Hyperstatic structure plane model being built by structural mechanics is studied. Space model precisely reflected in real stress of the structure is built by finite element method (FEM) analysis commerce software. M...Hyperstatic structure plane model being built by structural mechanics is studied. Space model precisely reflected in real stress of the structure is built by finite element method (FEM) analysis commerce software. Mapping model of complex structure system is set up, with convenient calculation just as in plane model and comprehensive information as in space model. Plane model and space model are calculated under the same working condition. Plane model modular construction inner force is considered as input data; Space model modular construction inner force is considered as output data. Thus specimen is built on input data and output dam. Character and affiliation are extracted through training specimen, with the employment of nonlinear mapping capability of the artificial neural network. Mapping model with interpolation and extrpolation is gained, laying the foundation for optimum design. The steel structure of high-layer parking system (SSHLPS) is calculated as an instance. A three-layer back-propagation (BP) net including one hidden layer is constructed with nine input nodes and eight output nodes for a five-layer SSHLPS. The three-layer structure optimization result through the mapping model interpolation contrasts with integrity re-analysis, and seven layers structure through the mapping model extrpulation contrasts with integrity re-analysis. Any layer SSHLPS among 1-8 can be calculated with much accuracy. Amount of calculation can also be reduced if it is appfied into the same topological structure, with reduced distortion and assured precision.展开更多
In this paper,we propose a graph model embedded with compact physical equations for modeling the thermal dynamics of buildings.The principles of heat flow across various components in the building,such as walls and do...In this paper,we propose a graph model embedded with compact physical equations for modeling the thermal dynamics of buildings.The principles of heat flow across various components in the building,such as walls and doors,fit the message-passing strategy used by Graph Neural networks(GNNs).The proposed method is to represent the multi-zone building as a graph,in which only zones are considered as nodes,and any heat flow between zones is modeled as an edge based on prior knowledge of the building structure.Furthermore,the thermal dynamics of these components are described by compact models in the graph.GNNs are further employed to train model parameters from collected data.During model training,our proposed method enforces physical constraints(e.g.,zone sizes and connections)on model parameters and propagates the penalty in the loss function of GNN.Such constraints are essential to ensure model robustness and interpretability.We evaluate the effectiveness of the proposed modeling approach on a realistic dataset with multiple zones.The results demonstrate a satisfactory accuracy in the prediction of multi-zone temperature.Moreover,we illustrate that the new model can reliably learn hidden physical parameters with incomplete data.展开更多
In this paper, we construct two models for the searching task for a lost plane. Model 1 determines the searching area. We predict the trajectory of floats generated after the disintegration of the plane by using RBF n...In this paper, we construct two models for the searching task for a lost plane. Model 1 determines the searching area. We predict the trajectory of floats generated after the disintegration of the plane by using RBF neural network model, and then determine the searching area according to the trajectory. With the pass of time, the searching area will also be constantly moving along the trajectory. Model 2 develops a maritime search plan to achieve the purpose of completing the search in the shortest time. We optimize the searching time and transform the problem into the 0-1 knapsack problem. Solving this problem by improved genetic algorithm, we can get the shortest searching time and the best choice for the search power.展开更多
The building sector significantly contributes to climate change.To improve its carbon footprint,applications like model predictive control and predictive maintenance rely on system models.However,the high modeling eff...The building sector significantly contributes to climate change.To improve its carbon footprint,applications like model predictive control and predictive maintenance rely on system models.However,the high modeling effort hinders practical application.Machine learning models can significantly reduce this modeling effort.To ensure a machine learning model’s reliability in all operating states,it is essential to know its validity domain.Operating states outside the validity domain might lead to extrapolation,resulting in unpredictable behavior.This paper addresses the challenge of identifying extrapolation in data-driven building energy system models and aims to raise knowledge about it.For that,a novel approach is proposed that calibrates novelty detection algorithms towards the machine learning model.Suitable novelty detection algorithms are identified through a literature review and a benchmark test with 15 candidates.A subset of five algorithms is then evaluated on building energy systems.First,on two-dimensional data,displaying the results with a novel visualization scheme.Then on more complex multi-dimensional use cases.The methodology performs well,and the validity domain could be approximated.The visualization allows for a profound analysis and an improved understanding of the fundamental effects behind a machine learning model’s validity domain and the extrapolation regimes.展开更多
