Municipal engineering construction site intelligent management system and method

Abstract

The invention relates to the field of municipal engineering construction, and particularly discloses a municipal engineering construction site intelligent management system and method.The municipal engineering construction site intelligent management method comprises the steps that firstly, construction noise data, collected by a sound sensor, of a plurality of preset time points, construction vibration data collected by a vibration sensor and a surrounding citizen feedback opinion table collected by a database are obtained; and then feature extraction and correlation analysis are performed on the three by using a deep learning technology, and finally whether the municipal engineering night construction scheme needs to be adjusted is judged through a classifier, so that the construction scheme is optimized, the resident satisfaction is improved, the resident disturbing problem is reduced, and the construction efficiency is improved. And therefore, the municipal engineering is continuously promoted on the premise of not influencing the living quality of surrounding residents.

Description

Technical Field

[0001] The present application relates to the field of municipal engineering construction, and more specifically, to an intelligent management system and method for a municipal engineering construction site. Background Art

[0002] Municipal engineering refers to a series of projects carried out for the construction and maintenance of urban infrastructure, such as roads, bridges, drainage systems, parks and lighting facilities. Since these projects are often located near residential areas, they will inevitably have a certain impact on the daily lives of citizens. With the development of social economy and the acceleration of urbanization, the increase in nighttime construction has particularly disturbed the lives of residents.

[0003] In order to minimize the impact of nighttime construction on residents, dedicated environmental monitoring personnel are usually assigned to each construction site to supervise and manage the construction environment. However, with the rapid advancement of urbanization, the traditional method of assigning full-time personnel to each construction site for nighttime patrols can no longer meet the needs of modern municipal engineering management.

[0004] Therefore, a municipal engineering construction site intelligent management system and method is desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a municipal engineering construction site intelligent management system and method, which first obtains construction noise data at multiple predetermined time points collected by a sound sensor, construction vibration data collected by a vibration sensor, and feedback from surrounding citizens collected by a database, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally uses a classifier to determine whether the nighttime construction plan of the municipal engineering project needs to be adjusted, thereby optimizing the construction plan to improve residents' satisfaction and reduce nuisance problems, thereby allowing the municipal engineering project to continue to advance without affecting the quality of life of surrounding residents.

[0006] According to one aspect of the present application, a municipal engineering construction site intelligent management system is provided, which includes: A municipal engineering construction information acquisition module is used to obtain construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by the database; A municipal engineering construction information processing module, used to extract construction multimodal information correlation feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; The night construction plan adjustment judgment module is used to judge whether the night construction plan of the municipal engineering project needs to be adjusted based on the construction multimodal information association feature vector and the text semantic feature vector of the surrounding citizen feedback opinions.

[0007] According to another aspect of the present application, a method for intelligent management of a municipal engineering construction site is provided, which includes: Obtaining construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by a database; Extracting construction multimodal information association feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; Based on the construction multimodal information association feature vector and the surrounding citizens' feedback text semantic feature vector, it is determined whether the municipal engineering night construction plan needs to be adjusted.

[0008] Compared with the prior art, the present application provides an intelligent management system and method for municipal engineering construction sites, which first obtains construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by a database, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a classifier is used to determine whether the nighttime construction plan of the municipal engineering project needs to be adjusted, thereby optimizing the construction plan to improve residents' satisfaction and reduce nuisance problems, thereby allowing the municipal engineering project to continue to advance without affecting the quality of life of surrounding residents. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0010] Figure 1 It is a block diagram of the intelligent management system for municipal engineering construction sites according to an embodiment of the present application.

[0011] Figure 2 It is a block diagram of the municipal engineering construction information processing module in the municipal engineering construction site intelligent management system according to the embodiment of the present application.

[0012] Figure 3It is a block diagram of a feature extraction unit for feedback from surrounding citizens in an intelligent management system for a municipal engineering construction site according to an embodiment of the present application.

[0013] Figure 4 It is a block diagram of a nighttime construction plan adjustment and judgment module in a municipal engineering construction site intelligent management system according to an embodiment of the present application.

[0014] Figure 5 It is a flow chart of the intelligent management method of the municipal engineering construction site according to the embodiment of the present application. DETAILED DESCRIPTION

[0015] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0016] Figure 1 Schematic diagram of the municipal engineering construction site intelligent management system of the present application embodiment. Figure 1 As shown, according to the embodiment of the present application, the municipal engineering construction site intelligent management system 100 includes: a municipal engineering construction information acquisition module 110, which is used to obtain construction noise data at multiple predetermined time points collected by the sound sensor, construction vibration data collected by the vibration sensor, and feedback from surrounding citizens collected by the database; a municipal engineering construction information processing module 120, which is used to extract construction multimodal information association feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; a nighttime construction plan adjustment judgment module 130, which is used to judge whether the municipal engineering nighttime construction plan needs to be adjusted based on the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector.

