Distribution network engineering construction safety protection device and intelligent monitoring system thereof

Abstract

The invention relates to a distribution network engineering construction safety protection device and an intelligent monitoring system thereof, the distribution network engineering construction safety protection device is composed of an intelligent sensing device, an intelligent monitoring system and a visual monitoring platform, the intelligent sensing device is responsible for accurately collecting data of a distribution network construction site in real time; and the intelligent monitoring system performs data preprocessing and data compression on the collected data of the intelligent sensing device, performs PCA (Principal Component Analysis) on the data to form a parameter feature data set, performs data analysis by applying an attention mechanism added in an improved LSTM (Long Short Term Memory) algorithm, and performs identification, fault positioning and fault removal and recovery on the feature data set. Accurate prediction and intelligent decision making are realized through an artificial intelligence algorithm, and the safety protection efficiency is remarkably improved. Various complex environments are flexibly handled through the positioning module, the maintenance cost is reduced, and the life cycle of the intelligent monitoring system is prolonged.

Description

Technical Field

[0001] The present invention relates to the field of safety protection technology, and in particular to a distribution network engineering construction safety protection device and an intelligent monitoring system thereof. Background Art

[0002] In the rapid development of power distribution network construction, safety assurance has always been a core issue throughout the entire project cycle. Traditionally, safety protection devices in this field mainly rely on manual intensive monitoring, physical barrier isolation measures and basic alarm devices. Although they have ensured construction safety to a certain extent, their limitations are becoming increasingly prominent.

[0003] In recent years, although a comprehensive monitoring system combining video surveillance technology with basic environmental sensors has emerged, it has improved the real-time visibility and environmental parameter monitoring capabilities of the construction site to a certain extent. The system can intuitively display the construction dynamics and trigger an alarm immediately when the environmental parameters are abnormal, effectively preventing the occurrence of some safety accidents. However, with the increasing complexity of application requirements, the comprehensive monitoring system still shows its shortcomings: the current system focuses on intuitive monitoring and immediate response, lacks in-depth data mining and analysis capabilities, fails to achieve intelligent prediction and early warning based on historical data and real-time information, and fails to achieve automated intervention and adjustment strategies, limiting its preventive and proactive safety protection capabilities in complex engineering environments. The failure to achieve seamless integration and deep data fusion between the video surveillance and environmental monitoring modules has led to serious information islands, making it difficult to form a comprehensive and integrated assessment of the safety risks of the construction site, reducing management efficiency and decision-making accuracy. Faced with the diverse construction environment, changing operating conditions and ever-increasing safety management requirements of distribution network projects, the existing system lacks sufficient flexible adjustment and optimization configuration capabilities, and it is difficult to quickly adapt to changes in safety requirements in different scenarios.

[0004] In order to overcome the deficiencies of the prior art and improve the safety protection level in the field of distribution network engineering construction, the present invention provides a distribution network engineering construction safety protection device and an intelligent monitoring system thereof, which significantly improves the intelligence level of the distribution network engineering monitoring system. Summary of the invention

[0005] The present invention provides a distribution network engineering construction safety protection device and an intelligent monitoring system thereof, which are used to solve the technical problems that the existing distribution network engineering construction safety protection device has low intelligence and integration and cannot quickly locate faults and protect distribution network engineering.

[0006] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present invention is as follows: a distribution network engineering construction safety protection device, comprising: a single-chip microcomputer, an intelligent sensing device, an intelligent monitoring system and a visual monitoring platform; wherein the intelligent sensing device is responsible for real-time and accurate data collection of the distribution network construction site, the intelligent monitoring system performs data preprocessing and data compression on the collected data of the intelligent sensing device, performs PCA dimensionality reduction processing on the data to form a parameter feature data set, and uses the attention mechanism added to the improved LSTM algorithm for data analysis, identifies the feature data set, and completes the positioning of the power grid fault in combination with the positioning algorithm, and determines the fault type. By judging the fault type and controlling the relevant single-chip microcomputer through the improved rANS decoding algorithm, the fault removal and recovery are completed, and then a safety report is generated, and the safety report is displayed through the visual monitoring platform.

