Mountain wind power plant blasting construction risk prediction system and method based on multi-source analysis

Abstract

The invention provides a mountain wind power plant blasting construction risk prediction system and method based on multi-source analysis. Key information related to mountain wind power plant blasting construction risks is extracted from multi-source heterogeneous data of a mountain wind power plant blasting construction site; extracting spatial and temporal distribution features, environmental influence features, construction parameter association features and risk inducement features of the blasting construction risk of the mountain wind power plant from the key information, and determining potential association relationships among the features; determining the dynamic influence weight of each feature on the blasting construction risk of the mountain wind power plant; and determining the risk grade distribution of the whole process of the blasting construction of the mountain wind power plant according to the potential association relationship and all the dynamic influence weights, and generating risk early warning information of the blasting construction of the mountain wind power plant based on the risk grade distribution. By adopting the scheme of the invention, the blasting construction risk of the mountain wind power plant can be dynamically predicted based on the incidence relation among the multi-source data in the blasting construction of the mountain wind power plant.

Description

Technical Field

[0001] This application relates to the field of risk prediction technology, and more specifically, to a risk prediction system and method for blasting construction in mountainous wind farms based on multi-source analysis. Background Technology

[0002] Risk prediction refers to a technical method that uses historical data, real-time dynamic data, and multi-dimensional feature information in a specific scenario to quantitatively predict and qualitatively assess the probability of occurrence, scope of impact, severity, and evolution trend of potential risk events in the future, through statistical analysis, machine learning, deep learning, and other algorithmic models.

[0003] As new energy development extends to mountainous areas, mountain wind farms have become an important development scenario due to their open terrain and stable wind speeds. Blasting is a core procedure in site leveling and foundation excavation. Mountainous terrain is rugged, with complex and variable geological conditions, coupled with fluctuating weather conditions, making blasting operations susceptible to multiple risks such as falling rocks, excessive vibration, and slope instability. Improper management can easily lead to safety accidents or environmental damage. Currently, blasting risk assessments largely rely on single monitoring data or empirical models, with data sources limited to construction parameters or static geological data, lacking a comprehensive consideration of dynamic factors such as the environment and terrain. The current method uses fixed weights to calculate the degree of risk impact, which is difficult to adapt to dynamic scenarios such as sudden changes in geological structure and short-term weather changes in mountainous environments. This results in incomplete risk identification and insufficient prediction accuracy. At the same time, the existing technology does not fully explore the intrinsic correlation between various risk influencing factors, making it difficult to achieve accurate risk prediction throughout the entire construction process. This often leads to delayed warnings or misjudgments, and fails to provide accurate risk management basis for on-site construction. Therefore, how to dynamically predict the risks of blasting construction in mountainous wind farms based on the correlation between multi-source data has become a problem faced by the industry. Summary of the Invention

[0004] This application provides a risk prediction system and method for blasting construction in mountainous wind farms based on multi-source analysis, which can dynamically predict the risk of blasting construction in mountainous wind farms based on the correlation between multi-source data in blasting construction.

[0005] Firstly, this application provides a method for predicting the risk of blasting construction in mountainous wind farms based on multi-source analysis, comprising the following steps: Collect multi-source heterogeneous data from blasting construction sites in mountainous wind farms; The multi-source heterogeneous data are correlated and fused to extract key information related to the blasting construction risks of mountain wind farms; From the key information, the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics of blasting construction risks in mountain wind farms are extracted, and the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk trigger characteristics is determined. Based on the terrain features and dynamic environmental changes at the mountain wind farm blasting construction site, the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm. Based on the potential correlations and all dynamic influence weights, risk prediction is performed on the entire process of blasting construction in mountainous wind farms to obtain the risk level distribution of the entire process of blasting construction in mountainous wind farms. Based on the risk level distribution, risk warning information for blasting construction in mountainous wind farms is generated.

[0006] In some embodiments, the multi-source heterogeneous data includes topographic elevation and geological structure data, real-time meteorological monitoring data, performance parameters of blasting materials, drilling and charging operation data, operating status data of construction equipment, distribution data of sensitive targets such as surrounding residential areas, roads, power transmission lines, and ecological protection areas, historical records of blasting risk accidents, and emergency resource allocation data.

[0007] In some embodiments, the correlation and fusion of the multi-source heterogeneous data to extract key information related to the risks of blasting construction in mountain wind farms specifically includes: The multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data; Based on the correlation features between the data in the preprocessed multi-source heterogeneous data, the multi-source heterogeneous data is correlated and fused to obtain correlated and fused data. To obtain risk characteristics related to blasting construction risks in mountainous wind farms; Based on the aforementioned risk characteristics, key information related to the blasting construction risks of mountain wind farms is extracted from the associated fusion data.

[0008] In some embodiments, extracting the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk precipitating characteristics of blasting construction risks in mountain wind farms from the key information specifically includes: From the aforementioned key information, we can extract the spatiotemporal distribution information, environmental impact information, construction parameter information, and risk precipitating information of blasting construction risks in mountain wind farms. Based on the aforementioned spatiotemporal distribution information, the spatiotemporal distribution characteristics of blasting construction risks in mountain wind farms are determined; Based on the aforementioned environmental impact information, determine the environmental impact characteristics of the blasting construction risks at the mountain wind farm; Based on the aforementioned construction parameter information, determine the construction parameter correlation characteristics of the blasting construction risk in mountain wind farms; Based on the aforementioned risk factor information, the risk factor characteristics of blasting construction risks in mountain wind farms are determined.

[0009] In some embodiments, determining the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk trigger characteristics specifically includes: The spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics are standardized to obtain standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics. Determine the dependencies and correlations among the standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk factor characteristics; Based on the dependencies and associations, potential associations are constructed among the spatiotemporal distribution features, environmental impact features, construction parameter association features, and risk trigger features.

[0010] In some embodiments, based on the terrain features and dynamic environmental changes at the mountain wind farm blasting construction site, the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk trigger features on the risk of mountain wind farm blasting construction are dynamically adjusted. Specifically, the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk trigger features on the risk of mountain wind farm blasting construction include: Obtain risk characterization indicators for blasting construction in mountainous wind farms; Based on the risk characterization index, determine the influence weights of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics on the blasting construction risk of mountain wind farms, respectively. Obtain historical risk case data for blasting construction in mountainous wind farms, as well as the terrain features and dynamic environmental changes at the blasting construction sites in mountainous wind farms; Construct a terrain-environment dynamic adjustment vector based on the terrain features and the dynamic environmental change features; Based on the historical risk case data and the terrain-environment dynamic adjustment vector, the initial influence weights are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution characteristics, the environmental influence characteristics, the construction parameter correlation characteristics, and the risk inducement characteristics on the blasting construction risk of mountain wind farms.

