Foundation pit safety early warning method based on multi-source monitoring data fusion
Through the fusion of multi-source monitoring data, the weight and credibility of foundation pit safety monitoring indicators are evaluated using the interquartile distance method and Chebyshev distance, combined with the D-S evidence theory, the problem of spatiotemporal differences and one-sided analysis of foundation pit safety monitoring information is solved, and the problem of spatial and temporal differences and one-sided analysis of foundation pit safety monitoring information is improved. The accuracy and robustness of safety warnings are achieved.
Patent Information
- Application Number
- CN202411191460.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-12
AI Technical Summary
The existing foundation pit safety monitoring technology has the phenomenon of "focusing on collection and neglecting analysis", which leads to significant and diverse temporal and spatial differences in monitoring information, and traditional data analysis methods are one-sided, resulting in poor accuracy of foundation pit safety warning.
The multi-source monitoring data fusion method is used to calculate the weight of the monitoring indicators through the interquartile distance method, and the credibility of the evidence is evaluated by using the Chebyshev distance, and the data fusion is combined with the D-S evidence theory to obtain a more accurate and robust foundation pit safety evaluation. grade.
It improves the accuracy and robustness of foundation pit safety assessment, ensures that the importance of different monitoring indicators is reasonably reflected, screens out reliable evidence sources, reduces the impact of misinformation, and improves the efficiency and accuracy of information fusion.
Smart Images

Figure CN118942229A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of foundation pit safety early warning, and more specifically, particularly relates to a foundation pit safety early warning method by fusing multi-source monitoring data. Background Art
[0002] As an important part of civil engineering construction, the safety of foundation pit is directly related to the life safety of construction workers and the stability of the surrounding environment. With the acceleration of urbanization, the scale and depth of foundation pit projects are increasing, and the safety of foundation pits is becoming increasingly prominent. Due to the influence of factors such as regionality, surrounding environment, and soil quality changes on the construction site, the internal force of the foundation pit support changes during the earth excavation process, which in turn causes the soil at the bottom of the foundation pit to bulge, road pipelines to deform, and uneven settlement and lateral displacement of surrounding buildings. When the deformation reaches a certain value, the support structure is damaged, the foundation pit becomes unstable and collapses, causing related economic losses and casualties.
[0003] Safety monitoring is one of the most commonly used and effective risk control methods in foundation pit engineering. However, the phenomenon of "focusing on collection and neglecting analysis" has always existed in actual engineering, resulting in the failure of monitoring work to play its due role. In addition, the monitoring information currently obtained from foundation pit engineering not only has significant temporal and spatial differences, but also has diverse types and a wide range of sources. Therefore, the characteristics of monitoring information present multi-source heterogeneous characteristics. Traditional data analysis methods mostly use threshold analysis methods of a single indicator (such as displacement or deformation rate of key points, etc.), which only reflects the local performance of the foundation pit from one aspect, which is one-sided and leads to poor accuracy of foundation pit safety warning.
[0004] In view of the above problems, the present invention proposes a foundation pit safety early warning method based on multi-source monitoring data fusion, which integrates and applies different monitoring data to comprehensively, comprehensively and reasonably evaluate the safety status of the foundation pit to avoid information blind spots and improve information integrity. Summary of the invention
[0005] In view of the problems existing in the prior art, the purpose of the present invention is to provide a foundation pit safety early warning method based on the fusion of multi-source monitoring data. By using the interquartile range method to calculate the weights of multi-source monitoring indicators and using the Chebyshev distance to evaluate the credibility of different evidence, the advantages of these two methods can be combined to improve the accuracy and robustness of the overall evaluation. Through weight distribution, it can be ensured that the importance of different monitoring indicators is reasonably reflected in subsequent analysis; and through credibility evaluation, reliable sources of evidence can be screened out to avoid the influence of erroneous or misleading information on the analysis results.
[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a foundation pit safety early warning method based on multi-source monitoring data fusion;
[0007] S1. According to the characteristics of foundation pit engineering and safety monitoring, a foundation pit safety evaluation index system is constructed to obtain the cumulative value and change rate of multi-source time series monitoring indicators;
[0008] S2. Determine the foundation pit safety level identification framework, use multi-source time series monitoring data as different evidence and perform normalization processing;
[0009] S3, determining the basic probability distribution values of different focal elements in the identification framework based on multi-source time series monitoring data;
[0010] S4. The interquartile range method is used to calculate the weights of multi-source monitoring indicators, and the Chebyshev distance is used to obtain the credibility of different evidences;
[0011] S5. According to the weights of multi-source monitoring indicators and the credibility of evidence, the basic probability distribution values of different focal elements are modified;
[0012] S6. Use the DS evidence theory synthesis rules to fuse the multi-source monitoring data of the foundation pit to obtain the foundation pit safety evaluation level.
[0013] Optionally, the construction of the foundation pit safety evaluation index system includes:
[0014] S11. Determine the evaluation index system and determine the key safety evaluation indicators based on the characteristics of the foundation pit project, such as geological conditions, support structure type, excavation depth, and surrounding environmental impact.
[0015] S12, data collection, real-time monitoring of various indicators through sensors to obtain time series data;
[0016] S13, data preprocessing, cleaning, denoising and interpolation of data to obtain cumulative values and change rates;
[0017] The determination index system is composed of surrounding environment monitoring indicators and monitoring indicators of multiple foundation pit sections;
[0018] The surrounding environment monitoring indicators include: cumulative value of building settlement, building settlement change rate, cumulative value of surface settlement, surface settlement change rate, cumulative value of pipeline settlement and pipeline settlement change rate;
[0019] The foundation pit profile monitoring indicators include: cumulative value of vertical displacement of wall top, rate of change of vertical displacement of wall top, cumulative value of horizontal displacement of wall top, rate of change of horizontal displacement of wall top, cumulative value of deep horizontal displacement, rate of change of deep horizontal displacement, cumulative value of support internal force, cumulative value of groundwater level and rate of change of groundwater level.
