Pouring environment monitoring system and method

Abstract

The invention discloses a pouring environment monitoring system and method, and relates to the field of pouring monitoring, and the pouring environment monitoring system comprises a management module which serves as a central interaction platform and provides interaction authority for each functional module after identity verification; the environment scanning module is used for deploying a sensor in a pouring area to periodically scan a pouring environment in real time and collecting area state data; the evaluation unit is used for analyzing and processing the temperature information in the area according to the area state data provided by the environment scanning module, and extracting a temperature index, a pollution index and a concrete solidification index; according to the method, the dynamic environment condition can be rapidly adjusted, prediction of the future pouring abnormal coefficient is more accurate, the safety of the construction process is improved, a scientific basis is provided for construction quality control, the accuracy and effectiveness of decision making are improved, and the safety and emergency processing efficiency of the construction process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pouring monitoring, and specifically provides a pouring environment monitoring system and method. Background Art

[0002] Construction projects involve a large number of materials and processes. As a key link, the quality of concrete pouring directly affects the structural safety and durability of buildings. Environmental factors, such as temperature, humidity, pollutants, etc., can significantly affect the curing process and final strength of concrete. Therefore, it is particularly important to monitor these factors in real time. With the continuous improvement of safety standards and quality requirements in the construction industry, traditional monitoring methods for the pouring process are difficult to meet the needs of modern projects. Building an advanced monitoring system to achieve real-time monitoring of the environment and data analysis has become an important means to ensure construction quality and safety;

[0003] Existing technologies often rely on simple data recording and analysis, lack deep learning and intelligent prediction capabilities, usually rely on manual monitoring or fixed-frequency data collection, cannot achieve real-time feedback on environmental changes, lack integrated analysis of temperature fluctuations, pollutant concentrations, and concrete hardening speeds, can only respond passively during anomaly detection, lack a flexible alarm mechanism, and do not fully consider the influence of multiple environmental factors. Summary of the Invention

[0004] (I) Technical Problems to be Solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a pouring environment monitoring system and method, which can effectively solve the problems of the existing technology.

[0006] (II) Technical Solutions

[0007] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0008] The present invention discloses a pouring environment monitoring system, including:

[0009] A management module, as a central interaction platform, provides interaction permissions for each functional module after authentication.

[0010] An environment scanning module, used to deploy sensors in the pouring area to periodically and real-time scan the pouring environment and collect area status data.

[0011] An evaluation unit, used to analyze and process the temperature information in the area according to the area status data provided by the environment scanning module, and extract the temperature index, pollution index, and concrete setting index.

[0012] Feature building module, which is used to identify temperature fluctuations, pollutant concentration change rate, and concrete hardening speed characteristics to form a multi-dimensional feature set for input into the prediction model;

[0013] The model building module is used to build a prediction model using a deep learning algorithm based on the feature set built by the feature building module, and train it based on the historical temperature index, pollution index and concrete solidification property characteristics to train and predict the casting abnormality coefficient within a specified period;

[0014] The abnormal output module is used to extract the prediction model trained by the model building module, input the current real-time temperature index, pollution index and concrete solidification property characteristics, analyze the influence of temperature and pollution status in the pouring area on concrete drying, and predict the pouring abnormal coefficient in the future specified period;

[0015] The alarm response module is used to trigger a real-time alarm when the abnormal coefficient submitted by the abnormal output module exceeds the preset calibration threshold, and transmit the warning information to the management end to remind or link the mechanical equipment to start and stop directly.

[0016] Furthermore, the evaluation unit is provided with submodules at the lower level, and the submodules include: a material evaluation module, a temperature evaluation module and a pollution evaluation module, wherein:

[0017] The material evaluation module is used to extract temperature data from the raw data and generate a temperature index according to the preset temperature index classification;

[0018] The temperature assessment module is used to extract the concentration data of harmful substances, dust and chemical pollutants in the air from the environmental sensor, normalize them according to the set standards, and generate the pollution index;

[0019] The pollution assessment module is used to evaluate the concrete solidification index during the pouring process. It outputs the comprehensive solidification property parameters of concrete by real-time monitoring of the temperature, humidity and hardening speed indicators inside and on the surface of the concrete.

