Visualized grouting quality online evaluation and self-correction system and method

Abstract

The present application relates to the technical field of foundation engineering, and particularly relates to a visual grouting quality online evaluation and self-correction system and method, which comprises a data acquisition module integrated with multiple sensors; a data processing module for generating standardized grouting process data from sensor data; a state analysis module for real-time evaluation of the standardized data, identification of grouting state and quality conditions by using preset atlas analysis algorithms and an expert rule base; a geological information module for generating a geological response atlas for assisting evaluation and control decision-making; and a control execution module for adjusting the operating parameters of grouting equipment, including valve opening and closing, grout flow rate and grouting sequence, according to the evaluation results output by the state analysis module, so as to realize adaptive correction control of the grouting process. The present application can effectively avoid under-grouting, over-grouting and abnormal grout escape, improve grouting uniformity and construction quality, reduce the risk of manual intervention and misjudgment, and significantly improve the safety, efficiency and engineering reliability of grouting operations.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of foundation engineering, in particular to a visual grouting quality online evaluation and self-correction system and method. BACKGROUND

[0002] Curtain grouting is a technique that injects cement slurry, chemical slurry and other materials into rock or soil through drilling to form a vertical or inclined continuous impermeable body (i.e. "curtain"), which is mainly used to block groundwater seepage, reinforce soft ground, and prevent the spread of pollutants.

[0003] In the prior art, for foundation curtain grouting, consolidation grouting and other processes: most solutions only conduct survey or spot check before and after grouting, and cannot evaluate the diffusion and consolidation state of slurry in the target soil during the grouting process; even if pressure, flow and other online monitoring are introduced, they are usually used for recording or manual experience judgment, without forming a quantifiable "over-grouting" and "under-grouting" evaluation index system; existing visualization, digital twin or geophysical imaging solutions mostly focus on the visualization and effect evaluation of the whole project or the whole section, lack device-level closed-loop control on the scale of a single hole or a single grouting section, and cannot automatically perform actions such as supplementary grouting and pressure relief backfilling for correction; grouting parameters (pressure curve, termination criteria, etc.) mostly rely on pre-design and manual experience, and cannot adaptively optimize over-grouting / under-grouting criteria and control parameters according to historical data as the project progresses.

[0004] Therefore, there is a need for a device and method that can obtain working conditions and soil responses in real time during foundation grouting, online evaluate grouting quality, and automatically correct over-grouting and under-grouting states, so as to improve foundation consolidation quality, reduce slurry waste and the risk of disastrous spread. SUMMARY

[0005] Based on the above description, the present application provides a visual grouting quality online evaluation and self-correction device, which effectively avoids under-grouting, over-grouting and abnormal slurry escape, improves grouting uniformity and construction quality, reduces manual intervention and misjudgment risk, and significantly improves the safety, efficiency and engineering reliability of grouting operations.

[0006] In one aspect, the technical solution of the present application to solve the above technical problems is as follows: a visual grouting quality online evaluation and self-correction system, comprising a data acquisition module, a data processing module, a state analysis module, a geological information module and a control execution module;

[0007] The data acquisition module integrates multiple sensors for real-time acquisition of pressure, flow, cumulative grouting volume, backfilling volume, slurry conductivity, and stratum vibration and pore pressure parameters during grouting;

[0008] The data processing module calibrates and filters the data of each sensor and synchronizes the time to generate standardized grouting process data;

[0009] The state analysis module uses a preset atlas analysis algorithm and an expert rule base to perform real-time evaluation on the standardized data, and identifies the grouting state and quality condition;

[0010] The geological information module associates historical geological survey data with real-time grouting data, and generates a geological response atlas for assisting evaluation and control decision-making;

[0011] The control execution module automatically adjusts the operating parameters of the grouting equipment, including valve opening and closing, slurry flow rate, and grouting sequence, according to the evaluation results output by the state analysis module, so as to realize adaptive deviation correction control of the grouting process.

[0012] The above technical solution integrates data acquisition, processing, analysis, geological correlation, and control execution modules to form a complete grouting quality evaluation and feedback closed loop, realizes intelligent process control, collects grouting and formation response data in real time, determines the grouting state in combination with atlas analysis and geological characteristics, automatically adjusts grouting parameters and processes, effectively avoids under-grouting, over-grouting, and abnormal slurry escape, improves grouting uniformity and construction quality, reduces manual intervention and misjudgment risk, and significantly improves the safety, efficiency, and engineering reliability of grouting operations.

[0013] On the basis of the above technical solution, the present application can also be improved as follows.

[0014] Further, the data acquisition module includes pressure sensors, flow meters, slurry concentration or conductivity sensors, back slurry metering devices installed at the grouting pipeline and the drill hole, and vibration acoustic sensors and pore water pressure sensors arranged in the grouting area, for acquiring multi-source data of grouting dynamics and formation response.

[0015] Through the above technical solution, the vibration acoustic sensor and the pore water pressure sensor are buried in the hole wall or the surrounding soil to monitor the formation filling reaction, capture information such as cavity closure, slurry penetration path, and structural response, and improve the identification ability of hidden problems such as under-grouting and slurry escape; this data acquisition scheme has the advantages of modularity, scalability, and strong adaptability, and can be applied not only to traditional single-hole grouting operations, but also to complex scenarios such as multi-hole synchronous grouting and diaphragm walls; at the same time, it is physically integrated with the grouting equipment, can realize integrated construction deployment, reduces wiring and misconnection risk, and significantly improves engineering practicability and system stability.

[0016] Further, the data processing module includes a signal conditioning circuit, an A / D converter, and a data synchronization unit, for amplifying and filtering, analog-to-digital conversion, and time alignment processing of the original signals of different sensors, and outputting the processed data in a predetermined format.

[0017] By the technical solution, the data synchronization unit ensures consistency of different parameters in time dimension based on the timestamp or trigger synchronization mechanism, which is beneficial for multi-parameter association analysis; the data output in unified format not only meets the input requirements of the subsequent state identification module, but also can realize high-order functions such as historical data comparison, atlas generation and machine learning modeling; the module can also support edge computing deployment, avoid excessive dependence on central server processing resources, enhance the system's independent operation capability, and is particularly suitable for complex construction sites or tunnel, foundation pit and other network signal limited areas, ensuring that the system has high stability and real-time response capability in actual engineering.

[0018] Further, the state analysis module has a built-in grouting process expert rule base, which contains threshold conditions and logical rules for identifying under-grouting, over-grouting, blockage and escaped grouting states; the state analysis module compares and operates the real-time monitored pressure-flow curve and grouting amount change curve with the rule base, and outputs the corresponding grouting state signal and alarm prompt when the conditions of a certain rule are met.

[0019] The above technical solution, through the preset pressure-flow-time and other multi-parameter threshold rules, the system can judge whether there are common problems such as under-grouting, over-grouting, blockage and escaped grouting in real time, avoiding dependence on the experience of operators; the state identification model takes the standard atlas curve and historical law as a reference, compares the real-time monitoring data with the rule base, quickly outputs the state label and gives a risk prompt, thereby realizing "early warning and real-time control"; in addition, the module has rule updateability, which can continuously optimize the parameter threshold based on historical data, enhancing the system's ability to adapt to different geological and grouting process conditions; this way greatly improves the identification efficiency and automatic response capability, especially in multi-hole grouting, staggered strata and long-time operation, which can significantly reduce the probability of construction accidents, material waste and grouting quality deviation, and is a key technical support for the intelligent evolution of engineering sites.

[0020] Further, the geological information module includes a database for storing geological profile information and a data fusion unit, the data fusion unit associates the geological parameters of the corresponding position of the drill hole with the real-time grouting data to generate a geological response graph; the state analysis module adjusts the identification criteria of the grouting state based on the geological response graph, and the control execution module optimizes the control strategy of the grouting pressure, flow or sequence accordingly.