[0011] In some embodiments, risk prediction is performed on the entire process of blasting construction for mountain wind farms based on the potential correlations and all dynamic influence weights, resulting in a risk level distribution for the entire process of blasting construction for mountain wind farms, specifically including: Identify all risk prediction units for the entire blasting construction process in mountainous wind farms; Based on the potential correlations and all dynamic influence weights, a risk assessment is conducted on the blasting construction of each risk prediction unit to obtain the risk assessment value of each risk prediction unit. Convert the risk assessment values ​​of each risk prediction unit into risk levels; The risk level distribution of the entire blasting construction process for mountain wind farms is determined based on all risk levels.

[0012] Secondly, this application provides a risk prediction system for blasting construction in mountainous wind farms based on multi-source analysis, including: The data acquisition module is used to collect multi-source heterogeneous data from blasting construction sites in mountainous wind farms. The processing module is used to correlate and fuse the multi-source heterogeneous data and extract key information related to the blasting construction risks of mountain wind farms; The processing module is also used to extract the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics and risk inducement characteristics of the blasting construction risk of the mountain wind farm from the key information, and to determine the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics and the risk inducement characteristics; The processing module is further configured to dynamically adjust the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the risk of blasting construction in mountainous wind farms based on the terrain features and dynamic environmental change features of the blasting construction site, thereby obtaining the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the risk of blasting construction in mountainous wind farms. The execution module is used to predict the risks of the entire process of blasting construction in mountainous wind farms based on the potential correlations and all dynamic influence weights, obtain the risk level distribution of the entire process of blasting construction in mountainous wind farms, and generate risk warning information for blasting construction in mountainous wind farms based on the risk level distribution.

[0013] Thirdly, this application provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting the risk of blasting construction in mountainous wind farms based on multi-source analysis.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The multi-source analysis-based risk prediction system and method for blasting construction in mountainous wind farms provided in this application avoids prediction bias caused by a single data dimension by collecting multi-source heterogeneous data; it correlates and fuses multi-source heterogeneous data and extracts key risk-related information, solving the problems of heterogeneous data formats and information redundancy, and accurately screening out core risk information; it extracts four types of risk features and clarifies their potential correlations, clearly outlining the risk impact logic behind the multi-source data, and locking in the core features and interactions affecting blasting construction risks; it adjusts the feature influence weights in conjunction with terrain features and dynamic environmental changes, breaking the limitations of fixed weights, making the weights under multi-source data correlation adaptable to the real-time on-site environment, and significantly improving the dynamic adaptability of risk prediction; it predicts risks throughout the entire process based on potential feature correlations and dynamic weights and generates early warning information, deeply integrating multi-source data correlations with dynamic weights, directly realizing the dynamic prediction of blasting construction risks in mountainous wind farms, and outputting risk level distribution and early warning information, making the dynamic prediction results applicable to practical guidance. By adopting the above scheme, the risk of blasting construction in mountainous wind farms can be dynamically predicted based on the correlation between multiple sources of data in the blasting construction. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis, according to some embodiments of this application. Figure 2 This is an exemplary flowchart illustrating the determination of potential relationships according to some embodiments of this application; Figure 3 This is an exemplary flowchart illustrating the determination of risk level distribution according to some embodiments of this application; Figure 4 This is a structural schematic diagram of a multi-source analysis-based risk prediction system for blasting construction in mountainous wind farms, as shown in some embodiments of this application. Figure 5 This is a schematic diagram of the structure of a computer device for implementing a method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis, according to some embodiments of this application. Detailed Implementation

[0017] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] refer to Figure 1 The figure is an exemplary flowchart of a method for predicting the risk of blasting construction in mountainous wind farms based on multi-source analysis, according to some embodiments of this application. The method for predicting the risk of blasting construction in mountainous wind farms based on multi-source analysis mainly includes the following steps: In step 101, multi-source heterogeneous data of the blasting construction site of the mountain wind farm are collected.

[0019] It should be noted that the multi-source heterogeneous data of the mountain wind farm blasting construction site in this application represents various data sets with different acquisition channels, data formats and structures. It reflects the multi-dimensional actual situation related to blasting construction safety, such as the terrain conditions, environmental dynamics, key parameters of the construction process, equipment operating status, distribution of surrounding sensitive targets, and historical risk events in the blasting construction area. The multi-source heterogeneous data includes topographic elevation and geological structure data, real-time meteorological monitoring data, performance parameters of blasting materials, drilling and charging operation data, operating status data of construction equipment, distribution data of sensitive targets such as surrounding residential areas / roads / transmission lines / ecological protection areas, historical blasting risk accident records and emergency resource allocation data.

[0020] In step 102, the multi-source heterogeneous data is correlated and fused to extract key information related to the blasting construction risks of mountain wind farms.

[0021] In some embodiments, the following steps can be used to correlate and fuse the multi-source heterogeneous data to extract key information related to the blasting construction risks of mountain wind farms: The multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data; Based on the correlation features between the data in the preprocessed multi-source heterogeneous data, the multi-source heterogeneous data is correlated and fused to obtain correlated and fused data. To obtain risk characteristics related to blasting construction risks in mountainous wind farms; Based on the aforementioned risk characteristics, key information related to the blasting construction risks of mountain wind farms is extracted from the associated fusion data.

[0022] In specific implementation, the multi-source heterogeneous data is preprocessed to obtain the preprocessed multi-source heterogeneous data. This can be achieved in the following way: for the multi-source heterogeneous data, the Grubbs criterion is used to remove outliers exceeding 3 times the standard deviation, missing data is filled with the median of the time series of the same type of data, and then the Min-Max normalization method is used to uniformly map data of different dimensions to the [0,1] interval to eliminate the difference in dimensions, thus obtaining the preprocessed multi-source heterogeneous data. Other methods can also be used in other embodiments, which are not limited here.

