Building equipment intelligent linkage system based on digital twinning

Abstract

The application relates to a building equipment intelligent linkage system based on digital twinning. The system comprises a data acquisition unit, a data processing unit, a virtual-real embedded modeling unit and a global intelligent linkage strategy control unit. The full-quantity operation parameter cluster of a building intelligent equipment and an environment space-time sensing element set are extracted to obtain an original data set. Data structure packaging and identification are performed to obtain a unique traceability embedded code communication data matrix. Space-time dimension fusion modeling and metadata completion are utilized to obtain building real-time intelligent analysis data with space-time depth coupling. The physical state of the building equipment is captured and digitized signal translation is performed to obtain equipment real-time digital state flow of the twinning model. A virtual operation scene is copied and multi-mode simulation operation is performed to obtain an energy-saving equipment energy consumption simulation analysis result set and an operation mode library. Energy-saving equipment operation data is extracted to predict the development trend of equipment faults, and an energy-saving control instruction set of the building equipment is generated.

Description

Technical Field

[0001] This invention belongs to the field of smart building technology, specifically relating to a building equipment intelligent linkage system based on digital twins. Background Technology

[0002] Currently, the following areas for improvement remain in the intelligent linkage of building equipment: With the continuous advancement of smart city construction, smart buildings have become the core direction of the construction industry, and the intelligent linkage management of building equipment is a key link in achieving building energy conservation and efficient operation and maintenance. Although current intelligent linkage systems for building equipment have achieved automated control of some equipment, there are still many problems to be solved in terms of data utilization, model adaptation, linkage decision-making, and energy-saving operation and maintenance, making it difficult to meet the refined, intelligent, and energy-saving operation needs of modern buildings.

[0003] Existing building equipment data acquisition processes often suffer from unstable multi-source data transmission, inconsistent data formats, and a lack of robust breakpoint resume and data verification mechanisms. This leads to data loss and noise interference, resulting in insufficient accuracy and completeness of the collected raw data. Furthermore, data processing is limited to basic cleaning, failing to extract features and fuse data across time and space to meet building energy-saving requirements. It also lacks effective data traceability management, making it impossible to provide high-quality, traceable datasets for subsequent equipment status analysis.

[0004] In terms of building equipment status perception and simulation, traditional systems have not established a digital twin model that accurately matches the physical building, making it difficult to achieve real-time status synchronization between physical equipment and virtual models. This makes it impossible to accurately simulate different operating modes of equipment and perform energy consumption analysis. Although some systems introduce simple virtual models, the model structure parameters are fixed and cannot be dynamically updated based on abnormal data from actual equipment operation. This leads to a gradual decrease in the fit between the model and the physical equipment, making it impossible to provide a reliable simulation basis for equipment operation decisions.

[0005] Furthermore, existing building equipment linkage is mostly based on simple trigger-based control with preset thresholds, lacking intelligent decision analysis capabilities. It does not combine equipment operating status prediction with energy-saving targets to formulate collaborative linkage strategies, making it difficult to achieve efficient collaborative energy saving for multiple types of equipment such as HVAC and lighting. At the same time, equipment maintenance is mostly carried out through reactive or periodic maintenance, lacking a predictive maintenance mechanism based on equipment operating data. This makes it easy for sudden equipment failures to affect the normal operation of the building and also increases equipment operation and maintenance costs. Moreover, after the linkage decision command is generated, there is a lack of a sound security verification and priority sorting mechanism, which can easily lead to command conflicts, illegal control and other problems, affecting the safety and stability of equipment operation.

[0006] In summary, current intelligent building equipment linkage systems suffer from technical shortcomings in areas such as data lifecycle management, digital twin model construction and dynamic updates, intelligent linkage decision-making, and predictive maintenance. These shortcomings result in low operating efficiency, high energy consumption, and high operation and maintenance costs for building equipment. There is an urgent need for an intelligent building equipment linkage system that can achieve accurate data collection and processing, real-time linkage between physical and virtual spaces, intelligent collaborative energy saving of multiple devices, and predictive maintenance capabilities, in order to improve the overall operation and management level of smart buildings. Summary of the Invention

[0007] To address the aforementioned problems in the existing technology, this invention provides a building equipment intelligent linkage system based on digital twins; The objective of this invention can be achieved through the following technical solutions: a data acquisition unit, a data processing unit, a virtual-real hybrid modeling unit, and a global intelligent interconnection and control unit; The data acquisition unit extracts the full set of operating parameters of the building's intelligent devices and the set of environmental spatiotemporal sensing elements to obtain the original dataset; The data processing unit encapsulates and identifies the data structure based on the original dataset through time-slice tracing and embedding, and obtains a unique tracing and embedding code for the transmission data matrix; it uses spatiotemporal dimension fusion modeling and metadata completion to obtain spatiotemporally deeply coupled real-time intelligent analysis data of buildings. The virtual-real hybrid modeling unit, based on the real-time intelligent analysis data of the building, uses the status digital of the building equipment to perform twin mapping, captures the physical state of the building equipment and translates the digital signals, and obtains the real-time digital state flow of the equipment in the twin model; it replicates the virtual operation scenario and performs multi-mode simulation operation to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment. The global intelligent control unit extracts energy-saving equipment operation data based on the energy consumption simulation analysis result set and operation mode library, predicts the development trend of equipment failure, and generates a set of energy-saving control instructions for building equipment.

