CN-122022351-A - Unmanned aerial vehicle carrying decision-making method and system based on multi-objective optimization

CN122022351ACN 122022351 ACN122022351 ACN 122022351ACN-122022351-A

Abstract

The invention relates to an unmanned aerial vehicle carrying decision method and system based on multi-objective optimization, wherein the unmanned aerial vehicle carrying decision method based on multi-objective optimization comprises the following steps of S1, collecting carrying data and carrying out modeling operation based on the carrying data to obtain modeling data, S2, carrying out three-level self-adaptive matching operation on the modeling data based on a preset carrying equipment library to obtain candidate scheme data, and S3, carrying out multi-index calculation and analysis operation on each candidate scheme in the candidate scheme data to obtain an optimal candidate scheme. The method carries out feasibility analysis through three-level self-adaptive matching, realizes comprehensive evaluation based on a full period through multi-index calculation, carries out balance optimization on candidate schemes through multi-objective optimization, completes scientific decision to obtain an optimal scheme under multiple objectives, and fuses multi-source heterogeneous data.

Inventors

WANG ZIYI
WAN CHENGKUAN
JING TIANQI
XIONG CHAO
BAI PU

Assignees

四川信天翁航空科技有限公司
西藏先锋绿能环保科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260205

Claims (10)

1. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization is characterized by comprising the following steps of: S1, collecting carrying data and carrying out modeling operation based on the carrying data to obtain modeling data; S2, performing three-level self-adaptive matching operation on modeling data based on a preset carrying equipment library to obtain candidate scheme data; and S3, multi-objective optimization, namely performing multi-index calculation and analysis operation on each candidate scheme in the candidate scheme data to obtain the optimal candidate scheme.
2. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 1, wherein the multi-index computing and analyzing operation computes each index and obtains a multi-index composite score by weighted summation.
3. The unmanned aerial vehicle delivery decision-making method based on multi-objective optimization of claim 1, wherein the metrics computed by the multi-metric computing and analysis operations include, but are not limited to, cost functions, time functions, and risk functions; the cost function comprises flight operation cost and site building reconstruction cost.
4. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 3, wherein the cost function has a calculation formula: C(s)=C flight +C infra C infra =δ(mode)×[max(0,S req -S avail )×P earth +N tree ×P comp ] Wherein C (S) is a cost function, C flight is the flight operation cost, C infra is the site foundation construction and transformation cost, S avail is the site available flat area, S req is the minimum required area of equipment, P earth is the site leveling unit price, N tree is the number of trees to be cut, P comp is the tree compensation unit price, and delta (mode) is a mode coefficient.
5. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 2, wherein the weighted summation is performed after the normalization processing operation is performed on each calculated index by the multi-index calculation and analysis operation.
6. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 2, wherein the calculation formula of the multi-index comprehensive score is as follows: Score=w c C′(s)+w t T′(s)+w r R′(s) In the formula, score is a multi-index composite Score, w c 、w t 、w r is the weight coefficient sum of the cost function, the time function and the risk function, 1, C '(s), T '(s), and R '(s) is the index obtained by performing normalization processing on the cost function C(s), the time function T(s) and the risk function R(s), respectively.
7. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 1, wherein the step S3 is followed by the steps of: and S4, generating a scheme, namely generating a final decision report according to the best candidate scheme.
8. The unmanned aerial vehicle carrying decision-making method based on multi-objective optimization of claim 1 is characterized in that the three-level self-adaptive matching operation sequentially comprises a first-level decision-making operation, a second-level decision-making operation and a third-level decision-making operation, wherein the first-level decision-making operation judges whether carrying can be carried in an unmanned aerial vehicle carrying mode or not; The unmanned aerial vehicle carrying operation mode comprises, but is not limited to, a light-load unmanned aerial vehicle cluster mode, a heavy-load unmanned aerial vehicle mode and a large-small machine type combination mode.
9. The unmanned aerial vehicle delivery decision method of claim 8, wherein the three-stage decision operation comprises the sub-steps of: s3.1, judging whether a plurality of delivery starting points or a plurality of delivery ending points exist or not, if so, performing special delivery route planning operation; S3.2, performing general delivery route planning operation; s3.3, generating candidate scheme data according to the transportation voyage information obtained by the special transportation route planning operation and the general transportation route planning operation; and in the transportation route planning operation, carrying out the sectional transportation route planning operation if the following three conditions occur: the number of delivery start points is below a preset threshold and the number of delivery end points is below a preset threshold; Dividing a transport start point with a material weight ratio higher than a preset duty ratio threshold value into key transport start points, and dividing a transport end point with a material weight ratio higher than the preset duty ratio threshold value into key transport end points; the unmanned aerial vehicle carrying operation mode is the heavy-load unmanned aerial vehicle mode; and in the transportation route planning operation, carrying out the route planning operation of the aerial delivery unmanned aerial vehicle if the following two conditions are met at the same time: the number of the conveying end points corresponding to one conveying start point is not lower than a preset threshold value; the unmanned aerial vehicle carrying operation mode is the light-load unmanned aerial vehicle cluster mode or the large-small machine type combination mode; And the aerial delivery unmanned aerial vehicle route planning operation is used for transporting the materials at the same delivery starting point in a mode of throwing one or more auxiliary airplanes in the middle of the transportation of the main airplane.
10. A multi-objective optimization-based unmanned aerial vehicle carrying decision system, applying the multi-objective optimization-based unmanned aerial vehicle carrying decision method of any one of claims 1 to 9, characterized by comprising: The data acquisition interface module is used for receiving the carrying data; The scene digital modeling module is used for performing modeling operation based on the carrying data to obtain modeling data; The database module is used for storing a preset carrying equipment library; The self-adaptive decision engine is used for carrying out three-level self-adaptive matching operation on the modeling data based on the carrying equipment library to obtain candidate scheme data; And the comprehensive evaluation module is used for carrying out multi-index calculation and analysis operation on each candidate scheme in the candidate scheme data to obtain the optimal candidate scheme.