[0023] Furthermore, in specific implementation, the multi-source heterogeneous data is correlated and fused based on the correlation characteristics between various data in the preprocessed multi-source heterogeneous data to obtain correlated and fused data. This can be achieved in the following way: Based on the basic attributes of the preprocessed multi-source heterogeneous data—time stamp, spatial coordinates, representational meaning, and numerical distribution—a time correlation window is determined according to the standard operation time of each process in the blasting construction of a mountain wind farm: drilling, charging, detonation, and post-blast treatment. Specifically, the corresponding time correlation window is set at 1.1-1.2 times the standard operation time, ensuring that the window completely covers the entire time period from start to finish of the process. The data acquisition cycle is defined using GIS spatial coordinates with a 1-meter accuracy standard to delineate the spatial association range. Combined with the actual logic of blasting construction: terrain conditions directly affect the selection of charge parameters, meteorological conditions relate to blasting safety control requirements, and equipment operating status corresponds to construction quality parameters. Semantic association rules are established, and potential linear associations are initially determined by comparing numerical change trends in structured data from multiple sources and heterogeneous datasets. This determines the association characteristics of temporal synchronization associations, spatial correspondence associations, semantic logical associations, and numerical correlation associations between data. Based on these association characteristics, association fusion is carried out, initially using timestamps and spatial coordinates as dual core benchmarks to integrate multiple... In heterogeneous data sources, topographic data, charge data, meteorological data, equipment operation data, and sensitive target data within the same time window and spatial range are initially matched and grouped. For data within each group, the Pearson correlation coefficient is used to calculate the linear correlation strength pairwise, setting a correlation coefficient absolute value ≥ 0.6 as the threshold for strong correlation, and retaining strongly correlated data pairs. For cross-type correlated data, feature mapping is first used to convert key information from unstructured / semi-structured data in multi-source heterogeneous data into quantifiable indicators. For example, topographic image elevation differences are converted into specific elevation differences, and fracture zone distribution characteristics are converted into fracture... The straight-line distance between the debris and the blast hole, and the equipment failure early warning information are converted into quantitative values ​​of the failure impact level. A one-to-one correspondence is established with the structured data fields. Then, the credibility weight is determined according to the data acquisition method. The weight of data collected by professional monitoring sensors that have been calibrated by metrology is set to 0.7, and the weight of data obtained by estimation based on empirical formulas or indirect observation is set to 0.3. The relevant data in the same association group are weighted and averaged according to the weights. Finally, the association fusion data containing multi-dimensional association logic, unified data format, and concentrated core information is formed. Other methods can be used for fusion in other embodiments, which are not limited here.

[0024] In addition, in specific implementation, the risk characteristics related to blasting construction risks in mountain wind farms can be obtained in the following way: by consulting the industry standard "Safety Regulations for Blasting" (GB6722) and sorting out historical risk cases of blasting construction in mountain wind farms, the specific indicators of risk characteristics can be identified. Risk characteristics include blasting vibration related characteristics, flyrock related characteristics, slope stability related characteristics, environmental impact related characteristics, and construction parameter compliance characteristics. Other methods can also be used in other embodiments, which will not be elaborated here.

[0025] In addition, in specific implementation, the key information related to the blasting construction risk of mountain wind farms can be extracted from the associated fusion data based on the risk characteristics in the following way: establish a field mapping relationship between risk characteristics and associated fusion data, such as peak particle velocity corresponding to vibration sensor monitoring data, and slope angle corresponding to slope data extracted from topographic images. Set safety thresholds for each risk characteristic according to industry standards, such as a safety threshold of 2.0 cm / s for peak particle velocity of blasting vibration. Retain data that exceeds the threshold or is in the critical range. At the same time, use Spearman rank correlation analysis to screen data items that are significantly related to the risk characteristics, with a significance level of α=0.05. Data greater than the significance level are considered significantly related. Remove redundant data that is not related to the risk characteristics. Finally, organize the key information into risk characteristic indicators, corresponding data values, sources of associated data, and degree of risk correlation. Other extraction methods can also be used in other embodiments, which are not limited here.

[0026] It should be noted that the correlation features in this application represent the degree of correlation between multi-source heterogeneous data in the blasting construction of mountain wind farms, reflecting the interaction and adaptation relationship of data from different sources and of different types in the construction scenario; the correlation fusion data represents the data set formed after matching, filtering, and quantitative integration of multi-source heterogeneous data, reflecting the comprehensive actual situation of the blasting construction area in terms of terrain conditions, construction operations, environmental conditions, equipment operation, and distribution of sensitive targets; the risk features represent specific indicators directly related to the risks of blasting construction, reflecting the key dimensions and judgment criteria that may cause blasting safety risks; and the key information represents the core data content extracted from multi-source heterogeneous data that is highly related to the risk features, reflecting the core causes, influencing factors, and quantitative representations directly related to the risks of blasting construction in mountain wind farms.

[0027] In step 103, the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics of the blasting construction risk of the mountain wind farm are extracted from the key information, and the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk inducement characteristics is determined.

[0028] In some embodiments, the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk precipitation characteristics of blasting construction risks in mountain wind farms can be extracted from the key information using the following steps: From the aforementioned key information, we can extract the spatiotemporal distribution information, environmental impact information, construction parameter information, and risk precipitating information of blasting construction risks in mountain wind farms. Based on the aforementioned spatiotemporal distribution information, the spatiotemporal distribution characteristics of blasting construction risks in mountain wind farms are determined; Based on the aforementioned environmental impact information, determine the environmental impact characteristics of the blasting construction risks at the mountain wind farm; Based on the aforementioned construction parameter information, determine the construction parameter correlation characteristics of the blasting construction risk in mountain wind farms; Based on the aforementioned risk factor information, the risk factor characteristics of blasting construction risks in mountain wind farms are determined.

[0029] In specific implementation, the extraction of spatiotemporal distribution information, environmental impact information, construction parameter information, and risk trigger information of blasting construction risks in mountain wind farms from the aforementioned key information can be achieved in the following way: The key information is categorized and sorted by field, and the spatiotemporal distribution information carrying timestamps, spatial coordinates, and risk-related data items is extracted. This spatiotemporal distribution information includes the construction period of each process, the time point of occurrence of risk events, the spatial location of the blast hole group, the radiation range of sensitive targets, and the latitude, longitude, and elevation information of risk data collection. Environmental impact information is extracted from meteorological monitoring data, geological condition data, and the distribution and impact data of sensitive targets. Construction parameter information is extracted from drilling parameters, charge parameters, detonation parameters, and equipment operating parameters. Risk trigger information is extracted from historical risk accident records, equipment failure logs, operational violation records, and environmental mutation data. Other extraction methods can also be used in other embodiments, which are not limited here.

[0030] It should be noted that the spatiotemporal distribution information in this application represents a set of various data related to time and space throughout the entire process of blasting construction in mountain wind farms; the environmental impact information represents a set of external environmental data related to blasting construction; the construction parameter information represents a set of various operation and equipment-related data generated during the blasting construction process; and the risk inducement information represents a set of various source data directly related to the occurrence of blasting construction risks.

[0031] In addition, in specific implementation, the spatiotemporal distribution characteristics of the risk of blasting construction in mountain wind farms can be determined based on the spatiotemporal distribution information in the following way: Subsequently, based on the spatiotemporal distribution information, the blasting construction area of ​​the mountain wind farm is divided into grids with a precision of 5 meters × 5 meters. The frequency and density of risk data in each grid are counted. At the same time, the construction process is segmented according to the time window, that is, each construction process time window is a time period. The distribution pattern of risk data in each time period is counted to determine the spatial clustering area and the high-incidence period of risk, forming a quantitative spatiotemporal distribution characteristic. Other methods can also be used to determine this in other embodiments, which are not limited here.