[0008] Specifically, the extraction of the full set of operating parameters of building intelligent devices and the spatiotemporal sensing element set of the environment is as follows: based on the full-domain deployment topology map of building intelligent devices, the operating parameters of intelligent devices and the spatiotemporal sensing data of the full-domain environment of the building are extracted simultaneously to obtain a multi-source hash data cluster; The validity of multi-source hash data clusters is verified by using heterogeneous data screening and identification technology to obtain the initial equipment environment data body; By integrating the equipment parameters and environmental data, a complete set of operational parameters for building intelligent devices and a set of spatiotemporal sensing elements for the environment are obtained.

[0009] Specifically, the process of data structure encapsulation and identification includes: using time-slicing frame segmentation technology to decompose continuous data into frames according to a preset time granularity to obtain a time-series data frame cluster; Frame data is implanted through on-chain traceability embedding to obtain time-series data frames; The data fields are standardized and formatted using a pre-defined structured format to obtain a unique traceable embedded code for the transmission data matrix.

[0010] Specifically, the process of modeling and metadata completion through spatiotemporal dimensional fusion includes: establishing a three-dimensional mapping relationship between data acquisition time, device spatial location, and feature parameters based on the communication data matrix through spatiotemporal dimension anchoring; Metadata topology completion supplements the full lifecycle metadata and environment-related metadata to obtain topological spatiotemporal related data; By using multi-dimensional data fusion to integrate and reconstruct equipment parameters, environmental data, and metadata, we can obtain real-time intelligent building analysis data with deep spatiotemporal coupling.

[0011] Specifically, the process of using the status data of building equipment for twin mapping includes: extracting the operating status features and physical attribute features of the building equipment to obtain the equipment status feature spectrum; By using digital twin mapping technology, a digital state mapping model that matches the physical device geometrically and has consistent attributes is constructed in virtual space to obtain the device's digital twin base model. By adapting the digital twin base model of the device to the twin model, the real-time mapping and linkage between the physical state of the building equipment and the digital model can be obtained.

[0012] Specifically, the process of capturing and digitizing signals includes: obtaining the physical state stream of the equipment by capturing the physical operating state of the building equipment and the interaction state with the environment; The analog physical state signal is converted into an initial digital signal that can be recognized by the digital twin model, thus obtaining the initial digital state flow of the device; The amplitude and frequency of the digital signal are calibrated, and the protocol adaptation of the twin model is performed simultaneously to obtain the device real-time digital state stream with deep twin model adaptation.

[0013] Specifically, the process of replicating and performing multi-mode simulation includes: obtaining a virtual operation scenario for the building equipment based on the real-time digital stream of the equipment and the real-time intelligent analysis data of the building; By presetting equipment energy-saving operation modes such as constant speed operation, variable frequency energy saving, and zone control through operation mode configuration technology, a multi-configuration energy-saving operation mode can be obtained; The corresponding modes are simulated in a virtual operation scenario to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment.

[0014] Specifically, the process of extracting energy-saving equipment operation data includes: obtaining operation parameters, energy consumption data and pattern matching features through data feature filtering; Simultaneously, data detection and logical contradiction verification are performed to obtain energy-saving equipment operation data; Principal component analysis algorithm is used to compress the dimensions and refine the features of the equipment operation data to obtain structured energy-saving equipment operation data.

[0015] Specifically, the process of predicting the development trend of equipment failures includes: integrating cross-dimensional deconstruction of historical failure records based on the energy-saving equipment operation data to obtain an energy-saving equipment full life cycle operation analysis library; By using a trend prediction mechanism to dynamically model operating parameters and extract fault correlation features, a fault dataset is obtained. The fault dataset is input into a pre-trained fault model to obtain prediction results.

[0016] Specifically, the specific process of the operation mode configuration technology includes: performing multi-dimensional deconstruction analysis of the operating characteristics of energy-saving equipment based on the virtual operation scenario of the building equipment, performing hierarchical decomposition and mapping of energy-saving targets, and obtaining the basis for mode configuration; By performing modular operations of constant speed operation, frequency conversion energy saving and zone control modes, the corresponding mode operation logic is defined in a closed loop, and the dynamic threshold of mode switching conditions is set to obtain the basic mode template. By using the multi-scenario calibration and optimization of the basic mode template line mode parameter thresholds, the pre-simulation results of the mode operation effect are verified, and a multi-configuration energy-saving operation mode is obtained.

[0017] Specifically, the process of generating energy-saving decision instructions includes: based on the linkage results between the twin model and the intelligent device, and combined with the predictive maintenance scheme, obtaining the energy-saving optimization target and operating constraints of the intelligent device; By introducing energy-saving operation modes and combining energy-saving strategies such as dynamic HVAC adjustment and on-demand lighting control, the energy-saving logic of collaborative linkage of intelligent devices is deduced. An energy-saving decision instruction set is generated based on the energy-saving logic.

[0018] Specifically, the specific process of HVAC dynamic adjustment includes: performing dynamic quantitative calculation of HVAC operating load based on the real-time intelligent analysis data of the building, and simultaneously making short-term predictions of load fluctuation trends to obtain load assessment results; Based on the load assessment results, the compressor frequency and fan speed are adjusted to adapt to the scenario of zoned temperature control logic. At the same time, the safety threshold of parameter adjustment range is limited to obtain a dynamic adjustment parameter set.

