A knowledge graph-based intelligent energy control method and system for dual-energy curtain walls

Abstract

This invention relates to the field of building energy regulation technology, and discloses a knowledge graph-based intelligent energy regulation method and system for dual-energy curtain walls. The system comprises an energy state perception module that acquires energy state perception values; a knowledge graph construction module that obtains a knowledge graph construction parameter set; an environmental correlation analysis module that generates an environmental correlation parameter set; a regulation threshold setting module that determines energy regulation thresholds; an energy trend prediction module that derives predicted energy distribution values; and a curtain wall feedback control module that formulates an intelligent energy balance regulation scheme for the dual-energy curtain wall. Through the synergistic effect of these modules, the system can comprehensively analyze environmental factors, energy state, and energy consumption data to achieve intelligent regulation of the dual-energy curtain wall's energy, optimize energy distribution, adapt to dynamically changing energy loads and environmental conditions, improve the energy utilization efficiency of the dual-energy curtain wall, and meet the needs of building energy conservation and indoor environmental comfort.

Description

Technical Field

[0001] This invention relates to the field of building energy regulation technology, specifically to a knowledge graph-based intelligent energy regulation method and system for dual-energy curtain walls. Background Technology

[0002] With the escalating global energy crisis and the deepening of the concept of sustainable development, the construction industry, as a key energy-consuming sector, has seen its energy-saving technology innovation and application become a focus of attention. Dual-energy curtain walls, as a new type of building envelope, integrate active and passive energy utilization technologies. Through rational design, they can achieve the collection and utilization of solar energy and the natural regulation of the indoor environment, demonstrating great potential in reducing building energy consumption. However, the energy regulation of dual-energy curtain walls still faces many challenges.

[0003] Traditional dual-energy curtain wall control methods rely heavily on preset fixed modes or simple sensor feedback, lacking comprehensive consideration of complex environmental factors and dynamic energy loads. In actual operation, due to fluctuations in internal building loads and real-time changes in external environmental conditions (such as light intensity, temperature, and wind speed), fixed control modes often fail to achieve optimal energy allocation. For example, operating according to the control parameters set in the morning during the midday sun may lead to indoor overheating, increasing the energy consumption of the air conditioning system; while on cloudy days or in the evening, over-reliance on active energy supply results in energy waste.

[0004] Existing control systems have significant shortcomings in data processing and analysis. Most systems can only collect and respond to single or a few environmental parameters, failing to establish correlations between multiple factors, leading to biased control decisions. Furthermore, the lack of effective utilization and knowledge accumulation of historical data makes it difficult to optimize current control strategies based on past operational experience. When faced with sudden changes in energy load, the systems often react slowly, unable to quickly adjust control modes to adapt to new energy demands, resulting in low energy efficiency.

[0005] Traditional control methods lack precision in predicting energy trends, making it difficult to accurately foresee changes in energy consumption over a future period, resulting in a lack of foresight in control measures. For example, in office buildings, there are significant differences in energy load between weekdays and holidays. If these changes cannot be predicted in advance and control strategies cannot be adjusted accordingly, a mismatch between energy supply and demand can easily occur. Furthermore, the setting of control thresholds is often based on empirical values ​​or static calculations, failing to incorporate real-time energy consumption data and dynamic updates to limit control precision and prevent the achievement of refined energy allocation.

[0006] Some studies have attempted to introduce intelligent algorithms to improve control effectiveness, but shortcomings remain in data correlation analysis and knowledge reasoning. Most systems fail to effectively integrate multi-dimensional data such as environmental factors, energy status, and control modes to form a systematic knowledge framework, resulting in a lack of a holistic perspective in control decisions. When faced with complex energy distribution conditions, it is difficult to quickly match the optimal control scheme, affecting the full realization of the energy-saving potential of dual-energy curtain walls. These problems have prevented dual-energy curtain walls from achieving the expected energy-saving effects in practical applications, hindering their large-scale promotion in the construction field. Summary of the Invention

[0007] The purpose of this invention is to provide a knowledge graph-based intelligent energy control method and system for dual-energy curtain walls, in order to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides a knowledge graph-based intelligent energy control system for dual-energy curtain walls, the system comprising:

[0009] The energy status sensing module analyzes the differences in regional energy parameters and loads based on the energy temperature, illuminance and energy consumption values ​​of each sub-region in the dual-energy curtain wall area, calculates the state differences between regions, integrates them into a state distribution parameter set, and obtains the energy status sensing value.

[0010] Based on the energy state perception value, the knowledge graph construction module extracts the parameter combination of curtain wall control mode and duration, selects the optimal mode and duration combination, and obtains the knowledge graph construction parameter set.

[0011] The environmental correlation analysis module constructs a parameter set based on the knowledge graph, extracts the current changes in environmental factors, analyzes the relationship between regulation modes and duration, matches environmental factors with regulation combinations, and obtains an environmental correlation parameter set.

[0012] The regulation threshold setting module extracts the current energy change value based on the environmental associated parameter set, combines it with real-time energy consumption data, allocates energy and load, sets a threshold, and applies the threshold to the regional energy allocation to obtain the energy regulation threshold.

[0013] The energy trend prediction module captures energy consumption data from energy sampling points based on the energy regulation threshold, performs knowledge reasoning on the energy consumption changes of energy sampling points using a knowledge graph, analyzes the energy consumption change trend corresponding to the energy load, categorizes and organizes the energy consumption change trend based on the reasoning results, and adjusts and analyzes the categorized data using a knowledge graph based on the energy consumption change information to obtain the predicted energy distribution value.

[0014] Based on the predicted energy distribution value, the curtain wall feedback control module analyzes the error value of energy and energy consumption through real-time energy and energy consumption data, and adjusts the curtain wall regulation in combination with the error value to obtain a dual-energy curtain wall intelligent energy balance regulation scheme.

[0015] Preferably, the energy status perception value includes a temperature parameter set, an illuminance parameter set, and an energy consumption difference parameter set; the knowledge graph construction parameter set includes a screening mode parameter and a duration parameter; the environmental association parameter set includes environmental factor change parameters and control mode duration matching parameters; the energy control threshold includes energy change parameters, energy consumption matching parameters, and threshold setting parameters; the energy distribution prediction value includes energy consumption trend analysis parameters and energy consumption relationship parameters; and the dual-energy curtain wall intelligent energy balance control scheme includes error analysis parameters and curtain wall adjustment parameters.

