The 2024 Top 10 Most Practically Valuable Technological Achievements by Beijing University of Civil Engineering and Architecture (BUCEA) have been unveiled. Out of 20 technological achievements, 10 are selected for their high value for practical application.

I. PS Series Building Measurement Robots

This achievement focuses on real-world indoor measurement during the pre-delivery inspection. The PS (Plane Segmentation) building measurement robots have been developed to automatically measure indoor spatial parameters. This series of measurement robots are designed to address various construction scenarios, including construction process monitoring, foundation pit deformation monitoring, road flatness measurement, and tunnel face measurement. They are mainly used for the automatic measurement of indoor spatial parameters, including room dimensions and area,  levelness and verticality of walls and columns, flatness of ceilings and walls, and the squareness of corners, typically for real-world indoor measurements. The robots can also be used during the construction process for real-time tracking of formwork positions, allowing for adjustments of spatial location. After completing the measurement, the PS series robots automatically upload the data to the cloud, generate statistical reports, display the results, and check whether the data meets the required standards based on relevant codes and regulations.

II. Key Technologies and Equipment for Smart Urban Pipeline Monitoring

This achievement involves three core technologies: First, Smart Digital Outlet System and Complete Sets of Equipment for Urban Pipeline Networks. Using IoT technology, this system connects hydrological and water quality data from the outlets with the control system in real time to improve the preciseness of management. Second, Smart Monitoring Equipment for Urban Drainage Networks: This system consists of underground main equipment and sensors that monitor water quality and quantity. It monitors the flow in both full and non-full pipelines and measures water quality indicators (ammonia nitrogen, conductivity, and redox potential) online. Third, AI-Based Smart Management System for Urban Drainage Networks: This system uses data-driven, rapidly reversible models, sparse data observation algorithms, and AI self-learning to conduct analysis and make decisions intelligently and comprehensively based on multi-source data. It identifies pollution sources in drainage basins and integrates three key functions for intelligent collaboration: water quality violation tracing, prediction, and intelligent decision-making. It is able to trace violations back to their sources within 30 minutes, with an accuracy of over 85%.

This technology has won several provincial and ministerial awards, including the 2023 Beijing Science and Technology Progress Award (Second Prize), the 2020 China Invention and Innovation Award (First Prize), and the 2018 Huaxia Construction Science and Technology Award (First Prize). It holds 8 patents (3 of which are U.S. patents) and has developed 3 systems with independent intellectual property rights and 2 sets of technical equipment.

III. High Missing Rate Image Restoration Method Based on Deep Learning

This achievement explores the principles of receptive field calculation and introduces a joint spatial-frequency restoration network. It uses a Fourier neural operator to build a Fourier-based global receptive field module that performs convolution in the frequency domain to capture the global receptive field in the spatial domain. This is then combined with spatial domain information to better preserve both global structures and local texture details. An image restoration method for missing regions is proposed based on wavelet domain prior probability sampling. It applies wavelet transforms to obtain two low-entropy components as network inputs, limiting the information processed by the network within its capacity and expanding the feature selection range based on probability distributions to improve the network’s fault tolerance.

The main goal of high missing rate image restoration is to restore damaged areas using the known parts of the image, ensuring that the restored sections blend with the intact areas as smoothly as possible to meet visual requirements. This method has significant potential for application in various areas, such as image editing, film production, and digital restoration of cultural heritage.

IV. Batch Synthesis Technology and Application of Metal-Organic Framework Derivatives

This achievement introduces a new, cost-effective process and equipment for mass production of metal-organic frameworks (MOFs). It also invents technologies for fast batch separation and purification of MOFs, with a single batch production capacity reaching kilogram-level yields and significant cost savings. By using waste PET plastic and metal-containing wastewater as raw materials, a series of high-value-added MOF-based materials are synthesized for wastewater purification, cutting costs by up to 99.8% compared to traditional commercial production processes. Using mass-produced MOFs as precursors, and leveraging self-developed new heating equipment, various MOF-derived functional materials have been created to efficiently degrade new pollutants in water. The mechanism behind MOF-derived functional materials’ ability to efficiently remove contaminants from wastewater has also been revealed. To address challenges of recycling and separation in the use of MOF-derived powder materials in practical water treatment, several types of loaded MOF-derived functional materials have been developed in a controllable manner, along with a suitable continuous operation device. By optimizing process parameters and conditions, continuous mineralization and detoxification of new pollutants in wastewater have been achieved, comprehensively improving water treatment efficiency.