[0032] In addition, in specific implementation, the environmental impact characteristics of the blasting construction risk in mountain wind farms can be determined based on the environmental impact information in the following way: Spearman's rank correlation analysis method is used to calculate the correlation strength between each data point in the environmental impact information and the risk data. A significance level of α=0.05 is used as the criterion to screen out core environmental data with a significant correlation where the P-value is less than α. Then, referring to the "Safety Regulations for Blasting" (GB6722) and relevant standards for environmental assessment of mountain wind power construction, and combining engineering practice experience, the impact threshold of each core environmental data point is clarified, and they are divided into three levels: slight, moderate, and severe, according to their degree of impact on the blasting construction risk. Finally, by statistically analyzing the probability of risk occurrence and the degree of impact corresponding to each level of environmental data in historical construction cases, the risk contribution of each level is quantified. For example, the contribution of slight impact level is 20%, moderate impact level is 50%, and severe impact level is 80%. The core environmental data, the level classification standard, and the corresponding risk contribution are integrated to form standardized environmental impact characteristics. Other methods can also be used in other embodiments, which are not limited here.

[0033] In addition, in specific implementation, the correlation characteristics of construction parameters for determining the risk of blasting construction in mountain wind farms based on the aforementioned construction parameter information can be achieved in the following way: The construction parameter information is divided into structured continuous parameters and discontinuous discrete parameters; then, for the combination of structured continuous parameters, the Pearson correlation coefficient formula is used to calculate the synergistic correlation strength; for the combination of discontinuous discrete parameters, a level value is first assigned according to the degree of risk impact, and then the Spearman rank correlation coefficient is used to calculate the correlation strength. Strongly correlated parameter pairs are selected based on the absolute value of the correlation coefficient ≥ 0.6; next, a scatter plot of the strongly correlated parameter combinations is plotted and a linear / nonlinear trend line is fitted, and the parameter combinations are analyzed in relation to risk characterization data such as peak blasting vibration, maximum distance of flyrock, and slope displacement. The study identifies the changing patterns of parameters and clarifies their impact on risk trends. Referring to the "Safety Regulations for Blasting" (GB6722) and the special plan for blasting construction in mountainous wind power areas, and combining historical safe construction cases, the study statistically analyzes the probability of risk occurrence corresponding to different parameter values ​​to determine reasonable threshold ranges for each key construction parameter. Subsequently, multiple parameter combinations are verified through on-site test blasts or numerical simulations of blasting. With safety and compliance as the core, while also considering construction quality and efficiency, the optimal parameter combination pattern for risk characterization is selected. Finally, the study integrates strongly correlated parameter pairs, the influence patterns of parameter combinations, the threshold ranges of key parameters, and the optimal combination pattern to form standardized construction parameter correlation characteristics for blasting construction risks in mountainous wind farms. Other methods can be used to determine these characteristics in other embodiments, which are not limited here.

[0034] Furthermore, in specific implementation, determining the risk causal characteristics of blasting construction risks in mountain wind farms based on the aforementioned risk causal information can be achieved in the following way: The risk causal information is categorized into five types according to causal type: sudden environmental changes, equipment failures, operational violations, unreasonable parameter settings, and incompatible geological conditions. Subsequently, a weighted statistical method is used to quantify the impact degree of each type of causal factor. The comprehensive impact value is calculated using the impact weight and the occurrence frequency weight. Combined with the "Blasting Safety Regulations" (GB6722) and the risk assessment standards for mountain wind power construction, a comprehensive impact value ≥0.7 is defined as a high-risk causal factor, and 0.4- A grading standard of 0.7 for medium-risk triggers and ≤0.4 for low-risk triggers was used to screen out high-frequency, high-impact core triggers. Then, by reviewing historical risk accident cases, the specific triggering conditions and impact range of each core trigger were clarified, and the correlation paths between the triggers and risk characterization data such as peak blasting vibration, flyrock distance, and slope instability were analyzed. Finally, the core trigger types, grading results, triggering conditions, frequency of occurrence, impact range, and risk correlation paths were integrated to form standardized and quantifiable risk trigger characteristics for blasting construction in mountain wind farms. Other methods can be used to determine these triggers in other embodiments, which are not limited here.

[0035] It should be noted that the spatiotemporal distribution characteristics in this application represent a quantitative characterization of the distribution patterns of blasting construction risks in mountain wind farms across time and space, reflecting the concentration and variation patterns of risks at different process stages and in different spatial regions; the environmental impact characteristics represent the degree of influence of environmental data on the blasting construction risks of mountain wind farms, reflecting the aggravating or mitigating effects of different environmental conditions on risks; the construction parameter correlation characteristics represent the characteristics of the correlation between construction parameters, reflecting the compliance boundaries of construction parameters, the adaptation logic between parameters, and the direct impact path on risks; and the risk trigger characteristics represent the characteristics of the high-frequency, high-impact core sources that trigger blasting construction risks, reflecting the root causes, triggering mechanisms, and transmission logic of risks, providing source guidance for targeted risk prevention and control.

[0036] In some embodiments, reference Figure 2 As shown, this figure is an exemplary flowchart for determining potential correlations in some embodiments of this application. In this embodiment, determining the potential correlations among the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk trigger characteristics can be achieved by the following steps: In step 1031, the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk cause characteristics are standardized to obtain standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk cause characteristics. In step 1032, the dependencies and correlations among the standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk factor characteristics are determined. In step 1033, potential correlations are constructed between the spatiotemporal distribution features, the environmental impact features, the construction parameter correlation features, and the risk trigger features based on the dependencies and the associations.

[0037] In specific implementation, the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk factor features are standardized to obtain standardized spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk factor features. This can be achieved in the following way: Standardize the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk factor features; map all numerical features to the [0,1] interval using the Min-Max normalization method; convert all categorical features to binary values ​​using one-hot encoding, for example, encoding high slope clusters as 100 and gentle slopes as 010; and use the Grubbs criterion to remove outliers that still exist after standardization to ensure the consistency and reliability of the feature data. This yields standardized spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk factor features. Other methods can also be used in other embodiments, which are not limited here.