[0016] Preferably, the energy status sensing module includes:

[0017] The data acquisition submodule collects temperature, illuminance and energy consumption values ​​in each sub-region of the dual-energy curtain wall area in a regionalized manner, locates invalid data, removes abnormal data, and arranges the extracted temperature, illuminance and energy consumption values ​​in regional order to generate a regional energy status dataset.

[0018] The state difference analysis submodule analyzes the energy parameters between regions based on the regional energy state dataset, calculates the ratio of changes in regional parameters, sorts the differences between regions by weight, marks regions with excessive fluctuation differences, and obtains regional energy state difference data.

[0019] The state distribution integration submodule, based on the regional energy state difference data, calls the regional energy difference values ​​to perform multi-dimensional summarization, screens regional energy value differences, classifies them according to the value size, and arranges the regional energy values ​​in an orderly manner to generate energy state perception values.

[0020] Preferably, the knowledge graph construction module includes:

[0021] Based on the energy status perception value, the parameter extraction submodule identifies the control status of the curtain wall in each area, records the control mode and control duration of the curtain wall, standardizes the recorded data, sorts the standardized data by mode and duration, and generates a curtain wall parameter dataset.

[0022] The graph construction and optimization submodule analyzes the pattern and duration values ​​in the curtain wall parameter dataset, filters parameter combinations with high matching degree with energy status, records the matching results through pattern matching, adjusts the parameter combinations, and generates parameter combination optimization results.

[0023] The parameter selection submodule retrieves the optimization results of the parameter combination, determines the optimal matching mode and duration combination, adjusts the curtain wall control parameters, inputs the control configuration, verifies the stability of the parameter set, and generates a knowledge graph to construct the parameter set.

[0024] Preferably, the environmental correlation analysis module includes:

[0025] The environmental factor analysis submodule constructs a parameter set based on the knowledge graph, collects key data through environmental monitoring, including wind speed data, humidity data and radiation intensity data, performs time series analysis on the data, removes outliers, partitions the remaining data, and obtains environmental factor analysis data.

[0026] The parameter matching submodule analyzes the environmental factors analysis data to analyze the impact of environmental variables on the curtain wall control mode and duration, calculates the degree of impact of each environmental factor change on parameter adjustment, determines the optimal matching parameter settings based on the degree of impact, iteratively adjusts parameters to capture the optimal combination, and obtains the parameter docking results.

[0027] The associated parameter integration submodule selects the combination of modes and durations that match the current environmental conditions from the parameter docking results, conducts parameter adjustment experiments, optimizes parameter settings through multiple adjustments and verifications, determines and solidifies the parameters as operating standards, and generates an environmental associated parameter set.

[0028] Preferably, the control threshold setting module includes:

[0029] Based on the environmental correlation parameter set, the energy extraction submodule locates energy change monitoring points, extracts energy change values ​​in the monitoring area, continuously records energy increase / decrease rates, extracts multiple key change nodes corresponding to the change rates, sorts the node values ​​in order, and obtains the current energy change characteristic value.

[0030] The energy consumption matching submodule analyzes the node change value and real-time energy consumption data based on the current energy change characteristic value, and performs calibration according to the predetermined matching criteria. It then calls the matching criteria to perform distribution redistribution within the energy consumption range to obtain the energy consumption matching structure.

[0031] Based on the energy consumption matching structure, the load distribution submodule uses a dynamic threshold adjustment method to calculate the distribution of energy consumption among energy change nodes, sets upper and lower limits for node thresholds, applies the thresholds to regional energy values, and distributes them to obtain energy regulation thresholds.

[0032] Preferably, the dynamic threshold adjustment method includes obtaining the allocated load value of the matching energy change node, setting the upper and lower limits of the node threshold, applying the threshold to the regional energy value and allocating it.

[0033] Preferably, the energy trend prediction module includes:

[0034] The energy data capture submodule, based on the energy regulation threshold, applies a knowledge reasoning algorithm to capture energy consumption data at sampling points, remove outliers and correct errors, stores the data in a hierarchical manner according to intervals, performs knowledge-based processing, and generates a knowledge-based energy consumption dataset.

[0035] The energy load analysis submodule, based on the knowledge-based energy consumption dataset, divides the energy load into intervals, extracts the changing trends and fluctuation characteristics, and generates a set of energy load and energy consumption change characteristics.

[0036] The trend distribution inference submodule adjusts the feature parameters and calibrates the trend data based on the energy load and energy consumption change feature set, extracts the distribution interval, and performs numerical prediction to obtain the predicted energy distribution value.

[0037] The knowledge reasoning algorithm includes calculating energy knowledge values ​​and generating a knowledge-based energy consumption dataset. The generated knowledge-based energy consumption dataset is processed by combining the energy consumption values, environmental values, and load factors of the sampling points.

[0038] Preferably, the curtain wall feedback control module includes:

[0039] The error analysis submodule extracts real-time energy and energy consumption data based on the energy distribution prediction value, analyzes the real-time energy consumption value and the prediction value, matches the energy consumption difference with the current energy information, and generates an energy consumption error value.

[0040] The parameter adjustment submodule sets the control and adjustment parameters of the curtain wall based on the energy consumption error value. It sets the adjustment range for areas with large errors and makes fine adjustments for areas with low errors. By comparing the control effects, it selects and integrates matching parameter sets to generate a curtain wall adjustment parameter set.

[0041] The regulation and control submodule applies the adjustment parameters at each curtain wall location based on the curtain wall adjustment parameter set, implements regulation operations item by item, monitors energy and energy consumption synchronously, gradually adjusts the order of regulation operations in each area, and generates a dual-energy curtain wall energy intelligent balance regulation scheme.

[0042] Preferably, the present invention also includes a knowledge graph-based intelligent energy control method for dual-energy curtain walls, applied to the aforementioned knowledge graph-based intelligent energy control system for dual-energy curtain walls, the method comprising the following steps:

[0043] Step 1: Energy Status Sensing Step. Based on the energy temperature, illuminance and energy consumption values ​​of each sub-region in the dual-energy curtain wall area, analyze the differences in regional energy parameters and loads, calculate the status differences between regions, integrate them into a status distribution parameter set, and obtain the energy status sensing value.

[0044] Step 2: Knowledge graph construction step. Based on the energy state perception value, extract the parameter combination of curtain wall control mode and duration, select the optimal mode and duration combination, and obtain the knowledge graph construction parameter set.

[0045] Step 3: Environmental correlation analysis step, constructing a parameter set based on the knowledge graph, extracting the current changes in environmental factors, analyzing the relationship between regulation mode and duration, matching environmental factors with regulation combinations, and obtaining the environmental correlation parameter set;

[0046] Step 4: Setting the control threshold. Based on the environmental correlation parameter set, extract the current energy change value, combine it with real-time energy consumption data, allocate energy and load, set a threshold, and apply the threshold to the regional energy allocation to obtain the energy control threshold.