[0007] Furthermore, the intelligent sensing device includes a temperature and humidity sensor, a gas concentration sensor, a vibration sensor and an image recognition sensor. The temperature and humidity sensor collects the ambient temperature and humidity of the construction area, the gas concentration sensor collects the concentration of harmful gases in the construction area, the vibration sensor collects the vibration acceleration and vibration frequency of the construction equipment, and the image recognition sensor collects construction site image data including pixel matrix, color information RGB value, brightness, and contrast.

[0008] Furthermore, the intelligent monitoring system includes a data processing and analysis module, an intelligent decision-making and execution module, and a positioning module.

[0009] Furthermore, the data processing and analysis module performs data preprocessing and data compression on the collected data of the intelligent sensing device, and performs PCA dimensionality reduction processing on the data to form a parameter feature data set.

[0010] Furthermore, the data cleaning includes three processes: using the standard deviation method to remove outliers, improving the Z-score standardization to standardize the data, and using the minimum-maximum normalization to normalize the data.

[0011] Furthermore, the data compression uses rANS asymmetric number system entropy coding compression algorithm to compress the data.

[0012] Furthermore, the process of performing PCA on the data dimensionality reduction to form a parameter feature data set is as follows: (1) Calculate the covariance matrix: Decoded data , data matrix , whose covariance matrix ; (2) Calculate the eigenvalues ​​of the covariance matrix and the eigenvector ; (3) Select the eigenvectors corresponding to the first k largest eigenvalues , the feature vectors constitute a new feature space.

[0013] (4) Project the original data into the feature space to obtain the reduced-dimensional data .

[0014] Furthermore, the intelligent decision-making and execution module uses the attention mechanism added to the improved LSTM algorithm to perform data analysis, identify the feature data set, and completes the positioning of the power grid fault in combination with the positioning algorithm, thereby determining the fault type.

[0015] Furthermore, the positioning module completes the positioning of the power grid fault in combination with the positioning algorithm and determines the fault type.

[0016] Furthermore, the visual monitoring platform builds a mobile application interface by integrating high-definition video monitoring images and real-time data charts, realizes the functions of remote real-time monitoring, historical data query, and security report generation, and supports multi-terminal access.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention is highly intelligent and can integrate environmental monitoring, video surveillance, data analysis and emergency response. It can also achieve precise positioning and control through artificial intelligence algorithms, significantly improve the efficiency of safety protection, realize single-chip microcomputer control, quickly locate the fault location, and achieve fault removal and recovery.

[0018] (2) The present invention can flexibly respond to various complex environments through the positioning module, reduce maintenance costs, and improve the life cycle of the intelligent monitoring system. At the same time, through big data analysis to identify faults, it can achieve a shift from passive response to active prevention, effectively reducing the occurrence of safety accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a structural diagram of the safety protection device for the distribution network engineering construction of the present invention; Figure 2 This is a structural diagram of the intelligent monitoring system of the present invention; Figure 3 It is a schematic diagram of the structure of the single chip microcomputer of the present invention; Figure 4 This is a workflow diagram of the intelligent decision-making and execution module of the present invention. DETAILED DESCRIPTION

[0020] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0021] like Figure 1 As shown, the distribution network engineering construction safety protection device is composed of a single-chip microcomputer, an intelligent sensing device, an intelligent monitoring system and a visual monitoring platform. The intelligent sensing device is responsible for real-time and accurate data collection of the distribution network construction site. The intelligent monitoring system performs data preprocessing and data compression on the collected data of the intelligent sensing device, performs PCA dimensionality reduction processing on the data to form a parameter feature data set, and uses the improved LSTM algorithm to add the attention mechanism for data analysis, identifies the feature data set, and completes the positioning of the power grid fault in combination with the positioning algorithm, and determines the fault type. By judging the fault type and using the improved rANS decoding algorithm to control the single-chip microcomputer, the fault removal and recovery are completed, and then a safety report is generated. The safety report is displayed through the visual monitoring platform.

[0022] Specifically, the intelligent sensing device includes temperature and humidity sensors, gas concentration sensors, vibration sensors, and image recognition sensors. The temperature and humidity sensors monitor the ambient temperature and humidity of the construction area to prevent equipment failure or material damage caused by abnormal temperature and humidity. Indicates temperature using Indicates humidity.