The achievement has been granted 3 national invention patents and 2 utility model patents. A total of 29 papers have been published in renowned journals at home and abroad, cited over 2,000 times. Among them, 16 papers were included in ESI highly cited papers, and 4 papers were included as hot papers. MOF-derived functional materials have been successfully applied to the purification of high-salinity dyeing wastewater, demonstrating outstanding decolorization and mineralization effects, which have been highly recognized by users. Cooperation agreements and contracts have been signed with multiple companies.

V. “Parent-Child Cycle” Product and Service Design to Effectively Alleviate Traffic Jams During School Commutes

The “Parent-Child Cycle” is an innovative product designed for addressing school commute needs at elementary schools and kindergartens. Following the principles of green, smart, and shared transportation, it encourages families to use bicycles instead of motor vehicles. In terms of product design, the lightweight 20-inch small-wheel frame with a replaceable shell makes it easy to handle and extends the bike’s lifespan. The frame is solid and stable, equipped with safety features such as a child safety seat, internally hidden chain, breathable seat, and front shock absorber to ensure safe rides for parents and children. In terms of service operation design, the electronic fence technology is used to designate parking and guiding zones, enabling intelligent management. A fee system is also set up based on the school commute zone to promote a shift in school commuting practices. Additionally, cooperation with schools, governments, and communities ensures smooth project implementation.

In terms of innovative advantages and revenue models, the product expands from basic bike functionality to parent-child interaction scenarios, meeting various travel needs and breaking the limitations of traditional bike-sharing businesses. Revenue is generated through the following models: Bike rentals, charged by riding time in different periods; APP revenue sharing, by working with suppliers to sell parent-child products through the APP; Ad income, by attracting family brands to place ads; Data income, by generating revenue from selling processed user data.

VI. Digital Twin Mapping AI-Powered Energy Management and Carbon Control Operation Platform

This achievement uses advanced technologies such as artificial intelligence (AI), geographic information systems (GIS), and the Internet of Things (IoT) to create a “GIS+BIM+AI+IoT” model that enables precise management and intelligent operations of buildings throughout their entire lifecycle.

In the construction of carbon-neutral parks: AI and GIS models are used to monitor energy consumption and carbon emissions in real time, helping to predict and optimize carbon reduction in industrial parks, cut energy costs, and improve efficiency. In wall health monitoring: AI models are used to detect leaks, cracks, and hidden voids in the side elevation of the walls, providing early warnings about the building’s health and supporting urban renewal to ensure resident and property safety. In the digital twin mapping of power stations: Digital twins of power station operations are created by combining GIS technology, real-world business scenarios, and individual equipment details to offer precise alerts, monitor carbon emissions and carbon sinks, enable AI-powered video surveillance, and serve the digital and intelligent development of power stations. The Mobile Geospatial Big Data Cloud Service Innovation Team (www.dxkjs.com) focuses on the research of building map operations to serve related application areas.

VII. Urban Renewal Large Model: Shiliuzi (Pomegranate Seed), Responsibility Planner Agent

This achievement is based on the AI language model technology and extensive experience from responsibility planners, integrating knowledge of urban planning, architecture, urban management, urban governance, and technical economics, to create two main intelligent roles: the Responsibility Planner Assistant and the Urban Construction Consultant.