[0020] Furthermore, the feature construction module is interactively connected to a data fusion module via a wireless network, and the data fusion module is used to preprocess, clean and fuse the original data of the evaluation unit, interpolate and patch missing data, and issue warning prompts for abnormal data, so as to form a high-quality data set for prediction model input.

[0021] Furthermore, the process of the feature construction module performing multi-dimensional feature recognition includes:

[0022] The temperature index time series data is processed by time domain analysis method, and the root mean square error of the temperature change rate within the set time window is calculated. When the fluctuation amplitude of three consecutive windows exceeds the historical threshold, it is marked as an abnormal temperature fluctuation event;

[0023] A multi-pollutant coupling analysis model is established for the pollution index. The sliding window difference algorithm is used to calculate the concentration gradient changes of harmful substances, dust, and chemical pollutants in the air. When the change rate of any pollution factor exceeds the preset safety curve, a corresponding concentration change rate eigenvector is generated.

[0024] Based on the synchronous monitoring data of the internal and surface sensors of the concrete, a three-dimensional mapping table of temperature, humidity, and hardening speed is established. By comparing the deviation of the real-time hardening speed from the material solidification theory curve, the hardening acceleration factor is calculated. When the deviation exceeds the material specification value, a hardening anomaly flag is triggered.

[0025] The above-mentioned fluctuation characteristics, change rate characteristics, and hardening anomaly flags are time-stamped and aligned, and a multi-dimensional feature matrix containing time correlation is generated through normalization processing as the standardized input of the prediction model.

[0026] Furthermore, the model construction module is connected to an input management module through wireless network interaction. The input management module is used to check whether the integrity and timeliness of the data submitted to the model construction module meet the preset integrity threshold and timeliness threshold, and cache the abnormal input when the sudden data fluctuation reaches the preset fluctuation threshold.

[0027] Furthermore, the expression of the working logic of the prediction model in the abnormal output module is:

[0028] ;

[0029] In the formula, represents the predicted value of the pouring anomaly coefficient in the future Δ time unit, ( ) represents the Sigmoid activation function, mapping the output to the interval (0, 1), represents the characteristic weight coefficient of the th historical time node, represents the number of historical time nodes, represents the temperature index at the historical moment , represents the environmental factor correction coefficient of the temperature index, represents the reference temperature value, represents the pollution index at the historical moment , represents the environmental factor correction coefficient of the pollution index, represents the reference pollution value, represents the derivative of the concrete hardening speed, represents the environmental factor correction coefficient of the derivative of the concrete hardening speed.

[0030] Furthermore, the alarm response module is interactively connected to a configuration module via a wireless network, and the configuration module is used to preset the alarm threshold of the alarm response module at multiple levels, and trigger an alarm strategy of corresponding level content according to the abnormal coefficient value output by the abnormal output module, and the alarm strategy includes: sound, text message and email.

[0031] Furthermore, the environment scanning module is interactively connected to the management module and the evaluation unit through a wireless network, the feature construction module is interactively connected to the evaluation unit and the model construction module through a wireless network, and the abnormal output module is interactively connected to the model construction module and the alarm response module through a wireless network.

[0032] A pouring environment monitoring method comprises the following steps:

[0033] Step 1: Perform user identity authentication on the central interaction platform and define the user's interaction rights;

[0034] Step 2: Deploy sensors in the pouring area to periodically scan the pouring environment in real time and collect regional status data related to temperature, contaminants, and concrete solidification;

[0035] Step 3: Analyze the collected regional status data and extract the temperature index, pollution index and concrete solidification index;

[0036] Step 4: Identify the characteristics of temperature fluctuation, pollutant concentration change rate and concrete hardening speed to form a multidimensional feature set;

[0037] Step 5: Using the deep learning algorithm, the constructed feature set is trained based on the historical temperature index, pollution index and concrete solidification property characteristics to generate a prediction model for the anomaly coefficient;

[0038] Step 6: Use the trained prediction model to take the current temperature index, pollution index and concrete solidification property characteristics as input to predict the pouring anomaly coefficient within the specified future period;

[0039] Step 7: When the predicted pouring abnormality coefficient exceeds the preset calibration threshold, the system automatically triggers a real-time alarm and transmits a warning message to the management end or links the mechanical equipment to start and stop processing;

[0040] Step 8: According to the abnormal coefficient value, according to the multi-level preset alarm threshold, the corresponding level of alarm strategy is triggered, and the alarm notification is carried out by sound, SMS and email.

