Flow-state solidified soil durability evaluation method and system based on data analysis
By defining dynamic active density and constructing a three-dimensional evaluation model, the problems of spatial representativeness and dynamic tracking in the durability evaluation of fluidized solidified soil are solved, enabling accurate evaluation and prediction of the durability of fluidized solidified soil and improving the accuracy and practicality of the evaluation.
Patent Information
- Application Number
- CN202511180604.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-22
- Publication Date
- 2025-10-24
AI Technical Summary
Existing technologies for assessing the durability of fluidized solidified soil suffer from problems such as insufficient spatial representativeness, weak dynamic tracking capability, single assessment index, and fixed prediction cycle, resulting in low assessment accuracy and difficulty in distinguishing between spurious durability and true durability.
By using data analysis methods, a dynamic activity density is defined, a three-dimensional evaluation model is constructed, and an activity conduction dataset is combined to dynamically track changes in active substances, optimize the prediction cycle, capture early degradation signals, and improve evaluation accuracy.
It enables precise three-dimensional assessment of the durability of fluidized solidified soil, distinguishes false durability, improves prediction success rate, shortens assessment cycle, and enhances assessment accuracy and practicality.
Smart Images
Figure CN120833876A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of flow state solidified soil, in particular to a flow state solidified soil durability evaluation method and system based on data analysis. BACKGROUND
[0002] Flow state solidified soil, as a new type of engineering material formed by mixing soil, cementing material (such as cement, slag) and water, is widely used in roadbed filling, slope reinforcement, underground engineering seepage prevention and other fields due to its good flowability and controllable strength. Its durability (such as long-term strength retention, impermeability, and erosion resistance) directly determines the service life of the engineering structure and is the core indicator of engineering safety. However, the existing flow state solidified soil durability evaluation technology has the following limitations: Insufficient spatial representativeness: Traditional methods rely on laboratory sampling or single-point sensor monitoring, which is difficult to reflect the non-uniformity of material properties in three-dimensional space (such as differences in active substance distribution and strength gradient), and is prone to ignore local deterioration.
[0003] Weak dynamic tracking capability: The durability of flow state solidified soil is closely related to the hydration reaction of active substances (such as cement clinker), and the consumption of active substances over time will lead to dynamic changes in performance. Existing technologies mostly use fixed period detection, which cannot capture the correlation between active substance consumption and performance degradation in real time.
[0004] Single evaluation index: Existing methods mostly use compressive strength and other macroscopic properties as the only evaluation index, without considering the remaining amount of active substances (the material basis of durability), making it difficult to distinguish between "false durability" (short-term strength meets standards but active substances are exhausted) and "true durability" (active substances are sufficient and strength is stable).
[0005] Fixed prediction period: Existing prediction models use a uniform evaluation period (such as once a month), without considering the differences in degradation rates in different regions, resulting in low prediction accuracy (such as missing the maintenance opportunity due to the long period in regions with accelerated degradation).
[0006] Therefore, there is an urgent need for a flow state solidified soil durability evaluation technology that can combine dynamic changes in active substances, achieve accurate evaluation in three-dimensional space, and adaptively predict the period. SUMMARY
[0007] The present application aims to provide a flow state solidified soil durability evaluation method and system based on data analysis to solve the problems raised in the background technology. Specifically, the present application is based on the three-dimensional coupling logic of "material basis - spatial distribution - time evolution", and the core technical idea includes: The dynamic activity density is defined as "total amount of active substances in the sample domain / (volume*reference active substance amount)", directly reflecting the material basis (active substances are the core of strength formation and maintenance) of the durability of the flow state solidified soil, solving the problem of disconnection between the traditional index and the material essence, and realizing physical correlation of the dynamic activity density; The dynamic activity density difference between the sample domains (spatial transmission vector) is quantified by the active transmission data set, a three-dimensional evaluation model is constructed, the spatial radiation range of the scale representing the active substances and the performance parameters is influenced, the regional transmission characteristic value (dynamic activity density weighted compressive strength average) can comprehensively reflect the performance correlation of the local and the surrounding, and spatial transmission modeling is realized; The change rate of the regional transmission characteristic value is used to divide the degradation state (slowdown, stability, acceleration), the whole cycle dynamic of the hydration (performance improvement) and the degradation (performance attenuation) of the active substances is captured, the one-sidedness of the static evaluation is avoided, and dynamic representation of the degradation state is realized; The prediction success rate of different periods is iteratively evaluated, the optimal period is selected, the prediction frequency is matched with the actual degradation speed (for example, the period is shortened in the accelerated degradation area), and the prediction practicability is improved.