[0023] Gas concentration sensors detect the concentration of toxic and harmful gases (such as methane and carbon monoxide) to ensure the safety of air quality in the construction area. Indicates the concentration of toxic and harmful gases.

[0024] Vibration sensors are installed on construction equipment to monitor their operating status and promptly detect and warn of potential mechanical failures. Vibration sensor data mainly includes vibration acceleration and vibration frequency .

[0025] Image recognition sensors (including cameras): Use AI technology to perform face recognition, behavior recognition, etc. to monitor whether personnel behavior complies with safety regulations, while recording construction site conditions for subsequent tracing.

[0026] Image data mainly includes pixel matrix , color information (RGB value) ,brightness , Contrast ,The above sensors transmit the collected data wirelessly or wired.

[0027] like Figure 2 As shown, specifically, the intelligent monitoring system includes a data processing and analysis module, an intelligent decision-making and execution module, and a positioning module.

[0028] The data processing and analysis module performs data preprocessing and data compression on the collected data of the intelligent sensing device, and performs PCA dimensionality reduction processing on the data to form a parameter feature data set.

[0029] Data cleaning includes three processes: using the standard deviation method to remove outliers, improving the Z-score standardization to standardize the data, and using the minimum-maximum normalization to normalize the data.

[0030] (1) Data cleaning algorithm: Use the standard deviation method to identify and process outliers. For each data point (Concentration of harmful gases , vibration acceleration , vibration frequency , pixel matrix , color information ,brightness , Contrast ) in is the normalized weight value.

[0031] Calculate the average value The difference , and with the standard deviation If the difference Exceeding the threshold (3 times the standard deviation), the data point is considered an outlier and is removed or replaced.

[0032] (2) Data standardization: Data from different sources and units are standardized so that they can be compared and analyzed on the same scale.

[0033] Method: Improve Z-score standardization,

[0034] Among them, x is the sensor data after data cleaning, μ is the mean of the data, and σ is the standard deviation of the data. After standardization, the mean of the data is 0 and the standard deviation is 1.

[0035] (3) Data normalization: Normalization converts data into dimensionless values ​​between 0 and 1, which facilitates processing and analysis by subsequent algorithms. Use minimum-maximum normalization.

[0036] Minimum-maximum normalization formula:

[0037] Among them, x is the sensor normalized data after data cleaning and standardization, x min and x max are the minimum and maximum values ​​in the data respectively. After normalization, the value range of the data is between 0 and 1.

[0038] Missing value filling: Use linear interpolation to fill in missing values , whose two non-missing values ​​are and ,and ,but .

[0039] Data compression: To reduce the space required for data transmission and storage while maintaining the original information of the data as much as possible, the present invention uses rANS (asymmetric number system) entropy coding compression algorithm to compress the data.

[0040] The algorithm flow is as follows: (1) Initial range setting: Usually a larger initial range is set. ,in is a sufficiently large integer.

[0041] (2) Symbol interval division: For each symbol ,in It is data that has been cleaned (data standardization, data normalization, missing value filling). According to its probability Divide the current range into several subranges, each corresponding to a symbol.

[0042] Assume the current range is ,symbol The subintervals of are calculated as follows: Subinterval starting point

[0043] Subinterval length

[0044] in, is the size of the current range, All probabilities are less than The sum of the probabilities of the symbols .

[0045] (3) Update scope: Encoding Symbols After that, update the current range to the sub-interval corresponding to the symbol.

[0046] Assume that the encoded symbol is , the new range will become ,in is the starting point of the subinterval, is the length of the subinterval.

[0047] The encoded symbol data , perform dimensionality reduction processing (PCA) to form a parameter feature data set: The specific calculation process of PCA involves covariance matrix, eigenvalues ​​and eigenvectors, but it is more complicated to give a direct formula. The main steps are as follows: (1) Calculate the covariance matrix: For the decoded data , data matrix , whose covariance matrix , (2) Calculate the eigenvalues ​​of the covariance matrix and the eigenvector .

[0048] (3) Select the eigenvectors corresponding to the first k largest eigenvalues (k is the target dimension to be reduced), these feature vectors constitute a new feature space.