[0017] In the above-mentioned municipal engineering construction site intelligent management system 100, the municipal engineering construction information acquisition module 110 is used to obtain construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by databases. It should be understood that municipal engineering refers to a series of projects carried out to build and maintain urban infrastructure, including roads, bridges, drainage systems, parks, and lighting facilities. Since these projects are often located near residential areas, they will inevitably have a certain impact on the daily lives of citizens. With the continuous development of social economy and the acceleration of urbanization, the interference with life caused by night construction has become particularly prominent. In order to minimize the impact of night construction on residents, special environmental monitoring personnel are usually arranged at each construction site to ensure the supervision and management of the construction environment. However, with the rapid development of urbanization, the traditional method of relying on equipping each construction site with full-time personnel for night patrols can no longer meet the growing needs of modern municipal engineering management. Therefore, in the technical solution of this application, the construction noise data collected by the sound sensor at multiple predetermined time points, the construction vibration data collected by the vibration sensor, and the feedback form of the surrounding citizens collected by the database are obtained, and combined with deep learning technology, it is determined whether it is necessary to adjust the nighttime construction plan of the municipal engineering to optimize the construction plan. This process aims to improve residents' satisfaction and reduce interference with surrounding residents, thereby ensuring the continuous advancement of municipal engineering without affecting the quality of life of residents.

[0018] Specifically, sound sensors can collect construction noise data at multiple predetermined time points. These data can reflect the noise intensity and frequency characteristics of construction activities. Vibrations during construction may not only affect the safety of buildings, but also cause trouble to the lives of surrounding residents. As the direct impactors of construction activities, the opinions and suggestions of citizens can provide valuable references for construction management. In this way, comprehensive consideration of construction noise, vibration data and citizen feedback can help to more comprehensively understand the impact of construction activities, and then formulate scientific and reasonable construction management plans. This comprehensive assessment not only helps to improve construction efficiency, but also effectively protects the quality of life of surrounding residents and promotes harmonious social development.

[0019] In the above-mentioned municipal engineering construction site intelligent management system 100, the municipal engineering construction information processing module 120 is used to extract the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database. In this way, the deep fusion of multimodal information can be achieved. By combining objective data with subjective feedback, the impact of construction on residents can be more comprehensively evaluated and a more complete construction impact model can be constructed. This not only helps to improve construction efficiency and reduce interference with residents, but also enhances the public's trust and support for construction management.

[0020] Figure 2 FIG. 1 is a block diagram of a municipal engineering construction information processing module in a municipal engineering construction site intelligent management system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the municipal engineering construction information processing module 120 includes: a construction noise data feature extraction unit 121, which is used to perform feature extraction on the construction noise data at multiple predetermined time points collected by the sound sensor to obtain a construction noise feature vector; a construction vibration data feature extraction unit 122, which is used to perform feature extraction on the construction vibration data collected by the vibration sensor to obtain a construction vibration time-frequency domain feature vector; a construction multimodal information feature association unit 123, which is used to associate the construction noise feature vector with the construction vibration time-frequency domain feature vector to obtain the construction multimodal information association feature vector; a surrounding citizen feedback feature extraction unit 124, which is used to perform feature extraction on the surrounding citizen feedback form collected by the database to obtain the surrounding citizen feedback text semantic feature vector.

[0021] It should be understood that construction noise is a common environmental pollution. Being in a high-noise environment for a long time will not only affect the quality of life of surrounding residents, but may also pose a potential threat to their health. Through feature extraction, complex noise data can be converted into more concise and efficient feature vectors, making these data easier to analyze and apply. This process can help professionals quickly identify noise characteristics, such as the average, maximum, and frequency distribution characteristics of noise, so as to quickly understand the noise level generated during the construction process. By extracting noise features at multiple predetermined time points, the noise features generated by different construction stages or types of operations can be identified. For example, some equipment may generate loud noise during a specific time period. Through these feature vectors, construction managers can more effectively formulate noise control measures, adjust construction time, select low-noise equipment, etc., thereby reducing interference to residents.

[0022] Furthermore, vibration sensors can accurately capture vibration signals generated during construction activities. By extracting the features of these signals, more detailed vibration feature information can be obtained, such as vibration amplitude, frequency components, vibration modes, etc. These feature vectors can help engineers better understand the nature and changing patterns of construction vibrations. By analyzing the vibration characteristics generated by different construction stages or types of operations, the potential impact of vibrations on surrounding buildings can be determined. For example, vibrations of different frequencies may have different effects on the structure of a building. Some high-frequency vibrations may cause structural fatigue, while low-frequency vibrations may cause resonance in the building. This provides important technical support for construction safety management and environmental protection.

[0023] Furthermore, construction noise and vibration are two major factors that affect the surrounding environment and residents' lives during the construction process. Noise is mainly transmitted through the air, while vibration is transmitted through the ground. The two often occur at the same time, which may have a cumulative effect on the quality of life of residents. Therefore, correlating these two eigenvectors can help to more comprehensively evaluate the overall impact of construction activities on the environment.

