Dynamic monitoring method and system for cement mixing pile based on ultrasonic detection

Abstract

This invention discloses a dynamic monitoring method and system for cement mixing piles based on ultrasonic testing. The invention relates to the technical field of ultrasonic testing. By identifying ultrasonic testing events, multiple forming stages of the cement mixing pile are determined. Based on the stage content, corresponding forming morphology, and corresponding ultrasonic data combinations of each forming stage, forming influencing events during the forming process are determined. Based on the forming history of the cement mixing pile and the corresponding forming influencing events, a review area for the cement mixing pile is determined. Based on the location, corresponding morphology, and corresponding forming stage of this review area, a dynamic monitoring system for ultrasonic testing is established, improving the accuracy of the dynamic monitoring system. Simultaneously, based on the project content, corresponding priority, and forming requirements of each forming optimization project, the optimization content of each forming stage is determined, and the forming process table of the cement mixing pile is dynamically updated.

Description

Technical Field

[0001] This invention relates to the technical field of ultrasonic testing, and in particular to a dynamic monitoring method and system for cement mixing piles based on ultrasonic testing. Background Technology

[0002] With the development of technology, cement mixing piles are a type of deep mixing piles. Cement mixing piles utilize specialized mixing machinery to forcibly mix cement slurry (or cement powder) with soft soil deep within the foundation. Through a series of physicochemical reactions between the cement and soft soil, the soft soil hardens into a cement-reinforced soil pile body with integrity, water stability, and a certain strength. Dynamic monitoring of cement mixing piles and control of each forming stage are crucial. Current technologies collect data combinations from each forming stage and determine corresponding forming events based on the identification of these data combinations. However, this ignores forming events that affect the cement mixing pile during the forming process, making it impossible to further control the re-inspection area of ​​the cement mixing pile. This leads to inaccuracies in the dynamic monitoring system of ultrasonic testing and affects the updating of the cement mixing pile forming process schedule. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a dynamic monitoring method and system for cement mixing piles based on ultrasonic detection.

[0004] This invention provides a dynamic monitoring method for cement mixing piles based on ultrasonic testing, comprising: During the molding process of cement mixing piles, multiple ultrasonic data are determined based on the detection of the mixing space between cement and soil. Ultrasonic detection events are determined based on the multiple ultrasonic data, the preset shape of the cement mixing pile, and the corresponding molding data. Multiple forming stages of cement mixing piles are determined based on the identification of ultrasonic detection events. The forming influencing events of cement mixing piles during the forming process are determined based on the stage content, corresponding forming morphology and corresponding ultrasonic data combination of each forming stage. The forming process of cement mixing piles is collected. Based on the forming process and corresponding forming influencing events, the re-inspection area of ​​cement mixing piles is determined. Based on the regional location, corresponding regional morphology and corresponding forming stage of the re-inspection area, a dynamic monitoring system for ultrasonic testing is determined. The re-inspection area inherits the main characteristics of the internal anomalies and is marked as a whole risk unit. The dynamic monitoring system defines the optimal technical parameter configuration for long-term or subsequent monitoring in the re-inspection area. In the dynamic monitoring system of ultrasonic testing, the final shape of the re-examination area is determined based on the identification of the dynamic monitoring system of ultrasonic testing, and the forming optimization event is determined based on the final shape of the re-examination area and the preset shape of the cement mixing pile. Based on the identification of the molding optimization event, multiple molding optimization projects are determined. The optimization content of each molding stage is determined according to the project content, corresponding project priority and molding requirements of cement mixing piles, so as to dynamically update the molding process table of cement mixing piles.

[0005] This invention provides a dynamic monitoring system for cement mixing piles based on ultrasonic detection, which is applied to the aforementioned dynamic monitoring method for cement mixing piles based on ultrasonic detection.

[0006] Compared with the prior art, the beneficial effects of the present invention are: (1) In the molding process of cement mixing piles, multiple ultrasonic data are determined based on the detection of the mixing space between cement and soil. Ultrasonic detection events are determined based on the multiple ultrasonic data, the preset shape of cement mixing piles and the corresponding molding data. Multiple molding stages of cement mixing piles are determined based on the identification of ultrasonic detection events. Molding impact events of cement mixing piles in the molding process are determined based on the stage content of each molding stage, the corresponding molding shape and the corresponding ultrasonic data combination. Ultrasonic detection events are introduced, which takes into account the stage content of each molding stage, the corresponding molding shape and the corresponding ultrasonic data combination, and improves the accuracy of molding impact events of cement mixing piles in the molding process.

[0007] (2) Collect the forming process of cement mixing piles, determine the re-inspection area of ​​cement mixing piles based on the forming process of cement mixing piles and the corresponding forming influencing events, and determine the dynamic monitoring system of ultrasonic testing based on the regional location, corresponding regional shape and corresponding forming stage of the re-inspection area, so as to further control the re-inspection area of ​​cement mixing piles, realize the overall consideration of the regional location, corresponding regional shape and corresponding forming stage of the re-inspection area, and improve the accuracy of the dynamic monitoring system of ultrasonic testing.

[0008] (3) In the dynamic monitoring system of ultrasonic testing, the final form of the re-examination area is determined based on the identification of the dynamic monitoring system of ultrasonic testing, and the forming optimization event is determined based on the final form of the re-examination area and the preset form of the cement mixing pile; multiple forming optimization projects are determined based on the identification of the forming optimization event, and the optimization content of each forming stage is determined based on the project content, corresponding project priority and forming requirements of the cement mixing pile, further controlling the forming optimization event, fully considering the optimization content of each forming stage, improving the accuracy of the optimization content of each forming stage, and dynamically updating the forming process table of the cement mixing pile. Attached Figure Description

[0009] Figure 1This is a flowchart illustrating the dynamic monitoring method for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Figure 2 This is a flowchart illustrating step S11 in the dynamic monitoring method for cement mixing piles based on ultrasonic detection in this embodiment of the invention. Figure 3 This is a flowchart illustrating step S12 in the dynamic monitoring method for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Figure 4 This is a flowchart illustrating step S13 in the dynamic monitoring method for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Figure 5 This is a flowchart illustrating step S14 in the dynamic monitoring method for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Figure 6 This is a flowchart illustrating step S15 of the dynamic monitoring method for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Figure 7 This is a schematic diagram of the structural composition of the dynamic monitoring system for cement mixing piles based on ultrasonic detection in an embodiment of the present invention. Detailed Implementation

[0010] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0011] Please see Figures 1 to 7 A dynamic monitoring method for cement mixing piles based on ultrasonic testing is proposed and applied to ultrasonic testing scenarios. The dynamic monitoring method for cement mixing piles based on ultrasonic testing includes: Step S11: During the molding process of cement mixing piles, multiple ultrasonic data are determined based on the detection of the mixing space between cement and soil. Ultrasonic detection events are determined based on the multiple ultrasonic data, the preset shape of cement mixing piles, and the corresponding molding data. Step S12: Based on the identification of ultrasonic detection events, determine multiple forming stages of cement mixing piles, and determine the forming influencing events of cement mixing piles during the forming process based on the stage content, corresponding forming morphology and corresponding ultrasonic data combination of each forming stage. Step S13: Collect the forming process of cement mixing piles, determine the re-inspection area of ​​cement mixing piles based on the forming process of cement mixing piles and the corresponding forming influencing events, and determine the dynamic monitoring system of ultrasonic testing based on the regional location, corresponding regional morphology and corresponding forming stage of the re-inspection area. Step S14: In the dynamic monitoring system of ultrasonic testing, the final shape of the re-examination area is determined based on the identification of the dynamic monitoring system of ultrasonic testing, and the forming optimization event is determined based on the final shape of the re-examination area and the preset shape of the cement mixing pile. Step S15: Based on the identification of the molding optimization event, determine multiple molding optimization projects. Based on the project content, corresponding project priority and molding requirements of each molding optimization project, determine the optimization content of each molding stage to dynamically update the molding process table of cement mixing piles.

[0012] refer to Figure 2 In step S11, the specific steps are as follows: S111: Real-time detection of the cement mixing pile forming process, marking the mixing position of cement and soil, determining the mixing space between cement and soil based on the tracing of the mixing position, performing ultrasonic detection on the mixing space, and determining multiple ultrasonic data during the ultrasonic detection process; S112: Collect the forming database of cement mixing piles, determine the preset shape of cement mixing piles based on the traversal of the forming database, determine the ultrasonic detection frame based on the preset shape of cement mixing piles and multiple ultrasonic data, and determine the ultrasonic detection event based on the ultrasonic detection frame and multiple forming data of cement mixing piles.

[0013] In the embodiments of this application, while the mixing pile drill bit is performing mixing and grouting operations, the system uses an ultrasonic probe array installed near the drill bit or inside the drill rod to perform high-frequency scanning of the surrounding medium; the system does not perform indiscriminate data acquisition across the entire depth, but rather accurately identifies the critical interface where the cement grout and the original soil undergo physical and chemical reactions by analyzing the echo characteristics of the ultrasonic waves in real time, including sound time, amplitude and spectral characteristics.

[0014] The system identifies differences based on the properties of the medium, calculates acoustic parameters, and determines the mixing location when a region with a gradient change in acoustic impedance is detected. Combined with the depth encoder and angle sensor on the drilling rig, the system maps the identified mixing location to a three-dimensional coordinate system in real time. As the drilling and hoisting process proceeds dynamically, the system continuously tracks these key points in the time domain, forming a dynamic mixing trajectory line.