[0041] Furthermore, the sensors in step 2 include: a temperature sensor DS18B20, deployed inside and on the surface of the pouring area, for periodically collecting ambient and internal concrete temperature data; a laser dust sensor PM2008M, deployed at the air circulation nodes in the pouring area, for real-time detection of PM2.5 and PM10 particle concentrations; a chemical pollutant detection module MQ-135, integrated at the edge of the pouring area, for monitoring ammonia, sulfide, and benzene volatile concentrations; a temperature and humidity composite sensor DHT22, embedded in the concrete surface layer and the support structure, for synchronously collecting humidity and temperature change data; and a piezoelectric hardening monitoring sensor PZT-5A, embedded at the key nodes inside the concrete, for inferring the hardening rate by detecting the change rate of acoustic wave propagation speed.

[0042] (III) Beneficial Effects

[0043] Adopting the technical solution provided by the present invention, compared with the known prior art, it has the following beneficial effects:

[0044] 1. Using a deep learning algorithm to construct a prediction model, a model capable of capturing complex non-linear relationships is trained based on historical data. Compared with traditional statistical modeling methods, it has higher accuracy and adaptability, can be quickly adjusted for dynamic environmental conditions, and makes the prediction of future pouring anomaly coefficients more accurate, which not only improves the safety of the construction process but also provides a scientific basis for construction quality control.

[0045] 2. Through the deployment of multiple sensors, it is possible to comprehensively and real-time monitor the temperature, pollutant concentration, and concrete solidification state in the pouring environment, providing the ability to perform correlation analysis on multiple key indicators, ensuring a comprehensive understanding of the real-time state of the pouring environment, and thus improving the accuracy and effectiveness of decision-making.

[0046] 3. By integrating a real-time alarm and an automated response mechanism, when the monitored anomaly coefficient exceeds the preset threshold, it can automatically trigger an alarm and link mechanical equipment for corresponding processing. The real-time response ability greatly improves the safety of the construction process and the emergency handling efficiency, reduces the need for manual intervention, and reduces the risks caused by human factors. Through the design of a multi-level alarm strategy, the system can also flexibly adjust response measures according to the degree of anomaly, further improving the intelligent level and practicality of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0048] Figure 1 It is a schematic diagram of the framework of the present invention;

[0049] Figure 2 It is a schematic diagram of the framework of the evaluation unit in the present invention.

[0050] The labels in the figure respectively represent: 1. Management module; 2. Environment scanning module; 3. Evaluation unit; 31. Material evaluation module; 32. Temperature evaluation module; 33. Pollution evaluation module; 4. Feature construction module; 5. Model construction module; 6. Abnormal output module; 7. Alarm response module; 8. Data fusion module; 9. Input management module; 10. Configuration module. Specific implementation manners

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] The present invention will be further described below with reference to the embodiments.

[0053] ① Embodiment 1

[0054] A pouring environment monitoring system in this embodiment, as Figure 1 - Figure 2 shown, includes:

[0055] The management module 1, as a central interaction platform, provides the interaction permissions of each functional module after authentication;

[0056] The environment scanning module 2 is used to deploy sensors in the pouring area to periodically perform real-time scanning on the pouring environment and collect area status data;

[0057] The evaluation unit 3 is used to analyze and process the temperature information in the area according to the area status data provided by the environment scanning module 2, and extract the temperature index, pollution index, and concrete setting index; Sub-modules are deployed under the evaluation unit 3, and the sub-modules include: a material evaluation module 31, a temperature evaluation module 32, and a pollution evaluation module 33, where:

[0058] The material evaluation module 31 is used to extract temperature data from the original data and generate a temperature index according to the preset temperature indicators for classification;

[0059] A temperature evaluation module 32 is configured to extract data on the concentrations of harmful substances, dust, and chemical pollutants in the air from environmental sensors, perform normalization processing according to set standards, and generate a pollution index.

[0060] A pollution evaluation module 33 is configured to evaluate the concrete setting index during the pouring process, output comprehensive concrete setting attribute parameters by monitoring in real time the temperature, humidity, and hardening speed indicators inside and on the surface of the concrete.