[0008] In order to solve the above technical problems, the technical scheme is provided as follows: A durability evaluation system of flow state solidified soil based on data analysis, the system comprises a data acquisition module, a data preprocessing module, a model construction module and an evaluation and prediction module; The data acquisition module collects material characteristic parameters through the durability detection sensor buried in the flow state solidified soil and uploads the material characteristic parameters to the data processing center; the monitoring area is divided into sample domains, and the material characteristic parameters in each sample domain are received; The data preprocessing module is used for calculating the dynamic activity density of the sample domain, generating the time-sharing activity density data set and the active transmission data set under the sampling time node based on the dynamic activity density; The model construction module is used for constructing a three-dimensional evaluation model, quantifying the influence scale based on the active transmission data set, and weightedly calculating the regional transmission characteristic value; The evaluation and prediction module analyzes the durability degradation state based on the regional transmission characteristic value, and iteratively evaluates the optimal prediction period.
[0009] Further, the data acquisition module comprises a parameter monitoring unit and a space division unit; The parameter monitoring unit collects material characteristic parameters through the sensor, and associates the sensor code with the spatial position; The space division unit is used for constructing a three-dimensional model of the monitoring area, dividing the sample domains, and synchronizing the material characteristic parameter acquisition and the sample domain attribution.
[0010] Further, the data preprocessing module comprises an activity density calculation unit and a gradient data set generation unit; The activity density calculation unit obtains a dynamic activity density by combining the total amount of active substances in the sample domain, the sample domain volume and the reference active substance amount; The gradient data set generation unit is configured to generate a time-division activity density data set at a sampling time node and construct an activity conduction data set based on the activity density difference between sample domains.
[0011] Further, the model construction module comprises a scale quantification unit and a characteristic value calculation unit; The scale quantification unit is configured to fit an influence scale of the sample domain with the dynamic activity density as the center; The characteristic value calculation unit is configured to calculate the compressive strength of the material in the circular domain as the regional conduction characteristic value by weighting with the influence scale as the radius.
[0012] Further, the evaluation and prediction module comprises a degradation state analysis unit and a cycle optimization unit; The degradation state analysis unit quantifies the durability degradation state based on the regional conduction characteristic value change rate; The cycle optimization unit is configured to generate a degradation state set, iteratively evaluate the optimal prediction cycle and output the evaluation result.
[0013] A flow state solidified soil durability evaluation method based on data analysis, the method comprising the following steps: Step S1: collecting material characteristic parameters of the flow state solidified soil through a durability detection sensor; dividing the three-dimensional space of the monitoring area into sample domains and receiving the material characteristic parameters collected by the sensor in the sample domains; Step S2: obtaining the dynamic activity density of the sample domain by calculating the total amount of active substances in the sample domain, the sample domain volume and the reference active substance amount, analyzing and generating the time-division activity density data set and the activity conduction data set at the sampling time node based on the dynamic activity density; Step S3: constructing a three-dimensional evaluation model, analyzing and quantifying the influence scale based on the activity conduction data set to generate the regional conduction characteristic value; Step S4: analyzing and characterizing the durability degradation state based on the regional conduction characteristic value, and analyzing the optimal prediction cycle based on the degradation state.
