Multi-target building performance design expert system and method based on sensitivity matrix

Abstract

The invention discloses a multi-target building performance design expert system and method based on a sensitivity matrix, which are used for assisting a designer in carrying out performance optimization-based rapid design on a scheme in a building scheme design stage. The method comprises the following steps: scheme information acquisition: reading an initial scheme of a user and an expected value of the user for a performance target; evaluating the performance level of the current scheme through simulation or prediction; a multi-objective optimization algorithm based on a sensitivity matrixis used, a database module is called to carry out rapid performance optimization based on the initial scheme, and performance simulation is carried out on a set of schemes obtained through optimization; a scheme modification suggestion and an optimization scheme are fed back to a designer through a user interaction interface and an interpreter module, and the designer can quickly select an optimization scheme with improved performance as a result according to the suggestion or repeat the optimization process. According to the system, multi-objective optimization is guided by using a sensitivity matrix, and multiple scheme performance objectives are brought into a design decision in the initial stage of design.

Description

technical field [0001] This application relates to the technical field of building performance-based design software development, in particular to a sensitivity matrix-based multi-objective building performance-based design expert system and method. Background technique [0002] With the continuous increase of the construction of green buildings in our country, the importance of comprehensive analysis of building sound, light, heat, wind, comfort and other performances to guide architectural design is increasing, but the conflict between performances brings architectural design problem. In multi-objective optimization problems, improving any objective function necessarily weakens at least one other objective function. How to balance multiple performance goals in early architectural design is undoubtedly a research content with great potential and practicality. [0003] At present, the common way to solve multi-objective problems is to use multi-objective evolutionary algor...

AI Technical Summary

Problems solved by technology

The former takes a long time, and the latter is difficult to be applied in actual engineering because the accuracy of the proxy model is difficult to guarantee.

At the same time, the results of both are completely digitally driven, and the operation is complicated and lacks interaction with designers, so it is difficult to promote among designers.

Images