[0024] In particular, citizen feedback is a direct reflection of the impact of construction activities, including residents' feelings and opinions on noise, vibration, construction time, construction methods, etc. These opinions are often subjective, but they can provide important qualitative information. By extracting features from feedback, text information can be converted into semantic feature vectors in numerical form, which is convenient for subsequent analysis and comparison. This feature vector can capture the emotional tendencies, concerns, and main themes in residents' opinions, allowing analysts to quickly identify which issues are of greatest concern to residents and which aspects need improvement. By analyzing the feature vectors of feedback in different time periods or different regions, it can be found that citizens' attitudes towards construction activities change over time and their reactions to different construction projects. For example, in the early stages of a construction project, citizen feedback may be relatively negative, while as construction management measures improve, feedback may tend to be positive. Through such analysis, construction managers can monitor residents' emotional changes in real time and adjust construction strategies in a timely manner, thereby effectively improving residents' satisfaction.

[0025] In a specific embodiment of the present application, the construction noise data feature extraction unit 121 includes: converting the construction noise data at multiple predetermined time points collected by the sound sensor to obtain a construction noise input vector; passing the construction noise input vector through a construction noise time series feature extractor to obtain the construction noise feature vector.

[0026] It should be understood that construction noise data itself is often a complex and large raw signal. Directly using this data for analysis may face problems such as low data processing efficiency and difficulty in extracting information. Through the transformation process, the original noise signal can be converted into a structured input vector, thereby simplifying subsequent data processing and analysis. This transformation usually includes preprocessing, feature extraction and standardization of noise data, making the data more comparable and usable, and facilitating further analysis and modeling. The construction noise input vector can effectively capture the key characteristics of noise, such as average noise level, peak noise, noise spectrum distribution, etc. These features can provide specific information about the noise and help engineers understand the noise characteristics generated by different construction stages or construction equipment. This characteristic representation enables managers to quickly evaluate the impact of noise on the surrounding environment and then formulate corresponding management measures, such as selecting low-noise construction equipment or adjusting construction time to reduce the impact on residents.

[0027] Furthermore, the construction noise input vector is usually the raw noise data collected by sensors at multiple time points. These data may contain a lot of information, but their complexity makes it difficult to directly analyze the raw data. Through the construction noise time series feature extractor, more representative features can be extracted from these input vectors. These features can include key information such as the mean, standard deviation, peak, frequency distribution, and time series change trend of the noise, thereby simplifying the complexity of the data and improving the interpretability of the data. Construction noise is often a dynamic process, affected by multiple factors such as construction methods, time, and weather. Through time series feature extraction, we can understand the changing patterns of noise in different time periods, such as the difference in noise levels during construction peaks and troughs, or the noise characteristics generated by specific equipment at different construction stages. This in-depth analysis can provide construction units with targeted management suggestions, such as adjusting construction plans to avoid residents' rest time, or adopting construction methods to reduce noise, thereby effectively reducing the impact on the surrounding environment. Specifically, the construction noise input vector is fully connected using the fully connected layer of the construction noise time series feature extractor to extract high-dimensional implicit features of the eigenvalues ​​at each position in the construction noise input vector; and the construction noise input vector is one-dimensionally encoded using the one-dimensional convolutional layer of the construction noise time series feature extractor to extract high-dimensional implicit correlation features of the correlation between the eigenvalues ​​at each position in the construction noise input vector.

[0028] In a specific embodiment of the present application, the construction vibration data feature extraction unit 122 includes: performing wavelet transform on the construction vibration data collected by the vibration sensor to obtain a construction vibration time-frequency domain input vector; passing the construction vibration time-frequency domain input vector through a construction vibration time-frequency domain feature encoder based on a convolutional neural network to obtain the construction vibration time-frequency domain feature vector.

[0029] It should be understood that construction vibration data is usually a continuous time series signal, containing complex frequency components and time-varying characteristics. During the construction process, vibration signals will continue to change due to a variety of factors such as equipment operation, construction methods and environmental factors. Direct analysis of these raw data may not capture the key information in the signal. As a powerful time-frequency analysis tool, wavelet transform can effectively decompose vibration signals into components of different frequencies while maintaining time information. This ability of time-frequency domain analysis makes wavelet transform particularly effective in processing non-stationary signals, revealing the changing characteristics of signals in time and frequency. Wavelet transform can provide multi-scale vibration feature analysis. By selecting different wavelet basis functions, the vibration signal can be decomposed at multiple levels to extract high-frequency components (such as instantaneous vibration and impact) and low-frequency components (such as long-term trends and periodic vibrations). This multi-scale analysis can help construction managers understand the impact of construction activities on the surrounding environment at different stages. For example, in some construction stages, short-term high-frequency vibrations may occur, which may have a significant impact on surrounding buildings and residents' lives. Through wavelet transform, these key features can be accurately identified and provide data support for subsequent management and intervention.

