Engineering exhibition digital exhibition space dynamic regulation and control method fusing environment perception

Abstract

The invention discloses an engineering exhibition digital exhibition space dynamic regulation and control method fused with environment perception, and particularly relates to the technical field of dynamic regulation and control, and the method comprises the steps: constructing an environment perception and control integrated architecture, and enabling a distributed sensor network to collect environment characteristic parameters and equipment operation state information; the control center carries out abnormal value correction, space-time alignment fusion data and space-time feature extraction, and generates a state deviation analysis result and a space dynamic feature map; outputting an environment comfort level, an equipment stability index and an environment-equipment coupling coordination degree based on a multivariable coupling analysis model of a graph convolutional network and an attention mechanism; obtaining final judgment results such as environmental risks and equipment hidden dangers through thresholding processing and rule matching; and generating a differential optimization scheme to trigger regulation and control, a closed-loop iterative optimization algorithm and an early warning threshold. According to the invention, by realizing environment-equipment cooperative regulation and control, the regulation and control precision and the intelligent level are improved, continuous optimization is guaranteed, the energy consumption is reduced, and the risk response capability is enhanced.

Description

Technical Field

[0001] This invention relates to the field of dynamic control technology, and more specifically, to a method for dynamic control of digital exhibition spaces in engineering exhibitions that integrates environmental perception. Background Technology

[0002] Current digital exhibition space control in engineering exhibitions largely adopts traditional methods: On the one hand, environmental parameter (such as temperature and humidity) control and equipment operation status (such as air conditioning pressure and projector speed) monitoring are independent of each other, and local adjustments are only achieved through manual inspection or single-parameter controllers, without considering the coupling relationship between the environment and equipment (such as air conditioning temperature adjustment without considering the impact of lighting heat dissipation), which easily leads to control lag or resource waste; on the other hand, data processing methods are simple, only performing threshold judgments on single parameters, without extracting spatiotemporal features and multivariate coupling relationships (such as the impact of temperature and humidity differences in different exhibition areas and the duration of equipment deviation on the environment), making it difficult to accurately reflect the overall operating status of the exhibition space, resulting in insufficient control precision; in addition, existing control lacks a closed-loop iteration mechanism, and the environmental response and target fit after control are not continuously monitored, making it impossible to dynamically optimize control algorithms and rules, and parameter drift is prone to occur in long-term operation; at the same time, sensors mostly use fixed sampling intervals (such as continuous 5-second sampling), without adaptive adjustment according to environmental stability, resulting in excessive energy consumption and data redundancy or missing data in some scenarios. These issues all affect the protection of exhibits in the exhibition space (such as the sensitivity of paper exhibits to humidity) and the comfort of visitors, and urgently require an integrated dynamic control method that incorporates environmental perception to solve them. Summary of the Invention

[0003] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception. The method addresses the problems of separation, low precision, and lack of closed-loop optimization in exhibition space control mentioned in the background art through the following solutions.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a dynamic control method for digital exhibition space of engineering exhibitions that integrates environmental perception, comprising: S1: constructing an integrated architecture of environmental perception and control, and collecting environmental characteristic parameters and equipment operating status information of the exhibition space in real time through a distributed sensor network, wherein the environmental characteristic parameters and equipment operating status information together constitute the raw data and are transmitted to the control center; S2: The control center performs fusion processing and feature extraction on the collected raw data; establishes a comparison model between the environment-equipment joint state and the preset standard, generates environmental deviation analysis results and equipment operation performance deviation results, and constructs a dynamic feature map of the exhibition space; S3: The control center receives the environmental deviation analysis results and the equipment operation performance deviation results. Based on the multivariate coupling analysis model constructed by graph convolutional network and attention mechanism, it models and performs time series analysis on the dynamic feature map of the exhibition space, and outputs the comprehensive operation status analysis results of the exhibition space. The comprehensive operation status analysis results include the environmental comfort level, the equipment operation stability index, and the environment-equipment coupling coordination degree. S4: Based on the comprehensive operation status analysis results, thresholding and rule matching are performed to generate the final judgment results, which include environmental risk judgment results, equipment hidden danger judgment results, and overall control status judgment results. S5: Generate an optimization plan based on the final judgment result and trigger corresponding dynamic adjustment actions, and archive the optimization plan to the operation log and evaluation report.

