Search

CN-121997589-A - Old city reconstruction green building optimization method based on multi-source data fusion

CN121997589ACN 121997589 ACN121997589 ACN 121997589ACN-121997589-A

Abstract

The invention relates to a green building reconstruction optimization method based on multi-source data fusion. The method utilizes oblique photography, laser scanning, thermal infrared imaging and sensor network to collect building data, and ensures the data quality through pretreatment such as time synchronization, spatial registration, data calibration and the like. And constructing a dynamic thermodynamic three-dimensional model, and updating model parameters in real time. And generating a green transformation scheme by adopting a multi-objective optimization algorithm, and performing simulation verification by using energy plus. And the closed loop iterative optimization is realized by monitoring the energy consumption and the environmental parameters in real time and comparing the energy consumption and the environmental parameters with the model prediction. The invention can intelligently and finely reform the existing building, improves the energy-saving benefit and the environmental comfort level, and has good application prospect.

Inventors

  • WANG WEI
  • SUN YANNAN
  • ZHU QINGMEI
  • WANG WENLI

Assignees

  • 青岛丰拓力行科技服务有限公司

Dates

Publication Date
20260508
Application Date
20260123

Claims (7)

  1. 1. The old city reconstruction green building optimization method based on multi-source data fusion is characterized by comprising the following steps of: (1) Collecting building geometry data, thermal performance data and environment monitoring data as a multi-source data set, preprocessing the multi-source data set, and outputting standardized data after preprocessing; (2) The standardized data are fused to construct a dynamic three-dimensional model with thermal properties, and parameters of the dynamic three-dimensional model are dynamically updated based on real-time environment monitoring data; (3) Generating a green transformation scheme through multi-objective optimization based on a dynamic three-dimensional model, simulating and verifying by utilizing a digital twin technology, and outputting the green transformation scheme passing verification; (4) And dynamically triggering parameter adjustment of data calibration and dynamic three-dimensional model updating and iteration of the green reconstruction scheme according to the difference between the environment monitoring data and the predicted value of the dynamic three-dimensional model after the environment monitoring data passes the verification implementation of the green reconstruction scheme.
  2. 2. The method according to claim 1, wherein in step 1: collecting geometric structure data by adopting a three-dimensional scanning technology, wherein the three-dimensional scanning technology comprises oblique photogrammetry and laser radar scanning; The thermal performance data are collected by adopting a thermal infrared imaging technology and are used for obtaining the temperature distribution of the outer surface of the building; the method comprises the steps of collecting environment monitoring data by adopting a distributed sensor network, wherein the sensor network comprises a temperature and humidity sensor, an illumination intensity sensor and a gas concentration sensor which are deployed on inner and outer facades of a building; The preprocessing includes time synchronization, spatial registration and noise filtering, and data calibration and normalization.
  3. 3. The method according to claim 2, wherein in step 1: The oblique photogrammetry adopts an unmanned plane to carry an oblique camera, and collects multi-angle images; the laser radar scanning adopts a ground three-dimensional laser scanner; The thermal infrared imaging technology adopts a thermal imager; the distributed sensor network realizes data synchronous transmission through a wireless ad hoc network protocol; The time synchronization adopts NTP protocol to uniformly calibrate the time stamp of the multi-source data; The spatial registration adopts a characteristic point matching algorithm to carry out spatial registration on the laser point cloud and the oblique photographic image; The noise filtering adopts a wavelet transformation and median filtering combined algorithm to separate and filter impulse noise and Gaussian noise in sensor data.
  4. 4. A method according to claim 3, wherein in step 2: The dynamic three-dimensional model construction comprises dividing standardized data into building structure units based on a semantic segmentation algorithm, wherein the building structure units comprise walls, roofs and doors and windows, endowing each building structure unit with thermal properties, wherein the thermal properties comprise heat conductivity coefficients, heat capacity and solar radiation absorptivity; The method for updating the dynamic three-dimensional model parameters comprises the steps of accessing a meteorological data interface in real time to obtain time-by-time meteorological parameters of a target area, dynamically adjusting thermal parameters of the three-dimensional model based on the meteorological parameters, coupling building energy consumption monitoring data, and updating equivalent thermal resistance coefficients in a model heat conduction equation based on a back propagation algorithm.
  5. 5. The method according to claim 4, wherein in the step 3: The method comprises the steps of generating a green transformation scheme, determining an optimization target, determining decision variables, and establishing a multi-target optimization model, wherein the optimization target comprises an energy saving rate, a carbon emission reduction rate and economy; the simulation verification comprises the steps of applying a generated green transformation scheme to a dynamic three-dimensional model to perform simulation, evaluating energy saving rate, carbon emission reduction rate and economic index, judging whether preset constraint conditions are met or not, and if the simulation result does not meet the constraint conditions, adjusting decision variables to perform optimization and verification again.
  6. 6. The method according to claim 5, wherein in step 4: Monitoring environment monitoring data of the transformed building in real time, wherein the environment monitoring data of the transformed building comprise temperature, humidity and energy consumption; Comparing the environmental monitoring number of the reconstructed building with a predicted value of the dynamic three-dimensional model, calculating the difference, setting a difference threshold value, and triggering iterative optimization when the difference exceeds the threshold value; The iterative optimization comprises the steps of adjusting multi-source data calibration parameters and dynamic three-dimensional model parameters according to difference types, carrying out green reconstruction scheme optimization and verification again to generate a new green reconstruction scheme, deploying the new green reconstruction scheme into an actual building, and continuing monitoring and iteration.
  7. 7. A green building optimization system for old city reconstruction based on multi-source data fusion using the method of any one of claims 1-6, comprising the following modules: the data acquisition and preprocessing module is used for acquiring building geometric structure data, thermal performance data and environment monitoring data, and preprocessing the multi-source data set to output standardized data; The dynamic three-dimensional model construction and updating module is used for fusing standardized data to construct a dynamic three-dimensional model with thermal properties and dynamically updating parameters of the dynamic three-dimensional model based on real-time environment monitoring data; The scheme generation and verification module is used for generating a green transformation scheme through multi-objective optimization based on the dynamic three-dimensional model, simulating and verifying by utilizing a digital twin technology, and outputting the green transformation scheme passing verification; The closed loop iteration optimization module is used for monitoring environment data after reconstruction implementation, and triggering data calibration parameter adjustment, three-dimensional model updating and green reconstruction scheme iteration according to the difference between actual monitoring data and a three-dimensional model predicted value.

