A stockyard safety situation assessment method based on multi-source perception data

Abstract

This invention discloses a method for assessing the safety status of a material yard based on multi-source sensing data, belonging to the field of data processing technology. The method includes acquiring multi-source sensing data from various location units within a continuous sampling time and constructing a data time series; establishing adjacency relationships between location units based on spatial coordinates; correcting the data based on the differences between adjacent time points and the spatial differences between adjacent location units to obtain a target data series; calculating spatial difference indices for each location unit under various data types based on the target data series; determining weight values ​​based on the magnitude of the spatial difference index differences; weighted fusion of the spatial difference indices to obtain the overall spatial status value of the material yard; and performing spatial difference concentration analysis based on this spatial status value to determine the safety status level. This invention, through time and spatial consistency correction and multi-source fusion analysis, achieves a quantitative assessment and graded output of the overall safety status of the material yard, improving the ability to identify local spatial anomalies and the overall risk characterization capability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method for assessing the safety status of a material yard based on multi-source sensing data. Background Technology

[0002] In large-scale industrial settings such as material yards, each unit continuously generates various types of sensor data during operation, including temperature, dust concentration, and equipment operating status. These multi-source sensor data not only exhibit continuous changes over time, but also show significant spatial correlations between different units due to their spatial adjacency, resulting in a clear spatiotemporal coupling characteristic.

[0003] Existing methods for assessing the safety status of material yards typically suffer from the following shortcomings: Firstly, most methods only analyze time-series data from a single location unit independently, or simply summarize the overall data using global statistics, making it difficult to effectively capture the spatial differences between different location units and the abnormal clustering and diffusion trends in local areas. Secondly, during the fusion of multi-source sensing data, there are significant differences in the magnitude of change and the weight of influence between different data types. If the spatial differences in the contributions of each data type are not reasonably distinguished, it can easily lead to unreasonable weight allocation, resulting in insufficient stability and reliability of the overall safety status assessment results.

[0004] For example, existing dam safety monitoring technologies (see patent publication number CN119646952A) mostly rely on single-modality sensor data, resulting in isolated data sources and a lack of comprehensive analysis of historical data and spatial dynamic correlations. This makes it impossible to effectively model the propagation path and local clustering characteristics of potential hazards in spatial structures. Similarly, multi-source data risk early warning methods for large-scale scenarios such as ports (see patent publication number CN121836405A) introduce multi-source information correlation feedback, but still focus on global summarization and simple indicator processing. This makes it difficult to achieve accurate correction and differentiated weight fusion based on time changes and spatial neighborhood relationships, resulting in insufficient sensitivity to local spatial anomalies. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a method for assessing the safety status of a material yard based on multi-source sensing data. This method corrects the multi-source sensing data based on temporal variation characteristics and spatial neighborhood relationships, extracts spatial difference features of each location unit under different data types, constructs an overall spatial status representation of the material yard through multi-data type fusion, and assesses the overall safety status of the material yard based on the degree of concentration of spatial differences.

[0006] To achieve the above objectives, this invention provides a method for assessing the safety status of a material yard based on multi-source sensing data, comprising the following steps: Acquire multi-source sensing data of each location unit in the material yard within a continuous time period, and construct the data time series of each location unit according to the acquisition time; Establish spatial adjacency relationships between each location unit based on its spatial coordinates; Based on the data time series of each location unit and the spatial adjacency relationship between each location unit, the data of the same data type of each location unit is corrected according to the data difference of the same data type of each location unit at adjacent times and the data difference of the same data type of adjacent location units at the same time, so as to determine the target data sequence of the corresponding data type of each location unit. Based on the target data sequence of the data type corresponding to each location unit and the spatial adjacency relationship between each location unit, calculate the neighborhood difference of each location unit and its neighboring location units of the same data type at the same time, and determine the spatial difference index of each location unit under each data type based on the neighborhood difference. Based on the spatial difference index corresponding to each location unit under each data type, the magnitude of the difference in spatial difference index between each location unit and its adjacent location units under the same data type is calculated, and the weight value of each location unit under each data type is determined according to the magnitude of the difference in spatial difference index. Based on the spatial difference indicators of each location unit under each data type and their corresponding weight values, the spatial difference indicators of each location unit under each data type are weighted and fused to determine the overall spatial situation value of the material yard. Based on the overall spatial situation value of the material yard, the overall safety status of the material yard is classified, the corresponding safety situation level is determined, and the assessment results are output.