[0021] By the technical solution, the real-time grouting parameters are bound with the geological characteristics of the section through the space matching of the grouting hole position or section data; in the state analysis process, the system can automatically correct the threshold judgment standard according to the geological response map, for example, relaxing the upper limit of the pressure in the high permeability section and limiting the grouting rate in the weakly cemented layer section, so as to avoid misjudgment or over-control and improve the identification accuracy; the control strategy can also be automatically optimized accordingly, such as preferentially encrypting the sealing section, switching the sequence to seal the high permeability section first and the like, which helps to improve the grouting uniformity and avoid slurry escape and non-design path injection; the module is particularly suitable for the complex areas with uneven stratification, interlaced rock and soil or developed fractures, and significantly improves the intelligent level and regional adaptation ability of the whole system.

[0022] Further, the control execution module includes an electric valve, an adjustable speed grouting pump, a grout mixing device controller and a sequence control unit.

[0023] When the state analysis module determines that the grouting state is normal, the control execution module operates according to the preset process parameters;

[0024] When an abnormal state is detected, the control execution module automatically executes a correction measure, including starting and stopping the grouting pump, opening or closing the valve to change the flow direction of the slurry, adjusting the slurry ratio or switching the grouting hole, so as to correct the abnormality and restore normal grouting.

[0025] In the second aspect, the technical solution of the present application for solving the above technical problem is as follows: a method for online evaluation and self-correction of grouting quality, comprising the following steps:

[0026] Multi-parameter data acquisition; during the grouting construction process, real-time grouting data are collected through the pressure, flow, back slurry amount, concentration, conductivity, acoustic vibration and hole pressure sensors arranged;

[0027] Real-time data processing; the collected multi-source data are synchronously and standardized processed, abnormal interference is filtered out, and the data are converted into a unified time sequence grouting parameter set;

[0028] State identification analysis; the processed data are input into an expert rule base for analysis and comparison, the current grouting state is identified in combination with the grouting map features, whether under-grouting, over-grouting, blockage or slurry escape occurs is judged, and an evaluation result and prompt information are generated;

[0029] Geological information fusion; the real-time evaluation result is associated with the pre-stored geological model and parameters, the response characteristics of the current hole section stratum to grouting are analyzed, the conclusion of the state identification is corrected or evidenced, and the evaluation threshold is adjusted if necessary;

[0030] Automatic deviation correction control; real-time adjustment of grouting construction parameters according to the evaluation results, automatic execution of corresponding deviation correction control measures when abnormal is judged, including pressure reduction, pump stop, change of slurry formula or adjustment of grouting sequence; if the evaluation is normal, continue grouting according to the optimized parameters until completion.

[0031] Compared with the traditional method of relying only on manual observation of pressure and flow, the method simultaneously introduces multiple monitoring quantities such as back slurry quantity, concentration, conductivity, acoustic vibration and hole pressure, so that the grouting process is improved from single "quantity and pressure" rough judgment to comprehensive evaluation of "slurry behavior + stratum response", greatly improving the sensitivity and reliability of problem identification; through data synchronization and standardized processing, the outputs of various sensors are ensured to participate in analysis under unified time sequence and unified dimension, avoiding misjudgment caused by different sampling frequencies and different ranges. Combined with expert rule base and atlas features for state identification, typical abnormal working conditions such as under-grouting, over-grouting, blockage and slurry escape can be found in real time during grouting, and evaluation results with semantic information are output; geological information fusion further links the identification results with specific stratum conditions, making the judgment more consistent with the actual geological environment and reducing "false abnormalities" or excessive control; finally, automatic deviation correction control converts the analysis results into real-time adjustment of construction parameters, realizes the whole process closed-loop control from "finding problems" to "automatic parameter adjustment", and significantly improves the intelligentization, stability and controllability of grouting quality.

[0032] Further, the real-time data processing specifically includes: using a bidirectional time window interpolation algorithm to perform timestamp alignment processing on the collected pressure, flow, back slurry quantity, conductivity multi-source data, to ensure that each parameter has a corresponding value at the same time;

[0033] Based on the sliding average filtering algorithm or the median filtering algorithm, the original signal is denoised, the abnormal fluctuations are smoothed, and the high-frequency interference is eliminated;

[0034] The jump data or invalid data appearing in the collection process are subjected to three-sigma rule outlier rejection processing;

[0035] A linear drift correction model is used to automatically correct the sensor data with slow drift, maintaining measurement stability;

[0036] Different physical quantities are normalized and transformed according to their normal working intervals, and the data dimension is unified, so as to facilitate subsequent atlas construction and state identification.

[0037] The technical scheme has the advantages that: the three-sigma rule is used for abnormal value elimination, which can automatically identify invalid or jump data points and prevent extreme error data from pulling the overall judgment; the linear drift correction model provides an online correction method for the sensor zero drift problem in long-term construction, so that the data in the long-term monitoring process remains stable and comparable; the normalization transformation processing of different physical quantities eliminates the dimensional difference, so that the subsequent spectrum analysis and algorithm identification can be carried out in a unified numerical space, greatly simplifying the complexity of the identification model and improving the algorithm convergence; as a whole, the technical scheme ensures that the data of the input state recognition and control decision module are "clean, aligned, stable, and comparable" through a complete algorithm process, and provides key bottom support for realizing high-precision online evaluation and reliable correction control.

[0038] Further, the grouting state recognition analysis comprises:

[0039] A pressure-flow-time-concentration four-dimensional spectrum matrix is generated by using the processed grouting parameters, and the matrix is generated by setting a sliding window accumulation with a sampling frequency of 10 Hz per second as the basis for time series analysis;

[0040] The system compares the spectrum feature curve with an embedded expert rule library, and the rule library includes a composite logic criterion established based on experience:

[0041] If the flow rate reduction rate AQ / At exceeds the set threshold Q t , and the pressure rise rate AP / At is greater than the threshold P t , it is determined that the pipeline or hole section is blocked.

[0042] If the pressure remains low for a long time and the cumulative grout volume is abnormally high compared to the expected reference value, it is determined to be over-grouting or grout escape;

[0043] If the pressure rises rapidly and the grouting volume is much lower than the standard section value, it is determined to be under-grouting or poor sealing;

[0044] An adaptive dynamic threshold recognition algorithm can be optionally introduced, which automatically fits the reasonable range of key recognition variables based on statistical historical grouting data of similar strata or hole sections, improving the adaptability and accuracy of the model to different geological conditions.

[0045] By the technical solution, the four-dimensional graph matrix is used as a basis for time series analysis, which is beneficial to reveal the coupling relationship between pressure and flow, the concentration evolution law in the grouting process and the comprehensive change trend with time, so as to form visual and quantifiable grouting behavior characteristics. By comparing the graph characteristics with the composite logical criteria in the built-in expert rule library, the typical working conditions such as blockage, over-grouting, under-grouting and poor sealing can be quickly and automatically identified based on the physical mechanism and engineering experience, and the change from “looking at the curve and relying on experience” to “looking at the rules and automatically judging” is realized. For the parameter distribution difference caused by different strata and different construction methods, an adaptive dynamic threshold identification algorithm is introduced, the threshold range of each type is automatically fitted by statistical history similar working condition data, so that the identification model no longer depends on the fixed threshold, but is self-adjusted according to the actual engineering and historical data, the adaptability and identification accuracy of the model under the conditions of multiple work points and multiple strata are significantly improved, the false alarm and missed alarm are reduced, and the engineering universality and long-term use value of the system are improved.

[0046] Further, the automatic deviation correction control includes the following implementable strategies:

[0047] If the state recognition module outputs the “under-grouting” label, the system increases the upper limit of the grouting pressure by 10%-30% or prolongs the grouting duration to 150% of the original design value according to the stratum permeability of the corresponding hole section and the design target, and simultaneously enables the pressure stabilization compensation mode;

[0048] If the state recognition is “over-grouting” or “slurry leakage”, the system automatically executes the pressure reduction and speed reduction logic, including:

[0049] reducing the grouting pump speed by 20%-50%,

[0050] reducing the water-cement ratio of the slurry to improve the adhesion,

[0051] and limiting the cumulative amount of single-section grouting to be not more than a safety threshold V m .

[0052] If the state recognition is “blockage”, the system starts the back pressure-unloading-pulsation dredging program, and sequentially executes the short-time reverse unloading, opens the bypass valve to reduce the pressure difference, and periodically switches the main pump output at a frequency of 1-3 Hz for pulsation excitation;

[0053] All deviation correction actions are driven by the embedded working condition matching control rule table, which selects the optimal control scheme in combination with real-time monitoring parameters and pre-stored geological response information, and records the deviation correction process in real time for subsequent analysis and model retraining.