[0014] Further, the specific implementation process of step S1 comprises: The sensor for durability detection is embedded in the flow state solidified soil structure, and the material characteristic parameters are uploaded to the data processing center through a wireless network mode; the sensor is set with a unique code, and a storage unit is allocated to the data processing center based on the sensor code, which is used to store the material characteristic parameters collected by the corresponding sensor, including dynamic active density and material compressive strength; The flow state solidified soil is laid in the monitoring area, a three-dimensional space model of the monitoring area is established, and the monitoring area is divided in volume to form a sample domain, wherein one three-dimensional volume corresponds to one sample domain, and the data processing center receives the material characteristic parameters collected by the sensor in the sample domain in real time and distributes them to the storage unit.
[0015] Further, the specific implementation process of step S2 includes: The sampling time nodes of all sensors are synchronized, and the dynamic active density is stored in time according to the sampling time nodes to form a time-sharing active density data set, denoted as , wherein, represents the dynamic active density of the i-th sample domain, I represents the total number of sample domains, t is the sampling time node number, and , represents the total amount of active substances in the i-th sample domain, represents the volume value of the i-th sample area, is the preset reference active substance amount per unit volume of flow state solidified soil; The i-th sample domain is selected as the monitoring center of the monitoring area, and the j-th sample domain and the i-th sample domain form a spatial conduction vector in the monitoring area, denoted as , all spatial conduction vectors of the i-th sample domain at the t-th sampling time node are collected to form an active conduction data set, denoted as , wherein, represents the spatial conduction vector corresponding density difference, and , wherein, represents the dynamic active density of the j-th sample domain, and max() represents the maximum value function.
[0016] Further, the specific implementation process of step S3 includes: When the i-th sample domain is taken as the monitoring center of the monitoring area, the influence scale of the i-th sample domain is quantified based on the active conduction data set , wherein, min() is the minimum value function, and ceil{} is the ceiling function; , wherein, min() is the minimum value function, and ceil{} is the ceiling function; The influence scale is taken as the monitoring area radius, and the i-th sample domain is taken as the monitoring area center, and the circular domain The material compressive strength collected by all sensors inside the region is combined with the dynamic activity density The weighted calculation region conduction characteristic value , wherein k is the number of the sample region, U is a circular region , wherein the sample region contained in the circular region constitutes a set, represents the material compressive strength of the kth sample region.
[0017] Further, the specific implementation process of the step S4 includes: Based on the region conduction characteristic value, the durability degradation state value of the ith sample region at the tth sampling time node is evaluated : If , then , which represents that the durability of the ith sample region at the tth sampling time node has a degradation slowing trend state; If , then , which represents that the durability of the ith sample region at the tth sampling time node has a degradation stable trend state; If , then , which represents that the durability of the ith sample region at the tth sampling time node has a degradation accelerating trend state; Wherein, , and are respectively preset degradation state threshold intervals; The durability degradation state values at each sampling time node are collected to form a degradation state set of the ith sample region , wherein T represents the current sampling time node; The evaluation cycle f is initialized, and the prediction success rate of the evaluation cycle f is quantified , wherein , if , then , otherwise ; Let f=f+1, the prediction success rate of the evaluation cycle f is iteratively quantified, and when f=F, the iteration stops, and the evaluation cycle corresponding to the maximum prediction success rate is selected as the optimal prediction cycle, and F represents a preset maximum value of the evaluation cycle.
[0018] Compared with the prior art, the present application has the beneficial effects that: The monitoring region is divided into three-dimensional sample regions, the performance correlation between the sample regions is captured in combination with the activity conduction data set, the region conduction characteristic value can represent the comprehensive state of the local and surrounding areas, so that the spatial evaluation error is reduced; Real-time tracking of active material changes through time-division active density data sets, combined with dynamic analysis of regional conduction eigenvalues, can capture early degradation signals (such as sudden increase in active material consumption rate), making the capture response speed of regional conduction characteristics improve; The regional conduction eigenvalue couples dynamic active density (material basis) with compressive strength (macroscopic performance), which can distinguish "false durability" to improve the evaluation accuracy; Iterative optimization of the prediction period shortens the evaluation period of the degradation acceleration region (such as from 30 days to 10 days), to improve the prediction success rate. BRIEF DESCRIPTION OF DRAWINGS
[0019] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.