[0061] A feature construction module 4 is configured to identify features such as temperature fluctuations, pollutant concentration change rates, and concrete hardening speeds, and form a multi-dimensional feature set for input into the prediction model. The feature construction module 4 is connected to a data fusion module 8 through wireless network interaction. The data fusion module 8 is configured to preprocess, clean, and fuse the original data of the evaluation unit 3, perform interpolation repair on missing data, and give a warning prompt for abnormal data, so as to form a high-quality data set for input into the prediction model.

[0062] A model construction module 5 is configured to construct a prediction model using a deep learning algorithm based on the feature set constructed by the feature construction module 4, and perform training according to historical temperature indices, pollution indices, and concrete setting attribute features to train and predict the pouring abnormality coefficient within a specified period. The model construction module 5 is connected to an input management module 9 through wireless network interaction. The input management module 9 is configured to check whether the integrity and timeliness of the data submitted to the model construction module 5 meet the preset integrity threshold and timeliness threshold, and cache abnormal inputs when the sudden data fluctuation reaches the preset fluctuation threshold.

[0063] An abnormality output module 6 is configured to extract the prediction model trained by the model construction module 5, input the current real-time temperature index, pollution index, and concrete setting attribute features, analyze the influence of the temperature and pollution status in the pouring area on the concrete drying, and predict the pouring abnormality coefficient within a future specified period.

[0064] The expression of the working logic of the prediction model in the abnormality output module 6 is:

[0065] ;

[0066] In the formula, represents the predicted value of the pouring abnormality coefficient for the future Δ time unit, represents the Sigmoid activation function, which maps the output to the interval (0, 1), represents the feature weight coefficient at the th historical time node, represents the number of historical time nodes, represents the historical moment 's temperature index, represents the environmental factor correction coefficient of the temperature index, represents the reference temperature value, represents the historical moment of the pollution index, represents the environmental factor correction coefficient of the pollution index, represents the reference pollution value, represents the derivative of the concrete hardening speed, represents the environmental factor correction coefficient of the derivative of the concrete hardening speed;

[0067] The alarm response module 7 is used to trigger a real-time alarm when the abnormal coefficient submitted by the abnormal output module 6 exceeds the preset calibration threshold, and transmit the warning information to the management end for reminder or directly link the mechanical equipment to start or stop. The alarm response module 7 is connected to the configuration module 10 through wireless network interaction. The configuration module 10 is used to preset the alarm threshold of the alarm response module 7 at multiple levels, and trigger the alarm strategy of the corresponding level content according to the abnormal coefficient value output by the abnormal output module 6. The alarm strategy includes: sound, text message and email.

[0068] The environment scanning module 2 is connected to the management module 1 and the evaluation unit 3 through wireless network interaction. The feature construction module 4 is connected to the evaluation unit 3 and the model construction module 5 through wireless network interaction. The abnormal output module 6 is connected to the model construction module 5 and the alarm response module 7 through wireless network interaction.

[0069] Compared with the prior art, through the central interaction platform and multi-level modular design, identity authentication and permission management are realized, the system security and flexibility are improved, the deployed sensors can perform periodic real-time scanning to ensure comprehensive monitoring of the pouring environment, and by extracting and analyzing environmental data, a comprehensive temperature index, pollution index and concrete setting index can be generated, providing a scientific basis for construction quality;

[0070] In addition, the system applies deep learning methods to build a prediction model, which can not only monitor the current state, but also predict the future pouring abnormal coefficient, effectively preventing potential risks. The real-time alarm function combined with the multi-level alarm strategy ensures that relevant personnel can respond in time, improving the safety and reliability of the construction process.

[0071] ② Embodiment 2

[0072] On other levels, this embodiment also provides another optimization mechanism based on Embodiment 1, specifically a pouring environment monitoring method, including the following steps:

[0073] Step 1: Perform user identity authentication on the central interaction platform and define the interaction permissions of the user;

[0074] Step 2: Deploy sensors inside the pouring area, periodically scan the pouring environment in real time, and collect regional status data related to temperature, pollutants and concrete setting;

[0075] Step 3: Analyze the collected regional status data and extract the temperature index, pollution index and concrete solidification index;

[0076] Step 4: Identify the characteristics of temperature fluctuation, pollutant concentration change rate and concrete hardening speed to form a multidimensional feature set;

[0077] Step 5: Using the deep learning algorithm, the constructed feature set is trained based on the historical temperature index, pollution index and concrete solidification property characteristics to generate a prediction model for the anomaly coefficient;