[0030] Furthermore, the time-frequency domain input vector of construction vibration is obtained through techniques such as wavelet transform, which contains rich time-frequency information and can reflect the dynamic characteristics of vibration signals in time and frequency. However, the dimensions of these input vectors are often high, and direct use may lead to increased computational complexity and difficulty in capturing the deep-level features of the signal. By using a feature encoder based on a convolutional neural network, these high-dimensional input vectors can be effectively converted into lower-dimensional and more representative feature vectors. This dimensionality reduction process not only reduces the computational cost, but also retains the most important information in the signal. Convolutional neural networks perform well in processing time series signals and image data. They have the characteristics of local connection and weight sharing, which enables the network to effectively identify local patterns in the input data. In the analysis of construction vibration, the convolution layer can automatically learn the time-frequency characteristics of the vibration signal and extract feature representations of different frequency bands and time windows. This feature makes CNN particularly suitable for mining useful information from complex vibration data, identifying potential vibration patterns and abnormal features, and providing strong support for construction managers. Specifically, each layer of the construction vibration time-frequency domain feature encoder based on the convolutional neural network is used to perform convolution processing, mean pooling processing based on the local feature matrix, and nonlinear activation processing on the input data in the forward transfer of the layer, so that the last layer of the construction vibration time-frequency domain feature encoder based on the convolutional neural network outputs the construction vibration time-frequency domain feature vector, wherein the input of the construction vibration time-frequency domain feature encoder based on the convolutional neural network is the construction vibration time-frequency domain input vector.

[0031] In a specific embodiment of the present application, the construction multimodal information feature association unit includes: obtaining a construction noise feature vector and a construction vibration time-frequency domain feature vector as input; adjusting the dimensions of the construction noise feature vector and the construction vibration time-frequency domain feature vector; designing a Filter mechanism, wherein the Filter mechanism includes a matrix multiplication operation unit for calculating the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector; using the matrix multiplication operation unit in the Filter mechanism to calculate the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector to obtain a construction multimodal information association feature vector; and outputting the construction multimodal information association feature vector.

[0032] It should be understood that the filter mechanism is used to extract or enhance specific features from input data. Its advantage is that it can selectively process different frequency components of the signal by designing different filters (such as low-pass, high-pass, and band-pass filters), thereby achieving purposes such as denoising, feature extraction, and data smoothing. In deep learning, the filter mechanism can be used in the convolutional layer of a convolutional neural network (CNN). The learned filter parameters can be used to automatically extract key features from images or time series data, thereby improving the performance and generalization ability of the model.

[0033] Part of the deployment code is shown below.

[0034] import numpy as np class FilterMechanism: def __init__(self, noise_feature_vector, vibration_feature_vector): self.noise_feature_vector = noise_feature_vector self.vibration_feature_vector = vibration_feature_vector def compute_multi_modal_feature(self): # Make sure the dimensions of the two feature vectors are compatible if self.noise_feature_vector.shape != self.vibration_feature_vector.shape: raise ValueError("Noise feature vector and vibrationfeature vector must have the same shape.") # Calculate the product of feature vectors multi_modal_feature_vector = np.multiply(self.noise_feature_vector, self.vibration_feature_vector) return multi_modal_feature_vector noise_feature_vector = np.array([0.1, 0.2, 0.3]) vibration_feature_vector = np.array([0.4, 0.5, 0.6]) filter_mechanism = FilterMechanism(noise_feature_vector, vibration_feature_vector) multi_modal_feature_vector = filter_mechanism.compute_multi_modal_feature() print("Multi-modal feature vector:", multi_modal_feature_vector) In this way, the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector is calculated, thereby obtaining the construction multimodal information association feature vector. In the initialization method __init__ of the FilterMechanism class, the construction noise feature vector and the construction vibration time-frequency domain feature vector are received as input. Then, it is ensured that the dimensions of the two feature vectors are compatible, that is, they have the same shape. If the dimensions are inconsistent, the code will throw an error, prompting the user to perform a dimension alignment operation. Next, a Filter mechanism is designed, the core of which is a matrix multiplication operation unit. This operation unit is used to calculate the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector. Next, in the compute_multi_modal_feature method, the matrix multiplication operation unit is used to calculate the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector, where the result of the product is a new feature vector, called the construction multimodal information association feature vector. Finally, the calculated construction multimodal information association feature vector is returned.

[0035] By calculating the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector, a comprehensive construction multimodal information association feature vector is generated, which helps to better capture the complex environmental information of the construction site, thereby improving the performance of the subsequent deep learning model. Specifically, by multiplying the feature vectors of two different modes, noise and vibration, the two types of information can be effectively fused to extract a richer feature representation. This fused feature vector can better reflect the actual situation of the construction site, such as the mutual influence and synergy of noise and vibration. By fusing the noise and vibration feature vectors, the robustness and accuracy of the model can be improved, and the information loss or noise interference problems that may exist in a single modal feature vector can be reduced. In addition, the implementation of the deployment code is simple and clear, easy to understand and expand. Users can adjust the steps of feature vector preprocessing, dimension alignment, etc. according to actual needs to adapt to different application scenarios.

[0036] Figure 3 FIG. 1 is a block diagram of a feature extraction unit for feedback from surrounding citizens in an intelligent management system for a municipal engineering construction site according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the surrounding citizen feedback feature extraction unit 124 includes: a surrounding citizen feedback convolutional coding subunit 1241, which is used to pass the surrounding citizen feedback form collected from the database through a surrounding citizen feedback semantic encoder to obtain multiple surrounding citizen feedback text feature vectors; a surrounding citizen feedback feature splicing subunit 1242, which is used to splice the multiple surrounding citizen feedback text feature vectors to obtain the surrounding citizen feedback text semantic feature vector.