[0007] The technical solution provided in this invention has at least the following technical effects or advantages: For complex sensing scenarios with multiple location units and multiple data types in a material yard, a unified data processing mechanism based on time-varying features and spatial neighborhood relationships is constructed to achieve spatiotemporal consistency correction and spatial difference feature extraction of multi-source sensing data. On this basis, a material yard overall spatial situation characterization result integrating multiple data types is formed, thereby realizing quantitative assessment and hierarchical judgment of the overall safety status of the material yard.

[0008] Specifically, By acquiring multi-source sensing data from each location unit within a continuous sampling time, and combining time series and spatial adjacency relationships to construct a unified data organization structure, the data changes between different location units and within the same location unit at different time stages are comparable and continuous, thereby improving the consistency of multi-source data expression in the spatiotemporal dimension. By correcting the original multi-source sensing data based on the difference between adjacent sampling times and the spatial difference between adjacent location units, a joint constraint mechanism of temporal variation features and spatial neighborhood deviation features is introduced, enabling the data to adaptively adjust between temporal variation trends and spatial consistency, thereby improving the stability and reliability of the target data sequence. By calculating the spatial difference index of each location unit under different data types based on the target data sequence, and combining the neighborhood relationship to quantify the spatial difference, the spatial anomaly is no longer limited to the fluctuation of single point data, but can reflect its spatial diffusion characteristics within the neighborhood range, thereby improving the sensitivity and distinguishing ability of spatial difference identification. By using a weighting mechanism based on the magnitude of the spatial difference index difference, the influence of different data types in different location units is allocated differently, so that different types of data can reflect their actual contribution to the formation of spatial differences during the fusion process, thereby improving the rationality and stability of the multi-source data fusion results. By weighted and fused spatial difference indicators of each location unit under different data types, the overall spatial situation value of the material yard is obtained, which enables multi-source information to be expressed in a unified scalar space, thereby achieving a comprehensive characterization of the overall spatial anomaly. Furthermore, based on spatial contribution analysis and spatial difference concentration assessment, the distribution of spatial differences among different location units is characterized, so that the overall spatial anomaly can not only be quantified, but also identified whether it is concentrated in a few key location units, thereby improving the ability to identify risks dominated by local anomalies.

[0009] Compared with existing technologies, this invention constructs a multi-source data collaborative processing mechanism that integrates time-varying characteristics and spatial neighborhood relationships. It realizes a complete evaluation link from data correction, spatial difference extraction, multi-data type fusion to spatial concentration analysis. It can not only quantitatively evaluate the overall safety status of the material yard, but also reflect the distribution structure characteristics of spatial anomalies, thereby improving the comprehensiveness and refinement of safety situation assessment. It is suitable for continuous monitoring and risk identification tasks in complex spatial scenarios. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart illustrating a method for assessing the safety status of a material yard based on multi-source sensing data. Detailed Implementation

[0012] This invention introduces a multi-source data collaborative processing mechanism that combines time-varying features with spatial neighborhood relationships, multi-source spatial difference feature extraction, and spatial difference concentration assessment method. It proposes a material yard safety status assessment method based on multi-source sensing data, which solves the problems in traditional material yard safety assessments, such as difficulty in simultaneously considering time-varying features and spatial neighborhood relationships, insufficient expression of spatial anomaly distribution features, and lack of differentiated weight support for multi-source data fusion. These problems lead to insufficient sensitivity of the overall safety status assessment results to local spatial anomalies and limited ability to identify risk clustering features.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., used in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, a method for assessing the safety situation of a material yard based on multi-source sensing data includes the following steps: S1. Acquire multi-source sensing data of each location unit in the material yard within a continuous time period, and construct the data time series of each location unit according to the acquisition time; Specifically, the material yard is divided into several location units, each corresponding to a fixed spatial area. Each location unit is assigned a unique number, and sensing points for collecting multi-source sensing data are deployed within each location unit. A unified time recording method is set for all sensing points, so that all location units use the same time reference when collecting data. The multi-source sensing data includes, but is not limited to, temperature data, dust concentration data, and equipment operating status data.