[0054] For the “over-grouting” or “slurry leakage” state, the grouting pump speed is reduced by 20%-50%, the water-cement ratio is adjusted to improve the adhesion of the slurry, and a safety threshold V mThe upper limit can effectively inhibit the invalid diffusion of slurry and the diffusion to non-target areas, reduce material waste and peripheral structure risk; for the "blockage" state, through the back pressure-unloading-pulsation dredging program, the online dredging of the pipeline or channel is realized, the process interruption caused by simple and rough stoppage is avoided, and the risk of equipment or hole wall damage caused by overpressure is reduced. All strategies are uniformly managed by the working condition matching regulation table, the most suitable control combination is automatically selected according to real-time monitoring data and geological response information, and the execution process is recorded for subsequent optimization and model retraining. Such a set of quantitative and adaptive deviation correction control mechanism not only significantly improves the speed and accuracy of the system in dealing with abnormal working conditions, but also ensures that the grouting process is realized within the safety boundary to achieve the optimization of quality and efficiency, which reflects the practical engineering value of the intelligent construction control of the method.

[0055] Compared with the prior art, the technical scheme of the present application has the following beneficial technical effects:

[0056] 1. Multi-source full-quantity perception and high-quality data unification of the grouting process are realized, a plurality of sensors such as pressure, flow, back slurry amount, concentration, conductivity, acoustic vibration and hole pressure are arranged in the grouting pipeline and the stratum, and a signal conditioning circuit, an A / D conversion and time synchronization unit are configured, so that the system can realize all-around and multi-scale real-time acquisition of the grouting process and the stratum response. On this basis, two-way time window interpolation, sliding average / median filtering, three-sigma abnormality rejection, linear drift correction and normalization transformation algorithms are introduced, so that data of different sources, different dimensions and different sampling frequencies are unified into a stable, clean and aligned time series parameter set. This series of measures significantly improves the accuracy, continuity and comparability of the data, provides reliable basic input for subsequent atlas construction, state recognition and control decision, and solves the key bottleneck of "disordered data that cannot be directly used" in traditional grouting monitoring;

[0057] 2. A high-precision grouting state recognition system integrating expert experience and geological information is constructed; by generating a pressure-flow-time-concentration four-dimensional atlas matrix and accumulating in a sliding window with a sampling frequency of ≥10Hz, the system can finely depict the dynamic evolution characteristics of the grouting process. Combined with the compound logic criteria for under-grouting, over-grouting, blockage and slurry escape in the built-in expert rule base, automatic recognition and typing of key abnormal states are realized. At the same time, an adaptive dynamic threshold recognition algorithm is introduced, a reasonable threshold range is automatically fitted based on historical similar hole segments or stratum data, and the real-time recognition result is associated with the pre-stored geological model and parameters to form a geological response graph, which corrects or corroborates the recognition conclusion. Through the combination of "atlas features + expert rules + dynamic thresholds + geological fusion", the system significantly improves the recognition accuracy and robustness in complex geological conditions and multi-working condition scenarios, reduces false positives and false negatives, and upgrades the grouting quality evaluation from experience-based judgment to computable and transferable intelligent judgment. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 The overall structure block diagram of the visual grouting quality online evaluation and self-correction system of the embodiment 1 of the present application;

[0059] Figure 2 The structure schematic diagram of the visual grouting quality online evaluation and self-correction system of the embodiment 1 of the present application;

[0060] Figure 3 The hardware structure block diagram of the data acquisition and data processing module of the embodiment 1 of the present application;

[0061] Figure 4 The software function structure schematic diagram of the state analysis module and the geological information module of the embodiment 1 of the present application;

[0062] Figure 5 The control execution module and the grouting flow path schematic diagram of the embodiment 1 of the present application;

[0063] Figure 6 The grouting quality online evaluation and self-correction method flow chart of the embodiment 2 of the present application;

[0064] Figure 7 The grouting state recognition and correction control logic schematic diagram of the embodiment 2 of the present application.

[0065] Reference signs: 101, vibration acoustic sensor; 102, pore water pressure sensor; 103, slurry concentration or conductivity sensor; 104, pressure sensor; 105, back slurry metering device; 106, back slurry channel; 107, main grouting pipeline. DETAILED DESCRIPTION

[0066] In order to facilitate the understanding of the present application, the present application will be described more fully below with reference to the accompanying drawings. The embodiments of the present application are shown in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein in the specification merely describe specific embodiments of the present application and are not intended to limit the present application.

[0068] Reference Figure 1 and Figure 5In the embodiment, the visual grouting quality online evaluation and self-correction system mainly comprises a data acquisition module, a data processing module, a state analysis module, a geological information module and a control execution module, each module is connected through an industrial Ethernet or a field bus, and online monitoring, intelligent evaluation and automatic correction control of the grouting construction process are realized.

[0069] I. Data acquisition module; the data acquisition module is arranged at a grouting pipeline, a borehole orifice and a grouting influence area, and is used for acquiring multi-source data of grouting dynamics and stratum response. Specifically, the data acquisition module comprises:

[0070] 1. A pressure sensor on the main grouting pipeline, which is respectively arranged at an outlet of a grouting pump, near the orifice and at a segmented pipeline if necessary, and is used for collecting instantaneous pressure P(t) in the grouting process, and the range is generally 0-5 MPa or 0-10 MPa according to the design pressure; preferably, a pressure transmitter with a flat diaphragm structure is used, for example, a WIKA S-11 type pressure sensor.

[0071] 2. A flowmeter on the main grouting pipeline, which can be an electromagnetic flowmeter or a volumetric flowmeter, and is used for outputting instantaneous flow Q(t) and obtaining cumulative grouting volume V(t) through integration, and the accuracy is preferably not less than ±1%; preferably, the main grouting flowmeter is an electromagnetic flowmeter, for example, an RF3301-SL type slurry electromagnetic flowmeter or a Promag P300 type electromagnetic flowmeter.

[0072] 3. A back grouting volume measuring device arranged in a back grouting channel, which can be a back grouting barrel with a weighing sensor or a volumetric measuring device, and is used for measuring back grouting volume R(t) and recording back grouting start time and flow change in the back grouting process; the back grouting volume measuring device can be a weighing type back grouting barrel, and a shear beam weighing sensor is arranged at the bottom of the barrel, for example, an SBK500-C6 type sensor.

[0073] 4. A slurry concentration / conductivity sensor arranged at an outlet of a slurry preparation device or on the main grouting pipeline, and is used for monitoring slurry conductivity C(t) or concentration in real time, so as to reflect changes in water-cement ratio, additive content and the like; preferably, the slurry concentration / conductivity sensor is an inductive conductivity sensor, for example, an Endress Hauser CLS54D type sensor; or an ultrasonic slurry densitometer, for example, a Rhosonics SDM type slurry densitometer.

[0074] 5. Vibration acoustic sensor, multiple vibration / acoustic emission sensors are arranged around the monitoring hole within the grouting influence range, to collect the response A(t) of ground vibration acceleration, frequency spectrum characteristics or acoustic emission count, to reflect the process of slurry penetration and fracture filling; the vibration acoustic sensor can use IEPE type piezoelectric accelerometer, with sensitivity of about 100 mV / g, frequency response of 0.5 Hz to 10 kHz or above, for example, A26D100 type or similar products.

[0075] 6. Pore water pressure sensor, pore water pressure gauges are arranged in the representative monitoring hole, observation hole or grouting hole section, to record the change of pore water pressure U(t), to determine the influence range and degree of seepage field and stress field; the pore water pressure sensor preferably uses a vibrating wire type seepage pressure gauge, for example, GEOKON 4500 series vibrating wire type pore water pressure gauge.

[0076] The above-mentioned various sensors are connected to the signal input end of the data processing module through shielded cables or field buses, and the sampling frequency can be set to 1-50 Hz, preferably not less than 10 Hz in the state identification stage, to meet the accuracy requirements of subsequent atlas construction and time series analysis.