Description

Old city reconstruction green building optimization method based on multi-source data fusion Technical Field The invention relates to the technical field of green building transformation and digital twin, in particular to a method for optimizing a green building by transforming old city based on multi-source data fusion Background In recent years, green building reconstruction technology is rapidly developed in the old city updating field, and multi-source data fusion becomes a core means for improving energy efficiency evaluation precision. The prior art mainly relies on static integration of a Building Information Model (BIM) and a Geographic Information System (GIS), and combines an Internet of things sensor network (such as temperature and humidity and energy consumption monitoring) to acquire real-time data. Deep learning algorithms (e.g., U-Net semantic segmentation) are used for building structure recognition, while multi-objective optimization (NSGA-II) supports energy-saving scheme generation. The digital twin technology is introduced to realize scheme simulation verification, and part of the system adjusts the transformation strategy through a feedback mechanism. However, the data layer is still mainly composed of isomorphic data sets, the dynamic environment response capability is limited, and the reconstruction of the closed loop depends on manual intervention. The prior art has three core defects that the multi-source data fusion depth is insufficient, and the existing method integrates BIM, GIS and sensor data, but does not solve measurement deviation caused by space-time registration errors and equipment drift of heterogeneous data. This results in low reliability of the model input. And after the optimization decision and implementation are disjoint, the traditional optimization algorithm (such as a static genetic algorithm) generates a scheme, and a dynamic environment coupling mechanism is lacked. When weather suddenly changes, the model can not update heat conduction parameters in real time, and the actually measured energy consumption deviation is high. Meanwhile, the scheme verification depends on fixed scene simulation, and does not incorporate randomness of resident behaviors (such as frequency of use of an air conditioner), so that social acceptance is insufficient. The closed loop feedback is lack of dynamic triggering, namely the existing feedback system mostly adopts periodic manual inspection, and can not respond to transformation failure events (such as empty insulating layer) in time. Even if automatic monitoring is deployed, only local adjustment (such as calibrating a sensor) is triggered, and a full chain iteration mechanism (calibration, model and scheme linkage) is not established. Disclosure of Invention The present invention has been made in view of the above-described problems. The invention solves the technical problems that the existing old city green building reconstruction method has model distortion caused by insufficient fusion precision of multi-source heterogeneous data, energy efficiency deviation caused by dynamic response lag in case of environment mutation, a reconstruction scheme and an implementation effect lack of a closed loop iteration mechanism, and the problem of how to realize full-flow dynamic self-optimization from data acquisition to scheme optimization. In order to solve the technical problems, the invention provides the technical scheme that the method for optimizing the green building of the old city reconstruction based on multi-source data fusion comprises the steps of collecting building geometric structure data, thermal performance data and environment monitoring data as a multi-source data set, preprocessing the multi-source data set, and outputting standardized data after preprocessing; the standardized data are fused to construct a dynamic three-dimensional model with thermal properties, and parameters of the dynamic three-dimensional model are dynamically updated based on real-time environment monitoring data; generating a green transformation scheme through multi-objective optimization based on a dynamic three-dimensional model, simulating and verifying by utilizing a digital twin technology, and outputting the green transformation scheme passing verification; and dynamically triggering parameter adjustment of data calibration and dynamic three-dimensional model updating and iteration of the green reconstruction scheme according to the difference between the environment monitoring data and the predicted value of the dynamic three-dimensional model after the environment monitoring data passes the verification implementation of the green reconstruction scheme. As a preferred scheme of the method for optimizing the green building of the old city transformation based on multi-source data fusion, the method comprises the steps of collecting geometric structure data by adopting a three-dimensional scanning technology, wherein the three-dimen