[0016] Within a defined continuous time period, each location unit is controlled to collect data according to a fixed sampling cycle. At each sampling moment, temperature data, dust concentration data, and equipment operating status data of each location unit are collected, and a collection time identifier is added to the acquired multi-source sensing data. The continuous time period includes at least three sampling moments to form a continuous data difference calculation relationship based on adjacent moments, thereby supporting continuous change analysis. To avoid information redundancy due to excessively high sampling density and information loss due to excessively low sampling density, while ensuring that the values ​​of the same data type change between adjacent sampling moments, the sampling cycle is set to 1 to 10 minutes.

[0017] For each location unit, the multi-source sensing data acquired within a continuous time period are sorted according to the acquisition time identifier and arranged sequentially in chronological order to form the data time series of that location unit. During the arrangement process, each time node contains the multi-source sensing data of that location unit at that moment. When there is a data gap at a certain acquisition time, the multi-source sensing data of that location unit at the previous acquisition time is used to supplement it, so that the data time series remains continuous in the time dimension, thereby completing the construction of the data time series of each location unit.

[0018] S2. Establish the spatial adjacency relationships between each location unit based on its spatial coordinates, including: Using the overall spatial range of the material yard as the coordinate calibration benchmark, a unified two-dimensional coordinate system is established, and all position units are calibrated under the same coordinate benchmark. Obtain the spatial range of each location unit and determine the center position of the spatial region of each location unit. Use the coordinates of the center position of the spatial region of each location unit as the two-dimensional coordinates of each location unit. For any given position cell, based on the two-dimensional coordinates of that position cell, coordinate offsets are performed in the horizontal and vertical directions according to the spatial spacing between adjacent units to determine candidate coordinate positions. Then, the position cell corresponding to the candidate coordinate position is found from all position cells and used as the adjacent position cell of that position cell. The adjacent position units of each position unit are merged to form the spatial adjacency relationship between each position unit.

[0019] Specifically, a unified two-dimensional coordinate system is established using the overall spatial range of the material yard as the coordinate calibration benchmark, and all location units are calibrated using the same coordinate benchmark.

[0020] Obtain the spatial region range corresponding to each position unit, calculate the midpoint position of the spatial region range of each position unit in the horizontal direction and the midpoint position in the vertical direction, determine the coordinates of the center position of the spatial region of each position unit, extract the coordinate value of the center position of the spatial region of each position unit, and use it as the two-dimensional coordinates of each position unit.

[0021] Based on the mesh generation results, the spatial spacing between adjacent positional units is obtained and used as the unit interval between positional units. For any positional unit, its two-dimensional coordinates are read, its horizontal coordinate value is kept unchanged, and its vertical coordinate value is added to and subtracted by one unit interval to determine two candidate vertical coordinate positions. The positional unit corresponding to the candidate vertical coordinate position is searched among all positional units, and the found positional unit is determined as its vertical adjacent positional unit. Simultaneously, its vertical coordinate value is kept unchanged, and its horizontal coordinate value is added to and subtracted by one unit interval to determine two candidate horizontal coordinate positions. The positional unit corresponding to the candidate horizontal coordinate position is searched among all positional units, and the found positional unit is determined as its horizontal adjacent positional unit, thus determining the adjacent positional units of this positional unit.

[0022] The adjacent position units determined by each position unit in the horizontal and vertical directions are merged and associated and stored according to the unique position unit number to form a set of adjacent position units corresponding to each position unit, thereby forming the spatial adjacency relationship between each position unit.

