to ensure the convergence efficiency of the algorithm.

[0023] In one scheme, the iterative loading comprises: after each loading is completed, removing the allocated materials from the set to be loaded, if the set is not empty, returning to step 4 to select a new transport vehicle, otherwise, summarizing the loading schemes of each transport vehicle and outputting the total barycenter deviation, weight utilization rate and priority satisfaction rate indicators.

[0024] The present application has the following advantages:

[0025] 1. Significantly improve the loading space utilization rate: by constructing a multi-objective and multi-constrained mathematical model, reasonably arranging the layout of power grid materials in the cargo hold of the transport tool, effectively avoiding space waste, and maximizing the space utilization rate. This not only increases the amount of materials transported each time, reduces the number of transportation times, but also reduces transportation costs.

[0026] 2. Optimization ability under multi-objective and multi-constraint: the present application comprehensively considers multiple factors such as the weight, volume, shape and priority of power grid materials, and can realize multi-objective optimization while ensuring loading safety and transportation economy. Effectively solves the problem that traditional methods cannot consider multiple demands, and provides a more comprehensive and reasonable loading scheme.

[0027] 3. High computing performance and fast response: the use of advanced intelligent optimization techniques such as genetic algorithms greatly improves the computing efficiency of loading scheme generation. Even when faced with large-scale complex problems, the optimized results can be quickly obtained in a short time, meeting the dynamic changes in the process of power grid material transportation, improving the response speed and flexibility of transportation planning.

[0028] 4. High degree of automation, reducing human intervention: the present application realizes the automation optimization of the loading scheme, reduces the dependence on manual planning and decision-making. Through the intelligent algorithm system, the probability of human operation error is reduced, the scientificity and rationality of the loading process are ensured, and the overall safety and reliability of the transportation process are improved. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 For the method flowchart of the present application;

[0030] Figure 2 Mixed encoding mode diagram. DETAILED DESCRIPTION

[0031] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the related drawings. The drawings show typical embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described in the present application. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0032] Unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used in the present application in the specification of the present application are only for the purpose of describing the specific embodiments and are not intended to limit the present application. In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the related drawings. The drawings show typical embodiments of the present application. However, the present application can be implemented in many different forms and is not limited to the embodiments described in the present application. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0033] As Figure 1 shown, a loading optimization method under a power grid material target intelligent algorithm

[0034] The power grid materials are mainly irregular goods such as cables and coils. According to the loading planning concept model, a multi-objective and multi-constraint mathematical model is constructed.

[0035] Step 1 explains the power grid material loading problem as follows:

[0036] 1) The materials are placed along the longitudinal or transverse direction of the vehicle;

[0037] 2) The structure of each part of the vehicle is different and needs to be analyzed separately;

[0038] For the to-be-loaded materials {C1, C2, …, C N} and the transport vehicle P j , the following definitions are made:

[0039] 1) Material definition: the weight of each material is {α1, α2, …, α N}; the geometric size of the i-th material is The center of gravity is The loading priority of the material is defined as {u1, u2, …, u N}, u i ∈{0, 1}, the priority loading value is 1, and vice versa. The unloading priority of the material is defined as {η1, η2, …, η N}, η i ∈{0, 1}, the priority unloading value is 1, and vice versa.

[0040] 2) Transport vehicle definition: the maximum load of the transport vehicle is W j ; the ideal center of gravity position of the transport vehicle is The center of gravity envelope is The geometric size of the front cargo compartment of the transport vehicle is The size of the rear cargo compartment is The cargo compartment pressure limit is K represents the number of cargo compartment partitions.

[0041] 3) Post-loading definition: the allocation array of the front cargo compartment of the vehicle is , which indicates that the material i is loaded in the front section of the vehicle, and 0 indicates that it is not loaded. The allocation array of the rear cargo compartment of the vehicle is , which indicates that the material i is loaded in the rear section of the vehicle, and 0 indicates that it is not loaded. R = [r1, r2, …, r L ], r i ∈{1, 2, …, Q}, r i = 1 indicates that the material i is placed in the first position. The actual center of gravity of the transport vehicle after loading is CGj, the total size of the materials in the front cargo compartment is CG j , the total size of the materials in the front cargo compartment is The total size of the materials in the rear cargo compartment is The pressure value of each zone of the cargo compartment is

[0042] The objective function is:

[0043]

[0044] The constraint conditions are:

[0045]

[0046] Vehicle geometric size limit, vehicle section pressure limit); to-be-loaded material information (quantity, type, weight, center of gravity, geometric size, loading priority, unloading priority).