[0078] Step 6: Use the trained prediction model to take the current temperature index, pollution index and concrete solidification property characteristics as input to predict the pouring anomaly coefficient within the specified future period;

[0079] Step 7: When the predicted pouring abnormality coefficient exceeds the preset calibration threshold, the system automatically triggers a real-time alarm and transmits a warning message to the management end or links the mechanical equipment to start and stop processing;

[0080] Step 8: According to the abnormal coefficient value, according to the multi-level preset alarm threshold, the corresponding level of alarm strategy is triggered, and the alarm notification is carried out by sound, SMS and email;

[0081] The sensors in step 2 include: temperature sensor DS18B20, deployed inside and on the surface of the pouring area, for periodically collecting environmental and internal concrete temperature data; laser dust sensor PM2008M, deployed at the air circulation node in the pouring area, for real-time detection of PM2.5 and PM10 particle concentrations; chemical pollutant detection module MQ-135, integrated at the edge of the pouring area, for monitoring the concentrations of ammonia, sulfide and benzene volatiles; temperature and humidity composite sensor DHT22, embedded in the concrete surface and supporting structure, for synchronous collection of humidity and temperature change data; piezoelectric hardening monitoring sensor PZT-5A, pre-buried in key nodes inside the concrete, and inferring the hardening rate by detecting the rate of change of sound wave propagation velocity.

[0082] ③Example 3

[0083] This embodiment provides a process for performing multi-dimensional feature recognition, including:

[0084] The temperature index time series data is processed by time domain analysis method, and the root mean square error of the temperature change rate within the set time window is calculated. When the fluctuation amplitude of three consecutive windows exceeds the historical threshold, it is marked as an abnormal temperature fluctuation event;

[0085] A multi-pollutant coupling analysis model is established for the pollution index. The sliding window difference algorithm is used to calculate the concentration gradient changes of harmful substances, dust and chemical pollutants in the air. When the change rate of any pollution factor exceeds the preset safety curve, a corresponding concentration change rate feature vector is generated;

[0086] Based on the synchronous monitoring data of sensors inside and on the surface of concrete, a three-dimensional mapping table of temperature, humidity and hardening speed is established. By comparing the deviation of the real-time hardening speed from the material solidification theoretical curve, the hardening acceleration factor is calculated. When the deviation exceeds the material specification value, a hardening anomaly flag is triggered;

[0087] Align the above fluctuation characteristics, change rate characteristics and hardening anomaly flags by timestamp, and generate a multi-dimensional feature matrix containing time correlation through normalization processing as the standardized input of the prediction model.

[0088] Compared with the prior art, the monitoring and analysis capabilities of the pouring environment are significantly improved through the multi-dimensional feature recognition process. The time domain analysis method and the root mean square error are used to calculate the temperature change rate, improving the recognition accuracy of abnormal temperature fluctuation events; Secondly, a multi-pollutant coupling analysis model is constructed for pollution indicators, which can monitor and analyze the concentration changes of multiple pollutants in real time to ensure environmental safety;

[0089] By synchronously monitoring the temperature, humidity and hardening speed of concrete, establishing a three-dimensional mapping table and calculating the deviation, a more accurate evaluation of the hardening state is provided. The generated multi-dimensional feature matrix containing time correlation provides standardized input for the prediction model, improving the accuracy and reliability of the prediction. This integrated monitoring and analysis method effectively reduces potential risks and improves the intelligent level of the construction process.

[0090] Working principle: When the system in the present invention is carried, the management module 1 verifies the identity and assigns role permissions. The environment scanning module 2 deploys sensors inside the pouring area to periodically collect environmental status data. The material evaluation module 31 extracts and classifies to generate a temperature index. The temperature evaluation module 32 normalizes to generate a pollution index. The pollution evaluation module 33 evaluates the concrete setting index. The data fusion module 8 preprocesses, cleans, interpolates, and performs anomaly warning on the original data to form a high-quality multi-dimensional feature set. Subsequently, the feature construction module 4 performs multi-dimensional feature extraction on this data to identify temperature fluctuations, pollutant changes, and concrete hardening characteristics. The model construction module 5 constructs a prediction model through deep learning methods to train the prediction ability of the pouring anomaly coefficient. The anomaly output module 6 then inputs the real-time data into the model to predict the future pouring anomaly coefficient. If the predicted anomaly coefficient exceeds the preset threshold, the alarm response module 7 will trigger an alarm through the wireless network and notify the management party. At the same time, a multi-level alarm strategy is implemented with the help of the configuration module 10. The input management module 9 monitors the data of the model construction module 5 to check its integrity and timeliness to ensure the data stability input into the model;