[0023] S3. Based on the data time series of each location unit and the spatial adjacency relationship between each location unit, and according to the data difference of the same data type in adjacent time periods of each location unit and the data difference of the same data type in adjacent location units at the same time period, the data of the corresponding data type of each location unit is corrected to determine the target data sequence of the corresponding data type of each location unit, including, For any location unit, based on the multi-source sensing data corresponding to each sampling time in the data time series of that location unit, under the same data type, data of the same data type at adjacent sampling times are obtained sequentially, and the difference between the two is calculated. This difference is used as the time change of the corresponding data type at the next sampling time. Based on the spatial adjacency relationship between location units, at any sampling time, the same type of data corresponding to the location unit and all its adjacent location units at that sampling time is obtained. The difference between the corresponding data of the location unit and each of its adjacent location units is calculated to obtain the spatial difference and generate a spatial difference set. The deviation value is calculated based on the time change of the data type corresponding to the sampling time and the spatial difference set, and the neighborhood deviation value is calculated based on the data of the data type corresponding to each adjacent position unit. The product of the deviation value and the neighborhood deviation value is used as the adjustment amount. Based on the neighborhood deviation value and adjustment amount, the data of the corresponding data type of each position unit is corrected to construct the target data sequence of the corresponding data type of each position unit.

[0024] Furthermore, based on the neighborhood deviation value and adjustment amount, the data of the corresponding data type for each location unit is corrected to construct the target data sequence of the corresponding data type for each location unit, including, For any location unit, at any sampling time, the correction direction is determined based on the relationship between the neighborhood deviation value and zero. When the neighborhood deviation value is greater than 0, the adjustment amount is subtracted from the data of the corresponding data type of the location unit at that sampling time. When the neighborhood deviation value is less than 0, the adjustment amount is added to the data of the corresponding data type of the location unit at that sampling time to obtain the target data of the corresponding data type at that sampling time. The target data of the corresponding data type of each location unit at each sampling time are arranged in chronological order to form the target data sequence of the corresponding data type of each location unit.

[0025] Specifically, for any given location unit, multi-source sensing data corresponding to each sampling time in the data time series of that location unit is acquired sequentially in chronological order. Under the same data type, data between adjacent sampling times is acquired sequentially, and the data difference between adjacent sampling times is calculated. This data difference is used as the time change of the data type corresponding to the next sampling time. The time change of the data type corresponding to the first sampling time is defined as 0. For each type of data, the time changes corresponding to each sampling time are arranged sequentially in chronological order to form a time change sequence of the data type corresponding to that location unit. For example, for a certain location unit, at continuous sampling times... If the temperature data obtained are 20℃, 22℃, and 21℃, then the numerical differences between adjacent sampling times are respectively... The temperature change at that moment is 2℃. The temperature change at time t is -1℃, where The time change of the data type corresponding to a given moment is defined as a zero change; after being arranged in chronological order, the temperature time difference sequence is {0℃, 2℃, -1℃}, and each item in this sequence corresponds to... Time and time.

[0026] For any given location unit, based on the spatial adjacency relationships between location units, at any sampling time, acquire the same type of data corresponding to that location unit and all its adjacent location units at that sampling time. Under the same data type, calculate the difference between the corresponding data of that location unit and each of its adjacent location units to obtain the spatial difference between that location unit and its adjacent location units. Then, summarize the spatial differences corresponding to all adjacent location units at that sampling time and under that data type to form a spatial difference set. For example, the data type is temperature data, and the sampling time is... Let the position unit be i, and its adjacent position units be j and k. Then, position unit i is in... The spatial difference set of temperature data at different times is {location unit i and location unit j at...} The numerical difference of temperature data at time intervals, between position unit i and position unit k. The numerical difference between the temperature data at different times.

[0027] For any location cell, at any sampling time, acquire the time change and spatial difference set of the corresponding data type at that sampling time. Average the spatial differences in the spatial difference set to obtain the spatial average difference. Then, calculate the absolute difference between the spatial average difference and the time change to obtain the deviation value. Simultaneously, the average value M is calculated based on the data of the corresponding data types of all adjacent data types of the current data type. Then, the difference between the data of the current data type and the neighborhood average value M is calculated to obtain the neighborhood deviation value. Neighborhood deviation value The adjustment amount A is obtained by multiplying the absolute value of the deviation value D1 by the deviation value. .