[0020] Optionally, the sensor includes: a displacement meter, a stress meter and a water level meter, which are respectively used to measure the displacement of the structure or soil, monitor the stress change inside the structure, and observe the fluctuation of the groundwater level;
[0021] The sensor is connected to a data acquisition unit for collecting sensor data in real time and storing it in a database for subsequent analysis and processing;
[0022] The data acquisition unit is provided with an alarm. By setting early warning and alarm thresholds, when a monitoring indicator of a certain measuring point exceeds the set threshold, the multi-source monitoring data of the measuring point is integrated and analyzed, that is, through the evidence of other monitoring indicators, the system can automatically determine whether it is a "false alarm" or a real alarm danger, to ensure that protective measures are taken in time.
[0023] The data preprocessing is used to remove noise, incomplete or inconsistent records in the data, improve data quality, and obtain the cumulative value and change rate of the monitoring indicators;
[0024] The cumulative value is the sum of data from a certain starting point to the current time point, calculated by an accumulation operation;
[0025] The change rate is the ratio of the data difference between adjacent times to the time interval.
[0026] Optionally, the determination of the foundation pit safety level identification framework includes:
[0027] S21. Define the identification framework: The focal elements in the identification framework are set according to the requirements of the foundation pit safety level. For example, according to the safety evaluation standards and specifications, the general identification framework of the foundation pit safety level is set to Θ = {V1, V2, V3, V4} = {safe, relatively safe, relatively unsafe, unsafe}. Let m be the power set 2 of the identification framework Θ Θ A mapping on the interval [0,1], V is a proposition in the recognition framework; if it satisfies: m(Φ) = 0 and, then m is called 2 Θ The basic probability distribution function of Θ; m(V) is the basic probability distribution value of proposition V, which indicates the degree of trust m has in proposition V; if proposition V satisfies m(V)>0, then V is called a focal element on Θ.
[0028] S22, normalization processing: convert the original data of different monitoring indicators into dimensionless quantities, use the collected multi-source time series monitoring data of foundation pit as different evidences, and perform normalization processing, where the formula is:
[0029]
[0030] Where: is the value of the i-th monitoring indicator at time t; are the maximum and minimum values of the i-th monitoring indicator in the time 0-t, respectively; is the normalized monitoring data.
[0031] Optionally, the step of S3 includes:
[0032] S31, using a cloud model to determine the basic probability distribution values of different focal elements in the recognition framework;
[0033] The cloud model is a mathematical model based on probability theory. The core of the cloud model is to convert qualitative concepts into quantitative values and describe the uncertainty of a concept by generating clouds.
[0034] The generated cloud is a series of numerical points generated according to the digital features of the cloud model, and the numerical points represent the specific form of the qualitative concept in the quantitative space;
[0035] Among them, cloud models are established for the safety level intervals of multi-source monitoring indicators, and the digital characteristics of the models can be calculated using the forward cloud generator;
[0036] The digital features include the expected E y 、Entropy E n and super entropy H e , respectively represent the central position of the qualitative proposition, the uncertainty degree of the proposition and the uncertainty degree of entropy, and the expression is:
[0037]
[0038] H e,ij =k
[0039] Where: Indicates the value range of a focal element in the identification framework corresponding to a monitoring indicator;
[0040] Where: k is a constant, with a value of 0.01;
[0041] i=1,2,…,M;
[0042] j=1,2,…,N;
[0043] M is the number of multi-source monitoring indicators;
[0044] N is the number of focal elements in the identification frame;
[0045] S32. According to the time series monitoring data and the digital characteristics of the cloud model in different monitoring index classification intervals, the certainty of the jth focal element under the i-th monitoring index is calculated and used as the basic probability distribution value of the focal element:
[0046]
[0047] Where: Ey,ij Is E n,ij is the mean value, H e,ij is a normal random variable with standard deviation .
[0048] Optionally, the step of S4 includes:
[0049] S41. Weight calculation: According to the importance and reliability of the monitoring indicators, the weight of each indicator is calculated using the interquartile range method;
[0050] S42, credibility assessment: Evaluate the reliability of each monitoring data source, consider equipment accuracy, environmental factors, etc., and assign a credibility value to each data source;
[0051] The weight is the importance of the indicator, where the expression is:
[0052] Q d =Q3-Q1
[0053]
[0054] Where: Q1 and Q3 are the first quartile and the third quartile of the measured data of a monitoring indicator in the time 0-t respectively;
[0055] Q d is the interquartile range, which is used to describe the degree of dispersion of data. The larger its value, the greater the variability and the more information it provides;
[0056] w i is the importance weight of the multi-source monitoring indicator, i=1,2,…,M.
[0057] Optionally, the step of S5 includes:
[0058] S61. Evidence synthesis: Use the synthesis rules of Dempster-Shafer (DS) evidence theory to integrate evidence from different monitoring indicators and data sources;
[0059] S62, safety level determination: determine the final safety evaluation level of the foundation pit according to the fused basic probability allocation value BPA;
[0060] The basic probability assignment value BPA is used in DS evidence theory to indicate the basic degree of trust of each piece of evidence on each proposition in the identification framework;
[0061] The focal element with the largest basic probability allocation value BPA is taken as the final judgment result.