[0077] II. Data processing module; in a preferred embodiment, the data processing module is integrated in the field data acquisition terminal or industrial controller, and is further divided into a signal conditioning sub-module, an analog-to-digital conversion (A / D) sub-module, a data synchronization and cache sub-module, and a communication management unit, for realizing unified collection, shaping, digitization and packaged transmission of multi-source sensor signals.

[0078] (1) Signal conditioning sub-module, the signal conditioning sub-module designs channel circuits for different types of sensor output forms, including current type, voltage type, frequency type and IEPE type vibration sensor, which can specifically include:

[0079] For the collection of 4-20 mA current signals (such as pressure, flow, pore water pressure sensors), a precision sampling resistor (such as a 250 Ω standard resistor) is preferably used to convert the current to a 1-5 V voltage, and then amplified by an instrument amplifier;

[0080] For 0-5V, 0-10V voltage signals (such as conductivity, displacement sensors), input protection circuits and adjustable gain amplification stages are set, and second-order or multi-order low-pass filters are configured, with a cutoff frequency of 10-50 Hz, which can be selected according to the sampling frequency, to suppress high-frequency noise and achieve anti-aliasing;

[0081] For IEPE type vibration / acoustic sensors, a constant current excitation source (such as 2-4 mA) is configured, and a high-pass / band-pass filter is set in the signal path, to extract the vibration components within the target frequency band;

[0082] For the vibrating wire pore pressure gauge or similar sensor, a special excitation and demodulation circuit can be configured to convert the vibrating wire frequency signal into an analog voltage or digital quantity proportional to the pressure;

[0083] All channels are provided with overvoltage protection and surge suppression devices (such as TVS tubes, surge absorbers, isolation amplifiers, etc.), and the electrical isolation between the signal and the high-voltage equipment on site is realized by optical or magnetic isolation to ensure the anti-interference ability and safety of the system.

[0084] After signal conditioning, the output of each channel is unified as a standard analog quantity such as 0-5V or 4-20mA, which is sent to the analog-digital conversion submodule.

[0085] (2) Analog-digital conversion (A / D) submodule, the analog-digital conversion submodule is used to convert the conditioned analog signal into a digital quantity, and a multi-channel, high-resolution A / D converter or multi-channel A / D module is preferably used. Specifically, it can include:

[0086] The A / D resolution is preferably not less than 16 bits, and a 24-bit Sigma-Delta type A / D conversion chip is further preferably used to improve the resolution of small changes in pressure, flow and pore pressure;

[0087] The sampling frequency is set according to system requirements, generally 1-50 Hz, and the sampling frequency of the key channel is set to ≥10 Hz in the grouting state identification stage;

[0088] The channel configuration can use multi-channel synchronous sampling or multi-channel fast scanning + time correction, and the key parameters (such as pressure and flow) can use synchronous sampling A / D to reduce phase difference;

[0089] The A / D module communicates with the field bus or CPU through SPI, I 2 C or parallel bus communication, and the instantaneous values of each channel are stored in the cache area after sampling is completed.

[0090] (3) Data synchronization and cache submodule, the data synchronization and cache submodule is used to unify the time reference of each channel, complete the timestamp marking and data buffering, and specifically includes:

[0091] A high-precision crystal oscillator is used as the system clock, which serves as the time reference for all acquisition tasks; if necessary, periodic time correction can be performed through NTP synchronization of the upper computer or GPS / Beidou time service module to ensure the stability of the clock during long-term operation;

[0092] A unified timestamp is generated at the beginning of each sampling period, and this timestamp is assigned to all channel data collected in this period;

[0093] For channels with different sampling periods or non-synchronous sampling, the data synchronization unit resamples or interpolates them to a unified time axis through interpolation and buffering mechanism; for a small number of missing data points, nearest neighbor or linear interpolation can be used for completion;

[0094] The collected data is cached through a ring buffer or double buffering mechanism. When the host computer or communication bus is busy for a short time, a certain length of data (e.g. 1-5 minutes) can be cached locally to avoid data loss.

[0095] While the data is being written to the cache, basic data validity checks and exception markers are performed, such as detecting out-of-range, fixed value lock, sudden abnormality, etc. to provide markers for subsequent software filtering and three-sigma exception rejection.

[0096] (4) Communication management and data packaging. The processed digital signals are packaged by the communication management unit according to the predetermined data structure to form standardized grouting process data frames. The preferred data structure is:

[0097] [timestamp] + [hole number / hole segment number] + [pressure] + [flow] + [cumulative grouting volume] + [back grouting volume] + [slurry conductivity or concentration] + [vibration amplitude / characteristic value] + [pore water pressure] + [status marker / quality flag, etc.]

[0098] The timestamp can be in UNIX time or "year-month-day hour: minute: second.millisecond" format. The hole number / hole segment number is used to distinguish different grouting locations. The status marker field is used to record information such as whether the data has been filtered, whether there are exceptions, etc.

[0099] The communication management unit can use an Ethernet controller or a fieldbus interface module to send the above data frames according to TCP / IP, Modbus TCP, Profinet or other industrial Ethernet protocols to the host computer or central server. For distributed monitoring points, data can also be uploaded through RS485 bus in Modbus RTU mode. To ensure transmission reliability, CRC check code or frame check field can be set in the data frame for error detection and retransmission control at the receiving end.

[0100] After the above processing and packaging, the standardized grouting process data can be directly used in the host computer for subsequent filtering and denoising, normalization, three-sigma exception rejection, drift correction, and pressure-flow-time-concentration four-dimensional graph matrix construction, providing a consistent and reliable data basis for online evaluation by the state analysis module.

[0101] III. State analysis module; in a preferred embodiment, the state analysis module is deployed in the host computer or industrial tablet, with the industrial computer as the running carrier. The module can be divided into the following software units: data access and cache unit, atlas construction and feature extraction unit, expert rule base and rule reasoning unit, adaptive threshold updating unit, and state output and alarm display unit, for realizing real-time atlas analysis and grouting state recognition of grouting process data.

[0102] (1) Data access and cache unit; the data access and cache unit communicates with the data processing module through industrial Ethernet or field bus, and receives standardized grouting data frames at a fixed period (for example, 0.1 s or 0.2 s).

[0103] Each data frame includes at least: time stamp t, hole number / hole segment number ID, pressure P(t), flow rate Q(t), cumulative grouting volume V(t), back grouting volume R(t), conductivity / concentration C(t), vibration characteristic value A(t), pore water pressure U(t), and data quality identifier.

[0104] The unit establishes a circular cache queue according to the hole number and hole segment number, and maintains a data window of a recent period (for example, 5-30 min) for each grouting hole segment, providing a data basis for subsequent sliding time window analysis.

[0105] When the received data quality identifier is abnormal (such as out-of-range, failure point, etc.), the point is marked as "to be excluded" or "corrected" in the cache for subsequent algorithm skipping or interpolation processing.

[0106] (2) Atlas construction and feature extraction unit; the atlas construction and feature extraction unit generates various time series curves and multi-dimensional atlases using cached data, and extracts key feature quantities.

[0107] ① Curve and atlas construction; time is taken as the horizontal axis, and the pressure-time curve P(t), flow-time curve Q(t), cumulative grouting volume-time curve V(t), back grouting volume-time curve R(t), conductivity / concentration-time curve C(t), vibration amplitude-time curve A(t), and pore pressure-time curve U(t) are plotted respectively.

[0108] The time axis is divided into several windows by a fixed length sliding time window, for example, every 10 s or 30 s as an analysis window, and a pressure-flow-time-concentration four-dimensional atlas matrix = [P, Q, C, t] is generated for each window.

[0109] The four-dimensional atlas can be displayed in a pseudo-color way by superimposing a three-dimensional surface on the time index internally, for intuitive presentation of the coupled variation patterns of pressure, flow rate, and concentration over time.