[0005] The technical effects and advantages of this invention are as follows: 1. By constructing an integrated architecture for environmental perception and control, and combining it with a multivariate coupling analysis model, we break the traditional model of separating environmental parameter control and equipment status monitoring in exhibition spaces. This effectively links the coupling relationship between environmental characteristics and equipment operation, avoiding inefficiency caused by imbalance in the control of the two. From an overall perspective, this ensures that the needs for exhibit protection and visitor comfort are met simultaneously in the exhibition space. 2. Based on graph convolutional networks and attention mechanisms, spatiotemporal features of exhibition space environment and equipment data are extracted and multivariate correlation analysis is performed. The overall operating status of the exhibition space is quantitatively characterized by environmental comfort level, equipment operation stability index, and environment-equipment coupling coordination degree. Combined with thresholding processing and multi-dimensional rule base matching, accurate judgment of environmental risks and equipment hazards is achieved. Compared with the traditional single-parameter control mode, the scientific nature and intelligence of control decision-making are significantly improved. 3. By continuously monitoring the environmental response after regulation through the control center, and combining feedback information to dynamically optimize the control algorithm parameters and regulation rule system, while using a dynamic calibration mechanism to update the abnormal warning threshold, the regulation failure caused by parameter drift during long-term operation is effectively avoided, ensuring that the regulation strategy is always adapted to the actual operating state of the exhibition space, reducing the frequency of manual intervention, and improving the long-term stability and adaptability of regulation. 4. The adaptive sampling mechanism of the distributed sensor network can dynamically adjust the data acquisition interval according to the stability of the environment, reducing unnecessary energy consumption while ensuring data validity. At the same time, the differentiated dynamic control strategy achieves accurate response to different scenarios through priority division. Especially in emergency scenarios, it can quickly trigger control actions, effectively curb the spread of risks, and ensure the safe and stable operation of the exhibition space. Attached Figure Description

[0006] Figure 1 This is a schematic diagram of the overall architecture of the present invention; Figure 2 This is a flowchart illustrating the overall method of the present invention; Figure 3 This is a flowchart illustrating the process of obtaining comprehensive operational status analysis results according to the present invention. Detailed Implementation

[0007] 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.

[0008] like Figures 1 to 3 The method for dynamic control of digital exhibition space of engineering exhibition with integrated environmental perception shown includes: S1: Constructing an integrated architecture of environmental perception and control, and collecting environmental characteristic parameters and equipment operating status information of the exhibition space in real time through a distributed sensor network. The environmental characteristic parameters and equipment operating status information together constitute the raw data and are transmitted to the control center. S2: The control center performs fusion processing and feature extraction on the collected raw data; establishes a comparison model between the environment-equipment joint state and the preset standard, generates environmental deviation analysis results and equipment operation performance deviation results, and constructs a dynamic feature map of the exhibition space; S3: The control center receives the environmental deviation analysis results and the equipment operation performance deviation results. Based on the multivariate coupling analysis model constructed by graph convolutional network and attention mechanism, it models and performs time series analysis on the dynamic feature map of the exhibition space, and outputs the comprehensive operation status analysis results of the exhibition space. The comprehensive operation status analysis results include the environmental comfort level, the equipment operation stability index, and the environment-equipment coupling coordination degree. S4: Based on the comprehensive operation status analysis results, thresholding and rule matching are performed to generate the final judgment results, which include environmental risk judgment results, equipment hidden danger judgment results, and overall control status judgment results. S5: Generate an optimization plan based on the final judgment result and trigger corresponding dynamic adjustment actions, and archive the optimization plan to the operation log and evaluation report.

[0009] It should be specifically explained that, firstly, an integrated architecture for environmental perception and control is constructed, which collects environmental characteristic parameters and equipment operating status information of the exhibition space in real time through a distributed sensor network and transmits them to the control center.

[0010] It should be further explained that the environmental characteristic parameters and equipment operation status information of the exhibition space specifically cover the following: Environmental characteristic parameters include temperature, humidity, and carbon dioxide concentration. Temperature reflects the thermal environment of the exhibition space, humidity affects the preservation conditions of exhibits and human comfort, and carbon dioxide concentration reflects the air quality and the density of people in the space. Equipment operation status information involves the current and voltage parameters of lighting equipment (used to determine whether the lighting equipment is working stably and its energy consumption), the temperature and pressure parameters of air conditioning equipment (reflecting the cooling / heating effect of the air conditioning and the stability of the system), and the internal temperature and fan speed of projection equipment (to assess the heat dissipation status and operational safety of the projection equipment). These parameters and information are obtained because environmental characteristic parameters are directly related to the protection of exhibits and the visitor experience, while equipment operation status information is the foundation for ensuring the exhibition effect and the safe operation of equipment.

[0011] It should be further explained that the selection and deployment of sensors for collecting environmental characteristic parameters of the exhibition space are as follows: For temperature parameters, platinum resistance temperature sensors are selected, with a measurement range of -20℃ to 60℃ and an accuracy of ±0.1℃. They are wall-mounted, with three sensors evenly deployed per 100㎡ of exhibition space, corresponding to heights of 0.8m, 1.5m, and 2.2m above the ground, forming vertical temperature gradient monitoring points to ensure comprehensive capture of spatial thermal environment differences. For humidity parameters, capacitive humidity sensors are selected, with a measurement range of 20%RH to 90%RH and an accuracy of ±2%RH. They are installed at the same point as the temperature sensors to form an integrated temperature and humidity monitoring point. The sensor housing adopts an anti-condensation design to avoid measurement errors in high humidity environments. For carbon dioxide concentration monitoring, infrared carbon dioxide sensors are used, with a measurement range of 0 to 5000ppm, an accuracy of ±50ppm, and a response time of <60s. They are deployed in areas with frequent personnel flow (such as entrances and passageway intersections) and densely populated exhibition areas, with one sensor per 50㎡, installed at a height of 1.2m above the ground, with the sensor air inlet facing the direction of personnel activity.