[0019] The beneficial effects of this invention are as follows: This system forms a standardized management system for the entire lifecycle from data acquisition to processing. Based on the topology map of the building's intelligent equipment deployment, it achieves synchronous extraction of equipment and environmental data. Combined with heterogeneous data verification and dimensional integration, it ensures the integrity and effectiveness of the original data. Furthermore, it completes data encapsulation and identification through time-slice tracing and embedding, and combines spatiotemporal dimensional fusion modeling and metadata completion to establish a three-dimensional spatiotemporal mapping relationship and achieve multi-dimensional data reconstruction, forming real-time intelligent analysis data with deep spatiotemporal coupling. This data processing method completely solves the problems of inconsistent data formats, poor traceability, and weak spatiotemporal correlation in traditional systems. It not only achieves traceable data management but also provides high-quality, highly adaptable datasets for digital twin modeling and intelligent equipment linkage, improving the accuracy of system operation from the source.

[0020] By leveraging real-time intelligent building data to construct a digital twin model that highly matches the geometry and attributes of physical equipment, and through physical state capture and precise translation of digital signals, real-time state linkage between physical equipment and the virtual model is achieved, creating a highly realistic virtual operation scenario. Simulation of multiple energy-saving operation modes generates an energy consumption simulation analysis result set and operation mode library, providing scientific reference for actual equipment operation. Simultaneously, based on structured energy-saving equipment operation data, a full lifecycle operation analysis library is integrated, combined with trend prediction mechanisms and pre-trained fault models, to accurately predict equipment fault development trends. This replaces traditional reactive maintenance, proactively mitigating the risk of sudden equipment failures, significantly reducing operation and maintenance costs, and making the selection of equipment operation modes more energy-efficient and scientific, effectively improving the stability and economy of building equipment operation.

[0021] Based on digital twin simulation results and fault prediction conclusions, and combined with predictive maintenance schemes, the system clarifies equipment energy-saving optimization goals and operational constraints. It integrates specific energy-saving strategies such as HVAC dynamic adjustment and on-demand lighting control, and deduces the energy-saving logic of multi-device collaborative linkage to generate an energy-saving control instruction set tailored to the actual operational needs of the building. Specifically, HVAC dynamic adjustment achieves dynamic adaptation of equipment parameters while ensuring operational safety through precise load calculation and short-cycle prediction. Various devices are controlled on demand and coordinate efficiently, significantly improving the overall energy efficiency of the building. Simultaneously, the system conducts comprehensive security verification and quantitative sorting of the instruction set, eliminating illegal instructions and avoiding instruction conflicts at the source, ensuring the orderly and rational execution of instructions. This allows the intelligent linkage of building equipment to possess both energy efficiency and intelligence, while also ensuring security and stability, comprehensively improving the operation and management level of smart buildings. Attached Figure Description

[0022] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0023] Figure 1 This is a schematic diagram of a building equipment intelligent linkage system based on digital twins according to the present invention; Figure 2 This is a flowchart of the virtual-real model operation in this invention. Detailed Implementation

[0024] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0025] Please see Figure 1-2 A building equipment intelligent linkage system based on digital twins includes: a data acquisition unit, a data processing unit, a virtual-real integration modeling unit, and a full-domain intelligent linkage control unit; The data acquisition unit extracts the full set of operating parameters of the building's intelligent devices and the set of environmental spatiotemporal sensing elements to obtain the original dataset; The data processing unit encapsulates and identifies the data structure based on the original dataset through time-slice tracing and embedding, and obtains a unique tracing and embedding code for the transmission data matrix; it uses spatiotemporal dimension fusion modeling and metadata completion to obtain spatiotemporally deeply coupled real-time intelligent analysis data of buildings. The virtual-real hybrid modeling unit, based on the real-time intelligent analysis data of the building, uses the status digital of the building equipment to perform twin mapping, captures the physical state of the building equipment and translates the digital signals, and obtains the real-time digital state flow of the equipment in the twin model; it replicates the virtual operation scenario and performs multi-mode simulation operation to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment. The global intelligent control unit extracts energy-saving equipment operation data based on the energy consumption simulation analysis result set and operation mode library, predicts the development trend of equipment failure, and generates a set of energy-saving control instructions for building equipment.

[0026] Specifically, the extraction of the full set of operating parameters of building intelligent devices and the spatiotemporal sensing element set of the environment is as follows: based on the full-domain deployment topology map of building intelligent devices, the operating parameters of intelligent devices and the spatiotemporal sensing data of the full-domain environment of the building are extracted simultaneously to obtain a multi-source hash data cluster; The validity of multi-source hash data clusters is verified by using heterogeneous data screening and identification technology to obtain the initial equipment environment data body; By integrating the equipment parameters and environmental data, a complete set of operational parameters for building intelligent devices and a set of spatiotemporal sensing elements for the environment are obtained.