[0028] Based on neighborhood deviation value The magnitude relationship between zero and the value of zero determines the correction direction. When the value is greater than 0, the adjustment amount A is subtracted from the data of the corresponding data type at that sampling time for that position cell. When the value is less than 0, an adjustment amount A is added to the data of the corresponding data type at the sampling time for that location unit to obtain the target data of the corresponding data type at that sampling time. The target data of the corresponding data type for each location unit at each sampling time are arranged in chronological order to form the target data sequence of the corresponding data type for each location unit. The deviation value is used to characterize the degree of inconsistency between temporal variation and spatial difference, and the neighborhood deviation value is used to characterize the deviation direction of the current position relative to the neighborhood. The two are used together to determine the correction magnitude and direction, so as to achieve adaptive correction of the data of the corresponding data type for that location unit.

[0029] S4. Based on the target data sequence of the corresponding data type for each location unit and the spatial adjacency relationship between each location unit, calculate the neighborhood difference of each location unit and its neighboring location units of the same data type at the same time, and determine the spatial difference index corresponding to each location unit under each data type based on the neighborhood difference, including, For any given location cell, under the same data type at the same time, calculate the target data difference between the location cell and its neighboring location cells, and perform absolute value processing on the target data difference to obtain the neighborhood difference between the location cell and its neighboring location cells. The neighborhood differences between the location cell and all its neighboring location cells are accumulated to obtain the total neighborhood difference of the location cell at the sampling time and under the data type. Under the same data type, the total neighborhood difference corresponding to each sampling time of each location unit is averaged to obtain the spatial difference index corresponding to each location unit under each data type.

[0030] Specifically, for any location unit, at any sampling time, under the same data type, the target data V(i,t,y) of the corresponding data type at that sampling time is obtained based on the target data sequence of the corresponding data type of the location unit, where i represents the location unit number, t represents the sampling time, and y represents the data type; Based on the spatial adjacency relationship between position units, all adjacent position units of the position unit at the sampling time are obtained, and the target data V(j,t,y) of each adjacent position unit at the sampling time and under the data type are obtained respectively, where j represents the adjacent position unit number; Under the same data type, calculate the target data difference between the location unit and its neighboring location units respectively, and perform absolute value processing on the target data difference to obtain the neighborhood difference d(i,j,t) between the location unit and its neighboring location units. Then, sum up the neighborhood differences to obtain the total neighborhood difference E(i,t) of the location unit at the sampling time. At the sampling time and under the given data type, the target data difference between the location unit and its neighboring location units is calculated respectively. The absolute value of the target data difference is then processed to obtain the neighborhood difference d(i,j,t,y) between the location unit and its neighboring location units. The neighborhood differences between the location unit and all its neighboring location units are then summed to obtain the total neighborhood difference E(i,t,y) of the location unit at the sampling time and under the given data type. Under the same data type, the total neighborhood difference of the location unit at each sampling time is averaged to obtain the spatial difference index F(i,y) of the location unit under the data type. F(i,y) represents the degree of spatial inconsistency of location unit i relative to its neighborhood under data type y. The greater the degree of difference, the higher the degree of abnormality of the location unit under the data type. Thus, the spatial difference index of each location unit corresponding to the data type is determined.

[0031] S5. Based on the spatial difference index corresponding to each location unit under each data type, calculate the magnitude of the spatial difference index difference between each location unit and its adjacent location units under the same data type, and determine the weight value of each location unit under each data type based on the magnitude of the spatial difference index difference, including... For any given location unit, calculate the spatial difference index difference between that location unit and its adjacent location units, and process the absolute value of the difference to obtain the magnitude of the difference between that location unit and its adjacent location units. For any location unit, under the same data type, calculate the spatial difference index difference between the location unit and its adjacent location units under the same data type, and process the absolute value of the spatial difference index difference to obtain the magnitude of the spatial difference index difference between the location unit and its adjacent location units under the same data type. The difference magnitudes between the location cell and all its neighboring location cells are summed to obtain the neighborhood difference intensity value of the location cell. The neighborhood difference intensity values ​​of each location unit under each data type are normalized to obtain the weight values ​​of each location unit under each data type.