[0015] To avoid treating key locations merely as geometric points, the system needs to extend outward to construct a three-dimensional volume encompassing the effective interaction range, representing the mixing space of the cement-soil mixing area. With the mixing location as the geometric center, the system uses a spatial reconstruction algorithm to calculate an acoustic detection field within a cylindrical or ellipsoidal range, based on preset mixing blade radii, slurry diffusion models, and the half-wave diffusion angle of ultrasonic waves. As the drill rod moves, the system superimposes and traces the continuous mixing space along the time axis, constructing a continuous dynamic monitoring volume.

[0016] The ultrasonic probe transmits and receives signals only in a defined mixed space. The system then performs noise reduction, gain compensation, and feature extraction on the acquired raw waveforms, covering indicators such as the first wave duration, amplitude attenuation, and frequency drift. The final output is structured ultrasonic data.

[0017] Specifically, assuming that cement mixing pile construction is underway, with a designed pile length of 15 meters and a pile diameter of 500 mm, and the main stratum being soft plastic silty clay, the drilling rig is currently in the drilling and mixing stage, at a depth of 8.5 meters underground; when drilling to a depth of 8.5 meters, the drill bit rotates at a specific speed to cut the soil and inject cement slurry, at which point the ultrasonic transducer integrated between the drill bit blades emits ultrasonic pulses with a frequency of 50 kHz to the side.

[0018] The echo signal received by the system shows that the sound velocity of the medium calculated by acoustic time is between that of pure water and undisturbed silt, and the amplitude attenuation characteristics show typical characteristics of the early stage of cement-soil hydration. Based on this, the system determines that the mixing position of cement and soil is 250mm lateral radius of the drill bit, and accurately marks this polar coordinate point on a plane at a depth of 8.5m. Based on this marked position, the system, combined with the preset 500mm pile diameter of the cement mixing pile and the diffusion angle of the ultrasonic wave in the medium, constructs a cylindrical mixing space with a diameter of about 550mm centered on the drill rod axis.

[0019] Within this space, ultrasonic waves performed a sector scan, and the system acquired key data such as the initial wave duration of 1550 m / s, the received wave amplitude attenuated by 15% compared to the standard cement-soil sample, and the dominant frequency drifting from 50 kHz to 45 kHz. These data indicate that there are local soil clumps that are not completely dispersed within this space, and that the density of the medium is increasing with the hydration reaction. The system locks this set of ultrasonic data with spatiotemporal tags (depth 8.5 m, timestamp T) as the basic data source for generating subsequent ultrasonic detection events.

[0020] Furthermore, the system utilizes a molding database that not only contains historical data of the current piles, but also integrates historical pile standard samples under the same site and similar geological conditions. By traversing these data, the system can extract the optimal feature model of piles under the geological conditions, i.e., the preset shape.

[0021] The system employs multi-source data fusion technology to integrate soil layer parameters from geological survey reports, pile length and cement content from design drawings, and standard values ​​of acoustic parameters from historical pile formation. Through feature extraction and matching, the system finds historical segments that match the current depth and soil properties, and generates standard acoustic feature vectors that should be present at the current depth based on the matching results. These vectors cover indicators such as standard sound velocity, standard amplitude, and standard density, while also establishing the standard geometric profile.

[0022] The system is dedicated to building a dynamic evaluation model, namely an ultrasonic testing framework, to define the boundaries of qualified signals and the deviation standards that need to be recorded under the current working conditions. The system implements dynamic threshold setting in the framework construction, abandoning the fixed threshold mode, and instead dynamically generates the upper and lower limits of acoustic parameters based on the standard values ​​in the preset form and the allowable range of current construction errors. At the same time, the system configures higher weighting factors for key parts or key parameters such as the interface between soft and hard soil layers, and subdivides the detection area into multiple three-dimensional grid units, assigning an independent detection strategy to each unit.

[0023] The system inputs the real-time ultrasonic data acquired by S111 into the constructed ultrasonic detection framework for calculation. When the measured data deviates from the dynamic threshold set by the framework or meets a certain abnormal pattern, the system triggers an ultrasonic detection event. In this stage, the system calculates the deviation rate between the measured ultrasonic data and the reference value in the detection framework in real time and uses multi-parameter coupled judgment logic. For example, when the sound velocity decreases and the amplitude attenuation increases while the framework conditions are met, a segregation risk event is triggered. When the acoustic time change meets the conditions, a void risk event is triggered. Finally, the system encapsulates the spatiotemporal location, deviation value, and the grid cell to which the triggered event belongs, forming a structured ultrasonic detection event.

[0024] Specifically, when cement mixing piles are constructed to a depth of 10.5 meters, which is precisely at the interface between the silty clay layer and the silty sand layer, the system traverses the cement mixing pile formation database and finds that in the historical construction records of similar strata at a depth of 10.5 meters, the standard pile formation is characterized by uniform cement-soil mixing, a stable average sound velocity of 1650 m / s, and a wave amplitude attenuation of less than 6 dB. The system accurately extracts these characteristics and establishes them as the preset form for the current depth of 10.5 meters.

[0025] Based on the preset shape and taking into full account the acoustic fluctuations at the stratum interface, the system dynamically generates a detection framework for this depth; specifically, the sound velocity alarm threshold is set to 1600m / s (3% lower than the standard value), the amplitude attenuation threshold is 8dB, and the detection area is divided into 12 sectors along the pile perimeter, clearly requiring that the data in any sector must not be continuously lower than the threshold, thereby establishing a dynamic evaluation model.

[0026] The system inputs real-time ultrasonic data from a certain location at a depth of 10.5 meters, acquired during phase S111, into the detection framework for calculation. Since the data shows a sound velocity of only 1520 m / s, below the framework threshold of 1600 m / s, and an amplitude attenuation of 10 dB, exceeding the framework threshold of 8 dB, the system determines that both indicators deviate beyond the acceptable range. Based on this, the system confirms that the physical properties of the area do not meet the preset morphological requirements, and the abnormal data characteristics are consistent with the model of soil agglomeration due to insufficient mixing. Therefore, the system officially generates and marks an ultrasonic detection event, defining it as local heterogeneity of the pile medium, and pinpoints its specific location at the 3 o'clock sector at a depth of 10.5 meters, serving as the basis for subsequent quality review and optimization.

[0027] refer to Figure 3 In step S12, the specific steps are as follows: S121: Dynamically identify ultrasonic detection events and determine multiple stage markers during the identification process. Based on the tracing of each stage marker, determine the corresponding forming stage to identify multiple forming stages of cement mixing piles and mark the corresponding stage types. S122: In multiple molding stages, the corresponding stage content is determined based on the identification of each molding stage, and the molding form and corresponding ultrasonic data combination in each molding stage are marked. The first molding influencing factor is determined according to the stage content and corresponding molding form of each molding stage. S123: Determine the second molding influencing factor based on the stage content of each molding stage and the corresponding ultrasonic data combination, and determine the molding influencing events of cement mixing piles in the molding process based on the mapping relationship table of the first molding influencing factor, the second molding influencing factor and molding influencing events.

[0028] In the embodiments of this application, the S11 output received by the system is not a static list, but a data stream that flows in continuously over time; dynamic identification refers to performing sliding window processing on the event stream within a time window to identify the density and trend of events, and marking nodes with temporal characteristics, i.e., stage markers, on the time axis; the system sets a sliding time window, counts the number and severity of ultrasonic detection events triggered within the window, and uses a clustering algorithm to group ultrasonic detection events that are close in time and similar in nature into a group; each cluster of events is then assigned a unique stage marker, which includes not only ID, but also timestamp, depth range, centroid coordinates of the event cluster, and statistical characteristics of the event cluster, playing a role in locating key nodes on a complex construction time axis.

[0029] The system integrates stage markers with construction logs and sensor data through multi-sensor fusion to determine the process stage represented by the current marker, i.e., the forming stage. The system aligns the timestamp of the stage marker with the data streams from the depth sensor, speed sensor, and flow meter integrated into the drilling rig at the nanosecond level. Simultaneously, using the built-in construction process state machine model, the system determines which state branch the stage marker falls on based on the aligned sensor data. For example, increasing depth without grouting indicates drilling and mixing, while decreasing depth with grouting indicates lifting and grouting. This tracing process is essentially a process of accurately mapping physical events to logical processes, thereby clearly defining whether the marker represents a specific forming stage such as pre-mixing and sinking, first grouting and lifting, or re-mixing.

[0030] After identifying the specific forming stage, it is necessary to classify and label the stage to define its physical properties in order to facilitate subsequent data processing and statistical analysis. The system classifies attributes according to stage characteristics. Common classifications include physical morphology (such as pile bottom forming stage and pile top forming stage), process action (such as jet mixing stage and empty drilling cutting stage), and material state (such as initial setting stage and final setting stage). The system assigns the determined category to the forming stage so that it can be quickly retrieved and called in subsequent algorithms. For example, only the data of the jet mixing stage is analyzed while the data of the empty drilling stage is ignored, thereby improving the focus and efficiency of the analysis.

[0031] Specifically, the drilling rig operation log shows that the current time point is T, the drill bit depth is 8.0 meters below the surface, and grouting is underway. Within a 10-second time window before and after T, the system detected 15 ultrasonic detection events of abnormal acoustic impedance type densely triggered by the ultrasonic sensor in the depth range of 8.0 meters to 8.2 meters. The system performed cluster analysis on these events and generated a stage marker (Tag-ID: P-Event-092) at T+5 seconds on the time axis. This marker is associated with the above 15 event groups and marks their average sound velocity deviation as -12% (abnormally high), thus establishing a key node on the time axis.