[0012] Further explanation is needed regarding the specific methods for collecting operational status information of the exhibition equipment: For lighting equipment, current parameters are collected using a switchable current transformer with a range of 0-5A and an accuracy of 0.5 class. This transformer is directly mounted on the phase line of the lighting circuit, eliminating the need for power disconnection. For voltage parameters, an AC voltage sensor is used with an input range of 0-250VAC and an output of a 4-20mA standard signal. This sensor is connected in parallel to the power input of the lighting equipment. The distance between the sensor and the lighting control module is ≤3m to minimize line loss. For air conditioning equipment, temperature parameters are monitored using a surface-mount temperature sensor with a measurement range of -30℃ to 80℃ and an accuracy of ±0.2℃. The sensor is installed at the evaporator inlet and outlet, adhered to the heat exchanger surface using thermally conductive silicone, and covered with an insulation layer to isolate ambient temperature interference and ensure accurate measurement. Accuracy: Pressure parameter acquisition uses a diffused silicon pressure sensor with a range of 0~3MPa and an accuracy of 0.25%. It is installed on the high-pressure pipeline of the air conditioning compressor, and the pressure tap is connected to the pipeline through a special connector. The connection is sealed with sealant to prevent refrigerant leakage. The internal temperature monitoring of the projection equipment uses an NTC thermistor sensor with a measurement range of -50℃~125℃ and an accuracy of ±1℃. It is embedded in the heat sink near the light source module, and the sensor lead uses high-temperature resistant wire. Fan speed monitoring uses a photoelectric speed sensor: it is fixed to the outside of the fan blades by a bracket, with a detection distance of 3~5mm, a sampling frequency of ≥10Hz, and a speed measurement accuracy of ±1r / min, ensuring accurate capture of speed fluctuations (such as the change in fan speed from 2000r / min to 1800r / min).

[0013] It should be noted that the distributed sensor network adopts a star topology: the industrial-grade gateway in the control center serves as the core node. Environmental characteristic parameter sensors and equipment status sensors establish communication connections with the core node via LoRa wireless modules, with the network communication rate set to 9.6kbps to ensure data transmission stability. The core node features a dual-antenna redundancy design, automatically switching to the backup antenna when the primary antenna's received signal strength is <-85dBm, ensuring communication continuity. When the sensor nodes are battery-powered, a sleep-wake mechanism is implemented, with the sampling interval adaptively adjusting the trigger conditions: when the environmental parameters (temperature / humidity / carbon dioxide) fluctuate within ≤±0.3℃ (temperature), ±2% RH (humidity), and ±50ppm (carbon dioxide) for three consecutive sampling periods, it is considered 'stable', and the sampling interval is extended to 60s; when the fluctuation is >±0.5℃, ±3%, or >±5%, the sampling interval is extended to 60s. When the RH is ±100ppm, it is judged as 'large fluctuation' and the interval is shortened to 10s; when it is between the two, the interval is maintained at 30s; the sensor nodes using wired power supply maintain continuous monitoring and the data upload interval is fixed at 5s; all nodes in the network are assigned a unique device ID, and the core node realizes regional management of data through the ID and physical location mapping table.

[0014] It should be specifically explained that the control center performs fusion processing and feature extraction on the collected raw data, establishes a comparison model between the environmental state and the preset standard, and generates state deviation analysis results and spatial dynamic feature maps.

[0015] It should be further explained that the raw data fusion processing method is as follows: First, the raw data collected by various sensors are preprocessed. For outliers in environmental characteristic parameters, a sliding window method is used for identification and correction. That is, a fixed time window (such as 30 seconds) is set. When the deviation of a data point in the window from the adjacent data exceeds the normal fluctuation range (determined based on the 95% confidence interval of historical data of the same period in the past 30 days, such as temperature fluctuation range ±0.5℃, humidity ±3%RH), it is judged as an outlier and replaced with the mean value in the window. For missing values ​​in the equipment operation status information, interpolation is performed using the operating data of the same type of equipment at the same time to ensure the continuity of the data sequence. Then, the multi-source data is spatiotemporally aligned. Based on the system time of the control center, the data collected by different sensors at the same time are correlated and matched to form a unified dataset indexed by the time axis. At the same time, the data is classified and integrated according to the physical partition of the exhibition space (such as exhibition area, aisle, rest area).