[0049] (4) Project the original data into this new feature space to obtain the reduced-dimensional data .

[0050] The intelligent decision-making and execution module uses the attention mechanism added to the improved LSTM algorithm to perform data analysis, identify the feature data set, and complete the positioning of the power grid fault in combination with the positioning algorithm, thereby determining the fault type: like Figure 4 As shown in Figure 1, an attention mechanism is added to LSTM. Combining the attention mechanism with LSTM can further improve the performance when processing complex time series data. The attention mechanism can help the LSTM model pay more attention to historical data points related to the current prediction point during the prediction process, thereby improving the accuracy and stability of the prediction. The specific process is as follows: Calculate the attention weight , and then the attention weights are compared with the corresponding values Weighted sum, where the query and key The attention weight , by taking the attention score function Maps two vectors to a scalar , and then through the softmax operation to get:

[0051] Then soft attention: .

[0052] The intelligent decision-making and execution module automatically generates safety warning information based on the results of the data processing and analysis module, and activates the emergency response mechanism when necessary.

[0053] The positioning module combines the positioning algorithm to complete the positioning of the power grid fault and determine the fault type.

[0054] For the above feature dataset Identify and determine the type of fault .

[0055] The positioning algorithm is as follows: (1) Initialization: Let the feature dataset , transform the feature data into position coordinate data. Initialize the positioning coordinate error , and set the initial value of the iteration , define an empty matrix ; (2) Input: Coordinate data after recognition , sensor coordinate matrix The sparsity is K.

[0056] (3) Iteration process: When the iteration reaches Next time, A. Find the supporting index: ; B. Find the most relevant sensor coordinate index value and add it to the index set ; C. Update positioning error ; D. , return to A, when the iteration meets the control condition, the iteration ends: ; (4) Output: Support index set , positioning position coordinates ; (5) Map fault types to different positioning position coordinates .

[0057] For fault type Decodes and converts it back into range information , mainly through the following steps to restore the original symbol: (1) Read the code: Read the encoded binary number from the storage or transmission medium and convert it back into range information .

[0058] (2) Determine the symbol: Based on the current scope The probability distribution of the symbol and the symbol corresponding to the current range .

[0059] This usually involves comparing the range to the cumulative probability of the symbols.

[0060] (3) Scope of restoration: Assume that the decoded symbol is , according to the probability of the symbol and the current scope , calculate the new range after decoding .

[0061] and The calculation usually depends on , and The value of .

[0062] The corresponding formula is as follows: For character sets , can have any probability distribution (as long as the character set is arbitrarily long), but in practice it is limited to a quantitative range, generally taken Range. n is the codeword length. Within this range, the symbol The number of occurrences is , then we can get ≈ , Based on this quantification, the following formula is used to express the encoding of rANS:

[0063] Corresponding decoding formula:

[0064] in , which is a cumulative distribution statistical operation.

[0065] like Figure 3 As shown in the figure, the decoded data is input into the single-chip microcomputer to complete the fault removal and recovery, ensuring the normal operation of the power grid. The decoded data is used as a positive pulse signal through I / OP1.2. When the switch is enabled, it means that the fault location is successful. The positive pulse input suddenly changes, the switch is disconnected, and the fault removal and recovery are completed. The position difference is output as a negative pulse.

[0066] As a bridge between the system and users, the visual monitoring platform realizes the core functions of remote real-time monitoring, historical data query, and safety report generation through the Web / mobile application interface. It supports multi-terminal access and greatly enhances the management personnel's control over the safety status of the construction site.

[0067] Real-time monitoring is carried out by integrating high-definition video surveillance images and real-time data charts. The video surveillance images can capture the dynamics of the scene, while the real-time data charts can show the changing trends of key safety indicators such as temperature, humidity, gas concentration, etc.

[0068] Build a data warehouse based on big data processing platforms (such as Hadoop and Spark) to store massive historical data; develop an efficient data retrieval engine to support rapid response to complex query conditions.

[0069] Automatically generate detailed safety reports covering safety hazard analysis, emergency response records, safety trend forecasts, etc.