The applications of 3D building models are limited as producing them requires massive labor and time costs as well as expensive devices.In this paper,we aim to propose a novel and web-based interactive platform,VGI3D,...The applications of 3D building models are limited as producing them requires massive labor and time costs as well as expensive devices.In this paper,we aim to propose a novel and web-based interactive platform,VGI3D,to overcome these challenges.The platform is designed to reconstruct 3D building models by using free images from internet users or volunteered geographic informa-tion(VGI)platform,even though not all these images are of high quality.Our interactive platform can effectively obtain each 3D building model from images in 30 seconds,with the help of user interaction module and convolutional neural network(CNN).The user interaction module provides the boundary of building facades for 3D building modeling.And this CNN can detect facade elements even though multiple architectural styles and complex scenes are within the images.Moreover,user interaction module is designed as simple as possible to make it easier to use for both of expert and non-expert users.Meanwhile,we conducted a usability testing and collected feedback from participants to better optimize platform and user experience.In general,the usage of VGI data reduces labor and device costs,and CNN simplifies the process of elements extraction in 3D building modeling.Hence,our proposed platform offers a promising solution to the 3D modeling community.展开更多
Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty rem...Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty remote sensing data for the reconstruction of 3-D models of terrain regions. A FeedForward neural network isused to fuse the intensity data sets with the spatial data set after learning its geometry. Results on real data arepresented. Human performance evaluation is assessed on several perceptual tests in order to evaluate the fusionresults.展开更多
Model predictive control is a promising approach to reduce the CO 2 emissions in the building sector.However,the vast modeling effort hampers the widescale practical application.Here,data-driven process models,like ar...Model predictive control is a promising approach to reduce the CO 2 emissions in the building sector.However,the vast modeling effort hampers the widescale practical application.Here,data-driven process models,like artificial neural networks,are well-suited to automatize the modeling.However,the underlying data set strongly determines the quality and reliability of artificial neural networks.In general,the validity domain of a machine learning model is limited to the data that was used to train it.Predictions based on system states outside that domain,so-called extrapolations,are unreliable and can negatively influence the control quality.We present a safe operation approach combined with online learning to deal with extrapolation in data-driven model predictive control.Here,the k-nearest neighbor algorithm is used to detect extrapolation to switch to a robust fallback controller.By continuously retraining the artificial neural networks during operation,we successively increase the validity domain of the artificial neural networks and the control quality.We apply the approach to control a building energy system provided by the BOPTEST framework.We compare controllers based on two data sets,one with extensive system excitation and one with baseline operation.The system is controlled to a fixed temperature set point in baseline operation.Therefore,the artificial neural networks trained on this data set tend to extrapolate in other operating points.We show that safe operation in combination with online learning significantly improves performance.展开更多
基金This project is supported by Provincial Natural Science Foundation of Shanxi, China (No. 20041074)Provincial Natural Science Youth Foundation of Shanxi, China (No. 20051030)Provincial Education Office Key Subject of Shanxi, China (No. 20045027-20045028)
文摘Hyperstatic structure plane model being built by structural mechanics is studied. Space model precisely reflected in real stress of the structure is built by finite element method (FEM) analysis commerce software. Mapping model of complex structure system is set up, with convenient calculation just as in plane model and comprehensive information as in space model. Plane model and space model are calculated under the same working condition. Plane model modular construction inner force is considered as input data; Space model modular construction inner force is considered as output data. Thus specimen is built on input data and output dam. Character and affiliation are extracted through training specimen, with the employment of nonlinear mapping capability of the artificial neural network. Mapping model with interpolation and extrpolation is gained, laying the foundation for optimum design. The steel structure of high-layer parking system (SSHLPS) is calculated as an instance. A three-layer back-propagation (BP) net including one hidden layer is constructed with nine input nodes and eight output nodes for a five-layer SSHLPS. The three-layer structure optimization result through the mapping model interpolation contrasts with integrity re-analysis, and seven layers structure through the mapping model extrpulation contrasts with integrity re-analysis. Any layer SSHLPS among 1-8 can be calculated with much accuracy. Amount of calculation can also be reduced if it is appfied into the same topological structure, with reduced distortion and assured precision.
文摘In this paper,we propose a graph model embedded with compact physical equations for modeling the thermal dynamics of buildings.The principles of heat flow across various components in the building,such as walls and doors,fit the message-passing strategy used by Graph Neural networks(GNNs).The proposed method is to represent the multi-zone building as a graph,in which only zones are considered as nodes,and any heat flow between zones is modeled as an edge based on prior knowledge of the building structure.Furthermore,the thermal dynamics of these components are described by compact models in the graph.GNNs are further employed to train model parameters from collected data.During model training,our proposed method enforces physical constraints(e.g.,zone sizes and connections)on model parameters and propagates the penalty in the loss function of GNN.Such constraints are essential to ensure model robustness and interpretability.We evaluate the effectiveness of the proposed modeling approach on a realistic dataset with multiple zones.The results demonstrate a satisfactory accuracy in the prediction of multi-zone temperature.Moreover,we illustrate that the new model can reliably learn hidden physical parameters with incomplete data.