[0016] It should be further explained that the specific method of feature extraction is as follows: For environmental feature parameters, time-dimensional features and spatial distribution features are extracted. In the time dimension, the changing trends (such as rising, falling, and stabilizing) and fluctuation frequency (such as the number of fluctuations per hour) of temperature, humidity, and carbon dioxide concentration are analyzed. In the spatial dimension, the distribution differences of each parameter in different regions are identified (such as temperature > 25℃ is defined as a high-temperature zone, humidity > 60%RH is defined as a high-humidity zone, and carbon dioxide concentration > 1200ppm is defined as a poor air quality zone), and the physical coordinate range of each abnormal zone is determined. For equipment operating status information, the focus is on extracting the operating stability features of the equipment (such as the fluctuation amplitude of the current of lighting equipment and the rate of change of air conditioning pressure) and correlation features (such as the synergistic relationship between the internal temperature of the projection equipment and the fan speed). Through the above feature extraction, the raw data is transformed into a more representative feature vector.

[0017] It should be further explained that the comparison model between environmental conditions and preset standards is established as follows: the preset standards include threshold ranges for environmental characteristic parameters (such as temperature 18-25℃, humidity 40%~60%, carbon dioxide concentration ≤1000ppm) and baseline ranges for equipment operating conditions (such as lighting equipment voltage 220±5V, air conditioner evaporator inlet and outlet temperature difference 8-12℃, etc.). These standards are determined based on exhibit protection requirements, human comfort standards, and equipment operation manuals. The comparison model adopts a hierarchical comparison architecture. The first layer directly compares the real-time environmental characteristic parameters with the corresponding threshold ranges. The second layer matches the equipment operating status information with the baseline ranges. The third layer analyzes whether the correlation between environmental parameters and equipment status conforms to preset logic (such as whether abnormal air conditioner temperature parameters are accompanied by corresponding changes in ambient temperature).

[0018] It should be further explained that the state deviation analysis results are generated as follows: based on the output of the comparison model, the degree of deviation of each parameter is quantitatively calculated. For environmental characteristic parameters, the difference between the actual value and the standard median and the degree of deviation from the standard are calculated. For equipment operating status information, it is determined whether it exceeds the benchmark range and the duration of the deviation is recorded. At the same time, the correlation of deviations is analyzed to identify independent deviations caused by a single factor (such as an abnormal temperature in a certain area caused only by an air conditioning failure in that area) and compound deviations caused by multiple factors (such as an increase in carbon dioxide concentration accompanied by multiple exhibition areas with dense crowds and insufficient ventilation). Finally, an analysis result containing deviation parameters, deviation degree, deviation type and scope of influence is formed.

[0019] It needs further explanation that the core of generating the state deviation analysis results is to compare the collected real-time data with preset standards to intuitively display abnormal conditions of the environment and equipment. The specific method is as follows: First, the real-time data of environmental characteristic parameters (temperature, humidity, carbon dioxide concentration) and equipment operating status information (lighting current and voltage, air conditioning temperature and pressure, projection temperature and speed) are compared with their respective preset standard ranges. For example, if the temperature standard is 18-25℃, and the real-time value exceeds this range, it is considered a deviation. Then, the specific degree of deviation is calculated. For environmental parameters, the specific deviation value is the difference between the real-time measured value and the value within the preset standard range. The degree of deviation from the standard is expressed as a percentage of this difference relative to half the standard range. For example, if the half-range of the temperature standard is 3.5℃ (half of 25-18℃), then the degree of deviation here is 4.5 ÷ 3.5 × 100% ≈ 128.6%. The specific deviation value of equipment parameters is the real-time value... The difference from the upper (or lower) limit of the benchmark range is calculated as follows: for example, if the standard lighting voltage is 220±5V (i.e., 215-225V), and the real-time voltage is 230V, the specific deviation value is +5V. The degree of deviation is calculated as a percentage of this difference to the allowable fluctuation value (5V) of the benchmark range, i.e., 5÷5×100%=100%. At the same time, the duration of this out-of-tolerance state is recorded (e.g., it has lasted for 2 minutes). Next, the type of deviation is determined. If a single parameter exceeds the standard, it is considered an independent deviation, such as high humidity in a certain exhibition area but other parameters are normal. If multiple parameters deviate simultaneously and are related, such as an increase in carbon dioxide concentration and abnormal air conditioning operating pressure, it is classified as a compound deviation. Finally, these deviation information are compiled into a state deviation analysis result, which includes: the name of the parameter with deviation, the specific deviation value, the degree of deviation from the standard, the type of deviation (independent or compound), and the range of exhibition space areas that may be affected by the deviation.

[0020] It should be specifically explained that the spatial dynamic feature map is generated as follows: based on the planar layout of the exhibition space, visualization technology is used to intuitively present the spatial distribution of environmental feature parameters, regional differences in equipment operating status, and spatial correlation of status deviations. The map is dynamically updated according to the time dimension (e.g., refreshed every 5 minutes), and the distribution characteristics of environmental parameters are displayed through color gradients (e.g., blue to red indicates temperature from low to high). The operating status of equipment is identified through icon status (e.g., green for normal, yellow or red for abnormal). The propagation path and impact range of deviations are marked by connecting lines or thermal areas, enabling the control center to quickly grasp the dynamic environmental conditions of the entire exhibition space.