[0020] Figure 1 is a step schematic diagram of a data analysis-based flow state solidified soil durability evaluation method of the present application. DETAILED DESCRIPTION
[0021] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.
[0022] In the first embodiment: a data analysis-based flow state solidified soil durability evaluation system is provided, which comprises a data acquisition module, a data preprocessing module, a model construction module and an evaluation and prediction module; The data acquisition module acquires material characteristic parameters through the durability detection sensor embedded in the flow state solidified soil and uploads them to the data processing center; the monitoring area is divided into sample domains, and the material characteristic parameters in each sample domain are received; The data acquisition module comprises a parameter monitoring unit and a space division unit. The parameter monitoring unit acquires material characteristic parameters through sensors and associates sensor codes with spatial positions; The space division unit is used to construct a three-dimensional model of the monitoring area, divide sample domains, and synchronize material characteristic parameter acquisition and sample domain attribution; The data preprocessing module is used to calculate the dynamic active density of the sample domain, generate time-division active density data sets and active conduction data sets under the sampling time node based on the dynamic active density; The data preprocessing module comprises an active density calculation unit and a gradient data set generation unit. The active density calculation unit obtains the dynamic active density by combining the total amount of active substances in the sample domain, the volume of the sample domain, and the reference active amount. The gradient data set generation unit generates a time-dependent active density data set at a sampling time node and constructs an active conduction data set based on the difference in active density between sample domains. The model construction module constructs a three-dimensional evaluation model, quantifies the influence scale based on the active conduction data set, and calculates the regional conduction characteristic value by weighting. The model construction module includes a scale quantization unit and a characteristic value calculation unit. The scale quantization unit fits the influence scale of the sample domain with the dynamic active density as the center. The characteristic value calculation unit calculates the compressive strength of the material in the circular domain as the regional conduction characteristic value by weighting with the influence scale as the radius. The evaluation and prediction module analyzes the durability degradation state based on the regional conduction characteristic value and iteratively evaluates to obtain the optimal prediction period. The evaluation and prediction module includes a degradation state analysis unit and a period optimization unit. The degradation state analysis unit quantifies the durability degradation state based on the change rate of the regional conduction characteristic value. The period optimization unit generates a degradation state set, iteratively evaluates the optimal prediction period, and outputs the evaluation result.
[0023] Please refer to Figure 1 In this embodiment two: a data analysis-based flow state solidified soil durability evaluation method is provided, which is applicable to the above-mentioned embodiment one, and the method includes the following steps: Step S1: Collect the material property parameters of the flow state solidified soil through the durability detection sensor; divide the sample domain in the three-dimensional space of the monitoring area, and receive the material property parameters collected by the sensor in the sample domain. For example, the durability detection sensor is buried in the flow state solidified soil structure, and the material property parameters are uploaded to the data processing center through a wireless network; a unique code is set for the sensor, and a storage unit is allocated to the data processing center based on the sensor code for storing the material property parameters collected by the corresponding sensor, including dynamic active density and material compressive strength. The flow state solidified soil is laid in the monitoring area, a three-dimensional space model of the monitoring area is established, and the monitoring area is divided in volume to form a sample domain, wherein one solid volume corresponds to one sample domain, and the data processing center receives the material property parameters collected by the sensor in the sample domain in real time and allocates them to the storage unit. For example, a certain highway subgrade project uses fluidified soil (cement content 10%) as the base filling material, the monitoring area is a 20m (long) x 10m (wide) x 1m (thick) subgrade section, and the durability degradation trend during the service period needs to be evaluated to guide maintenance decisions; Sensor deployment, 30 durability detection sensors are buried in the monitoring area (1 per 5m³), the sensors have unique codes (such as "C1-5" represents the 1st row, 5th column sample area), the collected parameters include: total amount of active substances (kg), compressive strength (MPa), and are uploaded to the data processing center through a wireless network; The monitoring area is divided into 40 three-dimensional sample areas (5m x 5m x 0.5m, volume 12.5m³ each), and the data processing center stores the parameters according to the sample area (such as the data of sensor "C1-5" belongs to the 3rd sample area).