[0037] It should be understood that the feedback from the surrounding citizens usually exists in the form of text. Although these text information can reflect the real feelings and needs of the citizens, directly analyzing these raw text data will face the challenges of semantic understanding and information extraction. Due to the diversity of citizens' expressions, word usage habits and feedback content, it is often difficult to fully capture the true meaning of the opinions by simply relying on traditional keyword retrieval methods. Through semantic encoders, text can be converted into vector form. This vector representation can map similar opinions into similar vector spaces, which is convenient for subsequent analysis and comparison. The semantic encoders for feedback from surrounding citizens are usually based on deep learning technology (such as long short-term memory networks LSTM or transformer models Transformer), which can deeply mine the contextual information and semantic relationships of the text. This modeling capability enables the encoder to better understand the context of citizen feedback, thereby extracting more representative feature vectors. Specifically, the surrounding citizen feedback opinion form collected by the database is segmented to obtain a word sequence; the embedding layer of the surrounding citizen feedback opinion semantic encoder is used to map each word in the word sequence into a word embedding vector to obtain a sequence of word embedding vectors; the converter-based Bert model of the surrounding citizen feedback opinion semantic encoder is used to perform global context semantic encoding on the sequence of word embedding vectors to obtain multiple surrounding citizen feedback opinion text feature vectors.

[0038] Furthermore, multiple citizen feedback text feature vectors are usually generated based on different feedback contents, and each feature vector represents the semantic information and sentiment tendency of a specific feedback. However, a single feature vector may not be sufficient to fully reflect the overall opinion trend or potential problems. By splicing these feature vectors, the individual feedback information can be merged into a unified representation, thus forming a more representative and informative semantic feature vector. This comprehensive representation can better capture the diversity and complexity of citizen opinions, facilitating subsequent analysis and application. The process of splicing feature vectors can make feedback information from different sources complement each other. For example, some citizens may be concerned about the problem of construction noise, while others may express concerns about construction safety. By splicing these feature vectors together, managers can get a full picture of public feedback, including the connection and interaction between different opinions. This multi-dimensional information integration provides a richer context for analysis, making subsequent sentiment analysis, topic modeling, and trend identification more accurate and comprehensive.

[0039] In the above-mentioned municipal engineering construction site intelligent management system 100, the night construction plan adjustment judgment module 130 is used to judge whether the municipal engineering night construction plan needs to be adjusted based on the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector. It should be understood that the construction multimodal information association feature vector usually contains information from multiple data sources, such as vibration monitoring, noise level, construction equipment operation status, etc. These data can provide an objective impact assessment on the surrounding environment during the construction process. The surrounding citizen feedback text semantic feature vector is extracted from the direct feedback of the citizens, reflecting the public's real feelings, needs and attitude towards construction. The combination of these two types of feature vectors can provide decision makers with a comprehensive perspective, enabling them to better understand the actual impact of construction. Using these two feature vectors for correlation analysis can reveal the impact pattern of construction on citizen feedback. For example, if the noise monitoring data shows that the night construction noise exceeds the environmental standard, and there are strong opinions about noise nuisance in citizen feedback, this consistency indicates that the night construction plan may have a significant impact on residents' lives. In this situation, managers should consider adjusting the construction plan, such as adjusting the construction time, optimizing the construction process, or taking noise reduction measures to reduce interference to citizens.

[0040] Figure 4 FIG. 1 is a block diagram of a nighttime construction plan adjustment and judgment module in a municipal engineering construction site intelligent management system according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the night construction plan adjustment judgment module 130 includes: a night construction feature fusion unit 131, which is used to fuse the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector to obtain a night construction plan adjustment judgment feature vector; a night construction feature optimization unit 132, which is used to perform feature modulation optimization on the night construction plan adjustment judgment feature vector based on the entropy degradation maximization simulation value to obtain a sparse optimized night construction plan adjustment judgment feature vector; a municipal engineering night construction plan adjustment judgment classification unit 133, which is used to pass the sparse optimized night construction plan adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to determine whether the municipal engineering night construction plan needs to be adjusted.

[0041] It should be understood that the fusion feature vector not only improves the depth and accuracy of data analysis, but also enhances the transparency of construction management. When citizens see that managers actively use multi-source data for analysis in the decision-making process and make adjustments accordingly, they will enhance their trust and satisfaction with the management. This interaction and feedback mechanism helps to form a good social atmosphere and promote a harmonious relationship between construction activities and the community. By fusing the construction multimodal information association feature vector and the surrounding citizens' feedback text semantic feature vector, the night construction plan adjustment judgment feature vector is generated, which can not only enhance the comprehensive understanding of the impact of night construction, but also provide important support for scientific decision-making. This process helps to optimize the construction plan, reduce the negative impact on residents' lives, and achieve the sustainability and social responsibility of construction management.