[0027] In this embodiment, a topology map of the entire building's intelligent equipment deployment is drawn. The map accurately marks the physical installation location, equipment model, communication interface, sensor node affiliation, and network communication links between each device. It also associates the spatial information of building floors and functional areas (office areas, meeting rooms, corridors, and computer rooms). Based on this map, a multi-protocol data acquisition gateway (compatible with industrial protocols such as Modbus, Profinet, and BACnet) is deployed. The gateway synchronously extracts the operating parameters of each intelligent device in real time (such as the compressor frequency and fan speed of HVAC, and the on / off status and power of lighting) and the environmental spatiotemporal sensing data of each area of ​​the building (such as the temperature and humidity and CO2 concentration of office areas on each floor, collected by time and region). This forms a multi-source, heterogeneous, and irregular multi-source hash data cluster.

[0028] Heterogeneous data screening technology is adopted. By writing data validity verification scripts and setting judgment rules for null values, mutation values, and duplicate values, multi-source hash data clusters are automatically verified. Invalid null values ​​generated during the collection process, abnormal mutation data that exceeds the normal operating range of the equipment, and duplicate and redundant data from the same collection node are directly removed. At the same time, heterogeneous data of different formats (such as numerical, character, and binary) are standardized and converted into numerical structured data to obtain the initial equipment environment data body.

[0029] By using data classification and integration tools, the initial equipment environmental data volume is integrated according to equipment type, environmental monitoring area, and data acquisition dimension. The operating parameters of the same type of equipment are grouped into one cluster, and the environmental sensing data of the same area are grouped into one set. Finally, a structured and clearly classified cluster of all operating parameters of building intelligent equipment and a set of environmental spatiotemporal sensing elements are formed as the system's native dataset.

[0030] Specifically, the process of data structure encapsulation and identification includes: using time-slicing frame segmentation technology to decompose continuous data into frames according to a preset time granularity to obtain a time-series data frame cluster; Frame data is implanted through on-chain traceability embedding to obtain time-series data frames; The data fields are standardized and formatted using a pre-defined structured format to obtain a unique traceable embedded code for the transmission data matrix.

[0031] In this embodiment, based on the real-time requirements of office building equipment management (second-level collection, minute-level analysis), a fixed time granularity of 1 minute is preset. Time slicing and frame segmentation technology is adopted. Through the streaming data framing tool of the data processing platform, the continuously collected time-series raw data is automatically segmented and decomposed into frames, splitting the uninterrupted data stream into several independent data frames with 1 minute as the unit. All data frames are arranged in order according to the collection time to form a time-series data frame cluster.

[0032] By using on-chain traceability embedding technology, a unique traceability embedding code is generated for each data frame (composed of collection gateway number + device number + collection timestamp + region code). The traceability embedding code is then embedded into the header field of the data frame using a data embedding tool, so that each data frame has a unique traceability identifier, forming a time-series data frame.

[0033] A unified structured data format is preset (fields are defined as: traceability code, device number, area code, acquisition time, data type, and data value). Through a data format standardization tool, all time-series data frames with traceability codes are batch-standardized, and the field names, field order, and data storage format are unified. Missing fields are padded with zeros to form a communication data matrix with a unique traceability code and a completely unified structure, which is then stored in a distributed database.

[0034] Specifically, the process of modeling and metadata completion through spatiotemporal dimensional fusion includes: establishing a three-dimensional mapping relationship between data acquisition time, device spatial location, and feature parameters based on the communication data matrix through spatiotemporal dimension anchoring; Metadata topology completion supplements the full lifecycle metadata and environment-related metadata to obtain topological spatiotemporal related data; By using multi-dimensional data fusion to integrate and reconstruct equipment parameters, environmental data, and metadata, we can obtain real-time intelligent building analysis data with deep spatiotemporal coupling.

[0035] In this embodiment, based on the communication data matrix, and using acquisition time and device spatial location as dual anchor points, a three-dimensional mapping relationship between acquisition time, device spatial location, and feature parameters is established in the data modeling platform through spatial and time indexes. The formula is as follows: , in, (Time-related weight) (Spatial correlation weight), W p Associate weights with parameter attributes; These are the weighting coefficients; This allows each feature parameter to be precisely associated with a specific physical location (such as the HVAC equipment in a conference room on a certain floor) and a specific collection time, achieving precise spatiotemporal anchoring of the data.

[0036] Metadata topology and completion work is carried out. Through manual input and automatic retrieval of equipment files, the full life cycle metadata of the equipment (including: equipment serial number, manufacturer, installation time, calibration record, historical operation and maintenance record, and fault record) and environmental related metadata (including: regional functional attributes, personnel flow patterns, sensor node installation height, environmental influencing factors such as window orientation and sunshine duration) are supplemented. The completed metadata is then linked and bound to the communication data matrix to obtain the spatiotemporal correlation data of the topology.

[0037] By utilizing multi-dimensional data fusion technology and employing data fusion algorithms (such as weighted fusion), the data processing platform performs deep fusion and reconstruction of equipment operating parameters, environmental sensing data, and completed metadata. This eliminates redundant information and disconnects between data, enabling the three types of data to form a strong spatiotemporal correlation, resulting in spatiotemporally deeply coupled real-time intelligent building analysis data.

[0038] Specifically, the process of using the status data of building equipment for twin mapping includes: extracting the operating status features and physical attribute features of the building equipment to obtain the equipment status feature spectrum; By using digital twin mapping technology, a digital state mapping model that matches the physical device geometrically and has consistent attributes is constructed in virtual space to obtain the device's digital twin base model. By adapting the digital twin base model of the device to the twin model, the real-time mapping and linkage between the physical state of the building equipment and the digital model can be obtained.