[0091] The present invention has the advantages of efficient real-time monitoring, accurate data analysis, and intelligent early warning functions. By deploying sensors to collect temperature, pollution, and concrete setting state data, the system uses deep learning technology to construct a prediction model, which can identify abnormal situations during the pouring process in advance and trigger an alarm. The multi-dimensional feature recognition and anomaly monitoring mechanism helps to ensure the safety and quality of concrete construction, reduce construction risks, and improve resource utilization efficiency. The modular design of the overall system enables it to have good scalability and flexibility to adapt to the needs of different construction environments.

[0092] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A pouring environment monitoring system, characterized in that: include: The management module (1) serves as a central interaction platform and provides interaction permissions to each functional module after identity authentication; An environment scanning module (2) is used to deploy sensors in the pouring area to periodically scan the pouring environment in real time and collect regional status data; An evaluation unit (3) is used to analyze and process the temperature information in the area according to the area status data provided by the environment scanning module (2), and to extract a temperature index, a pollution index and a concrete solidification index; A feature construction module (4) is used to identify the characteristics of temperature fluctuation, pollutant concentration change rate, and concrete hardening speed to form a multi-dimensional feature set for input into the prediction model; A model building module (5) is used to build a prediction model using a deep learning algorithm based on the feature set built by the feature building module (4), and train the model based on the historical temperature index, pollution index and concrete solidification property characteristics to train and predict the casting abnormality coefficient within a specified period; The abnormal output module (6) is used to extract the prediction model trained by the model building module (5), input the current real-time temperature index, pollution index and concrete solidification property characteristics, analyze the influence of the temperature and pollution status in the pouring area on the concrete drying, and predict the pouring abnormality coefficient in the future specified period; The alarm response module (7) is used to trigger a real-time alarm when the abnormal coefficient submitted by the abnormal output module (6) exceeds a preset calibration threshold, and transmit the warning information to the management end to remind or link the mechanical equipment to directly start or stop processing.

2. A pouring environment monitoring system according to claim 1, characterized in that: The evaluation unit (3) is provided with submodules at a lower level, and the submodules include: a material evaluation module (31), a temperature evaluation module (32) and a pollution evaluation module (33), wherein: A material evaluation module (31) is used to extract temperature data from the original data and generate a temperature index according to a preset temperature index classification; The temperature evaluation module (32) is used to extract the concentration data of harmful substances, dust and chemical pollutants in the air from the environmental sensor, perform normalization processing according to the set standard, and generate a pollution index; The pollution assessment module (33) is used to evaluate the concrete solidification index during the pouring process, and outputs the comprehensive solidification property parameters of the concrete by real-time monitoring the temperature, humidity and hardening speed index inside and on the surface of the concrete.

3. A pouring environment monitoring system according to claim 1, characterized in that: The feature construction module (4) is interactively connected to a data fusion module (8) via a wireless network. The data fusion module (8) is used to pre-process, clean and fuse the original data of the evaluation unit (3), interpolate and patch missing data, and issue warning prompts for abnormal data, so as to form a high-quality data set for input into a prediction model.

4. A pouring environment monitoring system according to claim 1, characterized in that: The process of the feature construction module (4) performing multi-dimensional feature recognition includes: The temperature index time series data is processed by time domain analysis method, and the root mean square error of the temperature change rate within the set time window is calculated. When the fluctuation amplitude of three consecutive windows exceeds the historical threshold, it is marked as an abnormal temperature fluctuation event; A multi-pollutant coupling analysis model is established for the pollution index, and a sliding window difference algorithm is used to calculate the concentration gradient changes of harmful substances, dust and chemical pollutants in the air. When the change rate of any pollution factor exceeds the preset safety curve, the corresponding concentration change rate characteristic vector is generated; Based on the synchronous monitoring data of the internal and surface sensors of the concrete, a three-dimensional mapping table of temperature, humidity and hardening speed is established. By comparing the deviation between the real-time hardening speed and the theoretical solidification curve of the material, the hardening acceleration factor is calculated. When the deviation exceeds the material specification value, the hardening abnormality mark is triggered. The above-mentioned fluctuation characteristics, change rate characteristics and hardening anomaly marks are timestamped and normalized to generate a multidimensional feature matrix containing time correlation as the standardized input of the prediction model.