[0032] The system traces Tag-ID:P-Event-092 and retrieves drilling rig sensor data on the same time axis. The depth data shows a change from 8.0m to 7.8m, indicating a decrease in depth, meaning the drill rod is moving upwards. The grout flow rate data shows a value of 30L / min, indicating pressurized flow in the pipe, meaning grouting is in progress. The motor rotation data shows a forward lifting mode. The system inputs these features into the process state machine for matching and determines that the time period represented by Tag-ID:P-Event-092 is in the critical process step of the first grouting lift, thus successfully identifying the corresponding forming stage.

[0033] Since this stage involves the active injection of cement grout and is the core link in the formation of the main body of the pile, the system marks this stage as the main body shotcrete forming stage and marks the physical attribute of this stage as the high hydration heat active period in the database; through the processing of S121, the dense ultrasonic anomaly that occurred at 8.0 meters of the cement mixing pile was accurately defined as the main body shotcrete forming stage event that occurred during the first shotcrete lifting stage.

[0034] Furthermore, in multiple molding stages, the corresponding stage content is determined based on the identification of each molding stage, and the molding form and corresponding ultrasonic data combination in each molding stage are marked. The first molding influencing factor is determined according to the stage content and corresponding molding form of each molding stage, which takes into account the overall consideration of the stage content and corresponding molding form of each molding stage, and ensures the accuracy of the first molding influencing factor.

[0035] At this point, the stage content does not simply refer to the name of the grouting or drilling, but rather to the collection of all key construction parameters within that stage. Based on the forming stage determined in S121, the system extracts the corresponding time slices from the construction data stream to obtain the specific process parameters for that stage. During this process, the system locks the corresponding start and end timestamps based on the stage markers (such as Tag-ID) in S121, and extracts real-time operating data of the drilling rig in parallel within the locked time interval, covering drill rod lifting or lowering speed, mixing head rotation speed, grouting pump pressure, grouting volume per unit time, and current load value. The system assembles these parameters into a structured stage content data package, for example, describing a stage as having a lifting speed of 0.8 m / min, a rotation speed of 60 r / min, and a grouting pressure of 25 MPa.

[0036] The system utilizes ultrasonic data for inversion imaging to construct the internal physical morphology of the pile and binds it with the original detection data to form a morphological-data dual-modal evidence chain. Based on multiple sets of ultrasonic data within this stage, the system uses ray tracing or back projection algorithms to reconstruct the two-dimensional or three-dimensional acoustic parameter distribution map of the corresponding pile cross-section at this stage. In the reconstructed acoustic distribution map, an image edge detection algorithm is applied to identify the interface between the pile and the surrounding soil. The system calculates the geometric features enclosed by the interface, including the pile cross-sectional area, maximum and minimum radii, roundness deviation rate, and centroid offset. Finally, the calculated geometric morphology is strongly bound and stored with the original ultrasonic waveform data that generated the morphology, forming an unalterable evidence combination.

[0037] The system performs a deviation analysis between the stage content, which is the expected process input, and the formed shape, which is the actual geometric output, to extract the first and most intuitive physical factors that cause abnormal shapes. The system compares the formed shape with the preset shape in the design drawings to calculate whether there are geometric defects, such as diameter expansion, diameter reduction, bulging, depression, or even discontinuity in the cross section. The system performs causal matching between geometric defects and process parameters in the stage content. For example, when the shape shows that the diameter is significantly smaller than the design value, and the stage content shows that the spraying volume is normal but the lifting speed is extremely fast, it is determined that the insufficient forming is caused by excessive speed. The factors extracted by the system only describe macroscopic phenomena at the physical level, such as geometric diameter reduction of the pile, diameter expansion of the pile, or distortion of the pile cross section.

[0038] Specifically, the system identified the construction log data corresponding to the depth stage of 8.0 to 8.2 meters. The parameters extracted by the system showed that the lifting speed at this stage was 1.5 m / min, while the design requirement was only 0.5-0.8 m / min, the rotation speed was 45 r / min, and the grouting flow rate remained stable. Based on this, the system determined the process characteristics of this stage as high-speed lifting grouting, that is, the drill bit traversed a relatively long distance through the soil layer per unit time, thus establishing the key input parameter background.

[0039] The system performed tomographic imaging based on the ultrasonic data from this stage. The analysis showed that the received wave duration was generally prolonged in this range, the initial wave amplitude was weak, and the low-frequency components of the waveform increased. The cross-sectional images reconstructed based on these data showed that the pile edge exhibited a clear inward contraction trend in the direction from 3 o'clock to 6 o'clock. Calculations showed that the effective pile diameter at this location was only 420 mm, which did not meet the design requirement of 500 mm, and the edge interface was blurred. The system combined the contraction pattern of the effective diameter of 420 mm with the prolonged acoustic duration and low-frequency ultrasonic data to form a solid evidence combination for this stage.

[0040] The morphological assessment results showed that the diameter was significantly smaller than the design value, which is a typical case of insufficient geometric dimensions. A causal correlation analysis based on the stage details revealed that the lifting speed of 1.5 m / min far exceeded the process standard. With a constant unit grout volume, this excessively rapid lifting speed resulted in a severe shortage of cement grout distributed per unit length of pile, and the shearing time of the mixing blades on the soil was shortened, failing to effectively expand the designed pile diameter. Therefore, the system determined that the primary influencing factor for this anomaly was the geometric shrinkage of the pile due to high-speed lifting. This conclusion successfully ruled out the possibility of material defects (such as cement deterioration), pinpointing the root cause to the mismatch between the macroscopic geometric shape and the construction speed.

[0041] Therefore, based on the stage content and corresponding ultrasonic data combinations of each molding stage, the second molding influencing factor is determined. Based on the mapping relationship table of the first molding influencing factor, the second molding influencing factor, and molding influencing events, the molding influencing events of the cement mixing pile during the molding process are determined. This approach incorporates the overall consideration of the mapping relationship table of the first molding influencing factor, the second molding influencing factor, and molding influencing events, ensuring the accuracy of the molding influencing events of the cement mixing pile during the molding process. Simultaneously, ultrasonic detection events are introduced, incorporating consideration of the stage content, corresponding molding morphology, and corresponding ultrasonic data combinations of each molding stage, further improving the accuracy of the molding influencing events of the cement mixing pile during the molding process.

[0042] At this point, the system analyzes the fine characteristics of ultrasonic data in the time, frequency, and energy domains to infer the physical and mechanical properties of the medium inside the pile, thereby uncovering the microscopic causes of the abnormal shape. The system performs Fast Fourier Transform (FFT) on the ultrasonic data to analyze its spectral characteristics. If the dominant frequency drifts significantly to lower frequencies, it usually means that there are large aggregate particles in the medium or that the sound scattering is caused by uneven mixing. If the frequency band narrows, it means that the water content of the medium is too high.

[0043] Meanwhile, the system calculates the energy attenuation rate of ultrasonic waves during propagation. An abnormally high attenuation rate usually indicates the presence of high impedance differences or unhydrated soil clumps within the medium. The system also detects whether the received waveform is distorted or has a tail to reflect the heterogeneity of the medium. The system combines the grouting volume and current value in the stage content for joint analysis. For example, if the current is normal but the sound velocity is extremely low, the second factor is marked as unbroken soil. If the sound velocity is extremely low and the attenuation is extremely high, it is marked as grout segregation or the presence of voids.

[0044] Single factors are often ambiguous. In order to accurately determine responsibility, the system uses dual factors to construct a unique identification key, which is then queried in the expert mapping table to finally determine the specific molding impact event. The system combines the first factor (macro geometry) and the second factor (micro material) into a two-dimensional fault feature vector, such as <geometric diameter reduction, porous medium>.

[0045] The system searches a built-in mapping table trained on engineering experience and historical data. This table defines the correspondence between factor combinations and specific construction faults. For example, <reduced diameter, high gas content in the medium> corresponds to blockage of the grouting pipeline, while <expanded diameter, extremely high medium resistance> corresponds to encountering boulders during drilling. The system searches the table for the entry with the highest matching degree, transforming the abstract factor combination into specific and actionable shaping events, such as grouting interruption, inadequate re-mixing, or grout delivery pipe rupture, thereby completing the accurate identification of faults.

[0046] Specifically, the system conducted an in-depth analysis of the ultrasonic data combination at this stage, focusing not on the cross-sectional size but on the material of the cross-section. The data showed that although the pile body was reduced in size, the ultrasonic wave velocity in the reduced diameter area was only 1450 m / s, close to the sound velocity of water or silt, and far lower than the 1800 m / s sound velocity that qualified cement-soil should have. At the same time, waveform spectrum analysis showed that the dominant frequency did not drift to lower frequencies, thus ruling out the presence of large soil particles, and the energy attenuation was minimal, indicating almost full transmission. Combined with the fact that the grouting pump pressure recorded at this moment fluctuated slightly but did not return to zero, the system determined that this characteristic of low sound velocity, low attenuation, and no high-frequency distortion, combined with the background of high-speed lifting, indicated that the cement grout in this area was diluted too quickly, or that there was no cement grout filling in some areas, and it was completely occupied by disturbed undisturbed soil and water. Therefore, the system determined that the second influencing factor for forming was the liquefaction of the medium caused by the lack of curing agent, that is, the area became thin mud rather than cement-soil.

[0047] The system queries the mapping table for this combination. According to the table entry logic, <geometric diameter reduction, high medium density> corresponds to soil compression effect, <geometric diameter reduction, high air bubble content in medium> corresponds to local blockage of the grouting nozzle, and <geometric diameter reduction, medium liquefaction / cure agent deficiency> corresponds to severely insufficient grouting volume or instantaneous leakage in the grouting pipeline. Combined with the fact that the lifting speed of 1.5m / min recorded in the stage content is seriously excessive, the system determines that the root cause of medium liquefaction is that the grouting volume cannot keep up with the lifting speed, resulting in severe grout shortage per unit volume. The system determines that the forming impact event of the cement mixing pile at 8.0 meters is insufficient grouting volume (under-grouting) caused by mismatch of process parameters. This conclusion not only indicates that the pile body is smaller, but also that the pile body has become thin mud, directly guiding the construction party to check the ash hopper inventory, adjust the drilling speed, or check the pumping system.