[0047] Step 5: Energy trend prediction step. Based on the energy regulation threshold, capture energy consumption data of energy sampling points, combine knowledge graph to perform knowledge reasoning on the energy consumption changes of energy sampling points, analyze the energy consumption change trend corresponding to energy load, classify and organize the energy consumption change trend according to the reasoning results, and combine energy consumption change information to adjust and analyze the classified data using knowledge graph to obtain the predicted value of energy distribution.

[0048] Step 6: Curtain wall feedback control step. Based on the predicted energy distribution value, the error value of energy and energy consumption is analyzed through real-time energy and energy consumption data. The curtain wall regulation is adjusted in combination with the error value to obtain the dual-energy curtain wall intelligent energy balance regulation scheme.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] This knowledge graph-based intelligent energy control system for dual-energy curtain walls comprehensively captures energy temperature, illuminance, and energy consumption values ​​of each sub-region within the dual-energy curtain wall area through an energy status sensing module. It analyzes regional energy parameters and load differences to form a status distribution parameter set, providing precise initial data support for subsequent control. The knowledge graph construction module extracts parameter combinations of curtain wall control modes and durations based on the sensed values ​​and filters for the optimal solution. It integrates historical control experience with real-time data to form a structured knowledge system, making control decisions systematic and predictable.

[0051] The environmental correlation analysis module combines knowledge graph construction parameters with current environmental factor changes to deeply analyze the relationship between control modes and duration, achieving precise matching between environmental factors and control combinations. This allows control strategies to be flexibly adjusted according to environmental changes, avoiding the limitations of traditional fixed modes. The control threshold setting module combines environmental correlation parameters with real-time energy consumption data to rationally allocate energy and load and set thresholds, ensuring a more balanced distribution of energy across regions and reducing energy waste and supply-demand imbalances.

[0052] The energy trend prediction module uses a knowledge graph to reason about energy consumption changes at energy sampling points, analyzes and categorizes energy consumption trends, and adjusts the knowledge graph based on energy consumption information. This improves the accuracy of energy distribution prediction, enabling the system to anticipate energy demand in advance and creating conditions for proactive regulation. The curtain wall feedback control module analyzes errors using real-time data and adjusts the control scheme, achieving a dynamic balance between energy and energy consumption. This ensures that the dual-energy curtain wall maintains high-efficiency operation under different working conditions.

[0053] The system's modules work together to form a closed-loop intelligent control process, from data perception, knowledge construction, environmental correlation, threshold setting, trend prediction to feedback control. This effectively solves the problems of response lag, poor adaptability, and unreasonable energy distribution in traditional dual-energy curtain wall control, and promotes the development of dual-energy curtain wall energy control towards intelligence and precision, adapting to the dual needs of modern buildings for energy conservation and comfort. Attached Figure Description

[0054] Figure 1 This is a schematic diagram illustrating the working principle of the dual-energy curtain wall intelligent energy control system based on knowledge graphs as described in this invention.

[0055] Figure 2 This is a schematic diagram of the working principle of the energy status sensing module.

[0056] Figure 3 A schematic diagram illustrating the working principle of the energy trend forecasting module;

[0057] Figure 4 This is a schematic diagram illustrating the working principle of a knowledge graph-based intelligent energy control method for dual-energy curtain walls. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] Please see Figures 1-4 This invention provides a knowledge graph-based intelligent energy control system for dual-energy curtain walls. The system includes: an energy status perception module, a knowledge graph construction module, an environmental correlation analysis module, a control threshold setting module, an energy trend prediction module, and a curtain wall feedback control module. The specific implementation steps are as follows:

[0060] The energy status sensing module analyzes the differences between regional energy parameters and loads based on the energy temperature, illuminance, and energy consumption values ​​of each sub-region in the dual-energy curtain wall area, calculates the status differences between different regions, integrates this information into a status distribution parameter set, and then obtains the energy status sensing value.

[0061] The knowledge graph construction module is based on energy status perception values, extracts parameter combinations of curtain wall control modes and durations, selects the optimal mode and duration combination, and forms a knowledge graph construction parameter set.

[0062] The environmental correlation analysis module uses knowledge graphs to construct parameter sets, extracts the changes in current environmental factors, analyzes the relationship between regulation patterns and duration, matches environmental factors with regulation combinations, and obtains environmental correlation parameter sets.

[0063] The regulation threshold setting module extracts the current energy change value based on the environmental correlation parameter set, combines it with real-time energy consumption data to allocate energy and load, sets the threshold and applies it to regional energy allocation to obtain the energy regulation threshold.

[0064] The energy trend prediction module captures energy consumption data from energy sampling points based on energy regulation thresholds, performs knowledge reasoning on energy consumption changes using a knowledge graph, analyzes the energy consumption change trends corresponding to energy loads, categorizes and organizes the trends based on the reasoning results, and adjusts and analyzes the knowledge graph in conjunction with energy consumption change information to obtain predicted energy distribution values.

[0065] The curtain wall feedback control module is based on the predicted energy distribution value. It performs error value analysis through real-time energy and energy consumption data, and adjusts the curtain wall regulation in combination with the error value to finally obtain the intelligent energy balance regulation scheme for dual-energy curtain walls.

[0066] Example 1:

[0067] The energy status perception values ​​cover temperature parameter sets, illuminance parameter sets, and energy consumption difference parameter sets. The knowledge graph construction parameter set includes screening mode parameters and duration parameters. The environmental correlation parameter set involves environmental factor change parameters and control mode duration matching parameters. The energy control threshold consists of energy change parameters, energy consumption matching parameters, and threshold setting parameters. The energy distribution prediction values ​​include energy consumption trend analysis parameters and energy consumption relationship parameters. The dual-energy curtain wall energy intelligent balance control scheme includes error analysis parameters and curtain wall adjustment parameters.