The product’s key features include the Community Expert, Policy Assistant, Evaluation Master, Plan Generator, and Toolbox. The Community Expert provides holistic data services to users such as government departments, responsibility planners, and residents. The Policy Assistant offers objective and accurate policy references and in-depth interpretations to government agencies, property owners, planning teams, and residents involved in urban renewal. The Evaluation Master targets government departments, responsibility planners, and planning teams, providing multi-dimensional evaluation and analysis of uploaded urban renewal plans, and generating professional evaluation and optimization reports. The Plan Generator is specifically designed to assist planning teams in early-stage project development. By allowing users to input simple project information, it guides them step by step to specify requirements and style preferences, and automatically generates multiple plan options.

VIII. A Dual V-Type Pipeline Robot

This achievement is specifically designed to efficiently maintain complex pipeline systems. The unique V-shaped cross structure enhances the driving force and adaptability in complex environments. Its spiral forward movement and independent drive mechanism achieve flexible and efficient pipeline operations. The robot features a modular, lightweight design, allowing it to operate autonomously in pipes ranging from 600mm to 1,400mm in diameter, suitable for horizontal, vertical, and complex curved pipelines. Combining a mobile parallel structure with smart robotic arm technology, the robot has excellent ability to adapt to different environments. It can overcome obstacles, adjust its sizes, and support a variety of functional modules such as detection, cleaning, and repairing equipment, allowing it to be used in many different situations. Its key strength lies in simplifying the drive system, improving fault detection, making it easier for future technology upgrades.

This achievement is in line with China’s national priorities of smart city development, green low-carbon efforts, and industrial automation, representing technological innovation in key areas supported by government policies. In sectors such as oil and gas, urban drainage, and industrial transportation, the robot can effectively address pipeline maintenance challenges and improve operational efficiency. With its modular design and intelligent expansion capabilities, it is highly competitive in fast-growing fields such as smart manufacturing and unmanned systems.

IX. Three-Phase Energy Storage System and Method Using a Cross-Linked Honeycomb Plate for Overflow Heat Exchange

This achievement addresses key technical challenges in solution absorption energy storage, such as low storage efficiency, crystallization blockages in pipelines, and imbalances in charging and discharging. It introduces a three-phase energy storage device using a cross-linked honeycomb plate for overflow heat exchange. The device consists mainly of an evaporator/condenser, absorber/generator, honeycomb plate overflow heat exchange unit, steam transport pipeline, refrigerant circulation pipeline, and solution circulation pipeline.

Key innovations of this technology include: (1) Using a honeycomb structure as fins, the heat exchanger’s heat transfer performance is enhanced by increasing the fluid disturbance inside the cells and expanding the heat exchange area. (2) The large energy storage container is divided into smaller units using a honeycomb structure. When heated with plate thermal fluid, the solution inside forms small crystals. During energy storage, it increases the crystallization rate within the honeycomb. During energy discharge, it overcomes the resistance created by the water film on the crystal surface, speeding up the dissolution of the crystals. (3) The crystals are neatly confined in specific spaces, effectively preventing blockages in pipelines caused by crystals moving with the solution.

X. Large-Scale Scene Reconstruction and Intelligent Analysis System for Smart Cities

This achievement uses multi-source data integration and AI technology to drive technology breakthroughs in the precise management, decision-making support, and scene perception of smart cities. Core technologies include large-scale scene reconstruction, multi-target detection, detailed 3D reconstruction, and large-model scenario-based Q&A systems. By combining data from LiDAR, satellite remote sensing, and drone imagery, high-precision 3D city models are created to enable the identification and analysis of multiple targets such as people, vehicles, and buildings within a city. It also uses large language models to provide real-time city scene information and decision support.

This achievement offers high accuracy in reconstruction and multi-target detection, as well as fast query response. It also supports the processing and analysis of large-scale city-level data, which can be widely applied in urban planning, traffic control, environmental monitoring, and disaster response. The technology is currently in the laboratory validation phase, with its core algorithms already quite mature. Future research will focus on technology integration, dynamic updates, and cross-domain applications to diversify application scenarios and promote industrial development of smart cities.