[0024] Step S2: Calculate the dynamic activity density of the sample area based on the total amount of active substances in the sample area, the volume of the sample area, and the reference amount of active substances, and based on the dynamic activity density, analyze and generate the time-sharing activity density data set and the activity conduction data set at the sampling time node; For example, synchronize the sampling time nodes of all sensors, and store the dynamic activity density according to the sampling time nodes to form a time-sharing activity density data set, denoted as , where represents the dynamic activity density of the i-th sample area, I represents the total number of sample areas, t is the sampling time node number, and , represents the total amount of active substances in the i-th sample area, represents the volume value of the i-th sample area, is the pre-set reference amount of active substances per unit volume of fluidified soil; Select the i-th sample area as the monitoring center of the monitoring area, and form a spatial conduction vector between the j-th sample area and the i-th sample area in the monitoring area, denoted as Collect all spatial conduction vectors of the i-th sample area at the t-th sampling time node to form an activity conduction data set, denoted as , where represents the spatial conduction vector corresponding to the density difference, and , where represents the dynamic activity density of the j-th sample area, and max() represents the maximum value function; For example, taking the 8th sample field (i = 8) as an example, at t = 30 days (sampling time node), the total amount of active substance M8 collected by the sensor is 250 kg, the volume V8 is 12.5 m³, the reference active amount A0 is 20 kg / m³ (design value), and the dynamic active density A8 is 250 / (12.5*20) = 1.0 (actual physical meaning: the current activity is 100% of the initial state, and the active substance is not significantly consumed). Taking i = 8 as the center, the spatial conduction vector g(ij) of the j = 9th sample field is calculated, where A9 = 0.9, then: g(ij) = (max(1.0, 0.9) - 0.9) / 1.0 = 0.1 (reflecting the activity difference of j field relative to i field).
[0025] Step S3: Constructing a three-dimensional evaluation model, based on the active conduction data set, analyzing and quantifying the influence scale to generate regional conduction characteristic values; For example, taking the i-th sample field as the monitoring center of the monitoring area, based on the active conduction data set , the influence scale of the i-th sample field is quantified , where min() is the minimum value function, and ceil{} is the rounding up function. Taking the influence scale as the monitoring area radius and the i-th sample field as the monitoring area center, all the material compressive strengths collected by the sensors within the circular domain are retrieved from the data processing center, and the dynamic active density is combined to calculate the regional conduction characteristic value , where k is the number of sample fields, U is the set of sample fields contained in the circular domain , and represents the material compressive strength of the k-th sample field. For example, based on G(A8), max(G) = 0.3 (difference with edge field), min(G) = 0.05 (difference with adjacent center field), and sample field total number I = 40, then: S(8) = ceil(40*(0.3-0.05) / (0.3+0.05)) = ceil(40*0.25 / 0.35) = ceil(28.57) = 29 (unit: sample field radius, representing the influence range). The circular domain with a radius of S(8) = 29 contains 12 sample fields (U = {8, 9, 10,..., 19}), and the dynamic active density and compressive strength data are as follows: Table 1 Dynamic active density and compressive strength data table
[0026] Summation Σ(A k ×Pk )= 32.4,∑A k = 10.8, then the area conduction eigenvalue: 3.0 MPa.