[0048] refer to Figure 4 In step S13, the specific steps are as follows: S131: Real-time monitoring of cement mixing piles and marking of the forming process of cement mixing piles. The forming process of cement mixing piles presents all the contents of the cement mixing pile in the forming process. Based on the forming process of cement mixing piles and the corresponding forming impact events, multiple abnormal forming contents are determined and the corresponding abnormal forming locations are marked. Based on each abnormal forming contents and the corresponding abnormal forming locations, the review area of ​​cement mixing piles is constructed and determined. S132: Further re-examine the identified re-examination area of ​​the cement mixing pile and mark the location of the re-examination area. Based on the matching between the re-examination area and the forming process of the cement mixing pile, determine the corresponding regional morphology and forming stage. Determine the dynamic monitoring system for ultrasonic testing based on the regional location of the re-examination area, the corresponding regional morphology and the corresponding forming stage.

[0049] In the embodiments of this application, cement mixing piles are monitored in real time, and the forming process of cement mixing piles is marked. The forming process of cement mixing piles presents all the contents of the cement mixing pile during the forming process. Based on the forming process of cement mixing piles and the corresponding forming influencing events, multiple abnormal forming contents are determined, and the corresponding abnormal forming locations are marked. Based on each abnormal forming contents and the corresponding abnormal forming locations, a review area for cement mixing piles is constructed and determined. This approach takes into account the overall consideration of constructing each abnormal forming contents and the corresponding abnormal forming locations, ensuring the accuracy of the review area for cement mixing piles.

[0050] At this point, a complete and high-precision digital twin model is established to record every moment of the pile from ground breaking to pile formation, ensuring that subsequent analysis is based on verifiable data. The system synchronously records the sensor data of the drilling rig (including depth, speed, flow rate, and torque) and the raw waveform data of the ultrasonic detection system at a high-frequency sampling rate. At the same time, a unified spatiotemporal coordinate system is established to perform nanosecond-level time synchronization and millimeter-level spatial calibration of data from different sources. Based on this, the system constructs a continuous data stream model containing event indexes, clearly recording at time T, depth Z, event E triggered, acoustic characteristics at that time X, and process parameters Y, thus providing a complete contextual environment for subsequent backtracking.

[0051] The system traverses the entire forming process, extracts all forming-affecting events determined by S12, and marks them as textual and coded according to the nature of the events (such as underspray, broken pile, and diameter expansion) to generate abnormal forming content. The timestamp corresponding to each abnormal forming content is combined with the depth and angle sensor data at that time to convert it into absolute three-dimensional coordinates of the pile body (using a cylindrical coordinate system: depth Z, angle θ, radius r). In addition, the system also assigns a severity level to each abnormal location, and marks the abnormality level by calculating the degree of deviation between the ultrasonic data and the standard value. The greater the deviation, the higher the level.

[0052] The system employs spatial clustering algorithms such as DBSCAN or K-Means to search for distribution patterns of outliers in three-dimensional space and calculate the Euclidean distance or axial distance between each outlier. By merging multiple outliers with a distance less than a set threshold into an outlier cluster, the system calculates the minimum bounding box or convex hull of the outlier cluster. To ensure coverage of edge effects during review, the system automatically expands the bounding box to form a closed geometry, i.e., the review area. This review area inherits the main characteristics of the internal outliers and is marked as a whole risk unit.

[0053] Specifically, the system retrieves complete construction data of cement mixing piles and generates a digital forming process. This process accurately records the construction status and ultrasonic echoes every second from 0 meters to 15 meters. For example, the process clearly records that at 10 minutes and 30 seconds, at a depth of 8.0 meters and a lifting speed of 1.5 m / min, a set of abnormal ultrasonic waveforms was captured, thus establishing a complete digital foundation. The system scans the forming process and extracts specific anomalies, including: anomaly A, located at a depth of 7.9 meters on the southeast side (45° azimuth), with the anomaly being poor bonding due to under-spraying; anomaly B, located at a depth of 8.1 meters on the south side (90° azimuth), with the anomaly being soil clump aggregation; anomaly C, located at a depth of 8.3 meters on the southeast side (60° azimuth), with the anomaly being poor bonding due to under-spraying; and anomaly D, located at a depth of 12.0 meters on the north side, with the anomaly being slight diameter enlargement. The system maps these discrete anomalies one by one to the three-dimensional coordinates of the pile body.

[0054] During the construction of the review area, the system performed spatial clustering analysis and found that anomalies A, B, and C were very close in axial distance (span of only 0.4m) and adjacent in radial angle. The system determined that these three points belonged to the same quality defect zone. Based on this, the system generated a cylindrical review area, marked as Review-Zone-01. The area was defined as having a depth of 7.7 meters to 8.5 meters and an azimuth angle of 30° to 100°. Based on the characteristics of the three anomalies within it, this review area was characterized as a weak continuous bonding area caused by insufficient shotcrete volume. In contrast, anomaly D did not constitute an area due to its greater distance and was only recorded as an isolated point. Through S131, the Review-Zone-01 review area was accurately divided on the cement mixing pile. The subsequent S132 will specifically develop an encrypted ultrasonic scanning plan for this specific space.

[0055] Furthermore, the identified re-inspection area of ​​the cement mixing pile was re-inspected, and the location of this area was marked. Based on the matching between the re-inspection area and the forming process of the cement mixing pile, the corresponding regional morphology and forming stage were determined. According to the regional location, corresponding regional morphology, and corresponding forming stage of the re-inspection area, a dynamic monitoring system for ultrasonic testing was determined. This system takes into account the overall consideration of the regional location, corresponding regional morphology, and corresponding forming stage of the re-inspection area, ensuring the accuracy of the dynamic monitoring system for ultrasonic testing. At the same time, the re-inspection area of ​​the cement mixing pile was further controlled, realizing the overall consideration of the regional location, corresponding regional morphology, and corresponding forming stage of the re-inspection area, and improving the accuracy of the dynamic monitoring system for ultrasonic testing.

[0056] At this point, the system needs to reschedule the ultrasonic probes to perform high-precision secondary data acquisition in a specific space and establish an accurate coordinate system index. The system controls the ultrasonic transducer array to point to the spatial coordinates of the area being examined. For shallow areas, a high-frequency focusing probe is used to improve resolution, while for deep areas, a low-frequency high-power probe is used to enhance penetration.

[0057] Meanwhile, during the re-examination phase, a smaller scanning step size is used for high-density data sampling. For example, the angular resolution is reduced from 10° to 2°, and the depth sampling interval is reduced from 10cm to 1cm, thereby obtaining high-definition acoustic images of the area. During the re-examination, the system combines high-precision positioning sensors to reconfirm the boundary of the re-examination area and marks the corrected precise area location in the three-dimensional model of the pile, covering the depth range and radial orientation.

[0058] The system utilizes high-density ultrasonic data obtained from the review to perform acoustic tomography (CT) inversion, calculating the sound velocity distribution cloud map and amplitude attenuation distribution map within the region, thereby determining the specific physical morphology of the region, such as the presence of low-velocity cavities, high-impedance inclusions, or whether the interface is clear. Simultaneously, the system traces the coordinates of the reviewed region back to the timeline of the construction process, retrieving the sequence of process parameters at the time of the region's formation. The system combines morphology and process history to determine the formation stage. For example, if the morphology shows a lack of grout and the process history shows that it was in the high-pressure grouting stage but the pressure suddenly dropped, it is characterized as a grout supply failure type defect. If the process history shows that it was in the overspeed lifting stage, it is characterized as a process runaway type defect.

[0059] The system selects high-frequency ultrasound for fine crack morphology and low-frequency ultrasound for large-area loose morphology. Based on the attenuation characteristics of the regional morphology, the system sets a time gain control curve (TVG) that varies with depth to ensure the visibility of far-field signals and establishes a dedicated alarm threshold model for this region. Instead of using a global average threshold, a local threshold is set based on the statistical characteristics of the normal part of the region. In addition, the system defines the monitoring time series and triggering mechanism, and encapsulates the above parameters into a dynamic monitoring system for this region. The dynamic monitoring system defines the optimal technical parameter configuration for long-term or subsequent monitoring in this re-examined region.

[0060] Specifically, Review-Zone-01 (characterized as a continuous weak cement bond area) was located within the 7.7-8.5 meter and 30°-100° azimuth range of the cement mixing pile. The system then proceeded to S132 for in-depth diagnosis and monitoring of this area. The system controlled the ultrasonic probe to initiate a fan-shaped scan of Review-Zone-01 and used an asymmetric scanning mode to perform a denser scan of the defect core area (around 60° azimuth). Data acquisition results showed that extremely high-resolution waveform data of the area was obtained, revealing that the seemingly continuous defect area in the original scan actually contained multiple high-velocity hard cores with a diameter of approximately 5 cm. Based on echo analysis, the system precisely corrected the boundary of the reviewed area to a depth of 7.75m-8.45m and an azimuth of 40°-95°, thus achieving precise locking of the geometric boundary of the risk area.