[0038] In addition, in specific implementation, the dependencies and correlations between the standardized spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk factor features can be determined in the following way: The dependencies and correlations between features are determined in two steps. First, the correlation strength is calculated. For linear correlations between numerical features, the Pearson correlation coefficient is used; for correlations between categorical features and numerical features, the Spearman rank correlation coefficient is used; for potential nonlinear correlations, mutual information entropy is used. Strongly correlated feature pairs are selected based on a correlation coefficient absolute value ≥ 0.5 and a mutual information entropy ≥ 0.3, indicating a significant correlation. Second, false correlations are eliminated. Considering the blasting construction logic of mountain wind farms, for example, if non-construction area features have no substantial correlation with construction parameters, the business rationality of strongly correlated sub-feature pairs is verified, and false correlations are eliminated. Then, based on construction sequence and causal logic, the dependency direction of effective strongly correlated sub-feature pairs is clarified: if A is a prerequisite for B, it is a one-way dependency; if they influence each other, it is a two-way dependency. In other embodiments, other methods can also be used to determine the dependencies, which are not limited here.

[0039] In addition, in specific implementation, the potential correlation between the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk trigger features based on the dependency and correlation relationships can be achieved in the following way: using specific sub-features of the four types of features as nodes, the correlation strength value in the correlation relationship as the weight of the edge, and the dependency relationship as the direction of the edge, a correlation relationship model is constructed in the form of a directed weighted graph. At the same time, the construction scenario explanations of each correlation are labeled. For example, the scenario corresponding to the high slope spatial cluster + rainstorm environmental level to vibration exceeding the standard trigger is that blasting after rainstorm in the high slope area is prone to vibration exceeding the standard. Integrating nodes, weights, directions, and scenario explanations, a visualized and quantifiable potential correlation relationship of the four types of features is formed. Other methods can also be used to construct it in other embodiments, which are not limited here.

[0040] It should be noted that the correlation relationship in this application represents the relationship of correlation strength among the four types of characteristics: spatiotemporal distribution, environmental impact, construction parameter correlation, and risk inducement, reflecting the correlation strength among the sub-characteristics; the dependency relationship represents the logical dependence among the four types of characteristics: spatiotemporal distribution, environmental impact, construction parameter correlation, and risk inducement; the potential correlation relationship represents the comprehensive overall relationship of mutual correlation and interdependence among the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics, reflecting the overall picture of comprehensive correlation and dependence among the sub-characteristics of the four types of characteristics.

[0041] In step 104, based on the terrain features and dynamic environmental changes of the mountain wind farm blasting construction site, the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm.

[0042] In some embodiments, the influence weights of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics on the risk of blasting construction in mountainous wind farms are dynamically adjusted based on the terrain features and dynamic environmental change characteristics of the blasting construction site. The dynamic influence weights of these characteristics on the risk of blasting construction in mountainous wind farms can be achieved through the following steps: Obtain risk characterization indicators for blasting construction in mountainous wind farms; Based on the risk characterization index, determine the influence weights of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics on the blasting construction risk of mountain wind farms, respectively. Obtain historical risk case data for blasting construction in mountainous wind farms, as well as the terrain features and dynamic environmental changes at the blasting construction sites in mountainous wind farms; Construct a terrain-environment dynamic adjustment vector based on the terrain features and the dynamic environmental change features; Based on the historical risk case data and the terrain-environment dynamic adjustment vector, the initial influence weights are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution characteristics, the environmental influence characteristics, the construction parameter correlation characteristics, and the risk inducement characteristics on the blasting construction risk of mountain wind farms.

[0043] It should be noted that the risk characterization indicators for mountain wind farm blasting construction in this application represent measurable and verifiable specific indicators of the degree of safety risk, quality control effectiveness, and environmental impact range during the blasting construction process. They are the core basis for determining the risk level and calculating the impact weights of the four types of characteristics, providing quantitative support for risk assessment, dynamic adjustment of weights, and generation of early warning information. The risk characterization indicators include blasting safety indicators, construction quality indicators, environmental impact indicators, and accident-related indicators. For example, blasting safety indicators include peak blasting vibration, maximum flying distance of flyrock, slope displacement rate, and exposure time of personnel / equipment within the safety warning range; construction quality indicators include the qualified rate of rock fragmentation, borehole utilization rate, and rework rate of secondary blasting; environmental impact indicators include dust concentration, noise decibel value, and compliance rate of safe distance between sensitive targets; and accident-related indicators include the incidence rate of risk accidents, frequency of equipment failures, and number of operational violations.

[0044] In specific implementation, the influence weights of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk causal characteristics on the blasting construction risk of mountain wind farms, based on the risk characterization indicators, can be achieved in the following way: A hierarchical analysis structure model is constructed, with the influence weights of the four types of characteristics on the blasting construction risk set as the target layer, and risk characterization indicators such as peak blasting vibration, maximum flying distance of flyrock, and slope displacement rate set as the criterion layer, and the four types of characteristics set as the scheme layer; referring to the 1-9 scaling method, where 1 represents equal importance of the two characteristics, 3 represents slightly important, 5 represents significantly important, 7 represents strongly important, and 9 represents extremely important, with 2 / 4 / 6 / 8 as transitional scales, and the reciprocal is taken for inverse comparison. For each criterion layer risk characterization indicator, the influence degree of the four types of characteristics in the scheme layer on the indicator is compared one by one, constructing a 4×4 pairwise comparison judgment matrix for each type of risk characterization indicator; then, the sum-product method is used to calculate the maximum eigenvalue (λmax) and initial eigenvector of the judgment matrix: first, the columns of the judgment matrix are normalized, and the elements are normalized... The normalized value is calculated as the sum of the values ​​of the elements in the column. Then, the arithmetic mean of the elements in each row of the normalized matrix is ​​calculated to obtain the initial weights of the four types of features, i.e., the initial feature vectors. The maximum eigenvalue is calculated using the formula λmax = Σ(judgment matrix × feature vector)i / (feature vector i × 4) (4 is the matrix order). Then, a consistency check is performed. First, the consistency index CI = (λmax - 4) / (4 - 1) is calculated. Then, the random consistency index RI is consulted (RI = 0.90 for a 4th order matrix). If the consistency ratio CR = CI / RI < 0.1, the judgment matrix meets the logical consistency requirements. If it fails, the scale value of the judgment matrix is ​​adjusted by experts until it meets the requirements. Finally, the initial feature vectors that pass the consistency check are normalized. The normalized weight is calculated as the sum of the initial weights of the four types of features / the sum of the initial weights of the four types of features, ensuring that the sum of the influence weights of the four types of features is 1. Finally, the influence weights of the four types of features on the blasting construction risk of the mountain wind farm are obtained. In other embodiments, other methods can be used to determine the weights, which are not limited here.