[0027] Step S4: Based on the area conduction eigenvalue, analyze and characterize the degradation state of durability, and based on the degradation state, analyze to obtain the optimal prediction period; For example, based on the area conduction eigenvalue, the durability degradation state value of the ith sample area at the tth sampling time node is evaluated : If , then let , which indicates that the durability of the ith sample area at the tth sampling time node has a degradation slowing trend; If , then let , which indicates that the durability of the ith sample area at the tth sampling time node has a degradation stabilizing trend; If , then let , which indicates that the durability of the ith sample area at the tth sampling time node has a degradation accelerating trend; Wherein, , and are preset degradation state threshold intervals; Collect the durability degradation state values at each sampling time node to form the degradation state set of the ith sample area , wherein T represents the current sampling time node; Initialize the evaluation period f, and quantify the prediction success rate of the evaluation period f , wherein , if , then let , otherwise let ; Let f = f + 1, iterate the prediction success rate of the evaluation period f, and when f = F, the iteration stops, and the evaluation period corresponding to the maximum prediction success rate is selected as the optimal prediction period, and F represents the maximum value of the preset evaluation period; For example, degradation state judgment: t = 30 days and t = 60 days, E30[8, 29] = 3.0 MPa, E60[8, 29] = 3.2 MPa, calculate the change rate: (3.2 / 3.0)-1≈0.067 (6.7%); Preset threshold interval: α1=(5%, +∞) (degradation slowing), α2=[-5%, 5%] (stable), α3=(-∞, -5%) (accelerating), so 6.7%∈α1, D30(8)=-1 (degradation slowing, due to the continuous hydration of the active substance, the strength is improved); For example, after the optimal prediction period f is selected, if the current time node is T, and there is a stable trend of degradation state in a sample field at the current time node, it can be predicted that the sample field at T+f steps is stable in the degradation state.
[0028] It should be noted that "false durability" refers to the phenomenon that the short-term macro performance (such as compressive strength) of the flow solidified soil meets the standard, but the active substance (the material basis of durability) has been largely consumed, resulting in rapid degradation of the later performance. The present application realizes accurate differentiation of "false durability" through three-layer logic design of the algorithm model, and the core relies on the coupling calculation of dynamic activity density and regional conduction characteristic value, as follows: The dynamic activity density directly reflects the relative remaining amount of active substance, and is the core index of durability "sustainability". If Ai is too low (such as <0.3), it means that the active substance is close to exhaustion, and even if the current strength meets the standard, it will also rapidly degrade in the later period due to the lack of hydration reaction support; The strength is weighted by the dynamic activity density, so that the characteristic value of the "high activity + high strength" region is significantly higher than that of the "low activity + high strength" region, which is reflected in: If a region Pk=3.5MPa (meets the standard), but Ak=0.2 (active is very low), then Ak×Pk=0.7, which has low weight in the regional characteristic value; If another region Pk=3.2MPa (slightly low), but Ak=0.9 (active is sufficient), then Ak×Pk=2.88, which has high weight in the regional characteristic value; Through this weighting, the regional conduction characteristic value can effectively distinguish between "false durability" (low characteristic value) of "high strength but active exhaustion" and "real durability" (high characteristic value) of "slightly low strength but active sufficient", solving the misjudgment problem of traditional single strength index; The change rate of the regional conduction characteristic value is used to judge the degradation state: If the "false durability" region (Ak is low but Pk is high), the subsequent E(t+1) << E(t) will appear due to active exhaustion, and the change rate falls into α3 (degradation acceleration interval), which is marked as Dt_i=1; If the "real durability" region (Ak is high and Pk is stable), then E(t+1) ≈ E(t) or slightly improves, and the change rate falls into α2 (stable) or α1 (slow down), which is marked as Dt_i=0 or -1; further expressing the unsustainability of "false durability", so as to improve the evaluation accuracy; The core of the iterative optimization prediction cycle is to automatically match the evaluation frequency and the area degradation speed through an algorithm. The degradation acceleration area will be given a shorter cycle in the iteration because of the fast state change. The specific dependence is the degradation state set and the prediction success rate calculation logic in step S4, such as the high frequency of "1" in Yi of the degradation acceleration area and the fast state switching (such as from 0→1 only 2 time nodes), and the long duration of "0" in Yi of the stable area; At the same time, if f is too long (such as 30 days), the difference between Dt and Dt+f is large (such as 0 at t and 1 at t+30), the proportion of Q=0 is high, and Pf is low; if f is shortened (such as 10 days), the difference between Dt and Dt+f is small (such as 0 at t and still 0 or just changed to 1 at t+10), the proportion of Q=1 is high, and Pf is significantly improved; for the degradation acceleration area, shorter f (such as 10 days) in the iteration will show higher Pf.
