The development and methods of smart shading control guided by daylighting optimization
Daylight holds significant importance in architecture,not only enhancing the visual and mental well-being of occupants within living spaces but also effectively reducing the energy consumption of artificial lighting within buildings.The exterior surface of a building plays a decisive role in creating the indoor lighting environment.It responds to changes in outdoor lighting and coordinates the interaction between indoor and outdoor light.However,traditional building surfaces have long relied on manual adjustments during operation,unable to dynamically respond to changes in outdoor lighting,hence struggling to maximize the benefits of daylight.To address this issue,recent academic efforts have combined intelligent technology,initiating research into smart shading control.Smart shading is a pivotal component within smart buildings.By integrating Building Automation Systems(BAS)and Building Information Models(BIM),it can intelligently regulate indoor daylight.Existing research emphasizing smart shading practices aligned with optimizing daylight predominantly focuses on the form of shading and innovative materials but lacks in-depth exploration of control methods.The control methods of smart shading significantly impact the performance of daylight utilization,directly determining the upper limit of smart shading performance.However,existing research on control methods still falls short in practical implementation,requiring improvement in efficiency,adaptability to various scenarios,precision in performance enhancement,synergy among multiple functionalities,and control cost reduction.Hence,there is a need to explore effective control strategies for optimizing indoor daylight within smart shading systems.This involves establishing universal principles for different smart shading control processes,incorporating new technologies and methodologies to innovate smart shading control methods.This approach aims to enhance existing smart shading control performance and overcome current limitations.The advent of digital technology has revolutionized the expansion of knowledge and technological upgrades in architecture.This evolution is evident not only in the generation of designs driven by digital performance but also in the intelligent construction and operation of built environments.Consequently,architectural design for living spaces has evolved from"form design"to"control design".On another front,advancements in digital technology,particularly artificial intelligence,continuously break application barriers,driving the innovation of high-precision,high-efficiency human-machine interactive dynamic control systems.The study outlines the developmental trajectory of smart shading control focused on optimizing daylight and summarizes key design points for smart shading control,proposing smart shading control methods integrating current machine learning technologies.By outlining this trajectory,it elucidates the trend of control development from physical measured closed-loop systems to predictive open-loop digital models and the inclination toward cross-innovation utilizing machine learning technologies in the context of the artificial intelligence era.Machine learning's capability to effectively uncover complex patterns in environmental light information and manage nonlinearities significantly aids in resolving issues such as multi-dimensional data reduction and unclear correlation mappings in smart shading control,thereby greatly advancing its practical application.However,constructing machine learning predictive models fundamentally entails a symbiotic process between data and algorithms.The ultimate accuracy and efficiency of these models rely not only on the architecture design of the algorithms but also on the quality,quantity,and attributes of the data itself.Effectively applying machine learning algorithms to daylight prediction and shading control necessitates a clear understanding of fundamental aspects in smart shading control design.This includes defining basic principles in control design processes,facilitating the effective construction of smart shading control systems in different architectural lighting environments.From the developmental perspective of smart shading control and irrespective of its branch type,it encompasses variables influencing daylight performance,optimizing daylight goals,and dynamic daylight control methods,which serve as inputs,outputs,and control components,each highlighting specific trends in technical content development.In response to these trends,this research proposes current design points for intelligent shading and daylight control,encompassing determination of daylight environmental parameters,mapping of daylight environmental indicators,and multi-performance optimization within daylight environments.In conclusion,through a systematic review,the study has identified key design aspects in intelligent shading and daylight control,further proposing intelligent shading and daylight control design methods augmented by machine learning-related technologies.The robust data analysis and underlying information mining abilities of machine learning empower technical enhancements in determining daylight environmental parameters,constructing daylight indicator mappings,and optimizing daylight control goals within smart shading control.To effectively formulate intelligent shading and daylight control design methods and aid strategy formulation for smart shading operation during the architectural design stage,this research presents information analysis methods based on feature selection,model prediction methods based on multiple algorithms in parallel,and objective optimization methods based on proxy models.These methods collectively support the technical,scientific,and feasible aspects of innovation in architectural intelligent shading control design.
building smart shadingcontrol systemphysical propertiesdigital technologyartificial intelligence
万方数据
维普