[0039] In this embodiment, based on real-time intelligent building data, a feature extraction algorithm is used to extract dynamic operating status features (such as HVAC operating load, number of start-stop cycles, and adjustment rate) from equipment operating data. At the same time, static physical attribute features (such as equipment model, rated power, geometric dimensions, and component composition) are extracted from the original equipment manufacturer's design drawings and technical files. The dynamic and static features are integrated according to the equipment number to form a structured equipment status feature spectrum.

[0040] Using the Unity digital twin platform and a 1:1 geometric modeling approach, a 3D model that perfectly matches the geometric dimensions and structural features of the physical equipment is constructed in virtual space. Simultaneously, the equipment's state characteristic spectrum is embedded into the attribute module of the virtual model, giving it the same operational and physical attributes as the physical equipment, resulting in a digital twin base model of the equipment. All the digital twin base models of the equipment are imported into the overall digital twin model of the office building. Using model adaptation tools, a real-time data transmission channel (based on 5G / Industrial Ethernet) is established between the physical equipment and the virtual model, with a data synchronization frequency set at the second level, enabling real-time mapping and linkage between changes in the physical equipment's state and updates in the virtual model's state.

[0041] Specifically, the process of capturing and digitizing signals includes: obtaining the physical state stream of the equipment by capturing the physical operating state of the building equipment and the interaction state with the environment; The analog physical state signal is converted into an initial digital signal that can be recognized by the digital twin model, thus obtaining the initial digital state flow of the device; The amplitude and frequency of the digital signal are calibrated, and the protocol adaptation of the twin model is performed simultaneously to obtain the device real-time digital state stream with deep twin model adaptation.

[0042] In this embodiment, high-precision sensors, status monitoring terminals and data acquisition modules deployed on the equipment are used to capture the physical operating status of building equipment (such as compressor speed, fan air volume and lighting brightness) and the interaction status between the equipment and the environment (such as heat exchange between HVAC and indoor environment and linkage changes between lighting and ambient light) in real time and continuously. The captured continuous physical status is formed into a physical status stream of the equipment.

[0043] Using an analog-to-digital converter (A / D) module, the captured analog physical state signals (such as voltage, current, and temperature analog signals) are accurately converted into binary initial digital signals that can be recognized by the digital twin model, thus obtaining the initial digital state flow of the device.

[0044] The initial digital state stream is calibrated in amplitude and frequency using signal calibration tools to eliminate noise interference and numerical deviations during signal transmission and conversion, ensuring that the digital signal accurately corresponds to the physical state. Simultaneously, the calibrated digital signal is adapted to the communication protocol (MQTT protocol) of the digital twin model, and the digital signal is encapsulated into data packets that the model can parse, resulting in a real-time digital state stream of the device that is deeply adapted to the twin model, realizing the translation of the physical state into digital signals.

[0045] Specifically, the process of replicating and performing multi-mode simulation includes: obtaining a virtual operation scenario for the building equipment based on the real-time digital stream of the equipment and the real-time intelligent analysis data of the building; By presetting equipment energy-saving operation modes such as constant speed operation, variable frequency energy saving, and zone control through operation mode configuration technology, a multi-configuration energy-saving operation mode can be obtained; The corresponding modes are simulated in a virtual operation scenario to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment.

[0046] In this embodiment, the real-time digital state stream of the equipment and the real-time intelligent analysis data of the building are synchronously imported into the overall digital twin model of the office building. The equipment operation scenario, environmental change scenario and equipment-environment interaction scenario of the physical building are replicated 1:1 in the virtual space, restoring the real-time operating status of each device and the real-time environmental parameters of each area, and constructing a virtual operation and simulation background of building equipment that is completely synchronized with the physical building.

[0047] Through operation mode configuration technology, three types of energy-saving operation modes are preset in the virtual operation platform: ① Constant speed operation mode (the equipment runs at the rated constant speed, which is suitable for equipment under low load and stable conditions); ② Variable frequency energy-saving mode (the equipment adjusts the operating frequency according to load changes, which is suitable for load fluctuation conditions); ③ Zonal control mode (the equipment is controlled in different areas according to the load differences in building areas, which is suitable for uneven load conditions in different areas). The operation logic, control parameters and applicable scenarios are defined for each mode to form a multi-configuration energy-saving operation mode.

[0048] In the virtual operation scenario, the three types of energy-saving operation modes are simulated in their entirety. The simulation shows the operating status of equipment, energy consumption, collaborative linkage effect and environmental adaptability under different modes. The simulation results are quantitatively calculated and analyzed by energy consumption analysis tools to form a set of energy consumption simulation analysis results for energy-saving equipment. At the same time, the operation modes that have been verified by simulation and meet the energy-saving effect standards are standardized and stored to form a library of operation modes that can be directly called.

[0049] Specifically, the process of extracting energy-saving equipment operation data includes: obtaining operation parameters, energy consumption data and pattern matching features through data feature filtering; Simultaneously, data detection and logical contradiction verification are performed to obtain energy-saving equipment operation data; Principal component analysis algorithm is used to compress the dimensions and refine the features of the equipment operation data to obtain structured energy-saving equipment operation data.