[0062] Optionally, the step of evidence fusion includes:
[0063] 1) Choose any two pieces of evidence Y i and Yj ;
[0064] 2) Let m i and m j are the basic probability assignment values in the corresponding recognition framework, i, j = 1, 2, ..., M;
[0065] 3) Treat the basic probability distribution values of different focal elements under the same evidence body as coordinate points;
[0066] 4) Calculate the Chebyshev distance between two evidence bodies;
[0067] d ij =max|m i -m j |,d ij ∈[0,1]
[0068]
[0069] Where: d ij Represents evidence body Y i With evidence body Y j The similarity between ij The larger the value, the greater the inconsistency between the two pieces of evidence. Conversely, the more similar the two pieces of evidence are. i With evidence body Y j The similarity between them is expressed as follows:
[0070] s ij =1-d ij
[0071]
[0072] Evidence Body Y i With evidence body Y j The credibility between them is expressed as:
[0073]
[0074] Where: Credibility C rd (m i )satisfy i,j=1,2,…,M;
[0075] Credibility indicates the degree to which a body of evidence is supported by other bodies of evidence. The higher the degree of support, the more credible the body of evidence.
[0076] Optionally, in step S6, the following general fusion coefficient is proposed according to the importance weight of the foundation pit multi-source monitoring indicators and the credibility of the evidence:
[0077] η i =α1w i+α2C rd (m i )
[0078] Where: α1, α2 are constants with values between 0 and 1, and satisfy α1+α2=1;
[0079] If α1>α2, the importance weight of the multi-source monitoring indicator is considered to be more important than the credibility between the evidences;
[0080] On the contrary, if α1 < α2, the credibility between evidences is considered to be more important than the importance weight of multi-source monitoring indicators;
[0081] If α1=1, it means that the fusion coefficient η i Only the importance of monitoring indicators is considered;
[0082] On the contrary, if α2 = 1, it means that the fusion coefficient η i Only the credibility of the evidence is considered and the importance of multiple source monitoring indicators is considered to be the same.
[0083] In summary, the values of constants α1 and α2 can be determined according to actual needs.
[0084] Optionally, using the fusion coefficient η i Modify the basic probability distribution values of different focal elements:
[0085]
[0086] Where: m i (V j ) is the lower focal element V of the i-th evidence body j The basic probability distribution value of , i = 1, 2, ..., M, j = 1, 2, ..., N;
[0087] According to the DS evidence theory synthesis rules, multiple evidence bodies are synthesized
[0088]
[0089] Where: Φ represents the empty set; K is the conflict coefficient between evidences, Indicates the degree of conflict between the evidence.
[0090] Based on the fusion results of multi-source monitoring data and the maximum attribute principle, the safety level of the foundation pit is determined and automatic early warning is achieved.
[0091] Technical effects and advantages of the present invention:
[0092] 1. By using the interquartile range method to calculate the weights of multi-source monitoring indicators and using the Chebyshev distance to evaluate the credibility of different evidence, the advantages of these two methods can be combined to improve the accuracy and robustness of the overall evaluation. By weight allocation, it can ensure that the importance of different monitoring indicators is reasonably reflected in subsequent analysis; and by credibility assessment, reliable sources of evidence can be screened out to avoid the impact of erroneous or misleading information on the analysis results.
[0093] 2. By adjusting the weights of multi-source monitoring indicators and the credibility of evidence, the basic probability distribution values of different focal elements are corrected. By introducing the weights of multi-source monitoring indicators, the importance and reliability of different data sources can be comprehensively considered, making the information fusion process more comprehensive and accurate; the fusion of multi-source data can reduce this bias and improve the accuracy of the overall information; through automated weight and credibility calculation methods, human intervention and subjective judgment can be reduced, the efficiency and accuracy of information fusion can be improved, and the revised BPA can be used as an important input of the intelligent decision-making system to support the system's automated decision-making and intelligent reasoning process.
[0094] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0095] Figure 1 It is a structural schematic diagram provided by the present invention;
[0096] Figure 2 It is a schematic diagram of a foundation pit safety level cloud model provided by an embodiment of the present invention;
[0097] Figure 3 It is a schematic diagram of the change of probability distribution value of propositions after fusion of different numbers of evidence bodies provided by an embodiment of the present invention;
[0098] Figure 4 It is a schematic diagram of a foundation pit safety evaluation index system provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0099] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0100] like Figures 1 to 4 As shown, an embodiment of the present invention provides a foundation pit safety early warning method by fusing multi-source monitoring data.
[0101] S1. According to the characteristics of foundation pit engineering and safety monitoring, a foundation pit safety evaluation index system is constructed to obtain the cumulative value and change rate of multi-source time series monitoring indicators;
[0102] S2. Determine the foundation pit safety level identification framework, use multi-source time series monitoring data as different evidence and perform normalization processing;
[0103] S3, determining the basic probability distribution values of different focal elements in the identification framework based on multi-source time series monitoring data;
[0104] S4. The interquartile range method is used to calculate the weights of multi-source monitoring indicators, and the Chebyshev distance is used to obtain the credibility of different evidences;
[0105] S5. According to the weights of multi-source monitoring indicators and the credibility of evidence, the basic probability distribution values of different focal elements are modified;
[0106] S6. Use the DS evidence theory synthesis rules to fuse the multi-source monitoring data of the foundation pit to obtain the foundation pit safety evaluation level.