[0032] Specifically, for any location unit, under the same data type, the spatial difference index F(i,y) corresponding to the location unit under the data type is obtained. At the same time, based on the spatial adjacency relationship between each location unit, each location unit adjacent to the location unit is obtained, and the spatial difference index F(j,y) corresponding to each location unit adjacent to the location unit under the data type is obtained respectively. Calculate the spatial difference index difference between the given location unit and its adjacent location units under this data type, and perform absolute value processing on the spatial difference index difference to obtain the amplitude G(i,j,y) of the spatial difference index difference between the given location unit and its adjacent location units under this data type. G(i,j,y) represents the spatial difference between adjacent position units i and j under the data type y; The summation of the spatial difference index differences between the given location unit and its neighboring location units under this data type yields the neighborhood difference intensity value W(i,y) for that location unit under this data type. W(i,y) represents the degree of imbalance in the spatial difference distribution of location unit i relative to the overall spatial difference distribution within its neighborhood under data type y; The neighborhood difference intensity value W(i,y) of the location unit under the data type is normalized to obtain the weight value P(i,y) of the location unit under the data type. The weight value P(i,y) represents the relative influence of location unit i on the overall spatial difference distribution under data type y. This completes the determination of the weight value of each location unit under each data type.

[0033] S6. Based on the spatial difference indicators of each location unit under each data type and their corresponding weight values, the spatial difference indicators of each location unit under each data type are weighted and fused to determine the overall spatial situation value of the material yard, including, For any given location unit, under the same data type, obtain the corresponding spatial difference index and weight value of that location unit under that data type; and calculate the weighted spatial difference value of that location unit under that data type. The weighted spatial difference values ​​of the location unit under each data type are summarized to obtain the comprehensive spatial difference value of the location unit. The overall spatial situation value of the material yard is obtained by summarizing the comprehensive spatial difference values ​​of each location unit.

[0034] Specifically, for any location unit, under the same data type, obtain the spatial difference index F(i,y) and the weight value P(i,y) of the location unit under that data type. Then, based on the spatial difference index F(i,y) and the weight value P(i,y) of the location unit under that data type, calculate the weighted spatial difference value R(i,y) of the location unit under that data type. ; The weighted spatial difference values ​​corresponding to the location unit under each data type are accumulated to obtain the comprehensive spatial difference value S(i) of the location unit. The comprehensive spatial difference value S(i) represents the overall degree of difference between the location unit and its neighborhood under the comprehensive data type. The spatial difference values ​​of all location units are summed to obtain the overall spatial situation value S of the material yard. The spatial situation value S represents the degree of spatial difference between the overall material yard and each location unit under the comprehensive data of multiple data types. The greater the degree of spatial difference, the higher the degree of anomaly of the overall material yard.

[0035] S7. Based on the overall spatial situation value of the material yard, classify the overall safety status of the material yard, determine the corresponding safety situation level, and output the assessment results, including, Based on the overall spatial situation value of the material yard and the comprehensive spatial difference value of each location unit, calculate the spatial contribution of each location unit; Based on the spatial contribution of each location unit, the average spatial contribution of all location units is calculated, and the location units with spatial contributions greater than the average spatial contribution are selected as significant contributing location units. The spatial contribution of significantly contributing location units is accumulated to obtain the spatial difference concentration index; Based on the spatial difference concentration index, the overall safety status of the material yard is classified and the corresponding safety status level is determined.

[0036] Specifically, the comprehensive spatial difference value of all location units and the overall spatial situation value of the material yard are obtained. For any location unit, the proportion of the comprehensive spatial difference value S(i) of that unit in the overall spatial situation value S of the material yard is calculated to obtain the spatial contribution Q(i) of that location unit. ; Sort all spatial contribution units from largest to smallest to obtain a spatial contribution ranking sequence. Calculate the average spatial contribution of all spatial contribution units in the spatial contribution ranking sequence. Select spatial contribution units from the spatial contribution ranking sequence whose spatial contribution is greater than the average spatial contribution and regard them as significant contribution spatial contribution units. The spatial contribution of significant contributing location units is accumulated to obtain the spatial difference concentration index C. The spatial difference concentration index C is used to characterize the degree of dominance and concentration of spatial differences among location units, that is, the degree to which spatial differences are dominated by significant contributing location units. The larger the C value, the more concentrated the spatial differences are by a few significant contributing location units. When there is only one significant contributing location unit, it means that the spatial differences are highly concentrated in a single location unit. The overall safety status of the material yard is classified based on the spatial difference concentration index C, specifically as follows: When C≥0.75, the spatial differences are highly concentrated in the dominant location unit, and the overall safety status of the material yard is determined to be high-risk. When 0.5 ≤ C < 0.75, the spatial difference shows a clear dominant concentration trend, and the overall safety status of the material yard is determined to be medium risk level; When 0.25≤C<0.5, the spatial difference is relatively weak, and the overall safety status of the material yard is determined to be low risk. When C < 0.25, the spatial difference distribution is uniform and there is no obvious dominant unit. The overall safety status level of the material yard is determined to be the safety level.