[0070] Use cluster analysis to conduct in-depth analysis of historical data; combine natural language processing technology (such as TextRank, BERT) to generate structured report content; finally, use template engines (such as Freemarker, Thymeleaf) to fill the analysis results into the preset report template to generate the final security report.

[0071] Self-learning: Supervised learning algorithms are used for classification and regression tasks, unsupervised learning algorithms are used for cluster analysis, and deep learning algorithms can handle complex nonlinear relationships. These algorithms work together in the system's self-optimization process, enabling the system to continuously improve its intelligence and automation level.

[0072] The present invention constructs an intelligent safety protection device and system to achieve comprehensive perception and intelligent analysis of the environment, equipment and personnel behavior, and improve the accuracy and efficiency of safety protection. At the same time, through the positioning module, the system's adaptability to different distribution network construction environments is enhanced.

[0073] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A distribution network engineering construction safety protection device, characterized in that: include: An intelligent sensing device, an intelligent monitoring system and a visual monitoring platform; wherein the intelligent sensing device is responsible for collecting data from the distribution network construction site in real time and accurately, the intelligent monitoring system performs data preprocessing and data compression on the collected data of the intelligent sensing device, performs PCA dimensionality reduction processing on the data to form a parameter feature data set, and uses the attention mechanism added to the improved LSTM algorithm for data analysis, identifies the feature data set, and completes the positioning of the power grid fault in combination with the positioning algorithm, and determines the fault type. By judging the fault type and controlling the relevant single-chip microcomputer through the improved rANS decoding algorithm, the fault removal and recovery are completed, and then a safety report is generated, which is displayed through the visual monitoring platform.

2. A distribution network engineering construction safety protection device according to claim 1, characterized in that: The intelligent sensing device includes a temperature and humidity sensor, a gas concentration sensor, a vibration sensor and an image recognition sensor. The temperature and humidity sensor collects the ambient temperature and humidity of the construction area, the gas concentration sensor collects the concentration of harmful gases in the construction area, the vibration sensor collects the vibration acceleration and vibration frequency of the construction equipment, and the image recognition sensor collects construction site image data including pixel matrix, color information RGB value, brightness, and contrast.

3. A distribution network engineering construction safety protection device according to claim 1, characterized in that: The intelligent monitoring system includes a data processing and analysis module, an intelligent decision-making and execution module, and a positioning module.

4. A distribution network engineering construction safety protection device according to claim 3, characterized in that: The data processing and analysis module performs data preprocessing and data compression on the collected data of the intelligent sensing device, and performs PCA dimensionality reduction processing on the data to form a parameter feature data set.

5. A distribution network engineering construction safety protection device according to claim 4, characterized in that: The data cleaning includes three processes: using the standard deviation method to remove outliers, improving the Z-score standardization to standardize the data, and using the minimum-maximum normalization to normalize the data.

6. A distribution network engineering construction safety protection device according to claim 4, characterized in that: The data compression uses rANS asymmetric number system entropy coding compression algorithm to compress the data.

7. A distribution network engineering construction safety protection device according to claim 4, characterized in that: The process of PCA to form a parameter feature data set by reducing the data dimension is as follows: (1) Calculate the covariance matrix: Decoded data , data matrix , whose covariance matrix ; (2) Calculate the eigenvalues ​​of the covariance matrix and the eigenvector ; (3) Select the eigenvectors corresponding to the first k largest eigenvalues , the feature vectors constitute a new feature space; (4) Project the original data into the feature space to obtain the reduced-dimensional data .

8. A distribution network engineering construction safety protection device according to claim 3, characterized in that: The intelligent decision-making and execution module uses the attention mechanism added to the improved LSTM algorithm to perform data analysis, identify the feature data set, and complete the positioning of the power grid fault in combination with the positioning algorithm, thereby determining the fault type.

9. A distribution network engineering construction safety protection device according to claim 3, characterized in that: The positioning module combines the positioning algorithm to complete the positioning of the power grid fault and determine the fault type.

10. A distribution network engineering construction safety protection device according to claim 1, characterized in that: The visual monitoring platform builds a mobile application interface by integrating high-definition video monitoring images and real-time data charts, realizes the functions of remote real-time monitoring, historical data query, and security report generation, and supports multi-terminal access.

Images