[0021] It should be specifically noted that after receiving the environmental deviation analysis results and the equipment operation performance deviation results, the control center processes the dynamic feature map of the exhibition space based on a multivariate coupled analysis model constructed using graph convolutional networks and attention mechanisms, and outputs the comprehensive operation status analysis results of the exhibition space. Among them, the environmental deviation analysis results reflect the deviation of environmental parameters such as temperature, humidity, and carbon dioxide concentration from preset standards; the equipment operation performance deviation results reflect the abnormal state of equipment parameters such as lighting equipment current and voltage, air conditioning equipment temperature and pressure, projection equipment internal temperature and fan speed.

[0022] It should be specifically explained that the construction idea of ​​the multivariate coupling analysis model is as follows: the exhibition space is abstracted into a graph structure, and each physical area (such as exhibition area, passage, rest area) is regarded as a spatial environment node. Each node is associated with environmental parameters such as temperature, humidity, and carbon dioxide concentration. Lighting equipment, air conditioning equipment, projection equipment, etc. are regarded as equipment status nodes, and each node carries operating parameters such as current, voltage, temperature, pressure, and speed. The nodes are connected by edges, and the relationship of the edges covers spatial adjacency (such as two exhibition areas being adjacent), the influence relationship between equipment and environment (such as the temperature regulation of the area by air conditioning), and the collaborative relationship between equipment (such as the linkage between lighting and projection equipment). At the same time, the time series data of the dynamic feature spectrum is divided into time windows (such as 10 minutes) to form a multidimensional data input that includes spatial graph structure and time series.

[0023] It should be further explained that the spatial feature extraction method using graph convolutional networks is as follows: a two-layer graph convolutional network is designed. The first layer uses a 3×3 convolutional kernel and the activation function is ReLU to aggregate the first-order features of adjacent nodes (such as the association between a certain exhibition area and adjacent air conditioners). The second layer uses a 5×5 convolutional kernel and the activation function is Sigmoid to integrate the second-order features across regions (such as the association between multiple exhibition areas and the central ventilation system), and finally outputs a spatial feature matrix.

[0024] It should be specifically explained that the process of introducing the attention mechanism for time series analysis is as follows: First, the spatial features output by the graph convolutional network are arranged in chronological order to form a time series feature sequence. Then, the weights of features at different time steps are calculated through the query-key-value (QKV) mechanism: where query (Q) is the spatial feature of the current time step (e.g., the temperature and humidity features of a certain exhibition area at time t), key (K) is the spatial feature of historical time steps (e.g., the features from time t-1 to t-10), and value (V) is the importance coefficient of each time step feature. The weight calculation results reflect the degree of influence of historical features on the current state (e.g., the correlation between the abnormal projection temperature at time t and the fan speed at time t-5), such as the correlation between the current abnormal temperature of the projection equipment and the changes in fan speed over a period of time in the past. Finally, the time series features are aggregated based on the calculated attention weights to highlight the abnormal features of key time steps, such as the features of a period when the equipment exceeds tolerance for a long time.

[0025] It should be further explained that the calculation method for the environmental comfort level in the comprehensive operation status analysis results is as follows: based on the real-time values ​​of temperature, humidity, and carbon dioxide concentration and the preset comfort standards (such as temperature 18-25℃, humidity 40%~60%, carbon dioxide ≤1000ppm), each parameter is divided into five levels: excellent, good, medium, poor, and inferior according to the degree of deviation. For example, a temperature of 18-22℃ is "excellent" and 22-25℃ is "good". Then, the comprehensive comfort level is obtained by weighted averaging (temperature weight 0.4, humidity weight 0.3, carbon dioxide weight 0.3). The result is also divided into five levels, from level 1 to level 5, corresponding to comfort levels from high to low.

[0026] Further explanation is needed regarding the humidity and carbon dioxide level classifications: "Excellent" humidity corresponds to 45%–55% RH, "Good" to 40%–45% RH or 55%–60% RH, "Medium" to 35%–40% RH or 60%–65% RH, "Poor" to 30%–35% RH or 65%–70% RH, and "Inferior" to <30% RH or >70% RH; "Excellent" carbon dioxide concentration corresponds to ≤800 ppm, "Good" to 800-1000 ppm, "Medium" to 1000-1200 ppm, "Poor" to 1200-1500 ppm, and "Inferior" to >1500 ppm. The individual parameter scores (1-5 points, 1=Excellent, 5=Inferior) and the comprehensive comfort level calculation formula (temperature score × 0.4 + humidity score × 0.3 + carbon dioxide score × 0.3) are also clearly defined, with the results rounded to the nearest integer.