文摘In this paper, we construct two models for the searching task for a lost plane. Model 1 determines the searching area. We predict the trajectory of floats generated after the disintegration of the plane by using RBF neural network model, and then determine the searching area according to the trajectory. With the pass of time, the searching area will also be constantly moving along the trajectory. Model 2 develops a maritime search plan to achieve the purpose of completing the search in the shortest time. We optimize the searching time and transform the problem into the 0-1 knapsack problem. Solving this problem by improved genetic algorithm, we can get the shortest searching time and the best choice for the search power.
基金the financial support by the Federal Ministry for Economic Affairs and Climate Action(BMWK),promotional reference 03EN1066A and 03EN3060Dfunding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No.101023666.
文摘The building sector significantly contributes to climate change.To improve its carbon footprint,applications like model predictive control and predictive maintenance rely on system models.However,the high modeling effort hinders practical application.Machine learning models can significantly reduce this modeling effort.To ensure a machine learning model’s reliability in all operating states,it is essential to know its validity domain.Operating states outside the validity domain might lead to extrapolation,resulting in unpredictable behavior.This paper addresses the challenge of identifying extrapolation in data-driven building energy system models and aims to raise knowledge about it.For that,a novel approach is proposed that calibrates novelty detection algorithms towards the machine learning model.Suitable novelty detection algorithms are identified through a literature review and a benchmark test with 15 candidates.A subset of five algorithms is then evaluated on building energy systems.First,on two-dimensional data,displaying the results with a novel visualization scheme.Then on more complex multi-dimensional use cases.The methodology performs well,and the validity domain could be approximated.The visualization allows for a profound analysis and an improved understanding of the fundamental effects behind a machine learning model’s validity domain and the extrapolation regimes.
基金supported by the National Natural Science Foundation of China(NSFC)under project[no.41771484].
文摘The applications of 3D building models are limited as producing them requires massive labor and time costs as well as expensive devices.In this paper,we aim to propose a novel and web-based interactive platform,VGI3D,to overcome these challenges.The platform is designed to reconstruct 3D building models by using free images from internet users or volunteered geographic informa-tion(VGI)platform,even though not all these images are of high quality.Our interactive platform can effectively obtain each 3D building model from images in 30 seconds,with the help of user interaction module and convolutional neural network(CNN).The user interaction module provides the boundary of building facades for 3D building modeling.And this CNN can detect facade elements even though multiple architectural styles and complex scenes are within the images.Moreover,user interaction module is designed as simple as possible to make it easier to use for both of expert and non-expert users.Meanwhile,we conducted a usability testing and collected feedback from participants to better optimize platform and user experience.In general,the usage of VGI data reduces labor and device costs,and CNN simplifies the process of elements extraction in 3D building modeling.Hence,our proposed platform offers a promising solution to the 3D modeling community.
文摘Multimodality image registration and fusion are essential steps in building 3-D models from remotesensing data. We present in this paper a neural network technique for the registration and fusion of multimodali-ty remote sensing data for the reconstruction of 3-D models of terrain regions. A FeedForward neural network isused to fuse the intensity data sets with the spatial data set after learning its geometry. Results on real data arepresented. Human performance evaluation is assessed on several perceptual tests in order to evaluate the fusionresults.
基金This project has received funding from the European Union’s Hori-zon 2020 research and innovation programme under grant agreement No.101023666.
文摘Model predictive control is a promising approach to reduce the CO 2 emissions in the building sector.However,the vast modeling effort hampers the widescale practical application.Here,data-driven process models,like artificial neural networks,are well-suited to automatize the modeling.However,the underlying data set strongly determines the quality and reliability of artificial neural networks.In general,the validity domain of a machine learning model is limited to the data that was used to train it.Predictions based on system states outside that domain,so-called extrapolations,are unreliable and can negatively influence the control quality.We present a safe operation approach combined with online learning to deal with extrapolation in data-driven model predictive control.Here,the k-nearest neighbor algorithm is used to detect extrapolation to switch to a robust fallback controller.By continuously retraining the artificial neural networks during operation,we successively increase the validity domain of the artificial neural networks and the control quality.We apply the approach to control a building energy system provided by the BOPTEST framework.We compare controllers based on two data sets,one with extensive system excitation and one with baseline operation.The system is controlled to a fixed temperature set point in baseline operation.Therefore,the artificial neural networks trained on this data set tend to extrapolate in other operating points.We show that safe operation in combination with online learning significantly improves performance.