[0061] The system analyzed the review data and reviewed the formation process. The CT-retrieved regional morphology showed that the main body of the area had a low sound velocity (approximately 1450 m / s, close to silt), but it contained multiple scatterers with extremely high sound velocity (>3000 m / s, similar to rocks). Therefore, the morphology was characterized as a mixed defect morphology of cement and soil encasing isolated boulders. The process matching showed that when the system traced back the formation process at a depth of 8.0 meters, it found that when the drilling rig passed through this depth, the torque sensor reading instantly spiked to full scale and then suddenly dropped. This confirmed that the drill bit cut into an underground obstacle (isolated boulder) at this point, causing the mixing blades to slip instantly and the grouting to be interrupted. Based on the above information, the system determined that the formation stage of this area was the stage of grouting interruption and uneven mixing caused by cutting into an obstacle.

[0062] The system is equipped with waveform processing technology and variable aperture focusing technology to suppress strong specular reflections caused by boulders and enhance penetration into the surrounding soft soil medium. In terms of spectrum analysis strategy, a dual-frequency monitoring method is adopted, simultaneously emitting low frequencies to check the overall cementation and emitting high frequencies to detect micro-cracks around the boulders. At the same time, a sound velocity threshold with higher tolerance is set for this area, and the focus is shifted to monitoring the amplitude uniformity as a quality criterion. The system generates an anti-interference-multi-frequency composite dynamic monitoring system for Review-Zone-01. Subsequent monitoring of this area will strictly follow this set of parameters to accurately assess whether the cement grout has ultimately wrapped and solidified these boulders.

[0063] refer to Figure 5 In step S14, the specific steps are as follows: S141: Real-time monitoring of the dynamic monitoring system of the ultrasonic testing, dynamic identification of the dynamic monitoring system of the ultrasonic testing, and identification of multiple sub-re-examination areas during the identification process; S142: Based on the identification of each sub-review area, the corresponding morphological features are determined, and the final morphology of the review area is determined according to the feature position of multiple morphological features, the corresponding characteristic morphology, and the priority of each morphological feature. S143: Compare the final form of the inspection area with the preset form of the cement mixing pile, and present multiple morphological differences during the comparison process. At the same time, determine the forming optimization system of the cement mixing pile based on the detection of the cement mixing pile forming database, and determine the corresponding forming optimization events according to each morphological difference, the forming process of the cement mixing pile, and the forming optimization system of the cement mixing pile.

[0064] In the embodiments of this application, the dynamic monitoring system of the ultrasonic detection is monitored in real time, the dynamic monitoring system of the ultrasonic detection is dynamically identified, and multiple sub-review areas are determined during the identification process, thus introducing the method of determining multiple sub-review areas during the identification process.

[0065] At this time, the system loads the dynamic monitoring configuration file for the area to be reviewed. For example, if S132 is determined to be a strong interference environment, the system will automatically enable the bandpass filter or time gain control (TVG) to suppress boulder scattering or background noise. The ultrasonic probe runs according to the set scanning path (such as spiral scanning or fan scanning) and returns high sampling rate acoustic waveform data in real time. At the same time, the system performs noise reduction and signal enhancement on the acquired raw waveform to ensure that the data used for identification has a high signal-to-noise ratio.

[0066] The system calculates acoustic feature vectors in real time for each sampling point, covering sound velocity (VP), amplitude attenuation coefficient, dominant frequency drift, and waveform similarity. It uses a sliding window algorithm or clustering algorithm to analyze abrupt changes in acoustic feature vectors in the spatial dimension. When the Euclidean distance of the feature vectors exceeds a set threshold, the system identifies an acoustic boundary line. Based on these acoustic boundary lines, the system logically divides the continuous scanning space into different acoustic medium blocks.

[0067] The system maps the identified acoustic boundaries back to the three-dimensional coordinate system of the pile and calculates the spatial envelope of each medium block. It assigns attribute labels to the medium blocks based on their acoustic characteristics, such as marking high-velocity areas as high-density solids, extremely low-velocity areas as voids or segregation, and mixed wave areas as transition zones. The system summarizes all the labeled spatial blocks to form a list containing multiple sub-review areas, each of which includes its geometric boundaries and acoustic attribute descriptions.

[0068] Specifically, Review-Zone-01 (a suspected weak cemented zone containing boulders) of the cement mixing pile is being monitored. The system has been equipped with an anti-interference, multi-frequency composite dynamic monitoring system according to S132. The system controls the ultrasonic probe to perform a spiral scan of Review-Zone-01 (depth 7.75m-8.45m). The probe emits dual-frequency pulse signals, using 50kHz to penetrate the whole and 100kHz to capture minute cracks. At the same time, the system applies the variable aperture focusing algorithm set in S132 to filter out strong specular reflection interference generated by the surface of the boulders in real time. As the probe moves, the system continuously receives high-fidelity waveform data of the area after signal enhancement, ensuring the input quality for subsequent analysis.

[0069] The system analyzes the data stream layer by layer to find abrupt changes in acoustic properties. Feature analysis shows that at a depth of 8.00 meters and an azimuth angle of 60°, the system detects a sudden drop in sound speed from the normal 1800 m / s to 1450 m / s, with a significant increase in amplitude attenuation. At the center of this low-sound-speed region, a tiny high-impedance core with a sound speed as high as 3500 m / s is detected. Based on this, the algorithm identifies a clear acoustic boundary between the high-impedance core and the surrounding low-sound-speed region, as well as a transition boundary between the low-sound-speed region and normal cement-soil, providing a basis for region segmentation.

[0070] Based on the above identification results, the system precisely divided the original Review-Zone-01 into three sub-review areas. The first is Sub-review Area-1 (Zone-A), located at a depth of 8.00m, an azimuth angle of 58°-62°, and a radius of 0.2m-0.25m. Its sound velocity is extremely high (>3500m / s), and its waveform is hard. It is marked as the core of the isolated rock. The second is Sub-review Area-2 (Zone-B), which surrounds Zone-A. It is about 3-5cm thick, has an extremely low sound velocity (<1500m / s), and its waveform is chaotic. It is marked as the mud segregation layer. Finally, there is Sub-review Area-3 (Zone-C), which occupies the remaining part of Review-Zone-01 except for Zones-A and B. Its sound velocity is slightly lower (around 1700m / s) but within an acceptable range. It is marked as the area affected by slight disturbance.

[0071] Furthermore, based on the identification of each sub-review area, the corresponding morphological features are determined. The final form of the review area is determined according to the feature position, corresponding characteristic form, and priority of each morphological feature. This comprehensive consideration of the feature position, corresponding characteristic form, and priority of each morphological feature ensures the accuracy of the final form of the review area.

[0072] At this point, the system calculates the three-dimensional geometric parameters of each sub-review region, including spatial envelope volume, surface area, centroid coordinates, principal axis direction, and shape factors (such as sphericity and flatness). Simultaneously, the system quantifies the acoustic statistical characteristics within the region, including average sound velocity, sound velocity variance representing uniformity, average amplitude attenuation rate, and dominant frequency center. The system combines the above geometric and acoustic parameters to generate a unique multidimensional feature vector for each sub-review region, such as described as [high sound velocity, high volume, high sphericity] for suspected boulders, or [low sound velocity, high surface area, low sphericity] for suspected fissures.

[0073] The system employs a priority-weighted strategy, assigning the highest severity weight to features with excessively low sound velocity or excessively large volume, a higher positional weight to features of key stress-bearing parts of the pile, and a higher morphological weight to features with strong connectivity than isolated features. The system uses a weighted morphological reconstruction algorithm, growing and expanding from high-priority sub-regions as seed points to surrounding low-priority regions, while eroding or removing low-priority regions identified as artifacts or minor disturbances. The system calculates the unified geometric boundary after fusion, generating the final shape of the review area. This shape is a standardized geometric model after cleaning and weighting, which can most accurately reflect the true quality status of the pile in this area.

[0074] Specifically, three sub-review areas were identified in Review-Zone-01 of the cement mixing pile: Zone-A (isolated boulder core), Zone-B (mud segregation layer), and Zone-C (slightly disturbed area). The system calculated the feature vectors of the three sub-regions respectively. Zone-A (isolated boulder core) is located at a depth of 8.00m, with the center off the outside of the pile body. Its characteristics are a sound velocity of 3500m / s (extremely high), a volume of 0.005m³, and a regular shape. Zone-B (segregation layer) is adjacent to the periphery of Zone-A. Its characteristics are a sound velocity of 1450m / s (extremely low, similar to silt), a volume of 0.002m³, and a thin shell-like coating. Zone-C (disturbed area) is located in the area extending outward from Zone-B. Its characteristics are a sound velocity of 1700m / s (slightly lower than the design value), a dispersed volume, and an irregular shape.

[0075] The system executed a priority determination and fusion algorithm. In the priority analysis, Zone-A (isolated rock), although a foreign object, is a hard entity and does not damage the structural strength of the pile, so its priority is set to low. Zone-C (disturbance) has a slight deviation in sound velocity and is on the edge of tolerance, so its priority is set to medium-low. Zone-B (segregation layer) has an extremely low sound velocity and, as an encapsulating layer, cuts off the effective bond between the isolated rock and the cement-soil, which is a fatal defect that leads to shear failure, so its priority is set to the highest.

[0076] During the morphological fusion process, the system uses Zone-B as the core seed point, considers Zone-A as an internal inclusion, and removes its high sound velocity attribute from the defect definition, treating it only as a physical occupant. Since the sound velocity of Zone-C is close to the acceptable value and it is located on the periphery, the system determines it as a transition zone and smooths it during fusion, excluding it from the core defect volume. The system generates the final morphology of Review-Zone-01, described as a low-strength cemented defect attached to the surface of a hard inclusion. Its main body presents an irregular curved shell with a thickness of about 3.5cm, located at a depth of 8.0m and covering the lower half of the hard core. This morphology is marked as the final evaluation object for subsequent optimization decisions in S14. Through S142, the review problem of the cement mixing pile is accurately characterized. The system has captured the critical final morphology of the segregated encapsulation layer, providing an accurate target for subsequent precise reinforcement.