[0068] The energy status sensing module includes: a data acquisition submodule, a status difference analysis submodule, and a status distribution integration submodule. The operation of the energy status sensing module is as follows: The data acquisition submodule focuses on the energy temperature, illuminance, and energy consumption values ​​of each sub-area within the dual-energy curtain wall area, conducting regionalized data acquisition. During the acquisition phase, based on the structural layout and functional zoning of the dual-energy curtain wall, the entire area is divided into several sub-areas. Multiple data acquisition points are set up in each sub-area to ensure comprehensive and accurate capture of the energy status information of that area. For temperature values, temperature sensors distributed at different heights and locations are used for real-time acquisition. The sensor density is determined based on the area and temperature field distribution characteristics to cover potential temperature differences within the area. For illuminance values, photosensitive sensors collect the light intensity formed by natural and artificial light in the curtain wall area. Multiple points are collected according to the regional division, recording illuminance changes at different times. Energy consumption values ​​are collected through metering devices connected to the curtain wall's energy-consuming equipment, including energy consumption data from heating, cooling, and lighting systems, and the energy consumption per unit time is statistically analyzed by region.

[0069] During data acquisition, data verification is performed simultaneously. The collected temperature, illuminance, and energy consumption values ​​are initially screened within preset reasonable ranges to identify abnormal data outside these ranges. Examples include temperature sensor malfunctions resulting in values ​​significantly higher or lower than normal ambient temperatures, and abnormally low or drastically fluctuating illuminance values ​​caused by shading of the illuminance sensor. Simultaneously, invalid data is located through data continuity checks, such as data stream interruptions or missing values ​​due to communication outages. These abnormal and invalid data are removed from the raw data, retaining only the valid data that meets the requirements. Then, according to the sub-region division order, the extracted valid temperature, illuminance, and energy consumption values ​​are organized and arranged to form a structured regional energy status dataset. Each sub-region corresponds to a complete set of temperature, illuminance, and energy consumption data.

[0070] The state difference analysis submodule uses the regional energy state dataset as a basis to compare and analyze the energy parameters of different sub-regions. It calculates the ratio of temperature parameter changes between each sub-region and other sub-regions, i.e., the ratio of the average temperature of one sub-region to the average temperature of another, reflecting the degree of temperature difference between regions. Similarly, it calculates the ratios of illuminance parameter changes and energy consumption parameter changes between regions, comprehensively measuring the differences in temperature, illuminance, and energy consumption among regions. Based on the functional requirements and energy control priorities of the dual-energy curtain wall areas, weight values ​​are assigned to temperature, illuminance, and energy consumption differences. For example, temperature differences may be given a higher weight in office areas, while illuminance differences may be more important in exhibition areas. The difference values ​​of each region with all other regions are weighted and calculated to obtain a comprehensive difference value, and all regions are sorted by the magnitude of the comprehensive difference values. Regions whose comprehensive difference values ​​exceed the preset fluctuation range are marked as regions with excessive fluctuations, and their locations and difference data are recorded in detail to form regional energy state difference data.

[0071] The state distribution integration submodule calls upon regional energy state difference data and performs summary analysis from multiple dimensions. In the temperature dimension, it summarizes temperature values ​​for each region and temperature differences between regions, and statistically analyzes the number and distribution of regions in different temperature ranges. Similarly, in the illuminance dimension, it summarizes illuminance values ​​and illuminance differences to analyze the regional distribution characteristics of light intensity. In the energy consumption dimension, it summarizes energy consumption values ​​and energy consumption differences to understand the energy consumption levels and differences in each region. The regional energy values ​​under each dimension are categorized in ascending order; for example, temperature values ​​are divided into low, medium, and high temperature ranges; illuminance values ​​are divided into low, medium, and high illuminance ranges; and energy consumption values ​​are divided into low, medium, and high energy consumption ranges. Then, the classification results of each region in the three dimensions are integrated, and the regional energy values ​​are arranged in an orderly manner according to the region number to form an energy status perception value that includes a temperature parameter set, an illuminance parameter set, and an energy consumption difference parameter set. The temperature parameter set includes the temperature value, temperature range classification, and temperature difference data between regions for each region. The illuminance parameter set includes the illuminance value, illuminance range classification, and illuminance difference data between regions for each region. The energy consumption difference parameter set includes the energy consumption value, energy consumption range classification, and energy consumption difference data between regions for each region.

[0072] The regional energy values ​​refer to specific numerical values ​​related to energy status obtained through data collection in the various sub-regions divided within the dual-energy curtain wall area. These values ​​mainly include temperature, illuminance, and energy consumption. The collection of these values ​​is regionalized according to the structural layout and functional zoning of the dual-energy curtain wall, with multiple collection points set up in each sub-region to comprehensively capture the energy status. Temperature values ​​are acquired in real time by temperature sensors distributed at different heights and locations, covering potential temperature differences within the coverage area; illuminance values ​​are collected by photosensitive sensors, reflecting the intensity of natural and artificial light on the curtain wall area; energy consumption values ​​are obtained through metering devices connected to relevant energy-consuming equipment, statistically analyzing the energy consumption per unit time. These collected values ​​are processed, invalid and abnormal data are removed, and then arranged according to the order of the sub-regions. By analyzing these values, parameter changes between regions are calculated, and differences in temperature, illuminance, and energy consumption values ​​between different regions are screened and classified and arranged in an orderly manner according to magnitude. This data forms the basis for subsequent analysis of regional energy parameters and load differences, and the integration of the status distribution parameter set.

[0073] Example 2:

[0074] The knowledge graph construction module operates around energy state perception values, forming a knowledge graph construction parameter set through three stages: parameter extraction, graph optimization, and parameter selection. The knowledge graph construction module includes: a parameter extraction submodule, a graph construction optimization submodule, and a parameter selection submodule.

[0075] After the parameter extraction submodule is activated, it first interfaces with the temperature parameter set, illuminance parameter set, and energy consumption difference parameter set contained in the energy status perception value, and identifies the current control status of each dual-energy curtain wall area. The control status includes specific operating parameters such as the curtain wall's opening angle, light transmittance level, and ventilation mode. These parameters are obtained through the real-time data interface of the curtain wall control system. During the identification process, the control mode of each area's curtain wall switching from the current state to that state, as well as the duration of that mode's operation, is recorded simultaneously. For example, when the temperature parameter set of a certain area shows a high temperature, the ventilation mode of the curtain wall and the time that mode has been running are recorded.

[0076] The raw data was standardized to unify the data format and units of measurement. For control modes, the same mode described in different ways was uniformly named, such as "natural ventilation mode" and "fresh air introduction mode," to avoid data confusion caused by naming differences. For control duration, it was uniformly converted to values ​​in minutes. After standardization, the data was classified according to the control mode category (such as heat preservation mode, ventilation mode, shading mode, etc.) and duration range (such as 0-30 minutes, 31-60 minutes, etc.) to form a curtain wall parameter dataset. Each category contains detailed records of the corresponding mode and duration.