5. A pouring environment monitoring system according to claim 1, characterized in that: The model building module (5) is interactively connected to an input management module (9) via a wireless network. The input management module (9) is used to check whether the integrity and timeliness of the data submitted to the model building module (5) meet preset integrity thresholds and timeliness thresholds, and to cache abnormal input when sudden data fluctuations reach preset fluctuation thresholds.

6. A pouring environment monitoring system according to claim 1, characterized in that: The expression of the working logic of the prediction model in the abnormal output module (6) is: ； In the formula, Represents the predicted value of the casting abnormality coefficient in the future Δ time unit, ( ) represents the Sigmoid activation function, which maps the output to the (0,1) interval. Representative The feature weight coefficient of each historical time node, Represents the number of historical time nodes, Representing historical moments The temperature index, Environmental factor correction factor representing the temperature index, Represents the reference temperature value, Representing historical moments The pollution index, The environmental factor correction coefficient representing the pollution index, represents the baseline pollution value, represents the concrete hardening rate derivative, Environmental factor correction factor representing the derivative of concrete hardening rate.

7. A pouring environment monitoring system according to claim 1, characterized in that: The alarm response module (7) is interactively connected to a configuration module (10) via a wireless network. The configuration module (10) is used to preset the alarm threshold of the alarm response module (7) at multiple levels, and trigger an alarm strategy of a corresponding level of content according to the abnormal coefficient value output by the abnormal output module (6). The alarm strategy includes: sound, text message and email.

8. A pouring environment monitoring system according to claim 1, characterized in that: The environment scanning module (2) is interactively connected to the management module (1) and the evaluation unit (3) via a wireless network; the feature construction module (4) is interactively connected to the evaluation unit (3) and the model construction module (5) via a wireless network; and the abnormal output module (6) is interactively connected to the model construction module (5) and the alarm response module (7) via a wireless network.

9. A pouring environment monitoring method, the method is an implementation method of a pouring environment monitoring system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Perform user identity authentication on the central interaction platform and define the user's interaction rights; Step 2: Deploy sensors in the pouring area to periodically scan the pouring environment in real time and collect regional status data related to temperature, contaminants, and concrete solidification; Step 3: Analyze the collected regional status data and extract the temperature index, pollution index and concrete solidification index; Step 4: Identify the characteristics of temperature fluctuation, pollutant concentration change rate and concrete hardening speed to form a multidimensional feature set; Step 5: Using the deep learning algorithm, the constructed feature set is trained based on the historical temperature index, pollution index and concrete solidification property characteristics to generate a prediction model for the anomaly coefficient; Step 6: Use the trained prediction model to take the current temperature index, pollution index and concrete solidification property characteristics as input to predict the pouring anomaly coefficient within the specified future period; Step 7: When the predicted pouring abnormality coefficient exceeds the preset calibration threshold, the system automatically triggers a real-time alarm and transmits a warning message to the management end or links the mechanical equipment to start and stop processing; Step 8: According to the abnormal coefficient value, according to the multi-level preset alarm threshold, the corresponding level of alarm strategy is triggered, and the alarm notification is carried out by sound, SMS and email.

10. A pouring environment monitoring method according to claim 9, characterized in that: The sensors in step 2 include: a temperature sensor DS18B20, which is deployed inside and on the surface of the casting area and is used to periodically collect environmental and internal concrete temperature data; a laser dust sensor PM2008M, which is deployed at the air circulation node in the casting area and is used to detect PM2.5 and PM10 particle concentrations in real time; a chemical pollutant detection module MQ-135, which is integrated at the edge of the casting area and is used to monitor the concentrations of ammonia, sulfide and benzene volatiles; a temperature and humidity composite sensor DHT22, which is embedded in the concrete surface and supporting structure and is used to synchronously collect humidity and temperature change data; a piezoelectric hardening monitoring sensor PZT-5A, which is pre-buried at key nodes inside the concrete and infers the hardening rate by detecting the rate of change of the sound wave propagation velocity.

Images