[0077] Therefore, the final form of the inspection area is compared with the preset form of the cement mixing pile, and multiple morphological differences are presented during the comparison process. At the same time, the forming optimization system of cement mixing pile is determined based on the detection of the cement mixing pile forming database. The corresponding forming optimization events are determined according to each morphological difference, the forming process of cement mixing pile, and the forming optimization system. This approach takes into account the overall consideration of each morphological difference, the forming process of cement mixing pile, and the forming optimization system, ensuring the accuracy of the corresponding forming optimization events.

[0078] At this point, the system will accurately register the 3D model of the reconstructed inspection area based on ultrasonic data with the CAD design model in the spatial coordinate system to eliminate systematic errors caused by pile tilting or rotation. The system will perform a Boolean difference operation between the preset form and the final form. If the difference result is not empty, it means that there is missing material (such as narrowing or voids). If the operation overflows, it means that there is redundant material (such as widening). For the difference area (i.e., the part with morphological difference), the system will extract key geometric descriptions, including volume deviation rate (percentage of missing volume), spatial distribution (located in the pile core or on the pile side), connectivity (whether it penetrates the pile body), and minimum wall thickness. The system will render the difference part on the interface through a chromatographic cloud map. For example, it will highlight the severely narrowed area in red and the slightly segregated area in yellow.

[0079] The system extracts fingerprints (such as depth, stratum type, and defect characteristics) of the current morphological differences and performs a K-nearest neighbor (KNN) search in the forming database. The database returns historical successful case data for resolving such defects, including the corrected construction parameters used at the time (such as re-mixing speed and grouting pressure increase) and the evaluation of the corrected effect. Based on the search results, the system dynamically generates a forming optimization system. This system includes an algorithm model for such differences, i.e., a correction coefficient matrix. For example, it is defined that when encountering segregation of isolated boulders, the mixing speed coefficient needs to be adjusted to 0.7 and the grouting volume coefficient needs to be adjusted to 1.2.

[0080] The system logically correlates the molding process in S131 (such as insufficient slurry spraying due to excessive lifting) with the morphological differences (volume loss) in S143 to verify the physical cause of the defect. Combining the correction coefficients provided by the molding optimization system, the system runs its internal decision tree. If the difference is local and minor, it outputs a mark observation instruction; if it is a structural defect, it outputs a reinforcement intervention instruction. The final generated molding optimization event includes specific action parameters, such as: fixed-point re-stirring (specifying depth range), slurry injection (specifying pressure and flow rate), or adjustment of subsequent process parameters.

[0081] Specifically, the final form of the cement mixing pile Review-Zone-01 has been determined to be a low-strength cementing defect (3.5cm thick irregular curved shell, 8.0m deep) attached to the surface of hard inclusions. The system compares this final form with the standard design model of the cement mixing pile (a cylinder with a diameter of 500mm). The system found a significant negative difference set at a depth of 8.00 meters and a southeast orientation of 45°-90°. That is, the actual effective pile cross-section is about 0.0025m³ less than the design cross-section, and the missing part is located around the isolated rock that should be a high-strength interface. The system renders this area as dark red on the 3D model and marks it as a "bonding failure zone," indicating that the shear strength of this area is expected to be less than 60% of the design value.

[0082] The system calls the established database to search for the feature of interface bonding failure around the boulder. The database matching results show that 10 sets of construction records under similar geological conditions were returned. The data shows that conventional mixing cannot solve the sliding effect on the surface of the boulder. The optimal solution is to reduce disturbance and perform high-pressure infiltration grouting. Based on this, the system loads an optimized algorithm for this type of defect. The system stipulates that for this type of interface failure, a fixed-point static pressure grouting strategy must be adopted, and an early-strength agent must be added to the grout to prevent further diffusion.

[0083] Based on the forming process of the cement mixing pile at that time (the high-speed lifting recorded in S131 led to insufficient grouting), the system comprehensively generated optimization events; the events were characterized as local defect events requiring immediate intervention; the system output specific forming optimization events—local re-mixing and fixed-point grouting reinforcement, including the following specific actions: Action 1 (re-mixing) is to instruct the drilling rig to re-drill down to 7.5m, and then lift it to 8.5m at a low speed (0.3m / min) to perform secondary mechanical crushing of the defect area; Action 2 (grouting) is to start the high-pressure grouting pump when the re-mixing passes the 8.0m position, and perform fixed-point injection at an instantaneous pressure of 1.5 times the design pressure (lasting 30 seconds), forcing the cement grout to penetrate into the low-strength coating layer; Action 3 (process adjustment) is for subsequent construction pile sections (below 8.5m), the system automatically generates optimization instructions, requiring the lifting speed to be forcibly locked below 0.8m / min to prevent under-grouting from occurring again.

[0084] refer to Figure 6 In step S15, the specific steps are as follows: S151: Identify the molding optimization event and output the molding optimization content. Based on the context analysis of the molding optimization content, determine multiple molding optimization projects. Based on the comparison of each molding optimization project, determine the corresponding project priority and mark the project content of each molding optimization project. S152: Collect the molding requirements of cement mixing piles, determine multiple molding standard parameters based on the identification of the molding requirements of cement mixing piles, and determine the multimodal data of each molding stage according to the multiple molding standard parameters and the project content of each molding optimization project. S153: Determine the optimization content of each molding stage based on the multimodal data of each molding stage and the project priority of each molding optimization project; determine the corresponding molding optimization part based on the optimization content of each molding stage and the molding process of cement mixing piles, and trigger the dynamic update of the molding process table of cement mixing piles based on the molding optimization part.

[0085] In the embodiments of this application, the molding optimization event is identified and molding optimization content is output. Multiple molding optimization items are determined based on the context analysis of the molding optimization content. The corresponding item priority is determined based on the comparison of each molding optimization item, and the item content of each molding optimization item is marked. This approach takes into account the overall consideration of comparing each molding optimization item and ensures the accuracy of the corresponding item priority.

[0086] At this point, the system reads the type label of the optimization event (such as reinforcement, adjustment, and shutdown) and identifies the process domain to which the event belongs through the event feature library; it extracts key control parameters from the event description, such as extracting the grouting volume and pressure target from grouting reinforcement, and extracting the depth range and speed requirements from re-stirring; the system converts the above unstructured or semi-structured instructions into a standardized molding optimization content data package, covering the action type, target object, and preset parameter values.

[0087] The system captures real-time contextual information from the site, including equipment status (whether the drilling rig is currently within the allowable re-drilling depth range, whether the grouting pump pipeline has pressure), resource constraints (whether the remaining material level of the cement truck is sufficient to support additional grouting), and spatiotemporal constraints (whether the current grout initial setting time is approaching). Based on this, the system atomizes the optimization content according to the construction process logic. For example, grouting reinforcement is broken down into specific projects such as pipeline cleaning, positioning and drilling, high-pressure jetting, and lifting and mixing. Each decomposed action is marked as an independent optimization project and assigned contextual attributes, such as it must be completed before the grout initial setting.

[0088] The system calculates a comprehensive priority score for each optimization project. The calculation formula typically includes a safety factor (whether failure will lead to a safety incident, which has the highest weight), a timeliness factor (whether there is a strict time window), and a cost factor (whether delayed execution will lead to greater rework costs). The projects are sorted in descending order based on the calculated scores. The projects with the highest scores are marked as critical path projects and must be executed immediately, while projects with lower scores can be appropriately postponed. The system generates a final project content label for each sorted project, which contains specific execution parameters.

[0089] Specifically, S143 has generated an optimization event for the segregation defect of isolated boulders at 8.0 meters in the cement mixing pile—local re-mixing and fixed-point grouting reinforcement. The system received the optimization event code issued by S143; the system identified its type as defect repair-grouting type, and parsed and output the specific optimization content description: within the depth range of 7.8m to 8.2m, re-mixing and grouting are carried out, with the goal of increasing the sound velocity in this area to above 1900m / s and correcting the subsequent lifting speed.

[0090] The system analyzed the current construction context of the cement mixing pile. The context showed that only 15 minutes had passed since the first grouting was completed, and the cement slurry was in the early stage of initial setting, which was the optimal window for re-mixing. At the same time, the drill rod was currently at the top of the pile (15m depth), requiring re-drilling. There was residual slurry in the grouting pump pipeline that could be used directly. Based on the re-mixing and grouting, the system identified the following optimization items: Item A: Drilling the drill rod to 7.8m (prerequisite action); Item B: Starting the high-pressure grouting pump and maintaining a pressure of 28MPa (core repair action); Item C: Slowly raising and re-mixing at a speed of 0.3m / min to 8.2m (core repair action); Item D: Updating the lifting speed parameter of subsequent pile sections to 0.6m / min (preventive action).

[0091] The system prioritized items A, B, C, and D. The priority determination results showed that item A (drilling) is the physical prerequisite for executing B and C, with a priority of Level 1 (highest). Item B (grouting) directly determines whether defects can be filled and is subject to the initial setting time of the grout, making it highly time-sensitive, with a priority of Level 1. Item C (lifting) is an auxiliary action to grouting, with a priority of Level 2. Item D (parameter update) targets unconstructed pile sections and has no time urgency, with a priority of Level 3 (lowest). Based on this, the system ultimately marked item A as emergency drilling positioning (TargetDepth: 7.8m), item B as high-pressure grouting reinforcement (TargetPressure: 28MPa, LimitTime: 30min), item C as slow re-stirring lifting (Speed: 0.3m / min, Range: 7.8m > 8.2m), and item D as subsequent process parameter locking (MaxSpeed: 0.6m / min for Depth < 8.2m).