[0110] ② Feature quantity calculation; in each analysis time window, the following feature quantities are calculated, for example:

[0111] Pressure growth rate: ; Flow rate change rate: ;

[0112] Cumulative grouting volume increment: ; Grouting return volume ratio: ;

[0113] where ε is a small constant to prevent the denominator from being zero; vibration energy or vibration RMS value: ; or integrate the energy of a specific frequency band in the frequency domain; hole pressure change rate: ;

[0114] In addition, auxiliary features such as pressure-flow correlation coefficient, pressure platform maintenance time, grouting return start time, grouting return peak, etc. can be calculated as needed to finely distinguish different abnormal working conditions.

[0115] (3) Expert rule base and rule inference unit; the expert rule base and rule inference unit are the core of the state analysis module, which is used to map the above-mentioned feature quantities to specific grouting states.

[0116] ① Rule base structure; the expert rule base can adopt a hierarchical structure, including:

[0117] Basic threshold table: according to the formation type (sand layer, clay layer, fractured rock layer, etc.), grouting purpose (curtain, anti-seepage reinforcement, etc.), and hole segment design parameters (design pressure, design grouting volume, etc.), the reference intervals of each feature quantity are preset, for example allowed range, allowed range, reasonable interval, ΔV reasonable interval, etc.

[0118] Compound logic rule set: composed of "IF-THEN" form rules, multiple feature quantities are combined to determine a state. For example: if ≤-Q t and , and is lower than the set value, it is inferred as "blockage"; if P is lower than for a long time and ΔV is significantly higher than the reference value in a given time, it is inferred as "overgrouting / leakage"; if P rises rapidly in a short time and ΔV is significantly lower than the standard grouting volume at a certain depth of the hole segment , it is inferred as "undergrouting";

[0119] Priority and conflict resolution rules: When multiple rules are triggered at the same time, conflict resolution is carried out by preset priority, confidence level or weight. For example, "overfilling" with higher safety risk takes priority over "overfilling", and "clogging" with higher equipment risk takes priority over "underfilling", etc.

[0120] Rule-based reasoning process

[0121] The state analysis module executes the following reasoning process within each analysis window:

[0122] Extract the feature vector corresponding to the time window from the cached data. ;

[0123] For the current formation type of the borehole segment, obtain the corresponding threshold set from the basic threshold table. , , , , wait);

[0124] The complex logic rule set is evaluated sequentially, rules that meet the conditions are marked, and the matching degree or confidence degree of each rule is calculated.

[0125] If only one rule is triggered, the state corresponding to that rule is directly output as the current state.

[0126] If multiple rules are triggered, a unique status label and its confidence level are obtained by combining the preset priority and matching degree.

[0127] Optionally, a simple scoring mechanism or fuzzy membership calculation can be superimposed during the rule reasoning process. That is, a membership function of "normal / under-irrigation / over-irrigation / blockage / slurry overflow" is constructed for each feature quantity. The membership degree of each state is calculated based on the actual value of the feature quantity. Finally, the state with the largest membership degree is selected as the discrimination result, and the membership degree value is used as the confidence level output.

[0128] (4) Adaptive threshold update unit; In order to enhance the adaptability of the system under different engineering and geological conditions, an adaptive threshold update unit is set in the state analysis module to dynamically correct some thresholds in the rule base during the engineering process.

[0129] The adaptive module periodically collects historical grouting data under the same stratum type and similar construction conditions, and calculates the statistical indicators (such as mean, standard deviation, quantiles, etc.) of each characteristic quantity under the condition that has been "verified as normal".

[0130] Fine-tune the initial empirical thresholds in the basic threshold table, for example, adjust the thresholds under normal operating conditions. , The 95th percentile value will be used as the new upper / lower limit reference. The stable range of the threshold value as the normal backfill ratio interval;

[0131] The threshold adjustment process is controlled by limiting the amplitude, that is, the amplitude of each adjustment does not exceed a certain proportion (such as 5%-10%), to prevent the threshold from drifting too much due to individual abnormal data;

[0132] The data segments determined by manual review as "typical underfilling" and "typical overfilling" and other working conditions can be used as training samples to correct the discrimination boundaries of the corresponding states.

[0133] Through the above adaptive update, the system can gradually converge to a discrimination threshold that better meets the characteristics of the project while ensuring safety margins, improving the accuracy and robustness of state recognition.

[0134] (5) State output and alarm display unit, the state output and alarm display unit is responsible for outputting the discrimination results in machine signals and visualized manner at the same time.

[0135] ① State signal output, for each grouting hole section, the state analysis module periodically outputs the current state label (such as "normal", "underfilling", "overfilling", "slurry escape", "blockage", etc.) and confidence level;

[0136] Send the state signal to the control execution module through the software interface or field bus, and the control execution module executes the corresponding correction logic (such as pressure adjustment, speed reduction, hole section switching, etc.) accordingly;

[0137] For states that trigger serious abnormalities (such as high-confidence slurry escape, severe blockage, etc.), the module simultaneously outputs an emergency stop or forced pressure reduction instruction flag for the control execution module to respond quickly.

[0138] ② Visual alarm and color identification, in the upper computer interface, the following methods can be used for visual prompt:

[0139] In the real-time curve graph of each grouting hole, the current state is marked with different color sections, for example: green for normal, yellow for underfilling, orange for overfilling, and red for slurry escape or severe blockage; In the hole site layout diagram or cross-section diagram, the current state of each hole is displayed in the form of points or small icons, and the icons flash or have a red border when an abnormal state occurs; A pop-up window is displayed when an anomaly occurs, containing: hole number, hole section, state type, trigger time, main characteristic parameters (such as current pressure, flow, backfill amount, and calculated characteristic value), which facilitates quick judgment by on-site technical personnel;

[0140] At the same time, all state change events and alarm events are recorded to generate a time-ordered event log for subsequent quality analysis and construction review.

[0141] Through the above structure and algorithm design, the state analysis module can perform real-time atlas analysis and quantitative feature extraction on multi-source monitoring data during the grouting construction process, accurately identify typical working conditions such as under-grouting, over-grouting, blockage, and escaped grouting using expert rule library and adaptive threshold, and provide analysis results in the form of standard signals and intuitive interfaces to the control execution module and field personnel, providing reliable decision basis for online evaluation and self-correction control of the grouting process.

[0142] Four, a geological information module; in a preferred embodiment, the geological information module is provided in the upper computer or the special server, mainly composed of a geological profile database unit, a spatial index and hole section matching unit, a geological-grouting data fusion and geological response graph generation unit, and a stratum classification and threshold correction unit, used to realize the organic combination of engineering investigation results and real-time grouting monitoring data, and provide the state analysis module with adaptive geological condition discrimination basis.

[0143] (1) Geological profile database unit, the geological profile database unit is used to store and manage various types of geological information obtained during the engineering investigation stage, and realizes the standardized and structured description of the geological information.

[0144] ①Data source and content: the database pre-imports geological data from the engineering investigation report, including but not limited to: drill column chart: including hole number, plane coordinates (X, Y), hole elevation Z0, total hole depth H, etc. of each exploration hole; stratification information: stratigraphic layer number, top plate depth , bottom plate depth of each drill hole according to depth; rock type: lithology name (such as silty clay, medium sand, gravel layer, strongly weathered granite, etc.), lithology code; permeability parameters: such as permeability coefficient k, gradient, permeability test results (Lugeon value, single-hole pumping test, etc.); mechanical indicators: such as standard penetration number N value, cohesion c, internal friction angle φ, saturated uniaxial compressive strength, etc.; structural characteristics: fault fracture zone position, fissure development degree, joint spacing, soft interlayer distribution, etc.; hydrogeological information: groundwater level depth, confined water head, aquifer, and aquifuge distribution; other auxiliary parameters: natural moisture content, porosity ratio, density, etc.

[0145] ②Database structure: in order to facilitate fast retrieval and connection with grouting hole data, the geological profile database can adopt a relational database structure or an object-oriented database structure, and the typical table structure includes:

[0146] “Drill basic information table”: records hole number, coordinates, elevation, total hole depth, exploration date, etc.

[0147] “Drill stratification information table”: fields include hole number, layer number, top / bottom depth, lithology code, k value, N value, fissure index, etc.

[0148] Fault fracture zone information table: records fault name, spatial position, dip angle, width, filling, etc.

[0149] Hydrogeological information table: records regional groundwater level, aquifer top and bottom, confined water head, etc.