[0077] After receiving the curtain wall parameter dataset, the graph construction and optimization submodule performs correlation analysis on the control modes and duration values ​​in the dataset. For each control mode, it statistically analyzes the changes in the energy state perception value under different durations, and calculates the degree of influence of the mode and duration combination on temperature, illuminance, and energy consumption differences. Using a preset matching degree algorithm, it quantitatively evaluates the matching degree between each parameter combination and the current energy state. The matching degree calculation covers multiple dimensions, including the adaptability of the mode to temperature regulation and the rationality of the duration to energy consumption control. Based on the calculation results, the parameter combinations with the highest matching degree are selected as candidate combinations.

[0078] During the screening process, a pattern matching algorithm is used to re-validate candidate combinations. These combinations are then applied to a simulated energy environment to observe their effects on energy parameter regulation and record the matching results. Based on the matching results, the parameter combinations are fine-tuned, for example, by adjusting the duration of a certain ventilation mode. The matching degree is then recalculated until a stable optimized parameter combination is generated. This result includes multiple combinations of modes and durations with different matching degrees and their corresponding impact data.

[0079] The parameter selection submodule performs a comprehensive search of the parameter combination optimization results, extracts the matching degree values ​​of all candidate combinations, and determines the combination of mode and duration with the highest matching degree as the optimal combination by sorting. Based on this optimal combination, the control parameters of the curtain wall are adjusted, including the operating parameters of the drive device and the monitoring frequency of the sensors, so that the actual operating parameters of the curtain wall are adapted to the optimal combination.

[0080] The adjusted control parameters are input into the curtain wall control system to load the control configuration, ensuring the system can operate according to the new parameter combination. The stability of the parameter set is verified by continuously monitoring the system's operating status at different time periods. This observes whether the parameter set can maintain effective control over the curtain wall when energy conditions fluctuate, without frequent parameter jumps or loss of control. After multiple rounds of verification confirming stability, the optimal mode and duration combination, along with its corresponding control parameters, are defined as the parameter set for constructing a knowledge graph. This parameter set includes specific screening mode parameters (such as the name and code of the optimal ventilation mode) and duration parameters (such as the optimal operating minutes for this mode).

[0081] Example 3:

[0082] The environmental correlation analysis module operates based on a knowledge graph to construct a parameter set, generating an environmental correlation parameter set through three processes: environmental factor analysis, parameter matching, and correlation parameter integration. The environmental correlation analysis module includes: an environmental factor analysis submodule, a parameter matching submodule, and a correlation parameter integration submodule.

[0083] After the environmental factor analysis submodule integrates the filtering mode parameters and duration parameters from the knowledge graph construction parameter set, it activates the environmental monitoring equipment deployed in the dual-energy curtain wall area. This equipment includes wind speed sensors distributed on the outside of the curtain wall, humidity detectors installed at different heights, and a radiation intensity meter for monitoring solar radiation. The wind speed sensors collect real-time wind speed data at fixed intervals, recording the speed and direction changes of airflow across the curtain wall surface; the humidity detectors simultaneously capture the relative humidity value in the air, reflecting the humidity level of the environment; and the radiation intensity meter continuously monitors the intensity of sunlight radiation on the curtain wall, including the radiation angle and energy density at different times.

[0084] Time series analysis was performed on the collected wind speed, humidity, and radiation intensity data. The data was sorted chronologically by collection time to form a continuous data stream. Outliers in the data stream were identified using a sliding window method, such as sudden peaks in wind speed data far exceeding the normal range, or sharp drops in radiation intensity under unobstructed conditions. These outliers were marked and removed from the data sequence. The remaining valid data was then partitioned according to the area division of the dual-energy curtain wall, linking the environmental data of each area with the corresponding filtering mode parameters and duration parameters. This resulted in environmental factor analysis data containing the time-varying changes in wind speed, humidity, and radiation intensity data for each area.

[0085] The parameter matching submodule analyzes environmental factor data and quantifies the impact of environmental variables on the curtain wall's control mode and duration. For wind speed, it analyzes the optimal operating time of the curtain wall ventilation mode under different wind speed ranges; for example, in high wind speed environments, the operating time of the same ventilation mode can be appropriately shortened. For humidity, it studies the matching relationship between humidity changes and the curtain wall sealing mode; in high humidity environments, the duration of the sealing mode may need to be extended. Radiation intensity is closely related to the control of the curtain wall's shading mode; the higher the radiation intensity, the higher the frequency and duration of shading mode activation may be.

[0086] To calculate the degree of impact of changes in each environmental factor on parameter adjustments, a formula for calculating the degree of impact is introduced:

[0087]

[0088] Where S represents the degree of overall impact, This indicates the change in wind speed. Indicates the amount of change in humidity. The value represents the change in radiation intensity. α, β, and γ are the influence weighting coefficients for wind speed, humidity, and radiation intensity, respectively, and their values ​​are determined based on the design characteristics of the dual-energy curtain wall and the functional requirements of the area. Based on the magnitude of the comprehensive influence S, the optimal matching parameter settings under different environmental conditions are determined. By cyclically adjusting the combination of modes and durations, the environmental adaptability under different parameter settings is tested, and the parameter combination with the highest score is captured as the parameter docking result.

[0089] The associated parameter integration submodule selects mode and duration combinations that match the current environmental conditions from the parameter docking results, and conducts parameter adjustment experiments in the test area of ​​the dual-energy curtain wall. During the experiment, the selected control mode and duration parameters are first applied to the test area, and the energy status changes in the area are continuously monitored, including real-time data on temperature, illuminance, and energy consumption. Based on the monitoring results, the parameters are fine-tuned. For example, if it is found that a certain shading mode causes low indoor illuminance under the current radiation intensity, the running time of the mode is appropriately shortened or the shading level of the mode is adjusted.

[0090] After multiple adjustments and verifications, the parameter settings were gradually optimized until the parameter combination could stably adapt to the current environmental conditions without causing drastic fluctuations in energy status. The finalized parameter combination was solidified as standard operating parameters, including the control mode code, runtime range, and adjustment trigger conditions corresponding to different environmental factor change ranges. These parameters together constitute an environmental-related parameter set, where the environmental factor change parameters cover the specific change ranges of wind speed, humidity, and radiation intensity, and the control mode duration matching parameters include the mode and duration settings corresponding to these change ranges.

[0091] Example 4:

[0092] The control threshold setting module operates based on a set of environmental parameters and generates energy control thresholds through three stages: energy extraction, energy consumption matching, and load allocation. The control threshold setting module includes:

[0093] Energy extraction submodule, energy consumption matching submodule, load distribution submodule.