[0092] Furthermore, the molding requirements of cement mixing piles are collected, and multiple molding standard parameters are determined based on the identification of the molding requirements of cement mixing piles. Multimodal data for each molding stage are determined according to the multiple molding standard parameters and the project content of each molding optimization project. This approach takes into account the overall consideration of multiple molding standard parameters and the project content of each molding optimization project, ensuring the accuracy of multimodal data for each molding stage.

[0093] At this point, the system integrates geometric information from the BIM model, mix proportion information from the bill of materials, and soil mechanical parameters from the geological survey report through an interface; the system identifies textual and numerical specification requirements and maps them to specific physical quantity domains, such as mapping the design strength to the standard value of compressive strength, and mapping uniform mixing to the grout density fluctuation range and mixing time; the system determines specific control thresholds and constructs a set of standard parameters covering geometric parameters (such as pile diameter deviation and verticality deviation), material parameters (such as water-cement ratio, cement admixture ratio, and grout flow rate), and process parameters (such as lifting speed, drilling speed, and grouting pressure).

[0094] The system establishes a mapping table between standard parameters and sensor modes. Mechanical modes correspond to drilling resistance and torque to reflect soil breaking and mixing resistance; fluid modes correspond to pumping pressure and flow meter readings to reflect grouting volume; and acoustic modes correspond to ultrasonic wave velocity, initial wave amplitude, and dominant frequency to reflect internal medium solidification and density. For the optimization items determined in S151, the system sets dynamic data thresholds for each construction stage (down drilling, grouting, and re-mixing). For example, during the grouting stage, the pressure threshold for the fluid mode should be higher than the conventional value; during the re-mixing acceptance stage, the wave velocity threshold for the acoustic mode should reach the design value. The system outputs a structured dataset containing data type, sampling frequency, and acceptable threshold range as input for the optimization execution in S153.

[0095] Specifically, S151 has formulated four optimization projects for the 8.0-meter section of cement mixing piles, and the system has read the detailed technical specifications of the project to which the cement mixing piles belong. For the current defect in the silty sand layer and boulder mixed stratum, the specification requires that the pile quality of cement mixing piles must reach an effective pile diameter of 500mm and an unconfined compressive strength of ≥2.0MPa after 28 days. Based on the above requirements, the system calculates the forming standard parameters that must be controlled during the construction process. Parameter P1 stipulates that the grout flow rate must be stable at 30-35L / min; parameter P2 stipulates that the grouting pump pressure must not be lower than 20MPa to ensure the splitting of the boulder coating layer; parameter P3 stipulates that the ultrasonic velocity of the pile body after re-mixing must be ≥1900m / s to meet the corresponding strength standard; parameter P4 stipulates that the re-mixing lifting speed must be strictly controlled at 0.3±0.1m / min.

[0096] The system combines the above standard parameters and the four optimization items of S151 to generate a multimodal data monitoring scheme. In the drilling positioning stage of Phase 1 (corresponding to Project A), the system focuses on monitoring the depth encoder data (position mode), requiring the depth error to be <5cm. At the same time, it monitors the drill rod torque (mechanical mode). If the torque suddenly increases (>60kN·m), it is determined that the boulder has been hit again, and the angle needs to be finely adjusted.

[0097] During the high-pressure grouting stage of Phase Two (corresponding to Project B), the system focuses on monitoring the electromagnetic flowmeter reading and the pressure sensor voltage (fluid mode). The flow rate must be locked at 32L / min, and the pressure waveform must be maintained in the peak range of 25-28MPa. If the pressure is instantaneously <20MPa, the system determines that there is a risk of pipeline blockage and triggers an alarm.

[0098] In the slow re-stirring and acceptance phase of stage three (corresponding to project C), the system focuses on monitoring the full wave train data of ultrasonic waves (acoustic modes), and performs high-density scanning on the 7.8m-8.2m area. It requires that the duration of the first wave of received ultrasonic waves be shortened, the calculated sound velocity be >1900m / s, and the amplitude gain of the first wave be >6dB, to indicate that the slurry has filled the coating layer.

[0099] In the subsequent construction phase of Phase 4 (corresponding to Project D), the system focuses on monitoring the pulse frequency (motion mode) of the lifting speed sensor, requiring that the calculated lifting speed at any time must not trigger the threshold lock of >0.6m / min, to ensure that subsequent pile segments will not be under-grouted due to excessive speed; through S152, the re-mixing grouting of cement mixing piles is strictly constrained by a series of specific sensor values, realizing digital closed-loop control of construction quality.

[0100] Therefore, the optimization content for each molding stage is determined based on the multimodal data of each molding stage and the project priority of each molding optimization project. The corresponding molding optimization part is determined based on the optimization content of each molding stage and the molding process of the cement mixing pile. This optimization part triggers a dynamic update of the cement mixing pile molding process table, accommodating the overall consideration of the optimization content of each molding stage and the molding process of the cement mixing pile. This ensures the accuracy of the corresponding molding optimization part. Simultaneously, molding optimization events are further controlled, fully considering the optimization content of each molding stage, improving the accuracy of the optimization content of each molding stage, and dynamically updating the cement mixing pile molding process table.

[0101] At this point, the optimization content is not a simple parameter list, but a dynamic control strategy that combines priority weights for a specific stage; the system calculates the specific control instruction set for each stage based on the multimodal data threshold set in S152 and the project urgency in S151.

[0102] The system performs data-driven parameter generation, using the multimodal data thresholds determined in S152 as constraints to solve for process parameters in reverse. For example, it sets the target sound velocity > the required cement slurry flow density > and determines the grouting pump speed setting. The system performs priority weighting allocation. For high-priority projects, it generates mandatory locking optimization content, where parameter settings cannot be manually modified and must reach the set threshold to proceed to the next stage. For low-priority projects, it generates suggested guidance optimization content, where parameter settings are usually a range, allowing for manual fine-tuning within the range. The system outputs structured optimization content for each construction stage, including input parameter settings, process control logic, and output acceptance criteria.

[0103] The system searches along the timeline of the forming process to locate the original process timestamp that caused the defect, and simultaneously locks the corresponding pile segment on the depth axis. Through correlation analysis, the system determines whether the current optimization content covers the original process or is added after the original process. If it is a corrective optimization (such as grouting), the optimized part corresponds to a specific depth segment in the original process. If it is a preventive optimization (such as speed adjustment), the optimized part corresponds to a future time period in the original process that has not been executed. In the process flow diagram, the system marks the corresponding process nodes as to be updated or to be inserted, generating virtual process patches.

[0104] The system writes the calculated optimization content into the marked optimization section and drives the MES or BIM5D platform through the API interface to realize the automatic redrawing of the process table and the distribution of equipment control parameters. The system adopts a differential update mechanism to retain unaffected process nodes and insert or replace the optimized parts into the original process. For example, it adds a defect repair process after the current active process or modifies the attribute fields of subsequent processes. The system directly sends the optimization parameter values ​​determined by S153 to the PLC controller. For high-priority projects, the parameter locking function on the controller is activated to block manual intervention. On the front-end interface, the process table is automatically refreshed, the modified processes are highlighted, the new repair processes are inserted into the flowchart, and the estimated completion time is updated synchronously.

[0105] Specifically, S152 has determined the multimodal data for cement mixing piles, and S151 has determined the project priority. The system generates specific construction instructions based on the priority and data thresholds. For the drilling positioning stage (priority: high), the system generates precise positioning instructions, sets the drill rod verticality correction angle to +1.5° to align with the center of the boulder, and locks the upper limit of the drilling speed to 1.0m / min to prevent further deviation. For the high-pressure grouting stage (priority: highest), the system generates forced grouting logic, sets the grouting pump oil pressure to 28MPa and activates the constant pressure mode. The logic is set so that if the real-time pressure is <25MPa, the system automatically alarms and stops lifting until the pressure recovers.

[0106] For the re-mixing acceptance stage (priority: high), the system generates an acoustic closed-loop acceptance command, sets the lifting speed to 0.3 m / min, and integrates an ultrasonic real-time analysis algorithm. If the sound velocity at the current depth is <1900 m / s, the lifting mechanism will automatically stop. For the subsequent pile segment construction stage (priority: medium), the system generates speed limit parameter suggestions and modifies the default lifting speed limit in the equipment parameter table for subsequent depths to 0.6 m / min.

[0107] The system analyzed the forming process of cement mixing piles; the system located the timestamp T_Process_01 (corresponding to a depth of 7.8m-8.2m), which was marked as completed but had quality defects in the original process; the system marked the spatial depth segment corresponding to T_Process_01 as the optimization part that needs to be forcibly covered, and marked the process corresponding to the subsequent depth as the optimization part of parameter correction, thus achieving accurate locking of the problem area.

[0108] The system automatically inserted a new process node—defect repair-fixed-point re-mixing grouting—into the process table. This node was marked as the critical path and was expected to take 15 minutes. At the same time, the system rewrote the parameter configuration table for subsequent processes and forcibly locked the speed field to 0.6. These updates were sent to the touch screen in the drilling rig cab in real time via the Industrial Internet of Things. The speed control knob on the control panel was blocked by the system software logic and could only operate at the optimized new speed, ensuring that the optimization measures could not be tampered with and were thoroughly implemented, and truly realizing the closed-loop control of intelligent construction.