[0047] Step 2: Rotate, merge, and other preprocessing of the to-be-reloaded materials to improve the utilization rate of the vehicle width.

[0048] Step 3: Determine the set of planned loading materials {C1, C2, …, C N} to be loaded.

[0049] Step 4: The set of available transport vehicles is {P1, P2, …, P M}, and the transport vehicle P j is automatically selected as the current transport vehicle to be loaded based on factors such as the state of the transport vehicle, the geometric and weight characteristics of the priority loading materials, etc.

[0050] Step 5: Material allocation is performed on the front section of the transport vehicle's cargo hold.

[0051] Step 5.1: Initialize the population The individual code is The code length is consistent with the number of materials.

[0052] Step 5.2: The objective function of this stage is:

[0053]

[0054] The constraints are:

[0055]

[0056] There are two objective functions, equation () represents maximizing the loading weight, and equation () represents maximizing the number of priority loading materials. The constraints consist of four expressions, equations () and () represent that the individual material and the total material should not exceed the maximum load of the transport vehicle, and equations () and () represent that the size of the individual material and the total material should not exceed the corresponding space of the cargo hold.

[0057] Step 5.3: Perform non-dominated sorting on the population to obtain After calculating the crowding degree, perform selection, crossover, and mutation operations to generate a new generation of population.

[0058] Step 5.4: Merge the new population with the parent population, perform Pareto level division and calculate the crowding degree, and execute the elite retention strategy on the merged population to generate a new population.

[0059] Step 5.5: If the new population meets the iteration termination condition, output the optimal solution and go to Step 6; otherwise, go to Step 5.3.

[0060] Step 6: Perform material allocation on the rear section of the transport vehicle's cargo hold. Since the rear cargo hold is shorter, the number of types and quantities of materials that can be loaded is limited, so it is appropriate to use a greedy algorithm to solve the allocation array Define the remaining maximum load of the transport vehicle Add the materials to be loaded one by one, and the objective function that meets the requirements is:

[0061]

[0062] and satisfy the relevant weight and size constraints.

[0063] Step 7: The vehicle position placement layout optimization of the allocated goods can be regarded as a single-objective optimization process, which is solved by using a genetic algorithm.

[0064] Step 7.1: Initialize the population, and the set of goods allocated to the front section of the vehicle is The set of goods allocated to the rear section is The hybrid coding mode as shown in Figure 2 is adopted.

[0065] Here represents the placement rank of the goods in the front section of the cargo hold, and dis fore represents the distance between the first goods and the front end of the vehicle. Similarly, the rear section represents the same.

[0066] Step 7.2: Calculate the fitness function of each individual, and the objective function is:

[0067]

[0068] The constraint condition is:

[0069]

[0070] Step 7.3: Perform selection, crossover, and mutation operations on the population, and the population is improved.

[0071] Step 7.4: Determine whether the termination condition is met. If yes, output the optimal individual; otherwise, go to step 7.2.

[0072] Step 8: Delete the above planned goods from the set of goods to be loaded, forming a new set of goods to be loaded {C1, C2, …, C N′}. Determine whether the set is empty. If yes, output the loading scheme; otherwise, go to step 4.

[0073] Example:

[0074] Application of intelligent loading optimization method for power grid goods

[0075] A power company needs to transport a batch of emergency power grid goods (cables, transformers, coils, etc.) from the warehouse to the disaster area. The transport fleet includes 3 different types of transport vehicles (T1, T2, T3), and needs to maximize the loading capacity and prioritize the transportation of high-priority goods under the premise of ensuring safety.

[0076] The specific data are as follows:

[0077] 1. Information on the goods to be loaded

[0078] Type and quantity of goods: a total of 15 goods, including:

[0079] Cables (C1-C5): length 5-8 meters, diameter 0.3-0.5 meters, weight 200-500 kg;

[0080] Transformers (T1-T3): size 2m x 1.5m x 1.8m, weight 800-1200 kg;

[0081] Coils (S1-S7): diameter 1.2-1.8 meters, height 0.5-1 meter, weight 300-600 kg.

[0082] Priority definition:

[0083] Loading priority: cables C1, C2 in the disaster area are in urgent need (P=1), the rest P=0;

[0084] Unloading priority: transformers T1, T2 need to be unloaded last (U=1).

[0085] Center of gravity position: the center of gravity of all goods is located at the geometric center.