[0027] It should be specifically noted that the calculation of the equipment operation stability index needs to consider multiple factors: On the one hand, the standard deviation of the real-time value of the parameter from the median of the benchmark interval is calculated to assess the short-term fluctuation range of the parameter. For example, the smaller the standard deviation of the current of lighting equipment, the higher its stability. On the other hand, the number of times the parameter exceeds the benchmark interval per unit time is counted, and combined with the deduction of points for the duration of the deviation, a deduction weight for the duration of the deviation is set (after normalization, 0.1 points are deducted for deviations of 1-5 minutes, 0.3 points for deviations of 5-10 minutes, and 0.5 points for deviations of >10 minutes); for example, air conditioners... If the pressure deviation persists for 7 minutes, 0.3 points are deducted after normalization and directly included in the stability index calculation. Finally, after normalizing the fluctuation range and deviation frequency of each equipment parameter, the fluctuation range and deviation frequency of each equipment parameter are linearly weighted according to the influence weight of the equipment on the exhibition effect (air conditioning equipment directly determines environmental comfort, weight 0.5; projection equipment affects the digital presentation of exhibits, weight 0.3; lighting equipment affects the basic visitor experience, weight 0.2), generating a stability index of 0-100. The higher the value, the more stable the equipment operation.

[0028] It should be further explained that the fluctuation amplitude and the frequency of deviation are processed by min-max normalization: the parameter value is mapped to the interval [0,1], and the formula is normalized value = (actual value - minimum value) / (maximum value - minimum value); where the "maximum value / minimum value" of the fluctuation amplitude is determined based on the statistical data of the equipment operation data of the past 30 days (e.g. the maximum value of the lighting current fluctuation amplitude is 0.5A and the minimum value is 0A), and the maximum value of the frequency of deviation is set to 5 times / hour and the minimum value is 0 times / hour.

[0029] It should be further explained that the environment-equipment coupling coordination degree is calculated as "rule matching score × 0.6 + correlation analysis score × 0.4" (total score normalized to 0-1). The rule matching score is determined as follows: "1 point is awarded for fully complying with the preset rules (e.g., air conditioner chilled water valve opening ≥ 70% when temperature > 25℃), 0.5 points are awarded for partially complying (opening 50%~70%), and 0 points are awarded for not complying". The correlation analysis score is determined as follows: "1 point is awarded for Pearson correlation coefficient |r| ≥ 0.7, 0.6 points are awarded for 0.5 ≤ |r| < 0.7, and 0 points are awarded for |r| < 0.5".

[0030] Further explanation is needed regarding the four-step training and application of the multivariate coupling analysis model: Data collection and labeling: Collect environmental and equipment data for scenarios such as normal operation and abnormal failures (e.g., air conditioning failure, projector overheating), label comfort levels (levels 1-5), equipment stability status (stable / basically stable / unstable), and coupling coordination results (meeting standards / not meeting standards), forming a training dataset; Loss function setting: Combine regression loss (comfort level prediction error, using MSE loss) and classification loss (equipment failure status classification error, using cross-entropy loss); Model training: Use the Adam optimizer (learning rate 0.001, batch size = 32, weight decay coefficient = 1e-4, number of iterations 100) for iterative training. When the validation set loss function does not decrease for 5 consecutive rounds, trigger the early stopping mechanism to avoid overfitting, and finally save the model parameters with the minimum loss; Model application: The control center inputs real-time dynamic feature maps into the trained model and outputs comprehensive operational status analysis results.

[0031] It should be specifically noted that, subsequently, thresholding and rule matching are performed based on the comprehensive operational status analysis results to generate the final judgment results, which include environmental risk judgment results, equipment hazard judgment results, and overall control status judgment results.

[0032] It should be further explained that the thresholding process for the comprehensive operational status analysis results is as follows: For the environmental comfort level, three threshold standards are preset: the first threshold corresponds to "excellent" (meeting the ideal range for the protection of all exhibits and human comfort), the second threshold corresponds to "good" (slightly deviating from the ideal range but without substantial impact), and the third threshold corresponds to "poor" (exceeding the safe range and requiring immediate intervention); the equipment operation stability index is set with a quantitative threshold of 0-100 points, with 80 points and above being "stable", 60-79 points being "basically stable", and below 60 points being "unstable"; the environment-equipment coupling coordination degree is divided into three threshold ranges: "high coordination", "medium coordination", and "low coordination". The interval boundaries are determined based on the linkage response efficiency between environmental parameters and equipment status (e.g., if environmental parameters recover to the standard range within 5 minutes after equipment adjustment, it is considered high coordination). By comparing the real-time analysis results with the above thresholds, the status level of each indicator is initially defined.

[0033] It needs further explanation that the specific method of rule matching is as follows: A multi-dimensional rule base is established, including environmental risk assessment rules, equipment hazard assessment rules, and overall control status assessment rules. The environmental risk assessment rules are related to the environmental comfort level and the scope of the deviation. For example, if the environmental comfort level is poor and the scope of the deviation covers 3 or more exhibition areas, it is judged as "high risk". If the level is good but the duration exceeds 1 hour, it is judged as "medium risk". The equipment hazard assessment rules combine the equipment operation stability index and abnormal correlation characteristics. For example, if the stability index is <60 points and accompanied by 2 or more parameters exceeding the tolerance, it is judged as "major hazard". If the index is 60-79 points but a single parameter exceeds the tolerance for 30 minutes, it is judged as "general hazard". The overall control status assessment rules integrate the relationship of the three factors. For example, "environmental comfort 'excellent' + equipment operation 'stable' + coupling coordination 'highly coordinated'" is judged as "no control required", and "environment 'poor' + equipment 'unstable' + coupling 'low coordination'" is judged as "emergency control".