[0107] Furthermore, the present invention constructs a foundation pit safety evaluation index system covering foundation pit engineering characteristics and safety monitoring, which can more accurately reflect the safety status of foundation pit engineering and provide a solid foundation for safety evaluation; when determining the foundation pit safety level identification framework, normalization processing is performed to ensure the comparability of multi-source time series monitoring data in different evidences, avoid errors caused by different data dimensions or magnitudes, and provide a reliable data basis for subsequent weight calculation and credibility evaluation;
[0108] The weights of multi-source monitoring indicators are calculated using the interquartile range method, which can better reflect the importance of each monitoring indicator in the overall evaluation system. At the same time, the credibility of different evidence is obtained through the Chebyshev distance, which further improves the accuracy of the evaluation, the rationality of the distribution and credibility evaluation, and makes the evaluation results more in line with the actual situation; the DS evidence theory synthesis rule is used to fuse the multi-source monitoring data of the foundation pit. This theory has significant advantages in dealing with uncertainty and conflicting information, and by fusing multi-source monitoring data, the impact of various factors on the safety of the foundation pit can be comprehensively considered, thereby obtaining a more comprehensive and accurate safety evaluation level;
[0109] The entire evaluation process is based on the cumulative value and change rate of the time series, which can reflect the changes in the safety status of the foundation pit project in real time. This real-time and dynamic nature enables the evaluation results to adjust and optimize the construction plan in a timely manner to ensure the safe progress of the foundation pit project; the final foundation pit safety evaluation level provides a scientific basis for project management and decision-making. Through the comprehensive evaluation results, managers can understand the safety status of the foundation pit project in a timely manner, take corresponding preventive measures and emergency responses, and ensure project quality and personnel safety.
[0110] The application of the above steps in foundation pit safety evaluation improves the accuracy and reliability of foundation pit safety evaluation by constructing a comprehensive evaluation index system, scientific data processing, reasonable weight distribution and credibility assessment, the integration advantages of evidence theory, and real-time and dynamic characteristics, providing strong support for the safety management of foundation pit projects.
[0111] Specifically, the construction of the foundation pit safety evaluation index system includes:
[0112] S11. Determine the evaluation index system and determine the key safety evaluation indicators based on the characteristics of the foundation pit project, such as geological conditions, support structure type, excavation depth, and surrounding environmental impact.
[0113] S12, data collection, real-time monitoring of various indicators through sensors to obtain time series data;
[0114] S13, data preprocessing, cleaning, denoising and interpolation of data to obtain cumulative values and change rates;
[0115] The determination index system is composed of surrounding environment monitoring indicators and monitoring indicators of multiple foundation pit sections;
[0116] The surrounding environment monitoring indicators include: cumulative value of building settlement, building settlement change rate, cumulative value of surface settlement, surface settlement change rate, cumulative value of pipeline settlement and pipeline settlement change rate;
[0117] The monitoring indicators of the foundation pit profile include: the cumulative value of the vertical displacement of the wall top, the rate of change of the vertical displacement of the wall top, the cumulative value of the horizontal displacement of the wall top, the rate of change of the horizontal displacement of the wall top, the cumulative value of the deep horizontal displacement, the rate of change of the deep horizontal displacement, the cumulative value of the internal force of the support, the cumulative value of the groundwater level and the rate of change of the groundwater level;
[0118] The sensors include displacement meters, stress meters and water level meters, which are used to measure the displacement of structures or soil, monitor stress changes inside structures, and observe fluctuations in groundwater levels.
[0119] The sensor is connected to a data acquisition unit for collecting sensor data in real time and storing it in a database for subsequent analysis and processing;
[0120] Furthermore, the sensor has the characteristics of high precision and high reliability, which can ensure the accuracy of monitoring data. This accuracy provides a guarantee for the accuracy of evaluation results and avoids evaluation deviations caused by data errors;
[0121] By using sensors to monitor various indicators in real time, real-time time series data can be obtained, so that the evaluation process can promptly reflect changes in the safety status of the foundation pit project, providing strong support for subsequent decision-making;
[0122] The use of automated monitoring reduces manual intervention and errors, improves monitoring efficiency and data quality. At the same time, automated monitoring can also achieve all-weather, uninterrupted monitoring, providing strong support for the safety management of foundation pit projects.
[0123] The data preprocessing is used to remove noise, incomplete or inconsistent records in the data, improve data quality, and calculate cumulative values and change rates;
[0124] The cumulative value is the sum of data from a certain starting point to the current time point, calculated by an accumulation operation;
[0125] The change rate is the ratio of the data difference between adjacent times to the time interval.
[0126] Furthermore, the accumulated values and rates of change are used in data analysis and engineering monitoring to provide key information about changes in the state of a system or structure;
[0127] The cumulative value represents the total change from the start of monitoring to the current time point, and is used to provide the overall deformation of the system during the entire monitoring period;
[0128] By observing the changing trend of the cumulative value, it is possible to determine whether the system or structure is in a stable state, or whether there are abnormal conditions such as accelerated deformation;
[0129] The rate of change indicates the amount of change of the monitoring indicator per unit time, providing direct information about the speed of change;
[0130] The magnitude of the rate of change is directly related to the stability of the system; a faster rate of change may mean that the system or structure is undergoing rapid deformation or instability;
[0131] By setting a reasonable change rate threshold, the deformation of the system or structure can be warned and monitored in real time. When the change rate exceeds the threshold, the alarm mechanism can be triggered through the fusion analysis of multi-source monitoring data to remind relevant personnel to take timely measures to prevent the situation from deteriorating;
[0132] In risk assessment, the rate of change is an important reference indicator. By comparing the rates of change in different time periods or locations, the risk level of the system or structure under different conditions can be evaluated and corresponding risk management strategies can be formulated.
[0133] Specifically, the identification framework for determining the foundation pit safety level includes:
[0134] S21. Define the identification framework: The focal elements in the identification framework are set according to the requirements of the foundation pit safety level. For example, according to the safety evaluation standards and specifications, the general identification framework of the foundation pit safety level is set to Θ = {V1, V2, V3, V4} = {safe, relatively safe, relatively unsafe, unsafe}. Let m be the power set 2 of the identification framework Θ Θ A mapping on the interval [0,1], V is a proposition in the recognition framework; if it satisfies: m(Φ) = 0 and, then m is called 2 Θ The basic probability distribution function of Θ; m(V) is the basic probability distribution value of proposition V, which indicates the degree of trust m has in proposition V; if proposition V satisfies m(V)>0, then V is called a focal element on Θ.