[0094] After the energy extraction submodule accesses the environmental factor change parameters and control mode duration matching parameters from the environmental correlation parameter set, it locates energy change monitoring points based on these parameters. The distribution of monitoring points corresponds to the zoning of the dual-energy curtain wall. Sensors are installed at key locations in each zone. These sensors can capture dynamic changes in energy in real time, including the rate of temperature rise and fall, the magnitude of illuminance changes, and energy consumption fluctuations. For example, temperature monitoring points are set in the sun-facing and shaded areas of the curtain wall, different illuminance monitoring points are set in the office area and corridor area, and energy consumption monitoring points are set near equipment with concentrated energy consumption.

[0095] Energy change values ​​are extracted from the monitoring area and continuously recorded to form time-series data, from which changes in the rate of energy increase or decrease are identified. Key change nodes are then selected from these changes; these nodes typically represent moments when the rate of energy change shows a significant turning point, such as the point where temperature changes from a slow rise to a rapid rise, or when energy consumption suddenly fluctuates dramatically from a stable state. These node values ​​are arranged in chronological order, and the specific values ​​of temperature, illuminance, and energy consumption corresponding to each node are labeled to form current energy change characteristic values. These characteristic values ​​clearly reflect the patterns of energy change over different time periods.

[0096] The energy consumption matching submodule correlates node change values ​​with real-time collected energy consumption data based on current energy change characteristics. Real-time energy consumption data comes from energy metering devices related to the curtain wall, including real-time energy consumption values ​​from air conditioning and lighting systems. According to preset matching criteria, the node change values ​​and real-time energy consumption data are calibrated to ensure consistency in both time and numerical dimensions. For example, the time point corresponding to a certain temperature change node must accurately correspond to the air conditioning energy consumption data at that moment; if there is a time discrepancy, synchronous correction is performed.

[0097] Based on the calibrated correspondence, matching criteria are invoked to redistribute energy consumption across different intervals. For example, when a temperature change node indicates that the temperature has risen to a certain value, the air conditioning energy consumption for that period is allocated to the corresponding temperature interval, creating a clear correspondence between energy changes and energy consumption. This results in an energy consumption matching structure that includes the energy consumption distribution under different energy change nodes.

[0098] The load allocation submodule is based on an energy consumption matching structure and employs a dynamic threshold adjustment method. This method first determines the allocated load value matching each energy change node based on the data in the energy consumption matching structure, i.e., the energy load demand corresponding to that node. For example, at a node where the temperature rises rapidly, the corresponding air conditioning load value will increase accordingly.

[0099] Each node is assigned an upper and lower threshold based on its allocated load value. The upper threshold is the maximum energy load that the node can withstand, and the lower threshold is the minimum energy load required to maintain basic operation. These thresholds are then applied to the allocation of regional energy values. For example, in areas with higher temperatures, the upper threshold for air conditioning energy allocation is appropriately increased to meet cooling needs; while in areas with suitable temperatures, a lower upper threshold is set to avoid energy waste.

[0100] During the allocation process, the application ratio of the threshold is dynamically adjusted by combining real-time energy consumption data of each region to ensure that the energy allocation of each region not only conforms to the threshold range but also matches the actual load demand. Through this allocation method, the energy regulation threshold is finally obtained, which includes energy change parameters (such as the energy change rate of each node), energy consumption matching parameters (such as the correspondence between energy and energy consumption data), and threshold setting parameters (such as the upper and lower limit values ​​of the threshold for each node).

[0101] Example 5:

[0102] The energy trend prediction module and the curtain wall feedback control module work together to generate predicted energy distribution values ​​based on energy regulation thresholds, and further form a smart energy balance regulation scheme for the dual-energy curtain wall. The energy trend prediction module includes: an energy data capture submodule, an energy load analysis submodule, and a trend distribution inference submodule. The curtain wall feedback control module includes: an error analysis submodule, a parameter adjustment submodule, and a regulation control submodule.

[0103] In the energy trend prediction module, the energy data capture submodule accesses energy regulation thresholds, which include energy change parameters, energy consumption matching parameters, and threshold setting parameters. The submodule applies a knowledge reasoning algorithm to capture energy consumption data from preset energy sampling points within the dual-energy curtain wall area. The sampling points cover key energy consumption areas of the curtain wall system, including energy consumption monitoring points for heating equipment, cooling devices, and lighting systems, ensuring a comprehensive reflection of the overall energy consumption situation. During data capture, an outlier is identified through a data verification mechanism. This includes instantaneous energy consumption values ​​that deviate significantly from historical data or energy consumption data exceeding the threshold setting parameter range. These outliers are removed, and the remaining data undergoes error correction to eliminate deviations caused by differences in the accuracy of monitoring equipment. The corrected energy consumption data is stored hierarchically according to time intervals and regional intervals. Time intervals can be divided into hourly segments, daily segments, etc., while regional intervals correspond to the functional zones of the curtain wall. Simultaneously, the real-time energy consumption values, environmental values ​​(such as temperature and humidity), and load factors (such as the load ratio corresponding to the population density in the area) of the sampling points are combined for knowledge processing. The processed data is then transformed into structured information that meets the requirements of knowledge graph data format, generating a knowledge-based energy consumption dataset.

[0104] During knowledge-based processing, the first step is to integrate the captured real-time energy consumption, environmental values, and load coefficients from sampling points across multiple dimensions. The real-time energy consumption values ​​of each sampling point are then correlated and matched with corresponding environmental values ​​(such as real-time temperature and humidity data of the area where the sampling point is located) and load coefficients (such as the load ratio parameter calculated from the current population density in the area), ensuring a one-to-one correspondence between each set of data in both time and space. Based on the pre-defined data structure and entity relationship model of the knowledge graph, the integrated raw data undergoes format conversion and semantic annotation. Real-time energy consumption values ​​are mapped to attribute values ​​of "energy consumption entities" in the knowledge graph, environmental values ​​correspond to attribute parameters of "environmental factor entities," and load coefficients are associated with descriptive information of "load characteristic entities." In this way, scattered numerical data is transformed into structured knowledge entries containing entities, attributes, and relationships, enabling the data to be effectively identified and invoked by the knowledge graph. This generates a knowledge-based energy consumption dataset that meets the needs of knowledge reasoning, providing structured knowledge support for subsequent energy load analysis and trend prediction.

