Structural health monitoring sensor layout optimization method for constraining gradient-free general solution

Abstract

The invention belongs to the technical field of structure detection and health monitoring, and discloses a structure health monitoring sensor layout optimization method for constraining gradient-free general solution, which comprises the following steps: determining a target structure sensor optimization layout criterion, and selecting a gradient-free optimization algorithm with boundary constraint; and constructing an objective function of sensor optimization layout, expanding an objective function optimization vector translation interface, and obtaining a sensor layout optimization combination by using constrained gradient-free general solution. The invention can effectively solve the problem of layout combination optimization of large-scale structure sensors. Under the condition of not changing codes of an optimization algorithm, a vector coding translation interface is expanded and optimized for a target function; therefore, all constraint gradient-free optimization algorithms including artificial intelligence type and classical gradient-free type optimization algorithms considering boundary constraints can be generally used for solving a combinatorial optimization problem of sensor optimization configuration.

Description

technical field [0001] The invention belongs to the technical field of structural detection and health monitoring, and in particular relates to a method and system for optimizing the layout of structural health monitoring sensors with constrained non-gradient general solution. Background technique [0002] At present, large-scale civil engineering structures will be affected by various static and dynamic loads, environmental erosion and other factors during their long service period, and their technical status and use functions will often deteriorate continuously, reducing their safety in use and even threatening the overall safety of the structure. In recent years, frequent structural safety accidents have seriously threatened national property and people's life and property safety. Therefore, it is necessary to effectively build a structural health monitoring system, and carry out structural inspection and health status assessment on this basis to ensure that the structure ...

AI Technical Summary

Problems solved by technology

Nevertheless, in the process of implementing specific algorithms for solving combinatorial optimization problems, it often takes a considerable amount of work to determine the coding scheme and design genetic operators (crossover, mutation, etc.) (such as the paper DOI: 10.3969 / j.issn .1000-3835.2008.03.021, DOI: 10.1177 / 1475921719877579)

However, it should be pointed out that the directional design and modification process of the above optimization algorithms can only be aimed at the specific algorithm under study, and for different algorithms, it is necessary to study the corresponding source code modification strategies in detail, but this process usually requires in-depth interpretation of various optimization algorithms At the same time, understand and master the theoretical background of various algorithms and the corresponding program implementation in detail. This process is heavy in workload, extremely cumbersome and error-prone

On the other hand, in the face of the effective solution requirements of high-dimensional combinatorial optimization problems for the optimal layout of large-scale structural sensors, it is necessary to ensure the efficiency and reliability of the optimization algorithm used. However, most of the optimization algorithms that are both efficient and reliable are commercial It is impossible to obtain the source code of the packaged program or other mature closed source program, which makes the strategy of trying to design and modify the source code of the algorithm to meet the mandatory constraints of sensor layout combination optimization unfeasible, and cannot be used effectively and fully developed. Mature commercial closed-source code has a strong advantage in solving large-scale and high-dimensional combinatorial optimization problems of large-scale engineering structures

[0007] Through the above analysis, the existing problems and defects of the existing technology are: the existing structural health monitoring sensor layout optimization technology that relies on directional modification or redesign of the optimization algorithm has poor adaptability and versatility, heavy workload, extremely cumbersome and error-prone; at the same time This directional modification code strategy is not conducive to the real scene application of large-scale engineering structure health monitoring and ultra-large-scale combination decision-making problems because of the inability to obtain efficient and reliable program codes

Images