[0042] In particular, in the technical solution of the present application, the night construction plan adjustment judgment feature vector is formed by the fusion of multiple information such as construction noise features, construction vibration features and citizen feedback features. The complexity of this multimodal feature makes the relationship between different features non-linear and difficult to directly map to the structure expected by the learning parameter matrix. This structural difference may cause the learning parameter matrix to fail to fully capture the important information in the night construction plan adjustment judgment feature vector. Furthermore, due to the high dimension of the feature vector, there may be multiple features that overlap or have similar information content. For example, there may be a correlation between construction noise and feedback from surrounding citizens, but if there is no reasonable feature selection and processing mechanism, the classifier may treat similar information as different features, resulting in redundancy. Redundant features not only increase the computational complexity, but may also cause the model to confuse important information during the training process, thereby affecting the accuracy of the classification results. In this way, the model may be overfitted, thereby affecting the accuracy of the classification results of whether the night construction plan of the municipal engineering needs to be adjusted. Therefore, in the technical solution of the present application, the night construction plan adjustment judgment feature vector is optimized by feature modulation based on the entropy degradation maximization simulation value to obtain a sparse optimized night construction plan adjustment judgment feature vector.

[0043] Among them, the feature modulation optimization based on the entropy degradation maximization simulation value of the night construction plan adjustment judgment feature vector is performed to obtain the sparse optimized night construction plan adjustment judgment feature vector, including: extracting each row vector of the learning parameter matrix in the classifier as a set of night construction learning parameter row vectors; calculating the night construction entropy degradation maximization simulation value between the night construction plan adjustment judgment feature vector and each night construction learning parameter row vector in the set of night construction learning parameter row vectors to obtain a set of night construction entropy degradation maximization simulation values; based on the comparison between each night construction entropy degradation maximization simulation value in the set of night construction entropy degradation maximization simulation values ​​and a preset threshold, the learning parameter matrix is ​​gated and pruned to obtain a night construction learning parameter entropy degradation optimization matrix; the night construction plan adjustment judgment feature vector is matrix multiplied with the night construction learning parameter entropy degradation optimization matrix to obtain the sparse optimized night construction plan adjustment judgment feature vector.

[0044] The optimization steps are specifically expressed as follows: ;in, The set of row vectors of learning parameters for nightly construction, is a set operation, , , , are the first, second, and third row vectors of the nighttime construction learning parameters. , The eigenvalues ​​of the row vectors of nighttime construction learning parameters, The first value in the set of simulation values ​​for maximizing entropy degradation of nighttime construction The maximum simulation value of entropy degradation of nighttime construction, is the maximum value function, Represents the calculation vector and The Euclidean distance between Adjust the judgment feature vector for the night construction plan, For the The transposed vector of the row vector of nightly construction learning parameters, is the length of the feature vector, is matrix multiplication, , , , The first, second, and , learning parameter mask values, For mask operation, Indicates the preset threshold value, Learning parameter entropy degradation optimization matrix for nighttime construction, Adjust the judgment feature vector for sparse optimization of nighttime construction scheme.

[0045] In the technical solution of the present application, it is taken into consideration that in the process of inputting the night construction plan adjustment judgment feature vector into the learning parameter matrix of the classifier to obtain the classification result of whether the night construction plan of the municipal engineering project needs to be adjusted, there is a mismatch between the structure of the learning parameter matrix itself and the night construction plan adjustment judgment feature vector, that is, the learning parameter matrix has structural redundancy and information confusion relative to the night construction plan adjustment judgment feature vector, which may cause feature redundancy or noise in the classification result of whether the night construction plan of the municipal engineering project needs to be adjusted, resulting in model overfitting.

[0046] Based on this, in the technical solution of the present application, the feature modulation optimization based on the entropy degradation maximization simulation value is performed on the night construction plan adjustment judgment feature vector, which uses the night construction plan adjustment judgment feature vector as a wandering node, and uses the feature distribution between the night construction plan adjustment judgment feature vector and each row vector in the learning parameter matrix to simulate the wandering mode, and uses the distance topology between the night construction plan adjustment judgment feature vector and each row vector in the learning parameter matrix as a wandering map. In this way, the wandering result of the wandering node in a specific wandering mode on the wandering map is used to obtain the night construction entropy degradation maximization simulation value. Furthermore, based on the comparison between each night construction entropy degradation maximization simulation value in the set of the night construction entropy degradation maximization simulation value and a preset threshold, the part of the learning parameter matrix that does not contribute significantly to the input data representation and may even introduce noise or redundant information is screened out. Specifically, based on the comparison result, the learning parameter matrix is ​​gated and pruned to obtain the night construction learning parameter entropy degradation optimization matrix. Finally, the structurally optimized learning parameter matrix is ​​used as a feature modulation unit to linearly modulate the nighttime construction plan adjustment judgment feature vector to obtain the sparsely optimized nighttime construction plan adjustment judgment feature vector.