[0037] After determining the overall safety status level of the material yard, the safety status level is output as the safety status assessment result of the material yard.

[0038] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for assessing the safety status of a material yard based on multi-source sensing data, characterized in that, Includes the following steps, Acquire multi-source sensing data of each location unit in the material yard within a continuous time period, and construct the data time series of each location unit according to the acquisition time; Establish spatial adjacency relationships between each location unit based on its spatial coordinates; Based on the data time series of each location unit and the spatial adjacency relationship between each location unit, the data of the same data type of each location unit is corrected according to the data difference of the same data type of each location unit at adjacent times and the data difference of the same data type of adjacent location units at the same time, so as to determine the target data sequence of the corresponding data type of each location unit. Based on the target data sequence of the data type corresponding to each location unit and the spatial adjacency relationship between each location unit, calculate the neighborhood difference of each location unit and its neighboring location units of the same data type at the same time, and determine the spatial difference index of each location unit under each data type based on the neighborhood difference. Based on the spatial difference index corresponding to each location unit under each data type, the magnitude of the difference in spatial difference index between each location unit and its adjacent location units under the same data type is calculated, and the weight value of each location unit under each data type is determined according to the magnitude of the difference in spatial difference index. Based on the spatial difference indicators of each location unit under each data type and their corresponding weight values, the spatial difference indicators of each location unit under each data type are weighted and fused to determine the overall spatial situation value of the material yard. Based on the overall spatial situation value of the material yard, the overall safety status of the material yard is classified, the corresponding safety situation level is determined, and the assessment results are output.

2. The method for assessing the safety status of a material yard based on multi-source sensing data according to claim 1, characterized in that, The spatial adjacency relationship between each location unit is established based on the spatial coordinates of each location unit. include, Using the overall spatial range of the material yard as the coordinate calibration benchmark, a unified two-dimensional coordinate system is established, and all position units are calibrated under the same coordinate benchmark. Obtain the spatial range of each location unit and determine the center position of the spatial region of each location unit. Use the coordinate values ​​of the center position of the spatial region of each location unit as the two-dimensional coordinates of each location unit. For any given position cell, based on the two-dimensional coordinates of that position cell, coordinate offsets are performed in the horizontal and vertical directions according to the spatial spacing between adjacent units to determine candidate coordinate positions. Then, the position cell corresponding to the candidate coordinate position is found from all position cells and used as the adjacent position cell of that position cell. The adjacent position units of each position unit are merged to form the spatial adjacency relationship between each position unit.

3. The material yard safety situation assessment method based on multi-source sensing data according to claim 2, characterized in that, The step involves correcting the data of the corresponding data type for each position unit based on the data difference of the same data type at adjacent time points and the data difference of the same data type for adjacent position units at the same time point, to determine the target data of the corresponding data type for each position unit, including: For any location unit, based on the multi-source sensing data corresponding to each sampling time in the data time series of that location unit, under the same data type, data of the same data type at adjacent sampling times are obtained sequentially, and the difference between the two is calculated. This difference is used as the time change of the corresponding data type at the next sampling time. Based on the spatial adjacency relationship between location units, at any sampling time, the same type of data corresponding to the location unit and all its adjacent location units at that sampling time is obtained. The difference between the corresponding data of the location unit and each of its adjacent location units is calculated to obtain the spatial difference and generate a spatial difference set. The deviation value is calculated based on the time change of the data type corresponding to the sampling time and the spatial difference set, and the neighborhood deviation value is calculated based on the data of the data type corresponding to each adjacent position unit. The product of the deviation value and the neighborhood deviation value is used as the adjustment amount. Based on the neighborhood deviation value and adjustment amount, the data of the corresponding data type of each location unit is corrected to construct the target data sequence of the corresponding data type of each location unit.