[0045] In addition, in specific implementation, historical risk case data of blasting construction in mountainous wind farms, as well as the terrain features and environmental dynamic change characteristics of the blasting construction site, are obtained. The construction of a terrain-environment dynamic adjustment vector based on these terrain features and environmental dynamic change characteristics can be achieved in the following way: Historical risk case data of blasting construction in nearly five similar mountainous wind farms are collected. This historical risk case data includes the terrain conditions, environmental data, four types of characteristic parameters, and risk occurrence status of each case. Real-time terrain features of the construction site are extracted using GIS technology and quantified according to impact level. Values ​​are assigned, for example, slope gradient ≤20° is assigned value 1, 20°-45° is assigned value 2, and >45° is assigned value 3. The dynamic environmental change characteristics are collected in real time by on-site sensors and mapped to the [0,1] interval after Min-Max normalization. The quantified terrain feature values ​​and the normalized environmental feature values ​​are combined in dimensional order to construct a terrain-environment dynamic adjustment vector with dimensions of quantified slope gradient value, quantified lithology value, quantified elevation difference value, normalized real-time wind speed value, and normalized cumulative rainfall value. Other methods can be used in other embodiments, which are not limited here.

[0046] It should be noted that the historical risk case data in this application represents a structured data set of topographic conditions, environmental data, construction parameters, four types of characteristic parameters, and risk occurrences recorded in similar mountain wind farm blasting construction projects over the past five years, reflecting the correlation between the four types of characteristics and construction risks under different working conditions; real-time topographic features represent the static topographic attributes of the mountain wind farm blasting construction site, including slope gradient, lithology, elevation difference, and topographic complexity of the blast hole area, reflecting the inherent impact of the construction site topography on blasting construction risks; and environmental dynamic change features include real-time wind speed, cumulative rainfall, and air humidity, reflecting the dynamic impact of environmental changes on blasting construction risks during construction.

[0047] In addition, in specific implementation, the dynamic adjustment of each initial influence weight based on the historical risk case data and the terrain-environment dynamic adjustment vector can be achieved by dynamically adjusting the spatiotemporal distribution characteristics, environmental influence characteristics, construction parameter correlation characteristics, and risk inducement characteristics on the blasting construction risk of mountain wind farms. This can be achieved in the following way: Based on historical risk case data, the actual influence of the four types of characteristics on the risk characterization indicators under different terrain-environment combinations is statistically analyzed. For example, under the high slope terrain + rainstorm environment, the actual risk contribution of the environmental influence characteristics increases by 30% compared with the initial weight. Linear regression analysis is used to calculate the weight adjustment coefficients corresponding to each dimension of the adjustment vector. For example, for every increase of 1 in the quantified value of the slope, the adjustment coefficient of the environmental influence characteristics increases by 0.15. The initial influence weights are weighted and calculated with the terrain-environment dynamic adjustment vector and the corresponding adjustment coefficients, i.e., dynamic influence weight = influence weight × (1 + sum of quantified values ​​of each dimension of the adjustment vector × corresponding adjustment coefficients). The dynamic influence weights of the four types of characteristics adapted to the real-time dynamic changes of terrain and environment are obtained. Other methods can also be used for adjustment in other embodiments, which are not limited here.

[0048] It should be noted that the influence weights in this application represent the quantitative values ​​of the proportion of the initial importance of spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics to the blasting construction risks of mountain wind farms; the topography-environment dynamic adjustment vector represents the combined state of the degree of adjustment of the influence weights by real-time topography and dynamic environment; and the dynamic influence weights represent the dynamic importance proportion of the four types of characteristics to the blasting construction risks under different working conditions, which can be used to predict the risks of the entire blasting construction process of mountain wind farms.

[0049] In step 105, risk prediction is performed on the entire process of blasting construction of mountain wind farms based on the potential correlations and all dynamic influence weights, and the risk level distribution of the entire process of blasting construction of mountain wind farms is obtained. Based on the risk level distribution, risk warning information for blasting construction of mountain wind farms is generated.

[0050] In some embodiments, reference Figure 3 As shown, this figure is an exemplary flowchart for determining the risk level distribution in some embodiments of this application. In this embodiment, risk prediction is performed on the entire process of blasting construction of a mountain wind farm based on the potential correlations and all dynamic influence weights. The risk level distribution of the entire process of blasting construction of a mountain wind farm can be obtained by the following steps: In step 1051, each risk prediction unit for the entire process of blasting construction in a mountain wind farm is determined; In step 1052, risk assessment is performed on the blasting construction of each risk prediction unit based on the potential correlation and all dynamic influence weights to obtain the risk assessment value of each risk prediction unit. In step 1053, the risk assessment value of each risk prediction unit is converted into a risk level; In step 1054, the risk level distribution of the entire process of blasting construction in mountainous wind farms is determined based on all risk levels.

[0051] In specific implementation, the risk prediction units for the entire blasting construction process of a mountain wind farm can be determined in the following way: the risk prediction units are divided according to a two-dimensional approach of spatial grid and construction procedure. Spatially, the blasting construction area is divided into several spatial grid units with a precision of 5 meters × 5 meters. Temporally, the entire construction process is divided into four time stages: drilling, charging, detonation, and post-blast treatment. Each risk prediction unit is clearly defined as a combination of spatial grid and time stage, such as grid number 3 in a high slope area plus the charging stage. This yields the risk prediction units for the entire blasting construction process of a mountain wind farm. Other methods can also be used to determine these units in other embodiments, which are not limited here.

[0052] In addition, in specific implementation, risk assessment of blasting construction in each risk prediction unit is conducted based on the potential correlations and all dynamic influence weights. The risk assessment value of each risk prediction unit can be obtained in the following way: For each risk prediction unit, a risk assessment is carried out. First, the quantitative data of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics within the unit are extracted, i.e., data standardized to the [0,1] interval. Then, based on the directed weighted graph of potential correlations, the correlation strength weights between each sub-feature are extracted. Combined with the dynamic influence weights, the unit risk assessment value is calculated using a weighted summation method: First, the risk contribution component of each sub-feature is calculated. For example, the formula risk contribution component = sub-feature quantitative value × corresponding dynamic influence weight × correlation strength weight can be used for calculation. The risk contribution components of all effective sub-features under the four types of features are directly summed to obtain the initial summary value. If the initial summary value > 1, it is normalized by dividing by the theoretical maximum value of all sub-feature contribution components to ensure that the result falls within [0,1]. The interval is calculated with a single-class feature dynamic weight of 0.4, a sub-feature quantification value of 1, and a correlation strength weight of 1. The theoretical maximum value of the four types of features is 0.4×1×1×4=1.6. If the initial summary value is ≤1, it is directly retained, and the summative value is used as the risk assessment value of the corresponding risk prediction unit. In other embodiments, other methods can also be used for assessment, which are not limited here.