[0047] Furthermore, the complex problem of nighttime construction impact can be simplified into a clear classification task by using a classifier to process the sparse optimization nighttime construction plan adjustment judgment feature vector. For example, by training a machine learning model (such as a support vector machine, a decision tree, or a random forest), the impact of the nighttime construction plan can be divided into two categories: "needs adjustment" or "no adjustment". The classifier learns based on historical data and known feature vectors, and can quickly make judgments when a new feature vector is input. This efficient processing method not only saves time, but also reduces the bias of human decision-making. By obtaining the classification results, the impact of nighttime construction on the surrounding environment and citizens' lives can be intuitively understood. For example, if the classification result shows "needs adjustment", managers can consider taking measures such as optimizing construction time, improving construction equipment, or adding noise reduction measures to reduce the impact on citizens. This data-driven decision-making method not only improves the scientific nature of construction management, but also enhances public satisfaction and trust.

[0048] In summary, the embodiment of the present application first obtains construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by a database, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a classifier is used to determine whether the night construction plan of the municipal engineering project needs to be adjusted, thereby optimizing the construction plan to improve residents' satisfaction and reduce nuisance problems, thereby allowing the municipal engineering project to continue to advance without affecting the quality of life of surrounding residents.

[0049] As described above, the municipal engineering construction site intelligent management system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the municipal engineering construction site intelligent management system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the municipal engineering construction site intelligent management system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the municipal engineering construction site intelligent management system 100 can also be one of the many hardware modules of the terminal device.

[0050] Alternatively, in another example, the municipal engineering construction site intelligent management system 100 and the terminal device may also be separate devices, and the municipal engineering construction site intelligent management system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0051] Figure 5 Flow chart of the municipal engineering construction site intelligent management method according to the embodiment of the present application. Figure 5As shown, according to the intelligent management method of the municipal engineering construction site of the embodiment of the present application, it includes: S110, acquiring construction noise data at multiple predetermined time points collected by the sound sensor, construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; S120, extracting the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; S130, judging whether the municipal engineering night construction plan needs to be adjusted based on the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector.

[0052] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned municipal engineering construction site intelligent management method have been referred to above. Figures 1 to 4 The invention has been introduced in detail in the description of the intelligent management system for municipal engineering construction sites, and therefore, its repeated description will be omitted.

[0053] The phrase "based on" used in this specification does not mean "only based on" unless otherwise specified in other paragraphs. In other words, the phrase "based on" means both "only based on" and "at least based on".

[0054] Any reference to a unit using the names "first", "second", etc. used in this specification does not fully limit the number or order of these units. These names may be used in this specification as a convenient method to distinguish between two or more units. Therefore, a reference to a first unit and a second unit does not mean that only two units can be used or that the first unit must precede the second unit in some form.

[0055] The term "determining" used in this specification sometimes includes a variety of actions. For example, with respect to "determining", calculating, computing, processing, deriving, investigating, looking up (e.g., searching in a table, database, or other data structure), ascertaining, etc. can be regarded as performing "determining". In addition, with respect to "determining", receiving (e.g., receiving information), transmitting (e.g., transmitting information), input, output, accessing (e.g., accessing data in a memory), etc. can also be regarded as performing "determining". In addition, with respect to "determining", resolving, selecting, choosing, establishing, comparing, etc. can also be regarded as performing "determining". That is, with respect to "determining", several actions can be regarded as performing "determining".

[0056] The terms "connected", "coupled" or any variation thereof used in this specification refer to any direct or indirect connection or combination between two or more units, and may include the following situations: there are one or more intermediate units between two units that are "connected" or "coupled" to each other. The combination or connection between units can be physical, logical, or a combination of the two. For example, "connect" can also be replaced by "access". When used in this specification, two units can be considered to be "connected" or "coupled" to each other by using one or more wires, cables, and / or printed electrical connections, and as several non-limiting and non-exhaustive examples, by using electromagnetic energy with wavelengths in the radio frequency region, microwave region, and / or light (both visible and invisible light) region, etc.

[0057] The present invention has been described in detail above, but it is obvious to those skilled in the art that the present invention is not limited to the embodiments described in this specification. The present invention can be implemented as a modification and variation without departing from the purpose and scope of the present invention as determined by the description of the claims. Therefore, the description of this specification is for the purpose of illustration and does not have any restrictive meaning for the present invention.

Claims

1. A municipal engineering construction site intelligent management system, characterized in that: include: A municipal engineering construction information acquisition module is used to obtain construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by the database; A municipal engineering construction information processing module, used to extract construction multimodal information correlation feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; The night construction plan adjustment judgment module is used to judge whether the night construction plan of the municipal engineering project needs to be adjusted based on the construction multimodal information association feature vector and the text semantic feature vector of the surrounding citizen feedback opinions.

2. The municipal engineering construction site intelligent management system according to claim 1 is characterized in that: The municipal engineering construction information processing module includes: A construction noise data feature extraction unit, used for extracting features from the construction noise data collected by the sound sensor at a plurality of predetermined time points to obtain a construction noise feature vector; A construction vibration data feature extraction unit, used for extracting features from the construction vibration data collected by the vibration sensor to obtain a construction vibration time-frequency domain feature vector; A construction multimodal information feature association unit, used to associate the construction noise feature vector with the construction vibration time-frequency domain feature vector to obtain the construction multimodal information association feature vector; The feature extraction unit for the feedback from the surrounding citizens is used to extract features from the feedback form from the surrounding citizens collected from the database to obtain a semantic feature vector of the text of the feedback from the surrounding citizens.