[0105] After receiving the knowledge-based energy consumption dataset, the energy load analysis submodule divides the data into multiple intervals based on the magnitude of the energy load, with each interval corresponding to a specific load range. For example, the energy load of the lighting system is divided into three intervals: low load, medium load, and high load, each corresponding to a different power range. Within each interval, the module extracts the energy consumption trend, such as the direction of continuous increase or decrease in load over a certain period, as well as fluctuation characteristics, such as the frequency and amplitude of load changes. For heating systems, the module analyzes the load fluctuation patterns under different outdoor temperatures; for cooling systems, it observes the load trend over different time periods. By processing this data, an energy load and energy consumption change feature set is generated. This feature set includes the load range, trend description, and fluctuation characteristic parameters for each interval.

[0106] The trend distribution inference submodule adjusts characteristic parameters based on the energy load and energy consumption change feature set. For example, it corrects load characteristic deviations caused by seasonal variations and calibrates trend data to ensure consistency with actual operating conditions. Based on the calibrated trend data, it extracts the distribution intervals of energy consumption data, clarifying the distribution probability and range of energy consumption within each interval. Through analysis of historical and current data, it predicts future energy consumption values ​​within each interval. Combining this with energy consumption change patterns under similar operating conditions stored in the knowledge graph, it obtains predicted energy distribution values. These predicted values ​​include energy consumption trend analysis parameters (such as the load change trend curve for the next 24 hours) and energy consumption relationship parameters (such as the proportion of energy consumption values ​​corresponding to different energy loads).

[0107] In the curtain wall feedback control module, the error analysis submodule extracts real-time energy data and energy consumption data for the dual-energy curtain wall area based on the predicted energy distribution values. Real-time energy data includes current temperature and illuminance, while real-time energy consumption data includes the instantaneous energy consumption values ​​of each device. The real-time energy consumption values ​​are compared with the predicted energy consumption values ​​in the energy distribution prediction, and the difference between the two is calculated; this is the energy consumption error value. This error value is then correlated with current energy information (such as real-time temperature and illuminance). For example, if the real-time energy consumption value of a certain area is higher than the predicted value, while the temperature of that area is lower than the predicted temperature, this correspondence is recorded, generating an energy consumption error value dataset.

[0108] The parameter adjustment submodule sets the control and adjustment parameters for the curtain wall based on the energy consumption error value dataset. For areas with large error values, such as areas where real-time energy consumption is significantly higher than the predicted value, a larger adjustment range is set, for example, increasing or decreasing the air conditioning cooling power in that area by a certain value. For areas with small error values, fine adjustments are made, such as slightly adjusting the lighting brightness. By comparing the control effects under different adjustment parameters, a parameter set that matches the current error situation is selected. These parameter sets are then integrated, and conflicting parameters are removed to form a unified curtain wall adjustment parameter set.

[0109] The control and regulation submodule applies the curtain wall adjustment parameter set to each curtain wall location, implementing control operations item by item, such as adjusting the angle of sunshades, changing the opening degree of ventilation openings, and adjusting the brightness of lighting fixtures. During the control process, the energy status and energy consumption data of each area are monitored simultaneously, and the order of control operations for each area is adjusted according to the monitoring results. For example, areas with larger error values ​​and wider impact ranges are processed first, followed by areas with smaller error values. Through orderly control operations, the energy status of each area is ensured to gradually approach the predicted equilibrium state, ultimately generating a dual-energy curtain wall intelligent energy balance control scheme. This scheme includes error analysis parameters (such as the magnitude and distribution of error values ​​in each area) and curtain wall adjustment parameters (such as specific control operations and parameter settings).

[0110] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A knowledge graph-based intelligent energy control system for dual-energy curtain walls, characterized in that, The system includes: The energy status sensing module analyzes the differences in regional energy parameters and loads based on the energy temperature, illuminance and energy consumption values ​​of each sub-region in the dual-energy curtain wall area, calculates the state differences between regions, integrates them into a state distribution parameter set, and obtains the energy status sensing value. Based on the energy state perception value, the knowledge graph construction module extracts the parameter combination of curtain wall control mode and duration, selects the optimal mode and duration combination, and obtains the knowledge graph construction parameter set. The environmental correlation analysis module constructs a parameter set based on the knowledge graph, extracts the current changes in environmental factors, analyzes the relationship between regulation modes and duration, matches environmental factors with regulation combinations, and obtains an environmental correlation parameter set. The regulation threshold setting module extracts the current energy change value based on the environmental associated parameter set, combines it with real-time energy consumption data, allocates energy and load, sets a threshold, and applies the threshold to the regional energy allocation to obtain the energy regulation threshold. The energy trend prediction module captures energy consumption data from energy sampling points based on the energy regulation threshold, performs knowledge reasoning on the energy consumption changes of energy sampling points using a knowledge graph, analyzes the energy consumption change trend corresponding to the energy load, categorizes and organizes the energy consumption change trend based on the reasoning results, and adjusts and analyzes the categorized data using a knowledge graph based on the energy consumption change information to obtain the predicted energy distribution value. Based on the predicted energy distribution value, the curtain wall feedback control module analyzes the error value of energy and energy consumption through real-time energy and energy consumption data, and adjusts the curtain wall regulation in combination with the error value to obtain a dual-energy curtain wall intelligent energy balance regulation scheme.

2. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The energy status perception values ​​include a set of temperature parameters, a set of illuminance parameters, and a set of energy consumption difference parameters. The knowledge graph construction parameter set includes a screening mode parameter and a duration parameter. The environmental association parameter set includes environmental factor change parameters and control mode duration matching parameters. The energy control threshold includes energy change parameters, energy consumption matching parameters, and threshold setting parameters. The energy distribution prediction values ​​include energy consumption trend analysis parameters and energy consumption relationship parameters. The dual-energy curtain wall intelligent energy balance control scheme includes error analysis parameters and curtain wall adjustment parameters.

3. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The energy status sensing module includes: The data acquisition submodule is used to collect temperature, illuminance and energy consumption values ​​in the dual-energy curtain wall area in a regional manner, locate invalid data, remove abnormal data, and arrange the extracted temperature, illuminance and energy consumption values ​​in the regional order to generate a regional energy status dataset. The state difference analysis submodule is used to analyze energy parameters between regions based on the regional energy state dataset, calculate the ratio of regional parameter changes, sort the inter-regional difference values ​​by weight, mark regions with excessive fluctuation differences, and obtain regional energy state difference data. The state distribution integration submodule is used to summarize the regional energy difference values ​​in multiple dimensions based on the regional energy state difference data, screen the regional energy value differences, classify them according to the value size, and arrange the regional energy values ​​in an orderly manner to generate energy state perception values.

4. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The knowledge graph construction module includes: The parameter extraction submodule is used to identify the control status of each area of ​​the curtain wall based on the energy status perception value, record the control mode and control duration of the curtain wall, standardize the recorded data, sort the standardized data by mode and duration, and generate a curtain wall parameter dataset. The graph construction and optimization submodule is used to analyze the pattern and duration values ​​in the curtain wall parameter dataset, filter parameter combinations with high matching degree with energy status, adjust parameter combinations and generate parameter combination optimization results by matching and recording the matching results through pattern matching. The parameter selection submodule is used to retrieve the optimization results of the parameter combination, determine the optimal matching mode and duration combination, adjust the curtain wall control parameters, input the control configuration, verify the stability of the parameter set, and generate a knowledge graph to construct the parameter set.

5. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The environmental correlation analysis module includes: The environmental factor analysis submodule is used to construct a parameter set based on the knowledge graph, collect key data through environmental monitoring, including wind speed data, humidity data and radiation intensity data, perform time series analysis on the key data, remove outliers, partition the remaining data, and obtain environmental factor analysis data. The parameter matching submodule is used to analyze the impact of environmental variables on the curtain wall control mode and duration through the environmental factor analysis data, calculate the degree of impact of each environmental factor change on parameter adjustment, determine the optimal matching parameter settings based on the degree of impact, cyclically adjust the parameters to capture the optimal combination, and obtain the parameter docking results. The associated parameter integration submodule is used to select a combination of modes and durations that match the current environmental conditions from the parameter docking results, conduct parameter adjustment experiments, optimize parameter settings through multiple adjustments and verifications, determine and solidify parameters as operating standards, and generate an environmental associated parameter set.

6. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The control threshold setting module includes: The energy extraction submodule is used to locate energy change monitoring points based on the environmental correlation parameter set, extract energy change values ​​in the monitoring area, continuously record energy increase / decrease rates, extract multiple key change nodes corresponding to the change rate, sort the node values ​​in order, and obtain the current energy change characteristic value. The energy consumption matching submodule is used to analyze the node change value and real-time energy consumption data based on the current energy change characteristic value, and to perform calibration according to a predetermined matching criterion. It then calls the matching criterion to perform distribution redistribution within the energy consumption range to obtain the energy consumption matching structure. The load distribution submodule is used to determine the energy regulation threshold based on the energy consumption matching structure.

7. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 6, characterized in that, Determining the energy regulation threshold based on the energy consumption matching structure includes: determining the energy regulation threshold using a dynamic threshold adjustment method based on the energy consumption matching structure; wherein the dynamic threshold adjustment method includes obtaining the allocated load value of the matching energy change node, setting the upper and lower limits of the node threshold, applying the threshold to the regional energy value and allocating it.

8. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The energy trend prediction module includes: The energy data capture submodule is used to capture energy consumption data of sampling points based on the energy regulation threshold, apply knowledge reasoning algorithms, remove outliers and correct errors, store the data in a hierarchical manner according to intervals, perform knowledge-based processing, and generate a knowledge-based energy consumption dataset. The energy load analysis submodule is used to divide the energy load into intervals based on the knowledge-based energy consumption dataset, extract the changing trends and fluctuation characteristics, and generate a set of energy load and energy consumption change characteristics. The trend distribution inference submodule is used to adjust the feature parameters and calibrate the trend data based on the energy load and energy consumption change feature set, extract the distribution interval, and perform numerical prediction to obtain the predicted energy distribution value. The knowledge reasoning algorithm includes calculating energy knowledge values ​​and generating a knowledge-based energy consumption dataset. The generated knowledge-based energy consumption dataset is processed by combining the energy consumption values, environmental values, and load factors of the sampling points.

9. The knowledge graph-based intelligent energy control system for dual-energy curtain walls according to claim 1, characterized in that, The curtain wall feedback control module includes: The error analysis submodule is used to extract real-time energy and energy consumption data based on the energy distribution prediction value, analyze the real-time energy consumption value and the prediction value, match the energy consumption difference with the current energy information, and generate an energy consumption error value. The parameter adjustment submodule is used to set the control and adjustment parameters of the curtain wall based on the energy consumption error value. It sets the adjustment range for areas with large errors and makes fine adjustments for areas with low errors. By comparing the control effects, it selects and integrates matching parameter sets to generate a curtain wall adjustment parameter set. The regulation and control submodule is used to apply adjustment parameters to each curtain wall location based on the curtain wall adjustment parameter set, implement regulation operations item by item, monitor energy and energy consumption synchronously, gradually adjust the order of regulation operations in each area, and generate a dual-energy curtain wall energy intelligent balance regulation scheme.

10. A knowledge graph-based intelligent energy control method for dual-energy curtain walls, applied to the knowledge graph-based intelligent energy control system for dual-energy curtain walls as described in any one of claims 1 to 9, characterized in that, The method comprises the following steps: Step 1: Energy Status Sensing Step. Based on the energy temperature, illuminance and energy consumption values ​​of each sub-region in the dual-energy curtain wall area, analyze the differences in regional energy parameters and loads, calculate the status differences between regions, integrate them into a status distribution parameter set, and obtain the energy status sensing value. Step 2: Knowledge graph construction step. Based on the energy state perception value, extract the parameter combination of curtain wall control mode and duration, select the optimal mode and duration combination, and obtain the knowledge graph construction parameter set. Step 3: Environmental correlation analysis step, constructing a parameter set based on the knowledge graph, extracting the current changes in environmental factors, analyzing the relationship between regulation mode and duration, matching environmental factors with regulation combinations, and obtaining the environmental correlation parameter set; Step 4: Setting the control threshold. Based on the environmental correlation parameter set, extract the current energy change value, combine it with real-time energy consumption data, allocate energy and load, set a threshold, and apply the threshold to the regional energy allocation to obtain the energy control threshold. Step 5: Energy trend prediction step. Based on the energy regulation threshold, capture energy consumption data of energy sampling points, combine knowledge graph to perform knowledge reasoning on the energy consumption changes of energy sampling points, analyze the energy consumption change trend corresponding to energy load, classify and organize the energy consumption change trend according to the reasoning results, and combine energy consumption change information to adjust and analyze the classified data using knowledge graph to obtain the predicted value of energy distribution. Step 6: Curtain wall feedback control step. Based on the predicted energy distribution value, the error value of energy and energy consumption is analyzed through real-time energy and energy consumption data. The curtain wall regulation is adjusted in combination with the error value to obtain the dual-energy curtain wall intelligent energy balance regulation scheme.

Images