4. The material yard safety situation assessment method based on multi-source sensing data according to claim 3, characterized in that, The process of correcting the data of corresponding data types for each location unit based on neighborhood deviation value and adjustment amount, and constructing the target data sequence of corresponding data types for each location unit, includes, For any location unit, at any sampling time, the correction direction is determined based on the relationship between the neighborhood deviation value and zero. When the neighborhood deviation value is greater than 0, the adjustment amount is subtracted from the data of the corresponding data type of the location unit at that sampling time. When the neighborhood deviation value is less than 0, the adjustment amount is added to the data of the corresponding data type of the location unit at that sampling time to obtain the target data of the corresponding data type at that sampling time. The target data of the corresponding data type of each location unit at each sampling time are arranged in chronological order to form the target data sequence of the corresponding data type of each location unit.

5. The material yard safety situation assessment method based on multi-source sensing data according to claim 4, characterized in that, The calculation of the neighborhood difference between each location unit and its neighboring location units of the same data type at the same time, and the determination of the spatial difference index corresponding to each location unit under each data type based on the neighborhood difference, includes: For any given location cell, under the same data type at the same time, calculate the target data difference between the location cell and its neighboring location cells, and perform absolute value processing on the target data difference to obtain the neighborhood difference between the location cell and its neighboring location cells. The neighborhood differences between the location cell and all its neighboring location cells are accumulated to obtain the total neighborhood difference of the location cell at the sampling time and under the data type. Under the same data type, the total neighborhood difference corresponding to each sampling time of each location unit is averaged to obtain the spatial difference index corresponding to each location unit under each data type.

6. The material yard safety situation assessment method based on multi-source sensing data according to claim 5, characterized in that, The calculation of the spatial difference index difference between each location unit and its adjacent location units under the same data type, and the determination of the weight value of each location unit under each data type based on the spatial difference index difference magnitude, includes: For any given location unit, calculate the spatial difference index difference between that location unit and its adjacent location units, and process the absolute value of the difference to obtain the magnitude of the difference between that location unit and its adjacent location units. For any location unit, under the same data type, calculate the spatial difference index difference between the location unit and its adjacent location units under the same data type, and process the absolute value of the spatial difference index difference to obtain the magnitude of the spatial difference index difference between the location unit and its adjacent location units under the same data type. The difference between the spatial difference index of the location unit and its neighboring location units under this data type is summed to obtain the neighborhood difference intensity value of the location unit under this data type. The neighborhood difference intensity values ​​of each location unit under each data type are normalized to obtain the weight values ​​of each location unit under each data type.

7. The method for assessing the safety status of a material yard based on multi-source sensing data according to claim 6, characterized in that, The weighted fusion processing of spatial difference indicators for each location unit under various data types to determine the overall spatial situation value of the material yard includes, For any given location unit, under the same data type, obtain the corresponding spatial difference index and weight value of that location unit under that data type; and calculate the weighted spatial difference value of that location unit under that data type. The weighted spatial difference values ​​of the location unit under each data type are summarized to obtain the comprehensive spatial difference value of the location unit. The overall spatial situation value of the material yard is obtained by summarizing the comprehensive spatial difference values ​​of each location unit.

8. The method for assessing the safety status of a material yard based on multi-source sensing data according to claim 7, characterized in that, The overall safety status of the material yard is classified based on the overall spatial situation value of the material yard, and the corresponding safety status level is determined, including... Based on the overall spatial situation value of the material yard and the comprehensive spatial difference value of each location unit, calculate the spatial contribution of each location unit; Based on the spatial contribution of each location unit, the average spatial contribution of all location units is calculated, and the location units with spatial contributions greater than the average spatial contribution are selected as significant contributing location units. The spatial contribution of significantly contributing location units is accumulated to obtain the spatial difference concentration index; Based on the spatial difference concentration index, the overall safety status of the material yard is classified and the corresponding safety status level is determined.

Images