[0086] 2. Parameters of transport vehicles

[0087] T1 transport vehicle:

[0088] Maximum load: 8000 kg;

[0089] Cargo compartment size: front section (length 6m x width 2.5m x height 2m), rear section (length 4m x width 2.5m x height 2m);

[0090] Ideal center of gravity: 3.5m from the front of the vehicle;

[0091] Center of gravity envelope: allowed deviation ±0.8m;

[0092] Pressure limit: the front section is divided into 3 zones (each zone pressure ≤3000 kg / m2), the rear section is divided into 2 zones (each zone pressure ≤2500 kg / m2).

[0093] T2 transport vehicle: load 6000 kg, smaller cargo compartment size (front section 5m x 2m x 2m, rear section 3m x 2m x 2m).

[0094] T3 transport vehicle: load 10000 kg, larger cargo compartment size (front section 8m x 3m x 2m, rear section 5m x 3m x 2m).

[0095] Implementation process of the embodiment

[0096] Step 1: Construction of multi-objective mathematical model

[0097] Constraints include:

[0098] Total weight does not exceed the transport vehicle load capacity;

[0099] Front / Back cargo compartment size limit;

[0100] Actual center of gravity needs to be within the [2.7m, 4.3m] interval;

[0101] Cargo compartment partition pressure limit ≤ 3000kg / m 2 (for the front section) and 2500kg / m 2 (for the back section).

[0102] Step 2: Material preprocessing

[0103] Geometric optimization for irregular materials:

[0104] 1. Rotation adjustment:

[0105] Cable C1-C5 originally placed longitudinally (length 8m), exceeding the T1 front section length (6m), so rotate it horizontally to a width of 0.5m, and adjust the length to fit the cargo compartment;

[0106] Coil S1-S7 diameter 1.8m exceeds cargo compartment width (2.5m), rotate it to vertical placement (height 1m, diameter 1.8m).

[0107] 2. Merge small materials:

[0108] Merge coils S4 (diameter 1.2m) and S5 (diameter 1.0m) into a combined body with dimensions 1.2m x 1.0m x 1m, total weight 900kg, reducing the gap.

[0109] 3. Minimum circumscribed rectangle:

[0110] Transformers T1-T3 are regular cuboids themselves and do not need to be simplified;

[0111] The circumscribed rectangle of the merged coil combination is 1.5m x 1.5m x 1m (including safety gap).

[0112] After preprocessing, the material sizes are all adapted to the cargo compartment width (≤2.5m) and meet the pressure partition requirements.

[0113] Step 3: Determine the set to be loaded

[0114] Select materials according to preprocessing results:

[0115] High-priority materials: cable C1 (500kg, size 0.5m x 0.5m x 6m), C2 (450kg, 0.5m x 0.5m x 5m);

[0116] Common goods: Transformer T1 (1200 kg), T2 (1000 kg), Coils S1-S7 (300-600 kg);

[0117] Total weight to be loaded: 15 goods totaling 8600 kg.

[0118] Step 4: Transport vehicle selection

[0119] Matching rules:

[0120] 1. Load limit: T1 (8000 kg) and T3 (10000 kg) can carry 8600 kg, T2 (6000 kg) is excluded;

[0121] 2. Cargo compartment size:

[0122] T3 front section length 8m can accommodate cables C1 (6m) and C2 (5m), but T1 should be used first to reduce transportation costs;

[0123] 3. Priority adaptation: T1 front section remaining space can load C1, C2, meeting priority requirements.

[0124] Final selection of T1 transport vehicle as the first choice.

[0125] Step 5: Cargo compartment front section allocation (multi-objective genetic algorithm)

[0126] 1. Encoding and population initialization:

[0127] Individual encoding: 15-bit binary sequence, 1 indicates allocation to the front section, 0 indicates no allocation;

[0128] Population size: 100, iterations 50 generations, crossover rate 0.8, mutation rate 0.1.

[0129] 2. Objective function calculation:

[0130] Loading weight: total weight of the front section must be ≤8000 kg, cargo compartment front section length ≤6m;

[0131] Priority loading quantity: C1, C2 must be loaded (P=1);

[0132] Center of gravity calculation: $CG_j=\frac{\sum(w_i\cdot x_i)}{\sum w_i}$, x_i is the longitudinal position of goods i in the front section.

[0133] 3. Constraint processing:

[0134] Individuals with excessive weight or length are directly eliminated;

[0135] Individuals with excessive center of gravity deviation are subject to a penalty function (fitness reduced by 50%).