[0050] In this embodiment, based on the energy consumption simulation analysis result set and operation mode library, and according to the needs of equipment fault prediction and energy-saving decision-making, a data feature filtering algorithm (random forest feature filtering method) is used. The formula is as follows: , Among them, VI i Let be the importance value of the i-th feature, T be the number of decision trees, and err be the value of the i-th feature. t Let err be the prediction error for the t-th tree. t,i The prediction error of the t-th tree after shuffling the i-th feature.

[0051] The core operational parameters (operating frequency, load rate), actual energy consumption data (power, electricity consumption) and pattern matching characteristics (load fluctuation range, regional adaptability) of the equipment are accurately extracted from the full data, and redundant data that is irrelevant to the core requirements are eliminated.

[0052] By using data detection tools and logic verification scripts, the filtered feature data is subjected to multi-dimensional detection and logical contradiction verification: the completeness of the data (no missing fields) and accuracy (values ​​within a reasonable range) are detected, the logical correlation between data is verified (the positive correlation between equipment operating frequency and energy consumption) is verified, and abnormal and distorted data and logically contradictory data are removed to obtain real and effective energy-saving equipment operation data.

[0053] Principal Component Analysis (PCA) algorithm is used to compress dimensions and refine features of energy-saving equipment operation data in the data analysis platform. The high-dimensional, highly correlated original data is transformed into low-dimensional, uncorrelated principal component data. The core feature dimensions in the data are extracted, the data structure is simplified, and structured, high-value energy-saving equipment operation data is obtained.

[0054] Specifically, the process of predicting the development trend of equipment failures includes: integrating cross-dimensional deconstruction of historical failure records based on the energy-saving equipment operation data to obtain an energy-saving equipment full life cycle operation analysis library; By using a trend prediction mechanism to dynamically model operating parameters and extract fault correlation features, a fault dataset is obtained. The fault dataset is input into a pre-trained fault model to obtain prediction results.

[0055] In this embodiment, structured energy-saving equipment operation data is integrated with historical fault records of office building equipment. By deconstructing historical fault records across dimensions, the system analyzes the related factors such as equipment operating parameters, environmental conditions, operating modes, and load changes at the time of the fault. The deconstructed historical fault information is then associated and bound with real-time equipment operation data according to the equipment number. This constructs a full lifecycle operation analysis library for energy-saving equipment covering the entire lifecycle of equipment procurement, installation, operation, and maintenance, and stores it in a data warehouse.

[0056] By utilizing trend prediction mechanisms and combining them with time series analysis, dynamic modeling of real-time equipment operating parameters is performed. By extracting characteristic indicators related to equipment failures (such as abnormal equipment operating temperature and excessive frequency fluctuations), abnormal trends in equipment operating parameters are identified, and abnormal characteristic data are integrated to form a fault dataset.

[0057] The fault dataset is input into a pre-trained fault prediction model (which has been trained and optimized with a large amount of equipment operation data and typical fault cases). The model analyzes the current operating status of the equipment through calculation and outputs accurate prediction results of the equipment fault development trend, including potential fault types, fault occurrence probability, possible time and fault impact, providing a basis for predictive maintenance.

[0058] Specifically, the specific process of the operation mode configuration technology includes: performing multi-dimensional deconstruction analysis of the operating characteristics of energy-saving equipment based on the virtual operation scenario of the building equipment, performing hierarchical decomposition and mapping of energy-saving targets, and obtaining the basis for mode configuration; By performing modular operations of constant speed operation, frequency conversion energy saving and zone control modes, the corresponding mode operation logic is defined in a closed loop, and the dynamic threshold of mode switching conditions is set to obtain the basic mode template. By using the multi-scenario calibration and optimization of the basic mode template line mode parameter thresholds, the pre-simulation results of the mode operation effect are verified, and a multi-configuration energy-saving operation mode is obtained.

[0059] In this embodiment, based on the virtual operation scenario of building equipment, the operating characteristics of various energy-saving devices are analyzed from multiple dimensions to clarify the operating efficiency, energy consumption characteristics, dynamic adjustment capabilities, load adaptability, and collaborative linkage characteristics of the equipment. Simultaneously, combined with the overall energy-saving goals of the office building, the overall goals are hierarchically decomposed and mapped according to equipment type, functional area, and operating period, decomposing them into specific energy-saving indicators for each device and area, forming the basis for the operation mode configuration. Modular operation design is implemented for three modes: constant speed operation, variable frequency energy saving, and zone control. Independent operating logic programs are written for each mode, defining closed-loop control processes (e.g., data acquisition-analysis-adjustment-feedback) and equipment linkage rules. Furthermore, based on equipment operating characteristics, environmental change characteristics, and energy-saving indicator requirements, dynamic thresholds for mode switching are set, forming a standardized basic mode template. Using a basic model template, and combining different operating scenarios of office buildings (weekday morning peak, weekday off-peak, rest day, and holidays), the threshold values ​​of the model parameters are calibrated and optimized for multiple scenarios. The calibrated model is then simulated in a virtual operation scenario to verify whether the simulation results meet the energy-saving goals and equipment safety operation requirements. Based on the verification results, the model parameters are adjusted and improved to obtain a multi-configuration energy-saving operation mode that is adapted to multiple scenarios.