[0135] S22, normalization processing: convert the original data of different monitoring indicators into dimensionless quantities, use the collected multi-source time series monitoring data of foundation pit as different evidences, and perform normalization processing, where the formula is:
[0136]
[0137] Where: is the value of the i-th monitoring indicator at time t; are the maximum and minimum values of the i-th monitoring indicator in the time 0-t, respectively; is the normalized monitoring data.
[0138] Furthermore, after obtaining the normalized multi-source time series monitoring data, it can be input into the foundation pit safety level identification model as different evidences. The model can calculate the probability distribution value of each safety level according to the input evidence and the predefined identification framework, and output the final foundation pit safety level;
[0139] It should be noted that the construction and training of the foundation pit safety level identification model is a complex process that needs to be based on a large amount of historical data and expert experience. In practical applications, it is also necessary to make appropriate adjustments and optimizations based on specific engineering conditions;
[0140] By determining the foundation pit safety level identification framework, collecting multi-source time series monitoring data and normalizing them, strong data support is provided for the identification and early warning of foundation pit safety levels.
[0141] Specifically, the step of S3 includes:
[0142] S31, using a cloud model to determine the basic probability distribution values of different focal elements in the recognition framework;
[0143] The cloud model is a mathematical model based on probability theory. The core of the cloud model is to convert qualitative concepts into quantitative values and describe the uncertainty of a concept by generating clouds.
[0144] The generated cloud is a series of numerical points generated according to the digital features of the cloud model, and the numerical points represent the specific form of the qualitative concept in the quantitative space;
[0145] Among them, cloud models are established for the safety level intervals of multi-source monitoring indicators, and the digital characteristics of the models can be calculated using the forward cloud generator;
[0146] It should be noted that in the foundation pit safety warning, many factors such as geological conditions, excavation depth, support structure performance, etc. have different degrees of randomness and fuzziness. The cloud model uses its unique digital characteristics (expected E y 、Entropy E n and super entropy H e ) can simultaneously express the randomness and fuzziness of these factors, making the warning results more consistent with the actual situation;
[0147] The cloud model converts qualitative concepts into specific numerical representations in a quantitative way, thereby reducing the subjectivity and randomness of human judgment. In foundation pit safety warning, the cloud model can be used to determine the basic probability distribution values of different focal elements (such as different risk levels), which are more objective and accurate. By comprehensively considering the probability distribution values of multiple focal elements, a more accurate warning result can be obtained.
[0148] The digital features include the expected E y 、Entropy E n and super entropy H e , respectively represent the central position of the qualitative proposition, the uncertainty degree of the proposition and the uncertainty degree of entropy, and the expression is:
[0149]
[0150] H e,ij =k
[0151] Where: Indicates the value range of a focal element in the identification framework corresponding to a monitoring indicator;
[0152] Where: k is a constant, with a value of 0.01;
[0153] i=1,2,…,M;
[0154] j=1,2,…,N;
[0155] M is the number of multi-source monitoring indicators;
[0156] N is the number of focal elements in the identification frame;
[0157] S32. According to the time series monitoring data and the digital characteristics of the cloud model in different monitoring index classification intervals, the certainty of the jth focal element under the i-th monitoring index is calculated and used as the basic probability distribution value of the focal element:
[0158]
[0159] Where: E y,ij Is E n,ij is the mean value, H e,ij is a normal random variable with standard deviation .
[0160] Specifically, the step of S4 includes:
[0161] S41. Weight calculation: According to the importance and reliability of the monitoring indicators, the weight of each indicator is calculated using the interquartile range method;
[0162] S42, credibility assessment: Evaluate the reliability of each monitoring data source, consider equipment accuracy, environmental factors, etc., and assign a credibility value to each data source;
[0163] The weight is the importance of the indicator, where the expression is:
[0164] Q d =Q3-Q1
[0165]
[0166] Where: Q1 and Q3 are the first quartile and the third quartile of the measured data of a monitoring indicator in the time 0-t respectively;
[0167] Q d is the interquartile range, which is used to describe the degree of dispersion of data. The larger its value, the greater the variability and the more information it provides;
[0168] w i is the importance weight of the multi-source monitoring indicator, i=1,2,…,M.
[0169] Furthermore, the interquartile range not only provides a quantitative indicator of the degree of data dispersion, but also implies the distribution of the data. Through the interquartile range, we can understand the distribution of data in different intervals, and then judge the importance of each monitoring indicator in the data set, so as to assign reasonable weights to them. It should be noted that the interquartile range method determines the degree of data dispersion by calculating the difference between the third quartile and the first quartile. This method is less sensitive to outliers. Compared with statistics such as the mean or standard deviation, the interquartile range can better reflect the main distribution of the data and will not produce significant deviations due to a few outliers.
[0170] Furthermore, the Chebyshev distance focuses on the maximum coordinate difference between two points in the vector space, which can capture the maximum deviation in the data and ensure that any extreme cases that may lead to significant errors are not ignored when evaluating the credibility of evidence.
[0171] Specifically, the step of S5 includes:
[0172] S61. Evidence synthesis: Use the synthesis rules of Dempster-Shafer (DS) evidence theory to integrate evidence from different monitoring indicators and data sources;
[0173] S62, safety level determination: determine the final safety evaluation level of the foundation pit according to the fused basic probability allocation value BPA;
[0174] The basic probability assignment value BPA is used in DS evidence theory to indicate the basic degree of trust of each piece of evidence on each proposition in the identification framework;
[0175] The focal element with the largest basic probability allocation value BPA is taken as the final judgment result.