[0136] 4. Pareto front generation:

[0137] Get 3 optimal solutions by non-dominated sorting:

[0138] Scheme 1: Load C1, C2, T1, S1-S3, total weight 7800kg, center of gravity 3.6m;

[0139] Scheme 2: Load C1, C2, T2, S4-S6, total weight 7600kg, center of gravity 3.3m;

[0140] Scheme 3: Load C1, C2, T1, T2, total weight 8200kg (overload, need to be removed).

[0141] 5. Final selection: Scheme 1 (minimum deviation of center of gravity and meet the load).

[0142] Step 6: Allocation of rear section of cargo hold (greedy algorithm)

[0143] Remaining supplies: T2 (1000kg), S4-S7 (2100kg in total).

[0144] 1. Sort by unloading priority: T2 (U=1) needs to be unloaded last, prefer to load to the rear section;

[0145] 2. Try to load:

[0146] T2 size 2m x 1.5m x 1.8m, remaining length of rear section 4m, can accommodate;

[0147] S4-S7 combined size 1.5m x 1.5m x 1m, total weight 900kg;

[0148] Total weight of rear section: 1000kg + 900kg = 1900kg, remaining load 8000kg - 7800kg = 200kg, not over limit;

[0149] 3. Final allocation: Load T2 and S4-S7 in the rear section, remaining 200kg load cannot load S7 (600kg).

[0150] Step 7: Layout optimization (single-objective genetic algorithm)

[0151] 1. Hybrid encoding:

[0152] Front section supplies order: [C1, C2, T1, S1, S2, S3] is arranged as encoding;

[0153] Position offset: Real number encoding represents the coordinates of each supply in the cargo hold (e.g. C1 start x=0m, S1 start x=2m).

[0154] 2. Fitness function: Centroid deviation $\|CG_j 3.5\|$ ≤ 0.8m;

[0155] 3. Constraint verification:

[0156] Collision detection: Material clearance ≥ 0.1m;

[0157] Pressure limit: Front section partition 1 (0-2m) withstands 2500kg / m 2 , partition 2 (2-4m) withstands 2800kg / m 2 , no over-limit.

[0158] 4. Optimization results:

[0159] Final layout coordinates:

[0160] C1: x=0m, occupies 0-0.5m width;

[0161] C2: x=0.5m, occupies 0.5-1m width;

[0162] T1: x=1.5m, centrally placed;

[0163] S1-S3: x=3m-5m, evenly distributed;

[0164] Actual centroid: 3.4m, deviation 0.1m, meets envelope requirements.

[0165] Step 8: Iterative loading and scheme output

[0166] 1. Loaded materials: T1 front section loads C1, C2, T1, S1-S3 (7800kg), rear section loads T2, S4-S7 (1900kg), total 9700kg (over T1 load 8000kg, needs correction).

[0167] 2. Error correction: rollback to step 5, reselect scheme 2 (7600kg), rear section can load T2 (1000kg) and S4-S5 (600kg), total weight 7600kg+1600kg=9200kg (still overloading).

[0168] 3. Final adjustment: only front section loads C1, C2, T1, S1 (7800kg), rear section gives up T2, loads S4-S5 (600kg), total weight 8400kg, centroid 3.5m, meets all constraints.

[0169] 4. Remaining materials: T2, S6-S7 are loaded by T3 transport vehicle twice.

[0170] Results

[0171] T1 transport vehicle:

[0172] Loading weight: 8400 kg (load utilization rate 105%, overload needs to be adjusted to 9200 kg, final load 8000 kg);

[0173] Priority satisfaction: C1, C2 successful loading;

[0174] Center of gravity position: 3.5 m (ideal value), none of the pressure zones exceeded the limit.

[0175] T3 transport vehicle: second loading of remaining materials T2, S6-S7 (total weight 2800 kg), center of gravity deviation 0.3 m.

[0176] Total transportation efficiency: 2 times of transportation complete all materials, weight utilization rate 98.5%, priority achievement rate 100%.

[0177] Key data analysis table

[0178] Indicator T1 transport vehicle T3 transport vehicle Maximum load (kg) 8000 10000 Actual load (kg) 8000 2800 Load utilization rate 100％ 28％ Front center of gravity deviation (m) 0 0.3 Priority material loading quantity 2 0 Maximum value of the pressure partition (kg / m 2 )]]> 2900 1800

[0179] This embodiment realizes the safe and efficient loading of power grid materials through the synergy of multi-objective genetic algorithm, greedy algorithm and layout optimization, verifies the feasibility of the method under complex constraints. Through dynamic adjustment of transport vehicle selection and iterative loading strategy, the transportation efficiency and priority satisfaction rate are significantly improved, providing reliable technical support for power emergency logistics.