[0034] It should be further explained that the environmental risk assessment results are generated as follows: based on the thresholded environmental comfort level, combined with the impact range and duration in the state deviation analysis results, the environmental risk assessment rules in the rule base are matched; if the environmental comfort level is "poor" and the deviation impact range involves important exhibit areas, or the carbon dioxide concentration exceeds the standard and the personnel density is high (based on the design capacity of the exhibition space, the number of people per square meter is >0.8, or the real-time number of people in the area counted by the entrance infrared counter is >80% of the design capacity), it is judged as "high risk"; if the level is "good" but the humidity parameter continues to deviate and involves the paper exhibit area, it is judged as "medium risk"; if the level is "excellent" or there is only a slight deviation and the impact range is limited, it is judged as "low risk"; the assessment results must clearly specify the risk type (such as temperature risk, humidity risk), risk level and specific affected area.

[0035] It should be further explained that the equipment hazard assessment results are generated as follows: based on the threshold range of the equipment operation stability index, the types and durations of out-of-tolerance parameters in the equipment operation performance deviation results are correlated, and the equipment hazard assessment rules are matched; if the stability index is <60 points and the air conditioning pressure parameter exceeds the tolerance accompanied by temperature regulation failure, or the internal temperature of the projection equipment exceeds the standard and the fan speed is abnormal, it is judged as a "major hazard"; if the index is 60-79 points and the lighting equipment current fluctuates but does not affect the lighting effect, it is judged as a "general hazard"; if the index is ≥80 points and all equipment parameters are within the benchmark range, it is judged as "no hazard"; the assessment results clearly specify the equipment type involved in the hazard, the hazard level, and the possible failure consequences.

[0036] It should be further explained that the overall control status determination result is generated as follows: the overall control status determination rule is matched by comprehensively considering the environmental risk determination result, the equipment hazard determination result, and the threshold range of the environment-equipment coupling coordination degree. If "high environmental risk", "major equipment hazard" and "low coupling coordination" coexist, it is determined as "emergency control". If there is "environmental risk" or "general equipment hazard" and the coupling coordination degree is "moderate coordination", it is determined as "routine control". If there is "low environmental risk", "no equipment hazard" and the coupling coordination degree is "high coordination", it is determined as "no control required". The determination result must clearly indicate the control priority (e.g., emergency, routine) and the key direction of recommended control (e.g., prioritize adjusting air conditioning equipment to improve environmental risk).

[0037] It should be noted that, finally, an optimization scheme is generated based on the final judgment result, and corresponding dynamic adjustment actions are triggered. The optimization scheme is then archived in the operation log.

[0038] It should be specifically noted that the optimization plan is generated as follows: differentiated plans are formulated based on the type and level of the final judgment result; for "emergency control" scenarios (such as high environmental risk and major equipment hazards coexisting), the optimization plan focuses on "rapid risk mitigation," including priority-based equipment adjustment instructions (such as immediately increasing air conditioning cooling capacity, temporarily activating backup ventilation systems), emergency notification mechanisms (pushing audible and visual alarms to maintenance personnel), and risk spread containment measures (such as temporarily isolating affected exhibition areas); for "routine control" scenarios (such as environmental risk or general equipment hazards), the plan focuses on "gradual optimization," including phased equipment parameter adjustment plans (such as gradually reducing lighting voltage to the standard range, adjusting by 1℃ / 10 minutes). The solution includes measures such as adjusting air conditioning temperature, coordinating strategies with related devices (e.g., optimizing ventilation frequency simultaneously when adjusting air conditioning), and an effectiveness verification cycle (evaluating the control effectiveness every 30 minutes). For scenarios where "no control is required," the solution focuses on "continuous monitoring," including only regular data collection frequency settings (e.g., maintaining device status sampling every 5 seconds) and an anomaly warning threshold update mechanism. The anomaly warning threshold is dynamically calibrated using a 7-day rolling average method: every Monday morning, the mean and standard deviation of environmental / equipment parameters over the past 7 days are automatically calculated, and the 'mean + 2 times the standard deviation' is set as the new warning upper limit, and the 'mean - 2 times the standard deviation' is set as the warning lower limit. After calibration, the control center synchronizes the data to all sensor nodes to ensure that the warning threshold matches the actual operating status.