[0053] In addition, in specific implementation, the risk assessment value of each risk prediction unit can be converted into a risk level in the following way: referring to the "Safety Regulations for Blasting" (GB6722) and the risk assessment standard for mountain wind power blasting construction, risk level thresholds are divided: 0-0.3 is low risk, 0.3-0.6 is medium risk, 0.6-0.8 is high risk, and 0.8-1.0 is extremely high risk. The risk assessment value of each unit is converted into the corresponding risk level by referring to the threshold.

[0054] In addition, in specific implementation, the risk level distribution of the entire blasting construction process of a mountain wind farm can be determined by the following method: the risk level of all risk prediction units is marked on a GIS visualization map according to their spatial grid location and time stage coordinates, forming a spatial risk level heat map. Different colors are used to distinguish low, medium, high, and extremely high risks. At the same time, the risk level distribution statistics of each stage are organized according to the construction process time axis. For example, the high-risk units account for 35% in the blasting stage. The spatial distribution heat map and the time stage statistics are used as the risk level distribution of the entire blasting construction process of the mountain wind farm, which intuitively presents the risk level differences at different times and spaces and the risk evolution law of the entire process. Other methods can also be used in other embodiments, which are not limited here.

[0055] It should be noted that, in this application, the risk prediction unit refers to the smallest spatial unit in the entire blasting construction process where risk assessment can be carried out independently, reflecting the construction scenario and operation content within a specific spatiotemporal range; the risk assessment value refers to the parameter value that quantifies the severity of risk in the risk prediction unit, reflecting the strength of risk under the combined effect of various risk factors within the unit; the risk level refers to the severity level of risk in the risk prediction unit, reflecting whether the risk in the unit exceeds the safety control range and the intensity of the prevention and control measures to be taken; the risk level distribution represents the distribution pattern and evolution trend of risks in different spatial areas and different process stages throughout the blasting construction process, reflecting the spatiotemporal concentration areas of high risk throughout the entire process and the characteristics of risk changes with the construction progress.

[0056] In some embodiments, generating risk warning information for blasting construction in mountainous wind farms based on the risk level distribution can be achieved through the following steps: Obtain the pre-set risk warning schemes for each risk prediction unit corresponding to the entire process of blasting construction in mountainous wind farms; The risk level distribution is matched with the risk warning scheme preset by each risk prediction unit to generate risk warning information for blasting construction in mountain wind farms.

[0057] It should be noted that the risk warning scheme in this application refers to the risk prediction unit for each spatial grid and construction procedure combination in the entire process of blasting construction of mountain wind farms. It reflects the hierarchical prevention and control logic and practical requirements under different risk scenarios. The risk warning scheme includes five key elements: risk level, warning level, triggering conditions, prevention and control measures, and responsibility loop. It clarifies that low / medium / high / extremely high risk levels correspond to four levels of warning: blue / yellow / orange / red. The triggering conditions need to be combined with the unit risk level and construction scenario. For example, the triggering condition for an orange warning is a high risk level + a safe distance of less than 50m from sensitive targets. The prevention and control measures are subdivided into technical categories, such as adjusting the charge quantity, increasing the frequency of slope displacement monitoring, and optimizing the layout of blast holes, and management categories, such as expanding the safety warning range, increasing the number of dedicated monitoring personnel, and suspending high-risk procedures. At the same time, it clarifies the time limit for the implementation of the measures, the responsible team, and the review standards to ensure that the risk of each risk prediction unit can be accurately matched with the corresponding warning response and prevention and control measures.

[0058] In specific implementation, the risk level distribution is matched with the pre-set risk warning schemes of each risk prediction unit to generate risk warning information for blasting construction in mountain wind farms. This can be achieved in the following way: The risk level distribution results and risk warning schemes are linked by a unique ID of the risk prediction unit. A related query statement is called, and the system automatically matches the schemes based on the unit ID, risk level, corresponding warning level, and extraction of all elements of the warning scheme. Simultaneously, a manual review step is added to perform secondary verification of the warning schemes for high / extremely high-risk units and units surrounding sensitive targets in the matching results. If a high-risk unit is combined with a complex risk scenario such as a rainstorm, a warning level escalation mechanism is triggered, for example, the original orange warning level is upgraded. The alert is upgraded to a red alert. Finally, standardized risk warning information is generated: a single unit warning information must clearly specify the warning unit, warning level, risk level, core risk cause, specific prevention and control measures, execution time limit, and responsible team. The full-process warning information integrates the warning results of each unit to form a combination of spatial warning heat map and time axis warning list. It is displayed in a pop-up window on the GIS visualization platform, pushed to the industrial control terminal, and sent to the responsible personnel via SMS. At the same time, it is linked to the construction site monitoring system. If the unit risk assessment value does not drop to the safe range after the warning, a secondary warning reminder is triggered to ensure that the warning information is accurately delivered and the prevention and control measures can be implemented. Other matching methods can also be used in other embodiments, which are not limited here.

[0059] It should be noted that the risk warning information in this application refers to the risk prevention and control requirements for the entire process of blasting construction in mountainous wind farms. It can be used to convey the evolution trend of risks with construction procedures and the impact on sensitive targets, providing on-site management personnel and work teams with accurate and actionable risk response basis, ensuring timely implementation of technical adjustments, management and control measures, and preventing risks from escalating into safety accidents.

[0060] Furthermore, in another aspect of this application, in some embodiments, this application provides a risk prediction system for blasting construction in mountainous wind farms based on multi-source analysis, with reference to... Figure 4 The figure is a schematic diagram of the structure of a multi-source analysis-based risk prediction system for blasting construction in mountainous wind farms, according to some embodiments of this application. The multi-source analysis-based risk prediction system 400 for blasting construction in mountainous wind farms includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to acquire multi-source heterogeneous data from the blasting construction site of the mountain wind farm. Processing module 402, in this application, is used to correlate and fuse the multi-source heterogeneous data to extract key information related to the blasting construction risks of mountain wind farms; It should be noted that the processing module 402 in this application is also used to extract the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics and risk inducement characteristics of the blasting construction risk of the mountain wind farm from the key information, and to determine the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics and the risk inducement characteristics; Additionally, it should be noted that the processing module 402 in this application is also used to dynamically adjust the influence weights of the spatiotemporal distribution features, the environmental impact features, the construction parameter correlation features, and the risk inducement features on the risk of blasting construction in the mountain wind farm based on the terrain features and environmental dynamic change features of the mountain wind farm blasting construction site, so as to obtain the dynamic influence weights of the spatiotemporal distribution features, the environmental impact features, the construction parameter correlation features, and the risk inducement features on the risk of blasting construction in the mountain wind farm. The execution module 403 in this application is mainly used to predict the risk of the entire process of blasting construction of mountain wind farms based on the potential correlation and all dynamic influence weights, obtain the risk level distribution of the entire process of blasting construction of mountain wind farms, and generate risk warning information for blasting construction of mountain wind farms based on the risk level distribution.