Description

Unmanned aerial vehicle carrying decision-making method and system based on multi-objective optimization Technical Field The invention relates to the technical field of unmanned aerial vehicle application, in particular to an unmanned aerial vehicle carrying decision method and system based on multi-objective optimization. Background The current unmanned aerial vehicle technology development is rapid, and various industries gradually adopt unmanned aerial vehicle carrying modes for cargo transportation. In electric power engineering, particularly high-voltage, ultra-high voltage and ultra-high voltage transmission lines, the transmission lines are usually built in mountain areas, forest areas, canyons or river zones with complex terrains and rare trails. During the "small-traffic" phase of line construction (i.e. "last kilometer" transport), routine maintenance or fault rush repair in these areas, material transport faces significant challenges. This stage is often the "throat" link of the whole project, directly restricting project cost, construction period and safety. With the development of aviation technology, light-load unmanned aerial vehicles, heavy-load unmanned aerial vehicles and helicopters are gradually applied to power construction. However, at present, in the choice of the delivery mode, there is a general lack of scientific evaluation standards and decision mechanisms, and the following technical problems mainly exist: The multi-source heterogeneous data is difficult to fuse, and a system is lacking to fuse and analyze discrete multi-dimensional data. Lack of full cycle assessment, failure to comprehensively consider closed loop logic from field exploration, technical feasibility demonstration to economic and construction period assessment, often ignores implicit costs such as site leveling. Therefore, there is a need for a method and a system for carrying decision-making by an unmanned aerial vehicle with multi-objective optimization, which realize comprehensive evaluation of the whole period, balance and optimize candidate schemes, and fuse multi-source heterogeneous data. Disclosure of Invention The invention aims to overcome the defects of the prior art, provides a multi-objective optimization unmanned aerial vehicle carrying decision method and a multi-objective optimization unmanned aerial vehicle carrying decision system, realizes comprehensive evaluation based on a full period through multi-index calculation, performs balance optimization on candidate schemes through multi-objective optimization, and fuses multi-source heterogeneous data. The aim of the invention is realized by the following technical scheme: an unmanned aerial vehicle carrying decision-making method based on multi-objective optimization comprises the following steps: S1, collecting carrying data and carrying out modeling operation based on the carrying data to obtain modeling data; S2, performing three-level self-adaptive matching operation on modeling data based on a preset carrying equipment library to obtain candidate scheme data; and S3, multi-objective optimization, namely performing multi-index calculation and analysis operation on each candidate scheme in the candidate scheme data to obtain the optimal candidate scheme. Further, the multi-index calculation and analysis operation calculates each index and obtains a multi-index comprehensive score through weighted summation. Further, the metrics of the multi-metric calculation and analysis operations include, but are not limited to, cost functions, time functions, and risk functions; the cost function comprises flight operation cost and site building reconstruction cost. Further, the calculation formula of the cost function is as follows: C(s)=Cflight+Cinfra Cinfra=δ(mode)×[max(0,Sreq-Savail)×Pearth+Ntree×Pcomp] Wherein C (S) is a cost function, C flight is the flight operation cost, C infra is the site foundation construction and transformation cost, S avail is the site available flat area, S req is the minimum required area of equipment, P earth is the site leveling unit price, N tree is the number of trees to be cut, P comp is the tree compensation unit price, and delta (mode) is a mode coefficient. Further, the multi-index calculating and analyzing operation performs the weighted summation after performing the normalization processing operation on each index obtained by calculation. Further, the calculation formula of the multi-index comprehensive score is as follows: Score=wcC′(s)+wtT′(s)+wrR′(s) In the formula, score is a multi-index composite Score, w c、wt、wr is the weight coefficient sum of the cost function, the time function and the risk function, 1, C '(s), T '(s), and R '(s) is the index obtained by performing normalization processing on the cost function C(s), the time function T(s) and the risk function R(s), respectively. Further, the step S3 is followed by the following steps: and S4, generating a scheme, namely generating a final decision report