[0039] It should be further explained that the triggering mechanism of the dynamic adjustment action is as follows: the control center parses the optimization scheme into executable equipment control commands and transmits them to the execution layer through the wireless communication module; the execution layer executes the commands in sequence according to their priority: emergency adjustment commands have the highest priority (response delay ≤ 1 second) and directly drive the equipment to perform forced adjustment (such as cutting off the power to the faulty projection equipment or starting the emergency mode of the air conditioner); regular adjustment commands are executed according to a preset time sequence (such as adjusting the lighting equipment during non-peak visit periods) and the execution progress is transmitted back in real time through the feedback channel (such as "the air conditioner temperature has dropped from 28℃ to 26℃"); when no adjustment is needed, an automatic inspection mechanism is triggered (a status snapshot of all equipment is taken every 2 hours) to ensure that no potential risks are overlooked.

[0040] It should be further explained that the optimization plan is archived to the operation log in the following way: The operation log adopts a structured data format, and the recorded content includes the generation time of the optimization plan, the triggering reason, the specific control instructions (such as "adjust the air conditioning pressure to 1.2MPa" and "increase the projector fan speed by 20%), the parameter comparison before and after execution (such as the ambient temperature dropping from 30℃ to 24℃), and the execution status (success / failure / partial execution); the log is automatically sorted by timestamp, supports retrieval by exhibition area, equipment type, risk level and other dimensions, and the storage period is not less than 90 days, which is convenient for tracing historical control records.

[0041] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other. In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception, characterized in that: include: S1: Construct an integrated architecture for environmental perception and control. Collect environmental characteristic parameters and equipment operating status information of the exhibition space in real time through a distributed sensor network. The environmental characteristic parameters and equipment operating status information together constitute the raw data and are transmitted to the control center. S2: The control center performs fusion processing and feature extraction on the collected raw data; Establish a comparison model between the combined environmental and equipment status and preset standards, generate environmental deviation analysis results and equipment operating performance deviation results, and construct a dynamic feature map of the exhibition space; S3: The control center receives the environmental deviation analysis results and the equipment operation performance deviation results. Based on the multivariate coupling analysis model constructed by graph convolutional network and attention mechanism, it models and performs time series analysis on the dynamic feature map of the exhibition space, and outputs the comprehensive operation status analysis results of the exhibition space. The comprehensive operation status analysis results include the environmental comfort level, the equipment operation stability index, and the environment-equipment coupling coordination degree. S4: Based on the comprehensive operation status analysis results, thresholding and rule matching are performed to generate the final judgment results, which include environmental risk judgment results, equipment hidden danger judgment results, and overall control status judgment results. S5: Generate an optimization plan based on the final judgment result and trigger corresponding dynamic control actions, and archive the optimization plan to the operation log and evaluation report.

2. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The environmental characteristic parameters include temperature, humidity, and carbon dioxide concentration; the equipment operating status information includes the current and voltage parameters of the lighting equipment, the temperature and pressure parameters of the air conditioning equipment, and the internal temperature and fan speed of the projection equipment.

3. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The distributed sensor network is constructed with an industrial-grade gateway in the control center as the core node, using a star topology. Sensors that collect environmental characteristic parameters and sensors that collect equipment operating status information are connected to the core node through wireless communication modules, and a dual-antenna redundancy switching mechanism and a node identifier mapping mechanism are configured.

4. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The fusion process includes preprocessing of the original data, spatiotemporal alignment, and multi-source integration; wherein, the preprocessing includes correcting outlier values ​​of environmental parameters and imputing missing values ​​of equipment operating status using the sliding window method; the feature extraction includes extracting time trend features and spatial distribution features from environmental parameters, extracting operating stability features and inter-equipment correlation features from equipment status information, and generating feature vectors.

5. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The comparison model is established by pre-setting environmental parameter threshold ranges and equipment operating baseline intervals, taking the real-time collected joint feature matrix of environment and equipment as input, and performing environmental threshold comparison, equipment baseline comparison and environment-equipment logical association comparison in layers, and outputting environmental deviation analysis results and equipment operating performance deviation results.

6. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The multivariate coupling analysis model is constructed by transforming the exhibition space into a graph structure containing environmental nodes and equipment nodes. Nodes are connected by edges through spatial adjacency, environmental influence, and equipment collaboration. Spatial correlation features are extracted using a graph convolutional network, and temporal features are aggregated by combining an attention mechanism.

7. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The comprehensive operational status analysis results are obtained based on the output of a multivariate coupling analysis model. Three indicators are calculated: environmental comfort level, equipment operation stability index, and environment-equipment coupling coordination degree. The environmental comfort level is classified according to the degree of deviation of environmental parameters, the equipment operation stability index is calculated based on the fluctuation range and frequency of equipment parameters, and the environment-equipment coupling coordination degree is determined based on the results of rule matching score and correlation analysis score.

8. The method for dynamic control of digital exhibition space in engineering exhibitions that integrates environmental perception as described in claim 1, characterized in that: The final judgment result is generated by thresholding and matching the comprehensive operation status analysis results. Thresholding is used to determine the level of each indicator. Rule matching uses preset environmental risk judgment rules, equipment hazard judgment rules, and overall control status judgment rules to make multi-dimensional judgments on the analysis results and output environmental risk judgment results, equipment hazard judgment results, and overall control status judgment results.

Images