[0109] Please see Figure 7 , Figure 7 This is a schematic diagram of the structural composition of a dynamic monitoring system for cement mixing piles based on ultrasonic detection according to an embodiment of the present invention; the dynamic monitoring system for cement mixing piles based on ultrasonic detection includes: The ultrasonic detection event module 21 is used to determine multiple ultrasonic data based on the detection of the mixing space between cement and soil during the molding process of cement mixing piles, and to determine ultrasonic detection events based on the multiple ultrasonic data, the preset shape of cement mixing piles and the corresponding molding data. The molding impact event module 22 is used to determine multiple molding stages of cement mixing piles based on the identification of ultrasonic detection events, and to determine the molding impact events of cement mixing piles during the molding process based on the stage content, corresponding molding form and corresponding ultrasonic data combination of each molding stage. The dynamic monitoring system module 23 is used to collect the forming process of cement mixing piles, determine the re-inspection area of ​​cement mixing piles based on the forming process of cement mixing piles and the corresponding forming influencing events, and determine the dynamic monitoring system of ultrasonic testing based on the regional location, corresponding regional morphology and corresponding forming stage of the re-inspection area. The molding optimization event module 24 is used to determine the final shape of the re-examination area based on the identification of the dynamic monitoring system of ultrasonic testing in the dynamic monitoring system of ultrasonic testing, and to determine the molding optimization event based on the final shape of the re-examination area and the preset shape of the cement mixing pile. The optimization module 25 is used to identify multiple molding optimization projects based on the identification of the molding optimization event, and to determine the optimization content of each molding stage based on the project content, corresponding project priority and molding requirements of the cement mixing pile, so as to dynamically update the molding process table of the cement mixing pile.

[0110] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A dynamic monitoring method for cement mixing piles based on ultrasonic testing, characterized in that, include: During the molding process of cement mixing piles, multiple ultrasonic data are determined based on the detection of the mixing space between cement and soil. Ultrasonic detection events are determined based on the multiple ultrasonic data, the preset shape of the cement mixing pile, and the corresponding molding data. Multiple forming stages of cement mixing piles are determined based on the identification of ultrasonic detection events. The forming influencing events of cement mixing piles during the forming process are determined based on the stage content, corresponding forming morphology and corresponding ultrasonic data combination of each forming stage. The forming process of cement mixing piles is collected. Based on the forming process and corresponding forming influencing events, the re-inspection area of ​​cement mixing piles is determined. Based on the regional location, corresponding regional morphology and corresponding forming stage of the re-inspection area, a dynamic monitoring system for ultrasonic testing is determined. The re-inspection area inherits the main characteristics of the internal anomalies and is marked as a whole risk unit. The dynamic monitoring system defines the optimal technical parameter configuration for long-term or subsequent monitoring in the re-inspection area. In the dynamic monitoring system of ultrasonic testing, the final shape of the re-examination area is determined based on the identification of the dynamic monitoring system of ultrasonic testing, and the forming optimization event is determined based on the final shape of the re-examination area and the preset shape of the cement mixing pile. Based on the identification of the molding optimization event, multiple molding optimization projects are determined. The optimization content of each molding stage is determined according to the project content, corresponding project priority and molding requirements of cement mixing piles, so as to dynamically update the molding process table of cement mixing piles.

2. The dynamic monitoring method for cement mixing piles based on ultrasonic testing according to claim 1, characterized in that, During the molding process of cement mixing piles, multiple ultrasonic data points are determined based on the detection of the mixing space between cement and soil. Ultrasonic detection events are then determined based on these multiple ultrasonic data points, the preset shape of the cement mixing pile, and the corresponding molding data. These events include: The forming process of cement mixing piles is monitored in real time, and the mixing position of cement and soil is marked. The mixing space between cement and soil is determined by tracing the mixing position, and ultrasonic testing is performed on the mixing space. Multiple ultrasonic data are determined during the ultrasonic testing process. A formation database of cement mixing piles is collected. Based on the traversal of this formation database, a preset shape of the cement mixing piles is determined. An ultrasonic testing framework is determined based on the preset shape of the cement mixing piles and multiple ultrasonic data. Ultrasonic testing events are determined based on the ultrasonic testing framework and multiple formation data of the cement mixing piles.

3. The dynamic monitoring method for cement mixing piles based on ultrasonic detection according to claim 1, characterized in that, The process involves identifying multiple forming stages of the cement mixing pile based on the recognition of ultrasonic detection events, and determining the forming influencing events of the cement mixing pile during the forming process based on the stage content, corresponding forming morphology, and corresponding ultrasonic data combination of each forming stage, including: The ultrasonic detection events are dynamically identified, and multiple stage markers are determined during the identification process. The corresponding forming stage is determined by tracing each stage marker, so as to identify multiple forming stages of cement mixing piles and mark the corresponding stage types. In multiple molding stages, the corresponding stage content is determined based on the identification of each molding stage, and the molding form and corresponding ultrasonic data combination in each molding stage are marked. The first molding influencing factor is determined according to the stage content and corresponding molding form of each molding stage. The second set of forming influencing factors is determined based on the stage content of each forming stage and the corresponding combination of ultrasonic data. Based on the mapping relationship table of the first set of forming influencing factors, the second set of forming influencing factors and forming influencing events, the forming influencing events of cement mixing piles in the forming process are determined.

4. The dynamic monitoring method for cement mixing piles based on ultrasonic testing according to claim 1, characterized in that, The process of collecting data on the formation of cement mixing piles is used to determine the re-inspection area of ​​the cement mixing piles based on the formation process and corresponding influencing events. Furthermore, a dynamic monitoring system for ultrasonic testing is determined based on the location, morphology, and formation stage of this re-inspection area, including: The cement mixing piles are monitored in real time, and their forming process is marked. This forming process presents all the contents of the cement mixing pile during the forming process. Based on the forming process and the corresponding forming impact events, multiple abnormal forming contents are identified, and the corresponding abnormal forming locations are marked. Based on each abnormal forming content and the corresponding abnormal forming location, the review area of ​​the cement mixing pile is constructed and determined.

5. The dynamic monitoring method for cement mixing piles based on ultrasonic detection according to claim 4, characterized in that, The process of collecting the cement mixing pile's formation history, determining the re-inspection area of ​​the cement mixing pile based on the formation history and corresponding formation influencing events, and determining the dynamic monitoring system for ultrasonic testing based on the regional location, corresponding regional morphology, and corresponding formation stage of the re-inspection area, also includes: The identified areas for review of cement mixing piles are further reviewed, and the location of these areas is marked. Based on the matching between the review area and the forming process of the cement mixing piles, the corresponding regional morphology and forming stage are determined. A dynamic monitoring system for ultrasonic testing is then established based on the location of the review area, the corresponding regional morphology, and the corresponding forming stage.

6. The dynamic monitoring method for cement mixing piles based on ultrasonic testing according to claim 1, characterized in that, In the dynamic monitoring system of ultrasonic testing, the final shape of the re-examination area is determined based on the identification of the dynamic monitoring system of ultrasonic testing, and the forming optimization event is determined based on the final shape of the re-examination area and the preset shape of the cement mixing pile, including: The dynamic monitoring system of the ultrasonic testing is monitored in real time, and the dynamic monitoring system of the ultrasonic testing is dynamically identified, and multiple sub-examination areas are identified during the identification process; Based on the identification of each sub-review region, the corresponding morphological features are determined, and the final morphology of the review region is determined according to the feature position of multiple morphological features, the corresponding characteristic morphology, and the priority of each morphological feature.

7. The dynamic monitoring method for cement mixing piles based on ultrasonic testing according to claim 6, characterized in that, In the dynamic monitoring system for ultrasonic testing, the final shape of the re-examination area is determined based on the identification of the dynamic monitoring system for ultrasonic testing, and the forming optimization event is determined based on the final shape of the re-examination area and the preset shape of the cement mixing pile, which also includes: The final form of the review area is compared with the preset form of the cement mixing pile, and multiple morphological differences are presented during the comparison process. At the same time, the forming optimization system of cement mixing pile is determined based on the detection of the cement mixing pile forming database, and the corresponding forming optimization events are determined according to each morphological difference, the forming process of cement mixing pile, and the forming optimization system of cement mixing pile.

8. The dynamic monitoring method for cement mixing piles based on ultrasonic testing according to claim 1, characterized in that, The process involves identifying multiple molding optimization projects based on the identified molding optimization event, determining the optimization content for each molding stage based on the project content, corresponding project priority, and molding requirements of the cement mixing pile, and dynamically updating the molding process table of the cement mixing pile, including: The molding optimization event is identified and the molding optimization content is output. Multiple molding optimization projects are determined based on the context analysis of the molding optimization content. The priority of each project is determined by comparing the molding optimization projects, and the project content of each molding optimization project is marked.

9. The dynamic monitoring method for cement mixing piles based on ultrasonic detection according to claim 8, characterized in that, The process of identifying multiple molding optimization projects based on the identification of the molding optimization event, and determining the optimization content of each molding stage based on the project content, corresponding project priority, and molding requirements of the cement mixing pile, to dynamically update the molding process table of the cement mixing pile, also includes: The molding requirements of cement mixing piles are collected, and multiple molding standard parameters are determined based on the identification of the molding requirements of cement mixing piles. Multimodal data of each molding stage are determined according to the multiple molding standard parameters and the project content of each molding optimization project. The optimization content for each molding stage is determined based on the multimodal data of each molding stage and the project priority of each molding optimization project; the corresponding molding optimization part is determined based on the optimization content of each molding stage and the molding process of cement mixing piles, and the dynamic update of the molding process table of cement mixing piles is triggered based on the molding optimization part.

10. A dynamic monitoring system for cement mixing piles based on ultrasonic detection, characterized in that, The ultrasonic-based dynamic monitoring system for cement mixing piles is applied to the ultrasonic-based dynamic monitoring method for cement mixing piles as described in any one of claims 1-9.

Images