[0061] In addition, this application also provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis.

[0062] In some embodiments, reference Figure 5The figure is a schematic diagram of the structure of a computer device implementing a multi-source analysis-based risk prediction method for blasting construction in mountainous wind farms, according to some embodiments of this application. The multi-source analysis-based risk prediction method for blasting construction in mountainous wind farms described in the above embodiments can be... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0063] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0064] The communication bus 502 can be used to transmit information between the aforementioned components.

[0065] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0066] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0067] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0068] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0069] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0070] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis.

[0071] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0072] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis, characterized in that, Includes the following steps: Collect multi-source heterogeneous data from blasting construction sites in mountainous wind farms; The multi-source heterogeneous data are correlated and fused to extract key information related to the blasting construction risks of mountain wind farms; From the key information, the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics of blasting construction risks in mountain wind farms are extracted, and the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk trigger characteristics is determined. Based on the terrain features and dynamic environmental changes at the mountain wind farm blasting construction site, the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the blasting construction risk of the mountain wind farm. Based on the potential correlations and all dynamic influence weights, risk prediction is performed on the entire process of blasting construction in mountainous wind farms to obtain the risk level distribution of the entire process of blasting construction in mountainous wind farms. Based on the risk level distribution, risk warning information for blasting construction in mountainous wind farms is generated.

2. The method as described in claim 1, characterized in that, The multi-source heterogeneous data includes topographic elevation and geological structure data, real-time meteorological monitoring data, performance parameters of blasting materials, drilling and charging operation data, operating status data of construction equipment, distribution data of sensitive targets such as surrounding residential areas, roads, power transmission lines, and ecological protection areas, historical records of blasting risk accidents, and emergency resource allocation data.

3. The method as described in claim 1, characterized in that, The multi-source heterogeneous data is correlated and fused to extract key information related to the blasting construction risks of mountain wind farms, specifically including: The multi-source heterogeneous data is preprocessed to obtain preprocessed multi-source heterogeneous data; Based on the correlation features between the data in the preprocessed multi-source heterogeneous data, the multi-source heterogeneous data is correlated and fused to obtain correlated and fused data. To obtain risk characteristics related to blasting construction risks in mountainous wind farms; Based on the aforementioned risk characteristics, key information related to the blasting construction risks of mountain wind farms is extracted from the associated fusion data.

4. The method as described in claim 1, characterized in that, The spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk precipitating characteristics of blasting construction risks in mountain wind farms extracted from the aforementioned key information specifically include: From the aforementioned key information, we can extract the spatiotemporal distribution information, environmental impact information, construction parameter information, and risk precipitating information of blasting construction risks in mountain wind farms. Based on the aforementioned spatiotemporal distribution information, the spatiotemporal distribution characteristics of blasting construction risks in mountain wind farms are determined; Based on the aforementioned environmental impact information, determine the environmental impact characteristics of the blasting construction risks at the mountain wind farm; Based on the aforementioned construction parameter information, determine the construction parameter correlation characteristics of the blasting construction risk in mountain wind farms; Based on the aforementioned risk factor information, the risk factor characteristics of blasting construction risks in mountain wind farms are determined.

5. The method as described in claim 1, characterized in that, Determining the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics, and the risk trigger characteristics specifically includes: The spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics are standardized to obtain standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk trigger characteristics. Determine the dependencies and correlations among the standardized spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk factor characteristics; Based on the dependencies and associations, potential associations are constructed among the spatiotemporal distribution features, environmental impact features, construction parameter association features, and risk trigger features.

6. The method as described in claim 1, characterized in that, Based on the terrain features and dynamic environmental changes at the mountain wind farm blasting construction site, the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk trigger features on the blasting construction risk of the mountain wind farm are dynamically adjusted. Specifically, the dynamic influence weights of these features on the blasting construction risk of the mountain wind farm include: Obtain risk characterization indicators for blasting construction in mountainous wind farms; Based on the risk characterization index, determine the influence weights of the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics, and risk inducement characteristics on the blasting construction risk of mountain wind farms, respectively. Obtain historical risk case data for blasting construction in mountainous wind farms, as well as the terrain features and dynamic environmental changes at the blasting construction sites in mountainous wind farms; Construct a terrain-environment dynamic adjustment vector based on the terrain features and the dynamic environmental change features; Based on the historical risk case data and the terrain-environment dynamic adjustment vector, the initial influence weights are dynamically adjusted to obtain the dynamic influence weights of the spatiotemporal distribution characteristics, the environmental influence characteristics, the construction parameter correlation characteristics, and the risk inducement characteristics on the blasting construction risk of mountain wind farms.

7. The method as described in claim 1, characterized in that, Based on the aforementioned potential correlations and all dynamic influence weights, risk prediction is performed on the entire process of blasting construction for mountain wind farms. The risk level distribution of the entire process of blasting construction for mountain wind farms specifically includes: Identify all risk prediction units for the entire blasting construction process in mountainous wind farms; Based on the potential correlations and all dynamic influence weights, a risk assessment is conducted on the blasting construction of each risk prediction unit to obtain the risk assessment value of each risk prediction unit. Convert the risk assessment values ​​of each risk prediction unit into risk levels; The risk level distribution of the entire blasting construction process for mountain wind farms is determined based on all risk levels.

8. A risk prediction system for blasting construction in mountainous wind farms based on multi-source analysis, characterized in that, include: The data acquisition module is used to collect multi-source heterogeneous data from blasting construction sites in mountainous wind farms. The processing module is used to correlate and fuse the multi-source heterogeneous data and extract key information related to the blasting construction risks of mountain wind farms; The processing module is also used to extract the spatiotemporal distribution characteristics, environmental impact characteristics, construction parameter correlation characteristics and risk inducement characteristics of the blasting construction risk of the mountain wind farm from the key information, and to determine the potential correlation between the spatiotemporal distribution characteristics, the environmental impact characteristics, the construction parameter correlation characteristics and the risk inducement characteristics; The processing module is further configured to dynamically adjust the influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the risk of blasting construction in mountainous wind farms based on the terrain features and dynamic environmental change features of the blasting construction site, thereby obtaining the dynamic influence weights of the spatiotemporal distribution features, environmental impact features, construction parameter correlation features, and risk inducement features on the risk of blasting construction in mountainous wind farms. The execution module is used to predict the risks of the entire process of blasting construction in mountainous wind farms based on the potential correlations and all dynamic influence weights, obtain the risk level distribution of the entire process of blasting construction in mountainous wind farms, and generate risk warning information for blasting construction in mountainous wind farms based on the risk level distribution.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for predicting the risk of blasting construction in mountain wind farms based on multi-source analysis as described in any one of claims 1 to 7.

Images