[0029] It should be noted that, in this document, the terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.
[0030] Finally, it should be noted that: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A data analysis-based method for evaluating the durability of a fluidized soil solidified soil, characterized by comprising: The method comprises the following steps: Step S1: collecting material characteristic parameters of the flow state solidified soil through the durability detection sensor; dividing the three-dimensional space of the monitoring area into sample domains, and receiving the material characteristic parameters collected by the sensor in the sample domains; Step S2: obtaining the dynamic activity density of the sample domain through the calculation of the total amount of active substances in the sample domain, the volume of the sample domain, and the reference active substance amount, and analyzing and generating the time-sharing activity density dataset and the activity conduction dataset at the sampling time node based on the dynamic activity density; Step S3: constructing a three-dimensional evaluation model, analyzing and quantifying the influence scale based on the activity conduction dataset, and generating the regional conduction characteristic value; Step S4: analyzing and characterizing the degradation state of the durability based on the regional conduction characteristic value, and analyzing the optimal prediction period based on the degradation state.
2. The method for durability evaluation of fluidized solidified soil based on data analysis according to claim 1, characterized in that, The specific implementation process of the step S1 comprises: The durability detection sensor is buried in the flow state solidified soil structure, and the material characteristic parameters are uploaded to the data processing center through a wireless network; a unique code is set for the sensor, and a storage unit is allocated to the data processing center based on the sensor code, which is used to store the material characteristic parameters collected by the corresponding sensor, including the dynamic activity density and the material compressive strength; The flow state solidified soil is laid in the monitoring area, a three-dimensional space model of the monitoring area is established, and the monitoring area is divided in volume to form sample domains, wherein one solid volume corresponds to one sample domain, and the data processing center receives the material characteristic parameters collected by the sensor in the sample domains in real time and distributes them to the storage unit.
3. The method for durability assessment of fluidized soil solidified by data analysis according to claim 2, characterized in that, The specific implementation process of the step S2 comprises: The sampling time nodes of the sensors are synchronized, and the dynamic activity density is stored according to the sampling time nodes to form a time-division activity density data set, denoted as wherein, represents the dynamic activity density of the i th sample field, I represents the total number of sample fields, t is the serial number of the sampling time node, and , represents the total amount of active substances in the i th sample field, represents the volume value of the i th sample field, is the preset reference active substance amount of the fluidified soil per unit volume. The i-th sample field is selected as a monitoring center of a to-be-monitored center in a monitoring area, and the j-th sample field and the i-th sample field form a spatial transmission vector in the monitoring area, denoted as All spatial transmission vectors of the i-th sample field at the t-th sampling time node are collected to form an active transmission data set, denoted as , wherein represents the spatial transmission vector corresponding density difference, and , wherein represents the dynamic activity density of the j-th sample field, and max() represents a maximum value function.
4. The method for durability evaluation of fluidized solidified soil based on data analysis according to claim 3, characterized in that, The specific implementation process of the step S3 comprises: For the i-th sample field as the monitoring area of the to-be-monitored center, based on the active conduction data set , quantifying the impact scale of the i-th sample field , wherein min() is a minimum value function, and ceil{} is a rounding up function. with the influence scale The radius of the monitoring area is taken as the scale, and the i-th sample area is taken as the center of the monitoring area circle The compressive strength of the material collected by all sensors in the circular area is called from the data processing center The weighted calculation area conduction characteristic value is calculated , where k is the number of sample areas, U is the set of sample areas contained in the circular area , and represents the compressive strength of the material of the k-th sample area.