3. The municipal engineering construction site intelligent management system according to claim 2 is characterized in that: The construction noise data feature extraction unit comprises: Converting the construction noise data collected by the sound sensor at a plurality of predetermined time points to obtain a construction noise input vector; The construction noise input vector is passed through a construction noise time series feature extractor to obtain the construction noise feature vector.

4. The municipal engineering construction site intelligent management system according to claim 3 is characterized in that: The construction vibration data feature extraction unit comprises: Performing wavelet transform on the construction vibration data collected by the vibration sensor to obtain a construction vibration time-frequency domain input vector; The construction vibration time-frequency domain input vector is passed through a construction vibration time-frequency domain feature encoder based on a convolutional neural network to obtain the construction vibration time-frequency domain feature vector.

5. The municipal engineering construction site intelligent management system according to claim 4 is characterized in that: The construction multimodal information feature association unit includes: Obtain construction noise feature vector and construction vibration time-frequency domain feature vector as input; Adjust the dimensions of the construction noise feature vector and the construction vibration time-frequency domain feature vector; Designing a filter mechanism, the filter mechanism comprising a matrix multiplication operation unit for calculating the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector; Using the matrix multiplication operation unit in the Filter mechanism, the product of the construction noise feature vector and the construction vibration time-frequency domain feature vector is calculated to obtain a construction multimodal information association feature vector; Output the construction multimodal information associated feature vector.

6. The municipal engineering construction site intelligent management system according to claim 5 is characterized in that: The surrounding citizen feedback feature extraction unit includes: A surrounding citizen feedback convolutional encoding subunit, used for passing the surrounding citizen feedback form collected from the database through a surrounding citizen feedback semantic encoder to obtain a plurality of surrounding citizen feedback text feature vectors; The feature splicing subunit of the feedback from the surrounding citizens is used to splice the multiple text feature vectors of the feedback from the surrounding citizens to obtain the text semantic feature vector of the feedback from the surrounding citizens.

7. The municipal engineering construction site intelligent management system according to claim 6 is characterized in that: The nighttime construction plan adjustment and judgment module includes: A nighttime construction feature fusion unit, used to fuse the construction multimodal information association feature vector and the surrounding citizen feedback text semantic feature vector to obtain a nighttime construction plan adjustment judgment feature vector; A nighttime construction feature optimization unit is used to perform feature modulation optimization on a nighttime construction plan adjustment judgment feature vector based on an entropy degradation maximization simulation value to obtain a sparsely optimized nighttime construction plan adjustment judgment feature vector; The municipal engineering nighttime construction plan adjustment judgment classification unit is used to pass the sparse optimized nighttime construction plan adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the municipal engineering nighttime construction plan needs to be adjusted.

8. The municipal engineering construction site intelligent management system according to claim 7 is characterized in that: The nighttime construction feature optimization unit comprises: Extracting each row vector of the learning parameter matrix in the classifier as a set of row vectors of nighttime construction learning parameters; Calculating the nighttime construction entropy degradation maximization simulation value between the nighttime construction scheme adjustment judgment feature vector and each nighttime construction learning parameter row vector in the set of nighttime construction learning parameter row vectors to obtain a set of nighttime construction entropy degradation maximization simulation values; Based on the comparison between each nighttime construction entropy degradation maximization simulation value in the set of nighttime construction entropy degradation maximization simulation values ​​and a preset threshold, the learning parameter matrix is ​​gated and pruned to obtain a nighttime construction learning parameter entropy degradation optimization matrix; The nighttime construction plan adjustment judgment feature vector is matrix-multiplied with the nighttime construction learning parameter entropy degradation optimization matrix to obtain the sparse optimized nighttime construction plan adjustment judgment feature vector.

9. A method for intelligent management of municipal engineering construction sites, characterized in that: include: Obtaining construction noise data at multiple predetermined time points collected by sound sensors, construction vibration data collected by vibration sensors, and feedback from surrounding citizens collected by a database; Extracting construction multimodal information association feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database; Based on the construction multimodal information association feature vector and the surrounding citizens' feedback text semantic feature vector, it is determined whether the municipal engineering night construction plan needs to be adjusted.

10. The municipal engineering construction site intelligent management method according to claim 9, characterized in that: Extracting construction multimodal information association feature vectors and surrounding citizen feedback text semantic feature vectors from the construction noise data at multiple predetermined time points collected by the sound sensor, the construction vibration data collected by the vibration sensor, and the surrounding citizen feedback form collected by the database, including: Performing feature extraction on the construction noise data collected by the sound sensor at a plurality of predetermined time points to obtain a construction noise feature vector; Performing feature extraction on the construction vibration data collected by the vibration sensor to obtain a construction vibration time-frequency domain feature vector; Associating the construction noise feature vector with the construction vibration time-frequency domain feature vector to obtain the construction multimodal information associated feature vector; Feature extraction is performed on the surrounding citizen feedback form collected from the database to obtain a semantic feature vector of the surrounding citizen feedback text.

Images