[0176] Specifically, the steps of evidence fusion include:
[0177] 1) Choose any two pieces of evidence Y i and Y j ;
[0178] 2) Let m i and m j are the basic probability assignment values in the corresponding recognition framework, i, j = 1, 2, ..., M;
[0179] 3) Treat the basic probability distribution values of different focal elements under the same evidence body as coordinate points;
[0180] 4) Calculate the Chebyshev distance between two evidence bodies;
[0181] d ij =max|m i -mj |,d ij ∈[0,1]
[0182]
[0183] Where: d ij Represents evidence body Y i With evidence body Y j The similarity between ij The larger the value, the greater the inconsistency between the two pieces of evidence. Conversely, the more similar the two pieces of evidence are. i With evidence body Y j The similarity between them is expressed as follows:
[0184] s ij =1-d ij
[0185]
[0186] Evidence Body Y i With evidence body Y j The credibility between them is expressed as:
[0187]
[0188] Where: Credibility C rd (m i )satisfy
[0189] Credibility indicates the degree to which a body of evidence is supported by other bodies of evidence. The higher the degree of support, the more credible the body of evidence.
[0190] It should be noted that in practical applications, there may be certain correlations or dependencies between different monitoring indicators and data sources. Therefore, it is necessary to evaluate the independence of the evidence before fusion and take appropriate methods to deal with non-independent evidence.
[0191] Furthermore, DS evidence theory can directly handle unknown states. By integrating evidence from different monitoring indicators and data sources, it can provide a more comprehensive understanding of the safety status of the foundation pit and reduce the uncertainty caused by a single data source. The integrated BPA provides a more accurate and comprehensive basis for the final safety evaluation level of the foundation pit. By integrating evidence from multiple data sources, the degree of dependence on a single data source can be reduced, thereby improving the robustness of the system. DS evidence theory can handle evidence from different monitoring indicators and data sources, including heterogeneous, incomplete or conflicting evidence. Through synthesis rules, these heterogeneous data can be integrated into a unified evaluation result, providing more comprehensive information support for foundation pit safety evaluation.
[0192] Specifically, in step S6, according to the importance weights of the multi-source monitoring indicators of the foundation pit and the credibility of the evidence, the following general fusion coefficient is proposed:
[0193] η i =α1w i +α2C rd (m i )
[0194] Where: α1, α2 are constants with values between 0 and 1, and satisfy α1+α2=1;
[0195] If α1>α2, the importance weight of the multi-source monitoring indicator is considered to be more important than the credibility between the evidences;
[0196] On the contrary, if α1 < α2, the credibility between evidences is considered to be more important than the importance weight of multi-source monitoring indicators;
[0197] If α1=1, it means that the fusion coefficient η i Only the importance of monitoring indicators is considered;
[0198] On the contrary, if α2 = 1, it means that the fusion coefficient η i Only the credibility of the evidence is considered and the importance of multiple source monitoring indicators is considered to be the same.
[0199] In summary, the values of constants α1 and α2 can be determined according to actual needs.
[0200] Specifically, using the fusion coefficient η i Modify the basic probability distribution values of different focal elements:
[0201]
[0202] Where: m i (V j ) is the lower focal element V of the i-th evidence body j The basic probability distribution value of , i = 1, 2, ..., M, j = 1, 2, ..., N;
[0203] According to the DS evidence theory synthesis rules, multiple evidence bodies are synthesized
[0204]
[0205] Where: Φ represents the empty set; K is the conflict coefficient between evidences, Indicates the degree of conflict between the evidence.
[0206] Based on the fusion results of multi-source monitoring data and the maximum attribute principle, the safety level of the foundation pit is determined and automatic early warning is achieved.
[0207] Furthermore, in order to verify the effectiveness of the improved evidence theory in this patent, evidence bodies under three working conditions are selected for fusion analysis. The basic probability distribution values of different evidence bodies are shown in Table 1. Evidence body Y2 is an interference item, which has a large conflict with other evidence bodies;
[0208] Table 1 Basic probability distribution values of evidence bodies under different working conditions
[0209]
[0210]
[0211] The traditional evidence theory, Murphy method, Deng method and the method of the present invention are used to perform data fusion analysis on Table 1. Assuming that the values of constants α1 and α2 are both 0.5, the fusion results obtained by different methods are shown in Table 2.
[0212] Due to its own defects, the traditional evidence theory is difficult to obtain fusion results that are consistent with the actual situation under the three working conditions, and the value of the conflict coefficient K gradually approaches 1, resulting in serious conflicts between evidence bodies. Therefore, the results calculated by the traditional evidence theory have large errors.
[0213] In addition, since m(V1)=0 in the body of evidence Y2, the "veto power" of traditional evidence theory means that even if the number of evidence bodies is increased in the subsequent fusion process, the support for proposition V1 will not increase.
[0214] The specific meaning of "veto power" is: when the basic probability allocation (BPA) of a certain body of evidence to a certain proposition (such as V1) is 0, that is, m(V1) = 0, then no matter how many bodies of evidence are added later, and these new bodies of evidence show that m(V1) ≠ 0, the support for proposition V1 will not increase;
[0215] Different from this, the fusion results obtained by the Murphy method, the Deng method and the method proposed in the present invention are all consistent with the actual results, and as the number of evidence bodies increases, the support of proposition V1 gradually increases.
[0216] Table 2 Comparison of fusion results obtained by different methods
[0217]
[0218]
[0219] The probability distribution value of proposition V1 after fusion of different numbers of evidence bodies changes as follows: Figure 3As shown. Under the same working conditions, the values of the conflict coefficient K obtained by the three methods gradually decrease, ensuring the accuracy of the fusion results. The above results show that the method proposed in the present invention is superior to other methods in terms of fusion effect, fast convergence speed and anti-interference ability.