5. The method for durability assessment of fluidized solidified soil based on data analysis according to claim 4, characterized in that, The specific implementation process of the step S4 comprises: Based on the regional conduction eigenvalue, the durability degradation state value of the i-th sample region at the t-th sampling time node is evaluated : If , let , represents the durability degradation slowing trend state of the i-th sample field at the t-th sampling time node. If , then let , indicates the i-th sample field in the t-th sampling time node under the durability degradation stability trend state; If , let , represents the durability degradation acceleration trend state of the i-th sample field at the t-th sampling time node. wherein, , and are preset degradation state threshold intervals, respectively. Collect the durability degradation state values at each sampling time node to form a degradation state set of the i-th sample domain wherein T represents the current sampling time node; initializing an evaluation period f, quantifying a prediction success rate of the evaluation period f , where , if , then let , otherwise let ; Let f=f+1, and iteratively quantify the prediction success rate of the evaluation period f, and when f=F, the iteration stops, and the evaluation period corresponding to the maximum prediction success rate is selected as the optimal prediction period, and F represents the maximum value of the preset evaluation period.
6. A data analysis-based flow state solidified soil durability evaluation system that executes the data analysis-based flow state solidified soil durability evaluation method according to any one of claims 1 to 5, characterized by, The system comprises a data acquisition module, a data preprocessing module, a model construction module, and an evaluation and prediction module; The data acquisition module collects material characteristic parameters through the durability detection sensor buried in the flow state solidified soil and uploads them to the data processing center; the monitoring area is divided into sample domains, and the material characteristic parameters in each sample domain are received; The data preprocessing module is used for calculating the dynamic activity density of the sample domain, and generating the time-sharing activity density dataset and the activity conduction dataset at the sampling time node based on the dynamic activity density; The model construction module is used for constructing a three-dimensional evaluation model, quantifying the influence scale based on the activity conduction dataset, and weightedly calculating the regional conduction characteristic value; The evaluation and prediction module analyzes the durability degradation state based on the regional conduction characteristic value, and iteratively evaluates the optimal prediction period.
7. The data analysis based flow state solidified soil durability evaluation system according to claim 6, wherein The data acquisition module comprises a parameter monitoring unit and a space division unit; The parameter monitoring unit collects material characteristic parameters through the sensor, and associates the sensor code with the spatial position; The space division unit is used for constructing a three-dimensional model of the monitoring area, dividing sample domains, and synchronizing material characteristic parameter collection and sample domain attribution.
8. The data analysis based flow state solidified soil durability evaluation system according to claim 6, wherein, The data preprocessing module comprises an activity density calculation unit and a gradient data set generation unit; The activity density calculation unit obtains the dynamic activity density by combining the total amount of active substances in the sample domain, the sample domain volume and the reference active substance amount; The gradient data set generation unit is used for generating a time-based activity density data set at a sampling time node and constructing an activity conduction data set based on the activity density difference between sample domains.
9. The data analysis based flow state solidified soil durability evaluation system according to claim 6, wherein, The model construction module comprises a scale quantification unit and a characteristic value calculation unit; The scale quantification unit is used for fitting the influence scale of the sample domain with the dynamic activity density as the centroid; The characteristic value calculation unit is used for calculating the compressive strength of the material in the circular domain as the regional conduction characteristic value by weighting with the influence scale as the radius.
10. The data analysis based flow state solidified soil durability evaluation system of claim 6, wherein, The evaluation and prediction module comprises a degradation state analysis unit and a cycle optimization unit; The degradation state analysis unit quantifies the durability degradation state based on the change rate of the regional conduction characteristic value; The cycle optimization unit is used for generating a degradation state set, iteratively evaluating the optimal prediction cycle and outputting the evaluation result.