[0060] Specifically, the process of generating energy-saving decision instructions includes: based on the linkage results between the twin model and the intelligent device, and combined with the predictive maintenance scheme, obtaining the energy-saving optimization target and operating constraints of the intelligent device; By introducing energy-saving operation modes and combining energy-saving strategies such as dynamic HVAC adjustment and on-demand lighting control, the energy-saving logic of collaborative linkage of intelligent devices is deduced. An energy-saving decision instruction set is generated based on the energy-saving logic.

[0061] In this embodiment, based on the real-time linkage results between the digital twin model and the physical equipment, combined with the prediction results of the equipment failure development trend and the formulated predictive maintenance plan, the energy-saving optimization goals of each intelligent device are clarified. At the same time, based on the equipment's safe operating parameters, service life, and maintenance cycle, the operating constraints of the equipment (such as the compressor frequency adjustment range and the lower limit of lighting brightness adjustment) are determined.

[0062] Import energy-saving operation modes adapted to the current operating scenario of the office building from the operation mode library. Combine them with special energy-saving strategies for dynamic adjustment of HVAC and on-demand control of lighting. Through intelligent decision-making algorithms (genetic algorithms), comprehensively analyze the real-time operating status, energy-saving potential and linkage relationship of each device, deduce the energy-saving logic of multi-device collaborative linkage, and clarify the control methods (such as frequency conversion and amplitude modulation), core control parameters, control sequence and collaborative linkage rules (such as the linkage adjustment of HVAC and fresh air system) of each device.

[0063] Based on the derived energy-saving logic, various control requirements are transformed into standardized and executable equipment control instructions (instruction format: equipment number + control type + control parameter + execution time). These instructions are then categorized and integrated according to equipment type and control area to form a set of energy-saving control instructions for building equipment. The instruction set must meet the equipment operating constraints and energy-saving optimization goals.

[0064] Specifically, the specific process of HVAC dynamic adjustment includes: performing dynamic quantitative calculation of HVAC operating load based on the real-time intelligent analysis data of the building, and simultaneously making short-term predictions of load fluctuation trends to obtain load assessment results; Based on the load assessment results, the compressor frequency and fan speed are adjusted to adapt to the scenario of zoned temperature control logic. At the same time, the safety threshold of parameter adjustment range is limited to obtain a dynamic adjustment parameter set.

[0065] In this embodiment, based on real-time intelligent building data deeply coupled with spatiotemporal data, and combined with the functional attributes of each area of ​​the office building, real-time personnel density, ambient temperature and humidity, CO2 concentration, heat load changes, and external meteorological conditions (such as outdoor temperature and solar radiation intensity), the real-time operating load of the central air conditioning system is dynamically quantified and calculated using an HVAC load calculation model (such as the cooling load coefficient method). At the same time, a load trend prediction algorithm is used to make short-term predictions of load fluctuation trends. The combined load calculation results and prediction results form a comprehensive HVAC operating load assessment result.

[0066] Based on the load assessment results, the HVAC intelligent controller dynamically and precisely adjusts the compressor operating frequency and fan speed of the central air conditioning system to ensure that the equipment operating load is highly matched with the actual heat load demand of the building. At the same time, according to the load differences, temperature and humidity control requirements and personnel distribution characteristics of each area, the zone temperature control logic is adapted to the scenario and the air volume, water supply temperature and temperature control threshold of each area are adjusted.

[0067] During parameter adjustment, a safety threshold for the adjustment range is set for the core parameters to avoid damage to the equipment due to sudden parameter changes or excessive adjustment range. The adjustment parameters, safety thresholds and zone temperature control logic are integrated to form a set of dynamic HVAC adjustment parameters that take into account load matching, zone adaptation, energy efficiency and equipment safety, and are then sent to the HVAC field controller for execution.

[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A building equipment intelligent linkage system based on digital twins, characterized in that, include: Data acquisition unit, data processing unit, virtual-real integration modeling unit, and global intelligent interconnection and control unit; The data acquisition unit extracts the full set of operating parameters of the building's intelligent devices and the set of environmental spatiotemporal sensing elements to obtain the original dataset; The data processing unit encapsulates and identifies the data structure based on the original dataset through time-slice tracing and embedding, and obtains a unique tracing and embedding code for the transmission data matrix; it uses spatiotemporal dimension fusion modeling and metadata completion to obtain spatiotemporally deeply coupled real-time intelligent analysis data of buildings. The virtual-real hybrid modeling unit, based on the real-time intelligent analysis data of the building, uses the status digital of the building equipment to perform twin mapping, captures the physical state of the building equipment and translates the digital signals, and obtains the real-time digital state flow of the equipment in the twin model; it replicates the virtual operation scenario and performs multi-mode simulation operation to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment. The global intelligent control unit extracts energy-saving equipment operation data based on the energy consumption simulation analysis result set and operation mode library, predicts the development trend of equipment failure, and generates a set of energy-saving control instructions for building equipment.

2. The system according to claim 1, characterized in that, The extracted full set of operational parameters of building intelligent devices and the set of environmental spatiotemporal sensing elements are as follows: Based on the topology map of the full-domain deployment of intelligent building devices, the operating parameters of intelligent devices and the spatiotemporal sensing data of the entire building environment are extracted simultaneously to obtain multi-source hash data clusters; The validity of multi-source hash data clusters is verified by using heterogeneous data screening and identification technology to obtain the initial equipment environment data body; By integrating the equipment parameters and environmental data, a complete set of operational parameters for building intelligent devices and a set of spatiotemporal sensing elements for the environment are obtained.