[0150] Each table is associated by hole number, layer number, spatial coordinates, etc. Key values are established and indexed to quickly query by hole number, depth or formation type.

[0151] (2) Spatial index and borehole-hole segment matching unit

[0152] The spatial index and matching unit is responsible for establishing spatial correlation between construction grouting holes and exploration hole databases, and realizing quick matching of "grouting hole-geological profile".

[0153] ① Grouting hole parameter input: in the construction preparation stage, the design information of each grouting hole is input into the system, including: grouting hole number, planar coordinates, hole elevation, design hole depth, hole segment division (start and end depth of each segment), etc.

[0154] ② Spatial matching method: according to the engineering precision requirement, the spatial matching can adopt one or a combination of the following methods:

[0155] Nearest neighbor borehole matching: taking the planar coordinates of the grouting hole as the center, searching for the nearest exploration hole, and taking the column chart of the exploration hole as the reference geological profile of the grouting hole; Linear interpolation matching: for grouting holes located between two exploration lines, linear interpolation or weighted average can be performed on the same stratum of adjacent two or more exploration holes to construct a virtual geological profile; Subarea matching: when the geological survey has divided into several geological subareas, the geological subarea where the grouting hole is located is first determined, and then a representative hole or combination is selected as the reference.

[0156] ③ Hole segment-stratum unit division: after the association of grouting holes with exploration holes or virtual profiles is completed, "hole segment-stratum unit" division is performed for each grouting hole segment (such as 2 m-5 m segment):

[0157] For a hole segment depth interval [D1, D2], find one or more stratum horizons in the corresponding geological profile within the depth range;

[0158] If the hole segment falls completely within a single stratum, the hole segment stratum code is the lithology code of the layer and the corresponding parameters;

[0159] If the hole segment crosses multiple strata, the hole segment is divided into several subunits according to the length or weight calculation method, or the main stratum (the largest stratum) is taken as the main stratum with secondary stratum information.

[0160] Finally, one or more "formation unit" records are established for each grouting hole section, containing parameters such as depth, lithology, permeability coefficient, fracture degree, etc., as static input for subsequent fusion operations.

[0161] (3) Geological-grouting data fusion and geological response map generation unit

[0162] This unit is used to associate real-time grouting data with the above static geological information, and to construct a geological response map reflecting the "formation-grouting parameter-formation response" relationship.

[0163] ① Real-time data association: During grouting, each real-time grouting data frame contains hole number, hole section number or current grouting depth information. The geological information module associates these identifiers:

[0164] Locate the geological profile corresponding to the grouting hole according to the hole number;

[0165] Find the corresponding formation unit record according to the current grouting depth or hole section number;

[0166] Bind the grouting parameters at this moment (such as P, Q, V, R, C, A, U, etc.) with the static parameters in the formation unit (lithology, k value, N value, fracture index, etc.) to form a "geological-grouting" fusion record.

[0167] ② Geological response map generation, at the end of grouting or during construction, the system can generate geological response maps of different scales in the following ways:

[0168] In-hole longitudinal geological response map: with depth as the vertical axis, the distribution of each formation is plotted along the hole section direction, and grouting pressure, cumulative grouting volume, backflow ratio, hole pressure change, etc. are superimposed on each hole section to form a comprehensive profile of "depth-formation-grouting effect" with different colors or symbols;

[0169] Curtain line profile geological response map: along the curtain axis direction, distance as the horizontal axis, depth as the vertical axis, the formation distribution of multiple grouting holes is spliced into a 2D profile, and grouting parameter statistical values (such as unit length grouting volume, final pressure, backflow, etc.) are superimposed at the corresponding hole section position, used to evaluate the overall continuity and density of the curtain;

[0170] Formation category response map: according to the formation category, the average pressure-flow curve, average unit grouting volume, typical hole pressure response curve, etc. of each grouting hole section in the formation category are calculated to form a "typical geological response template" for the formation category.

[0171] These geological response maps can be displayed in graphical form on the host computer interface, or provided in data table form to the state analysis module as the basis for threshold adjustment and model training.

[0172] (4) Formation classification and threshold correction unit, which is used to convert the geological information into classification labels and parameter correction coefficients that can be directly used by the state analysis module, so as to realize the adaptive adjustment of the state identification criteria to the geological conditions.

[0173] ①Formation classification rule; according to the information of lithology, permeability coefficient k, fracture index, etc. in the geological profile database, the formation is divided into several standard categories, for example: high permeability sandy gravel layer: k≥k1, loose structure, good pore connectivity; medium permeability sandy soil layer: k2≤k<k1; low permeability clay layer: k<k2, or high N value and small pore ratio; fractured hard rock layer: lithology is rock and fracture index is high, joint development; fault fracture zone and soft interlayer: extremely broken structure, which may be a high permeability but unstable formation.

[0174] Each type of formation corresponds to a set of "recommended grouting parameter range" and "state identification sensitivity parameter", such as the upper limit of allowable pressure, unit length target grouting volume, allowable back slurry ratio range, pressure / flow rate change rate threshold, etc.

[0175] ②Threshold correction mechanism; when the state analysis module identifies the state of a hole section, the geological information module provides the formation category and the corresponding threshold correction coefficient of the hole section, for example:

[0176] For high permeability sandy gravel layer, appropriately relax the upper limit threshold of cumulative grouting volume , allow relatively higher grouting volume; reduce the sensitivity to flow rate drop, but increase the monitoring weight of back slurry ratio and slurry escape signs;

[0177] For low permeability clay layer or weakly weathered rock layer, appropriately lower the pressure upper limit and pressure rise rate threshold P t , so that the system responds to pressure surge earlier to prevent overpressure and local damage;

[0178] For fractured hard rock layer or fault fracture zone, more reliance is placed on vibration energy, acoustic response and change of adjacent hole pressure, and the weight of these indicators is increased to judge the risk of fracture penetration and slurry escape.

[0179] The threshold correction can be realized by using multiplicative coefficient or additive offset, for example:

[0180] ; wherein is the correction coefficient determined by the formation category.

[0181] Through this mechanism, the state analysis module uses different discrimination criteria in different formation conditions, avoiding misjudgment of "normal large grouting volume in high permeability layer" as overgrouting, and avoiding missing detection of "slight pressure anomaly in low permeability layer".

[0182] (5) Provide support for state analysis module and adaptive algorithm; the geological information module not only provides the formation category and correction threshold for the state analysis module in single discrimination, but also provides statistical basis for subsequent adaptive algorithm.

[0183] The adaptive threshold updating unit (located inside the state analysis module) can group by formation category when performing statistics on historical data, and respectively statistics the distribution characteristics of indicators such as pressure growth rate, flow change, unit length grouting volume, and back grouting ratio in normal working conditions of each type of formation, to correct the corresponding basic threshold table of each type of formation;

[0184] The confirmed working conditions such as "typical under-grouting", "typical over-grouting / leakage", and "typical plugging" can also be archived according to formation category to form an abnormal sample set with geological labels, which can be used as training data set for subsequent optimization of expert rules or introduction of machine learning assisted identification;

[0185] The "typical geological response template" formed in the geological response diagram can be used as a reference for quick comparison during construction. When the real-time grouting response of a hole section deviates significantly from the typical template of the same type of formation, it can be considered as a potential anomaly, triggering an early warning.

[0186] Through the above structure and function design, the geological information module closely combines static engineering geological survey results with dynamic grouting monitoring data, converts complex formation differences into quantifiable and callable parameters and classification information, provides the state analysis module with adaptive discrimination basis for geological conditions, and ensures that under-grouting, over-grouting, plugging, and leakage can be more accurately identified and distinguished under different formation conditions, thereby significantly improving the reliability and engineering applicability of the grouting quality online evaluation and self-correction system.

[0187] Five, control execution module; in a preferred embodiment, the control execution module is installed near the grouting equipment electric control cabinet and the field valve group, which is composed of grouting pump frequency control unit, valve and flow direction switching unit, grout mixing control unit, sequence control and working condition management unit, and safety interlocking and manual intervention unit, etc., for receiving the grouting state signal output by the state analysis module, automatically adjusting the pressure, flow, proportioning and grouting sequence of the grouting process, and realizing real-time self-correction control of the grouting process.