[0180] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM) or a random access memory (RAM), etc.

[0181] It should be understood that the above detailed description of the technical solutions of the present application by means of preferred embodiments is illustrative rather than limiting. A person of ordinary skill in the art can modify the technical solutions recorded in each embodiment on the basis of the present application, or make equivalent substitutions for part of the technical features; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for optimizing loading of power grid materials under an intelligent algorithm, characterized by: The method comprises the following steps: A multi-objective, multi-constraint mathematical model was constructed. The objective functions were to maximize the load weight, maximize the number of prioritized materials, and minimize the distance between the actual and ideal center of gravity. Constraints included the transport vehicle load, cargo hold dimensions, center of gravity envelope, and pressure limits. Rotate and merge irregular materials such as cables and coils to be loaded, thus improving the utilization rate of vehicle width; Determine the collection of materials to be loaded based on the weight, size and priority of the pre-processed materials; Select a transport vehicle that matches the materials to be loaded based on the transport vehicle's status, remaining load, and cargo hold size; A multi-objective genetic algorithm is used to distribute materials in the front section of the transporter's cargo hold, including initializing the binary coding population, generating the Pareto frontier through non-dominated sorting, crossover and mutation operations, and elite retention. Use the greedy algorithm to allocate the remaining materials to the rear section of the cargo hold, fill it in descending order of unloading priority and verify the constraints; Optimize cargo hold material layout through a single-objective genetic algorithm, adjust material order and position through hybrid coding, minimize center of gravity deviation, and verify collision and pressure constraints; Iteratively remove the loaded materials and repeat steps 4-7 until all loading is completed, and output the final plan.

2. The method according to claim 1, characterized in that The preprocessing includes: rotating the materials so that they are placed longitudinally or transversely along the vehicle, merging multiple small-sized materials into regular combinations to reduce gaps; using the minimum circumscribed rectangle to geometrically simplify irregular materials, and dynamically adjusting the merging strategy based on the width of the vehicle cargo hold to ensure that the size of the preprocessed materials is adapted to the pressure limit of the cargo hold partition.

3. The method according to claim 1, characterized in that The transport vehicle selection criteria include: the remaining load of the transport vehicle must be greater than the total weight of the materials to be loaded, the remaining space in the front and rear sections of the cargo hold is respectively greater than the maximum length and width combination of the pre-processed materials, and the transport vehicle status parameters give priority to transport vehicles that can accommodate high-priority materials.

4. The method according to claim 1, wherein The multi-objective genetic algorithm includes: individual encoding is a binary sequence consistent with the quantity of materials, indicating whether the materials are allocated to the front end; the crossover operation is a single point or uniform crossover, and the mutation operation is a random bit flip; the constraint processing adopts feasibility screening or penalty function to eliminate overloaded or oversized individuals.

5. The method according to claim 1, wherein The greedy algorithm includes: sorting the remaining materials from high to low according to the unloading priority, and trying to load them to the back section in turn. If the current material causes the total weight, size or pressure to exceed the limit, it will be skipped until it is impossible to continue loading or all materials have been traversed, and finally outputting the back section allocation result.

6. The method according to claim 1, characterized in that The hybrid coding includes: the order of the front and rear materials is represented by permutation coding to indicate the loading position, and the position offset is represented by real number coding to indicate the coordinates of the materials in the cargo hold; the fitness function is the Euclidean distance between the actual center of gravity and the ideal center of gravity, and a collision detection algorithm is used to verify that the materials do not overlap.

7. The method according to claim 1, characterized in that Genetic algorithms all adopt an elite retention strategy, which directly retains the non-dominated individuals with high ranking in each generation of the population to the next generation, and accelerates population evaluation through parallel computing to ensure the convergence efficiency of the algorithm.

8. The method according to claim 1, characterized in that The iterative loading includes: after each loading is completed, removing the allocated materials from the to-be-loaded set; if the set is not empty, returning to step 4 to select a new transport vehicle; otherwise, summarizing the loading plans of each transport vehicle and outputting the total center of gravity deviation, weight utilization rate, and priority satisfaction rate indicators.

Images