[0220] The main features of the present invention are: on the basis of making full use of the measured monitoring data of the foundation pit, the importance of different monitoring indicators and the credibility of evidence are taken into account at the same time, which ensures the objectivity of the identification of the safety status of the foundation pit and avoids the interference of subjective human factors. It can effectively overcome the following defects of the traditional DS evidence theory: it cannot solve the situation of serious and complete conflict of evidence; the evidence has a veto power over the proposition; it is sensitive to the basic probability distribution and has poor anti-interference ability. Therefore, not only can the overall evaluation results of the safety status of the foundation pit be obtained, but also the local safety levels of different monitoring sections of the foundation pit and the surrounding environment can be obtained, providing a more reasonable way and more accurate evaluation results for the safety early warning of the foundation pit;
[0221] In summary, the present invention adopts the interquartile range method to calculate the weights of multi-source monitoring indicators, and uses the Chebyshev distance to evaluate the credibility of different evidences, which can combine the advantages of these two methods and improve the accuracy and robustness of the overall evaluation;
[0222] Through weight allocation, we can ensure that the importance of different monitoring indicators is reasonably reflected in subsequent analysis; and through credibility assessment, we can screen out reliable sources of evidence to avoid the impact of erroneous or misleading information on the analysis results;
[0223] By modifying the basic probability distribution values of different focal elements through the weights of multi-source monitoring indicators and the credibility of evidence, and introducing the weights of multi-source monitoring indicators, the importance and reliability of different data sources can be comprehensively considered, making the information fusion process more comprehensive and accurate; the fusion of multi-source data can reduce this deviation and improve the accuracy of the overall information;
[0224] Through the automated weight and credibility calculation method, human intervention and subjective judgment can be reduced, the efficiency and accuracy of information fusion can be improved, and the revised BPA can be used as an important input of the intelligent decision-making system to support the system's automated decision-making and intelligent reasoning process.
[0225] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A foundation pit safety early warning method based on multi-source monitoring data fusion, characterized by: The steps include: S1. According to the characteristics of foundation pit engineering and safety monitoring, a foundation pit safety evaluation index system is constructed to obtain the cumulative value and change rate of multi-source time series monitoring indicators; S2. Determine the foundation pit safety level identification framework, use multi-source time series monitoring data as different evidence and perform normalization processing; S3, determining the basic probability distribution values of different focal elements in the identification framework based on multi-source time series monitoring data; S4. The interquartile range method is used to calculate the weights of multi-source monitoring indicators, and the Chebyshev distance is used to obtain the credibility of different evidences; S5. According to the weights of multi-source monitoring indicators and the credibility of evidence, the basic probability distribution values of different focal elements are modified; S6. Use the DS evidence theory synthesis rules to fuse the multi-source monitoring data of the foundation pit to obtain the foundation pit safety evaluation level.
2. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The foundation pit safety evaluation index system comprises: S11. Determine the evaluation index system and determine the key safety evaluation indicators based on the characteristics of the foundation pit project, such as geological conditions, support structure type, excavation depth, and surrounding environmental impact; S12, data collection, real-time monitoring of various indicators through sensors to obtain time series data; S13, data preprocessing, cleaning, denoising and interpolation of data to obtain cumulative values and change rates; The determination index system is composed of surrounding environment monitoring indicators and monitoring indicators of multiple foundation pit sections; The surrounding environment monitoring indicators include: cumulative value of building settlement, building settlement change rate, cumulative value of surface settlement, surface settlement change rate, cumulative value of pipeline settlement and pipeline settlement change rate; The foundation pit profile monitoring indicators include: cumulative value of vertical displacement of wall top, rate of change of vertical displacement of wall top, cumulative value of horizontal displacement of wall top, rate of change of horizontal displacement of wall top, cumulative value of deep horizontal displacement, rate of change of deep horizontal displacement, cumulative value of support internal force, cumulative value of groundwater level and rate of change of groundwater level.
3. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 2 is characterized by: The sensors include displacement meters, stress meters and water level meters, which are used to measure the displacement of structures or soil, monitor stress changes inside structures, and observe fluctuations in groundwater levels. The sensor is connected to a data acquisition unit for collecting sensor data in real time and storing it in a database for subsequent analysis and processing; The data acquisition unit is provided with an alarm. By setting the early warning and alarm thresholds, when a monitoring indicator of a certain measuring point exceeds the set threshold, after the multi-source monitoring data of the measuring point is integrated and analyzed, that is, through the evidence of other monitoring indicators, the system can automatically determine whether it is a "false alarm" or a real alarm danger, to ensure that protective measures are taken in time; The data preprocessing is used to remove noise, incomplete or inconsistent records in the data, improve data quality, and obtain the cumulative value and change rate of the monitoring indicators; The cumulative value is the sum of data from a certain starting point to the current time point, calculated by an accumulation operation; The change rate is the ratio of the data difference between adjacent times to the time interval.
4. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The framework for determining the foundation pit safety level includes: S21. Define the identification framework: The focal elements in the identification framework are set according to the requirements of the foundation pit safety level. For example, according to the safety evaluation standards and specifications, the general identification framework of the foundation pit safety level is set to Θ = {V1, V2, V3, V4} = {safe, relatively safe, relatively unsafe, unsafe}. Let m be the power set 2 of the identification framework Θ Θ A mapping on the interval [0,1], V is a proposition in the recognition framework; if it satisfies: m(Φ) = 0 and, then m is called 2 Θ The basic probability distribution function of ; m(V) is the basic probability distribution value of proposition V, which indicates the degree of trust m has in proposition V; if proposition V satisfies m(V)>0, then V is called a focal element on Θ; S22, normalization processing: convert the original data of different monitoring indicators into dimensionless quantities, use the collected multi-source time series monitoring data of foundation pit as different evidences, and perform normalization processing, where the formula is: Where: is the value of the i-th monitoring indicator at time t; are the maximum and minimum values of the i-th monitoring indicator in the time 0-t, respectively; is the normalized monitoring data.
5. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The steps of S3 include: S31, using a cloud model to determine the basic probability distribution values of different focal elements in the recognition framework; The cloud model is a mathematical model based on probability theory. The core of the cloud model is to convert qualitative concepts into quantitative values and describe the uncertainty of a concept by generating clouds. The generated cloud is a series of numerical points generated according to the digital features of the cloud model, and the numerical points represent the specific form of the qualitative concept in the quantitative space; Among them, cloud models are established for the safety level intervals of multi-source monitoring indicators, and the digital characteristics of the models can be calculated using the forward cloud generator; The digital features include the expected E y 、Entropy E n and super entropy H e , respectively represent the central position of the qualitative proposition, the uncertainty degree of the proposition and the uncertainty degree of entropy, and the expression is: H e,ij =k Where: Indicates the value range of a focal element in a certain monitoring indicator corresponding to the identification framework; Where: k is a constant, with a value of 0.01; i=1,2,…,M; j=1,2,…,N; M is the number of multi-source monitoring indicators; N is the number of focal elements in the identification frame; S32. According to the time series monitoring data and the digital characteristics of the cloud model in different monitoring index classification intervals, the certainty of the jth focal element under the i-th monitoring index is calculated and used as the basic probability distribution value of the focal element: Where: E y,ij Is E n,ij is the mean value, H e,ij is a normal random variable with standard deviation .
6. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The steps of S4 include: S41. Weight calculation: According to the importance and reliability of the monitoring indicators, the weight of each indicator is calculated using the interquartile range method; S42, credibility assessment: Evaluate the reliability of each monitoring data source, consider equipment accuracy, environmental factors, etc., and assign a credibility value to each data source; The weight is the importance of the indicator, where the expression is: Q d =Q3-Q1 Where: Q1 and Q3 are the first quartile and the third quartile of the measured data of a monitoring indicator in the time 0-t respectively; Q d is the interquartile range, which is used to describe the degree of dispersion of data. The larger its value, the greater the variability and the more information it provides; w i is the importance weight of the multi-source monitoring indicator, i=1,2,…,M.
7. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The step of S5 comprises: S61. Evidence synthesis: Use the synthesis rules of Dempster-Shafer (DS) evidence theory to integrate evidence from different monitoring indicators and data sources; S62, safety level determination: determine the final safety evaluation level of the foundation pit according to the fused basic probability allocation value BPA; The basic probability assignment value BPA is used in DS evidence theory to indicate the basic degree of trust of each piece of evidence on each proposition in the identification framework; The focal element with the largest basic probability allocation value BPA is taken as the final judgment result.
8. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: The steps of evidence fusion include: 1) Choose any two pieces of evidence Y i and Y j ; 2) Let m i and m j are the basic probability assignment values in the corresponding recognition framework, i, j = 1, 2, ..., M; 3) Treat the basic probability distribution values of different focal elements under the same evidence body as coordinate points; 4) Calculate the Chebyshev distance between two evidence bodies; d ij =max|m i -m j |,d ij ∈[0,1] Where: d ij Represents evidence body Y i With evidence body Y j The similarity between ij The larger the value, the greater the inconsistency between the two pieces of evidence. Conversely, the more similar the two pieces of evidence are. i With evidence body Y j The similarity between them is expressed as follows: s ij =1-d ij Evidence Body Y i With evidence body Y j The credibility between them is expressed as: Where: Credibility C rd (m i )satisfy Credibility indicates the degree to which a body of evidence is supported by other bodies of evidence. The higher the degree of support, the more credible the body of evidence.
9. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 7 is characterized by: In step S6, according to the importance weights of the multi-source monitoring indicators of the foundation pit and the credibility of the evidence, the following general fusion coefficient is proposed: or i =α1w i +α2C rd (m i ) Where: α1, α2 are constants with values between 0 and 1, and satisfy α1+α2=1; If α1>α2, the importance weight of the multi-source monitoring indicator is considered to be more important than the credibility between the evidences; On the contrary, if α1 < α2, the credibility between evidences is considered to be more important than the importance weight of multi-source monitoring indicators; If α1=1, it means that the fusion coefficient η i Only the importance of monitoring indicators is considered; On the contrary, if α2 = 1, it means that the fusion coefficient η i Considering only the credibility of the evidence, the importance of multiple monitoring indicators is considered to be the same; In summary, the values of constants α1 and α2 can be determined according to actual needs.
10. The foundation pit safety early warning method based on multi-source monitoring data fusion according to claim 1 is characterized by: Using the fusion coefficient η i Modify the basic probability distribution values of different focal elements: Where: m i (V j ) is the lower focal element V of the i-th evidence body j The basic probability distribution value of , i = 1, 2, ..., M, j = 1, 2, ..., N; According to the DS evidence theory synthesis rules, multiple evidence bodies are synthesized Where: Φ represents the empty set; K is the conflict coefficient between evidences, Indicates the degree of conflict between the evidence; Based on the fusion results of multi-source monitoring data and the maximum attribute principle, the safety level of the foundation pit is determined and automatic early warning is achieved.
Citation Information
Patent Citations
Target identification method based on time domain evidence fusion
CN116484311A
Dam safety assessment method based on mass monitoring inspection information fusion model
CN117574321A
Intelligent construction site safety risk evaluation method based on adaptive weighted fusion
CN117893010A
Cited By
Comprehensive pipe gallery underground diaphragm wall deformation risk early warning system
CN119287995A
An early warning system for the deformation risk of the diaphragm wall of an integrated utility tunnel
CN119287995B
Multi-source information fusion early warning system and method for unfavorable geology of tunnel
CN119692765A
Traffic network congestion identification method based on data fusion
CN119811096A
Coal mine monitoring system measuring point compliance analysis method and device and electronic equipment
CN120687927A