[0188] (1) Grouting pump and frequency control unit: the grouting pump and frequency control unit is used to realize fine adjustment of slurry delivery pressure and flow.

[0189] The grouting pump can use high-pressure slurry pumps such as piston pump or screw pump, and the rated pressure and flow are selected according to the engineering scale;

[0190] The pump motor uses a three-phase asynchronous motor or a permanent magnet synchronous motor, and a frequency converter is used to realize stepless speed control;

[0191] The frequency conversion control unit receives a given signal (such as 0-10V or communication setting frequency) from the internal controller of the control execution module, and realizes continuous adjustment of the pump rotating speed within the set range;

[0192] A simple closed-loop control loop is constructed through the feedback signal of the pressure sensor: under normal working conditions, pressure closed-loop control is adopted to stabilize the orifice pressure at the target pressure given by the state analysis module or the process parameters nearby;

[0193] When the flow rate needs to be controlled as the main control quantity, the flow rate closed-loop or the dual-constraint control mode of "pressure limit + flow rate priority" can be switched; under different conditions such as under-filling, over-filling and blockage, the sequential control unit issues the speed-up, speed-down or stop command to the frequency converter to realize the automation of pump start-stop and rotating speed adjustment.

[0194] To ensure safety, when the state analysis module outputs "severe overpressure" or "severe blockage" state and the pressure exceeds the safety upper limit , the frequency conversion control unit has a hardware emergency stop input, which cuts off the pump motor operation instruction through a relay to realize fast pump stop.

[0195] (2) Valve and flow direction switching unit: The valve and flow direction switching unit is arranged on the main grouting pipeline, branch and bypass back grouting pipeline, and is used to realize grouting flow direction, flow distribution and rapid pressure relief.

[0196] An electric regulating valve is arranged on the main grouting pipeline for fine flow regulation and pressure fine adjustment of the main flow rate;

[0197] An electric switching valve (electric ball valve, butterfly valve or stop valve) is arranged before each grouting hole or each hole segment branch to select the current grouting hole or hole segment, realize automatic switching of multiple holes and multiple segments;

[0198] A bypass back grouting electric valve is arranged between the main grouting main pipe and the back grouting pipeline, which can be opened to guide part of the grout back to the back grouting container or circulating tank when pressure reduction, unloading or blockage dredging program needs to be executed;

[0199] The valve actuator is preferably an electric actuator or an electric / pneumatic integrated actuator, which supports on-off quantity control and analog quantity (opening degree) control, and has a valve position feedback signal;

[0200] The actions of the valve and the pump are coordinated by the sequential control unit to ensure that "pressure switching" or impact and mispraying caused by misoperation is avoided during the switching process.

[0201] In the "overfilling / leakage" state, the control logic can execute: appropriately close the main regulating valve, open the bypass backfill valve, or reduce the grouting time of a certain hole section. In the "blockage" state, the downstream valve can be closed for a short time, the bypass unloading valve is opened, and the frequency converter is used to execute the pulsating pressure sequence to reduce the risk of blockage and perform dredging.

[0202] (3) Grout mixing control unit: The grout mixing control unit is responsible for the automatic control of the mixing of slurry, water-cement ratio, and admixture dosage, and can dynamically adjust the formula according to the state analysis results.

[0203] The grout mixing device includes a clean water tank, a water metering unit, a cement (or powder) metering unit, an admixture metering unit, and a mixing device. Each metering unit is controlled by an electric valve, a variable frequency screw conveyor, or a metering pump as an actuator;

[0204] The grout mixing controller calculates the water-cement ratio, cement dosage, and admixture dosage according to the preset formula table and sends control commands such as adding water, adding powder, and adding admixture according to the set formula sequence;

[0205] When the state analysis module determines that it is "underfilling" and needs to improve permeability, it can instruct the grout mixing controller to switch to a formula with higher permeability (such as moderately increasing the water-cement ratio or adjusting the type and dosage of admixture);

[0206] In the "overfilling / leakage" state, it can switch to a formula with higher viscosity or faster setting (reduce water-cement ratio, use quick-setting or high-binder), to enhance the plugging ability and quickly form a grout stop band;

[0207] The key parameters (water addition, powder addition, mixing time) during the grout mixing process can be uploaded to the upper computer for comparison with the conductivity / concentration sensor data to ensure the accuracy of formula execution.

[0208] Through the linkage of the grout mixing control unit and the state analysis module, the system not only adjusts the grouting pressure and flow, but also performs "soft adjustment" at the mixing level, achieving more essential control of grouting effect.

[0209] (4) Sequence control and working condition management unit: The sequence control and working condition management unit is the core logic layer of the control execution module, which is used to manage the construction sequence of multi-hole and multi-section grouting, and coordinate the cooperative action of pumps, valves, and grout mixing.

[0210] ① Construction sequence management: The grouting hole list, hole section division, and design sequence are set in the system in advance, such as from upstream to downstream, from shallow to deep, or according to the symmetric sequence;

[0211] The sequence control unit dynamically updates the to-be-grouted queue according to the completed hole section state (normal end, underfilling supplement, overfilling early termination, etc.);

[0212] For the hole section that needs to be re- grouted or supplemented, the corresponding "re- grouting task" can be inserted into the queue, and the execution is automatically arranged at the appropriate position of the main sequence.

[0213] ②Working mode management, the sequence control unit supports multiple working modes, including: automatic mode: driven by the state analysis module, the system automatically completes the whole process of starting the pump, opening the valve, grouting, stopping the pump, recording, etc. according to the preset strategy; Semi-automatic mode: key switching actions (such as hole section switching, formula switching) need to be confirmed manually, and other parameters are adjusted automatically by the system; Manual mode: only provides state monitoring and auxiliary protection, directly controlled by the operator Pump and valve, suitable for debugging or special working conditions.

[0214] In automatic mode, the sequence control unit executes the following logic cycle:

[0215] Select the current target hole section → Check the corresponding valve status and grout preparation → Start the grouting pump and open the corresponding branch valve → Grout according to the design parameters → Adjust the parameters in real time according to the output of the state analysis module and determine the termination condition → After grouting, automatically stop the pump, close the valve and record the results → Jump to the next hole section or execute the re- grouting task.

[0216] (5) State linkage and deviation correction control logic, the control execution module maps the state tags (normal, under- grouting, over- grouting, slurry escape, blockage, etc.) output by the state analysis module into specific control action combinations.

[0217] Normal state: the grouting pump runs in closed loop according to the target pressure or flow; the valve maintains the current opening and flow direction; the grout control maintains the current formula; the sequence control advances according to the design rhythm until the designed grouting volume or termination criteria are reached.

[0218] Under- grouting state: issue a speed up or pressure upper limit command to the frequency converter, increase the target pressure upper limit by 10% ~ 30% (within the safety range); extend the grouting time of this hole section or set a new target cumulative grouting volume; can instruct the grout control to switch to a formula with higher permeability; if it is still determined to be under- grouting after multiple adjustments, a "re- grouting task" can be automatically added and re- grouted in subsequent construction.

[0219] Over- grouting / slurry escape state: issue a speed down command to the frequency converter, reduce the pump speed by about 20% ~ 50%, reduce the flow and pressure growth rate; appropriately close the main regulating valve or close the current hole section branch valve for a short time; open the bypass back- grouting valve as appropriate to release part of the pressure and limit the single- section cumulative grouting volume of this hole section to not exceed the safety threshold ; instruct the grout control to switch to a high viscosity or quick setting formula to promote rapid plugging of potential slurry escape channels.

[0220] Blockage state: immediately execute the "anti-pressure-unloading-pulsating dredging program": short-term pump stop or pressure reduction, close the downstream valve, open the bypass unloading valve, and reduce the pipeline pressure to a safe level; then periodically switch the pump output state (for example, switch between setting high pressure / low pressure) at a frequency of 1-3 Hz to form a pressure pulsation to impact the blocked site; after a limited number of program executions (such as 3-5 cycles), the state is re-evaluated, and if the blockage is removed, the normal working condition is restored, and if there is still serious blockage, the pump is stopped and a manual maintenance prompt is sent, and the hole section grouting is not automatically restored.