3. The system according to claim 1, characterized in that, The specific process of data structure encapsulation and identification includes: By using time-slicing frame segmentation technology, continuous data is decomposed into frames according to a preset time granularity to obtain time-series data frame clusters; Frame data is implanted through on-chain traceability embedding to obtain time-series data frames; The data fields are standardized and formatted using a pre-defined structured format to obtain a unique traceable embedded code for the transmission data matrix.

4. The system according to claim 1, characterized in that, The specific process of spatiotemporal dimension fusion modeling and metadata completion includes: Based on the aforementioned communication data matrix, a three-dimensional mapping relationship between data acquisition time, device spatial location, and characteristic parameters is established through spatiotemporal anchoring. Metadata topology completion supplements the full lifecycle metadata and environment-related metadata to obtain topological spatiotemporal related data; By using multi-dimensional data fusion to integrate and reconstruct equipment parameters, environmental data, and metadata, we can obtain real-time intelligent building analysis data with deep spatiotemporal coupling.

5. The system according to claim 1, characterized in that, The specific process of using the status data of building equipment for twin mapping includes: Extract the operational status characteristics and physical attribute characteristics of building equipment to obtain the equipment status characteristic spectrum; By using digital twin mapping technology, a digital state mapping model that matches the physical device geometrically and has consistent attributes is constructed in virtual space to obtain the device's digital twin base model. By adapting the digital twin base model of the device to the twin model, the real-time mapping and linkage between the physical state of the building equipment and the digital model can be obtained.

6. The system according to claim 1, characterized in that, The specific process of capturing and digitizing signals includes: By capturing the physical operating state of building equipment and its interaction with the environment, the physical state flow of the equipment is obtained; The analog physical state signal is converted into an initial digital signal that can be recognized by the digital twin model, thus obtaining the initial digital state flow of the device; The amplitude and frequency of the digital signal are calibrated, and the protocol adaptation of the twin model is performed simultaneously to obtain the device real-time digital state stream with deep twin model adaptation.

7. The system according to claim 1, characterized in that, The specific process of replicating and simulating multi-mode operation includes: Based on the real-time digital state stream of the equipment and the real-time intelligent analysis data of the building, a virtual operation scenario of the building equipment is obtained; By presetting equipment energy-saving operation modes such as constant speed operation, variable frequency energy saving, and zone control through operation mode configuration technology, a multi-configuration energy-saving operation mode can be obtained; The corresponding modes are simulated in a virtual operation scenario to obtain the energy consumption simulation analysis result set and operation mode library of energy-saving equipment.

8. The system according to claim 1, characterized in that, The specific process for extracting operating data from energy-saving equipment includes: By filtering data features, operational parameters, energy consumption data, and pattern matching features can be obtained. Simultaneously, data detection and logical contradiction verification are performed to obtain energy-saving equipment operation data; Principal component analysis algorithm is used to compress the dimensions and refine the features of the equipment operation data to obtain structured energy-saving equipment operation data.

9. The system according to claim 1, characterized in that, The specific process for predicting the development trend of equipment failure includes: Based on the operating data of the energy-saving equipment, the cross-dimensional deconstruction and integration of historical fault records are used to obtain the energy-saving equipment full life cycle operation analysis library; By using a trend prediction mechanism to dynamically model operating parameters and extract fault correlation features, a fault dataset is obtained. The fault dataset is input into a pre-trained fault model to obtain prediction results.

10. The system according to claim 1, characterized in that, The specific process of the operation mode configuration technology includes: Based on the virtual operation scenario of the building equipment, a multi-dimensional deconstruction analysis of the operating characteristics of the energy-saving equipment is performed, and the energy-saving target is hierarchically decomposed and mapped to obtain the mode configuration basis; By performing modular operations of constant speed operation, frequency conversion energy saving and zone control modes, the corresponding mode operation logic is defined in a closed loop, and the dynamic threshold of mode switching conditions is set to obtain the basic mode template. By using the multi-scenario calibration and optimization of the basic mode template line mode parameter thresholds, the pre-simulation results of the mode operation effect are verified, and a multi-configuration energy-saving operation mode is obtained.

11. The system according to claim 1, characterized in that, The specific process for generating energy-saving decision instructions includes: Based on the linkage results between the twin model and the intelligent device, and combined with the predictive maintenance scheme, the energy-saving optimization target and operating constraints of the intelligent device are obtained. By introducing energy-saving operation modes and combining energy-saving strategies such as dynamic HVAC adjustment and on-demand lighting control, the energy-saving logic of collaborative linkage of intelligent devices is deduced. An energy-saving decision instruction set is generated based on the energy-saving logic.

12. The system according to claim 1, characterized in that, The specific process of HVAC dynamic adjustment includes: Based on the real-time intelligent analysis data of the building, the HVAC operating load is dynamically quantified and calculated, and the short-term trend of load fluctuation is predicted to obtain the load assessment result. Based on the load assessment results, the compressor frequency and fan speed are adjusted to adapt to the scenario of zoned temperature control logic. At the same time, the safety threshold of parameter adjustment range is limited to obtain a dynamic adjustment parameter set.

Images