[0221] All the above actions are accompanied by state recording before and after execution. The control execution module uploads the time of each deviation correction, control instructions and execution results to the host computer log for later analysis and rule library optimization.

[0222] (6) Safety interlocking and manual intervention unit, considering the complexity and safety of the construction site, the control execution module also includes multi-stage safety interlocking and manual intervention mechanism: hardware interlocking: set emergency stop button, mechanical limit switch, pressure limit switch, etc. When detecting extreme overpressure, pump overload, motor overtemperature, etc., the running instructions of the pump and key valves can be directly cut off by bypassing the host computer logic; Software interlocking: multiple condition judgments are set in the control program, such as when the state analysis is a serious abnormality and the pressure exceeds or the hole pressure exceeds the design limit, even if the operator does not issue a stop command, the system will forcibly execute the pump stop and pressure relief action; Manual priority mechanism: when the operator selects to enter manual or semi-automatic mode, manual operation instructions have priority over automatic control logic, but key safety interlocks are not bypassed to avoid accidents caused by misoperation; Authority and record: for sensitive operations such as changing the formula and adjusting the upper limit of the pressure, permission management can be set, and the key operations of each operator are recorded for quality traceability and responsibility division.

[0223] Through the above structure and control logic, the control execution module can form a closed loop with the state analysis module without changing the traditional grouting process, realize the whole process automatic adjustment of grouting pressure, flow, slurry ratio and construction sequence, and on the premise of ensuring safety, timely and targeted self-correction control for under-grouting, over-grouting, slurry leakage and blockage, etc. Abnormal conditions, greatly improving the controllability of grouting quality and the automation level of engineering construction.

[0224] Example 2

[0225] This embodiment gives a method and process of online grouting quality evaluation and self-correction based on the above system, taking single-hole or single-section grouting as an example.

[0226] The method comprises the following steps:

[0227] Step S1, multi-parameter data acquisition: Before the grouting operation starts, the operator inputs the design parameters into the upper computer, including the design grouting pressure range, the design grouting volume, the allowed back grouting volume, the target permeation index, etc., and the system retrieves the stratum type and parameters corresponding to the hole section from the geological information module.

[0228] During the grouting process, the data acquisition module continuously acquires the grouting orifice pressure P(t), the instantaneous flow rate Q(t), the cumulative grouting volume V(t) and the back grouting volume R(t) integrated therefrom, the grout conductivity or concentration C(t), the stratum vibration and acoustic response A(t), and the pore water pressure U(t) at a sampling frequency not less than 10 Hz.

[0229] The above real-time data are sent to the data processing module in time sequence.

[0230] Step S2, real-time data processing: the data processing module performs synchronous and standardized processing on the collected multi-source data, specifically including:

[0231] 1. Time stamp alignment, using a bidirectional time window interpolation algorithm to perform time correction on the discrete sampling points of different channels; for a target time point t0, a time window is formed by taking several sampling points before and after t0, and the consistent value of each parameter at t0 is obtained by linear or spline interpolation, so as to ensure that the pressure, flow rate, back grouting volume, conductivity, etc. have corresponding data at the same time.

[0232] 2. De-noising and smoothing, using a moving average filter or median filter algorithm on the original signals of each channel to eliminate high-frequency noise and occasional spikes. For example, the arithmetic mean value of the last N sampling points is used to replace the current value, or the median value is selected to suppress isolated abnormal points.

[0233] 3. Abnormal value elimination, in each time window, the mean and standard deviation of the signal are calculated, and the three-sigma rule (three times standard deviation method) is used to eliminate jump data and invalid data points that deviate significantly from the normal range, and interpolation or adjacent valid values are used for completion.

[0234] 4. Drift correction, a linear drift correction model is established for sensor data with slow drift (such as pressure and flow rate), for example, a periodic zero calibration or static load process is used to determine the reference, and the form is used to compensate for the slow drift, so as to maintain the stability of long-term monitoring.

[0235] 5. Normalization processing, each physical quantity is normalized according to its normal working interval, such as using the interval [0, 1] or [-1, 1] mapping, converting pressure, flow rate, back grouting volume, conductivity, vibration amplitude, pore pressure, etc. into dimensionless quantities, which facilitates the construction of a unified scale grouting atlas and state recognition.

[0236] ​After the processing is completed, a unified timing grouting parameter set is formed as the input of subsequent steps S3 and S4.

[0237] Step S3: Grouting state recognition analysis

[0238] In step S3, the state analysis module performs atlas construction and state judgment based on the processed data.

[0239] 1. Four-dimensional atlas matrix construction. Taking time as the main axis, key variables such as pressure P, flow rate Q, and concentration / conductivity C are selected, accumulated according to the set sliding time window (for example, every 10 s or 30 s), and a four-dimensional atlas matrix of pressure-flow rate-time-concentration is generated. This matrix not only retains the time evolution information, but also presents the joint change pattern of multiple parameters within the same time window.

[0240] 2. Rule base comparison. The system compares the atlas feature curve and the change rate of the key variables with the expert rule base. Typical composite logic criteria include: blockage discrimination: if the flow rate decrease rate AQ / At exceeds the set threshold Q t , and the pressure rise rate AP / At is greater than the threshold P t at the same time, and the grout return amount does not increase significantly, it is determined that the pipeline or hole section is blocked; over-grouting / leakage discrimination: if the pressure is maintained at a relatively low level or far below the lower limit of the design pressure for a long time, and the cumulative grout amount is significantly higher than the target reference value, and the grout return amount increases or the adjacent hole pressure abnormally rises, it is determined that the grouting is over or the grout leaks; under-grouting discrimination: if the pressure rises rapidly and approaches or exceeds the upper limit, and the cumulative grouting amount is significantly lower than the corresponding design section value, and the stratum vibration or hole pressure response is weak, it is determined that the grouting is insufficient or the curtain is poorly sealed;

[0241] Combined with acoustic vibration and hole pressure characteristics, complex situations such as sudden fracture penetration and long-distance grout leakage are comprehensively discriminated.

[0242] 3. Adaptive dynamic threshold. To improve the recognition accuracy under different geological conditions, an adaptive dynamic threshold recognition algorithm can be optionally introduced: according to the grouting data samples in the same or similar strata (provided by the geological information module) in the historical project, the statistical distribution of key parameters such as AQ / At, AP / At, cumulative grouting amount V, and grout return amount R is analyzed, and the reasonable threshold interval for discrimination is automatically fitted. The system dynamically adjusts the threshold values of Q t , P t , V reference value, etc. during the recognition process to improve the adaptability and accuracy of the model to different strata and different hole sections.

[0243] Through the above rule comparison, the state analysis module derives the current grouting state label (such as "normal", "under-grouting", "over-grouting", "slurry escape", "blockage", etc.) and the corresponding confidence level in each time window, and outputs the evaluation results and prompt information.

[0244] Step S4: geological information fusion and threshold correction, the geological information module fuses the geological profile and parameters of the position where the borehole is located with the above evaluation results:

[0245] If the current hole section is a high-permeability sandy gravel layer, the system allows a larger cumulative grouting volume and a higher flow upper limit, and accordingly relaxes the "over-grouting / slurry escape" threshold value;

[0246] If it is a low-permeability clay layer or weakly weathered rock layer, the threshold value of "over-grouting / slurry escape" is appropriately tightened, and the sensitivity to pressure surge is improved;

[0247] If it is a hard rock layer with developed fissures, more attention is paid to vibration energy, acoustic emission count and borehole pressure response to judge the fissure filling condition and curtain integrity.

[0248] In this way, the real-time evaluation result is not only based on the current monitoring data, but also combines the pre-stored geological model to correct or support the original identification conclusion, and automatically adjusts the evaluation threshold value when necessary, so that the identification result is more consistent with the actual geological environment.

[0249] Step S5: automatic deviation correction control; in step S5, the control execution module automatically adjusts the grouting construction parameters according to the final evaluation result to realize self-correction control. The following strategies can be used:

[0250] 1. Under-grouting correction strategy, when the state analysis module outputs the "under-grouting" label, the system automatically executes according to the stratum permeability of the corresponding hole section and the design target:

[0251] The upper limit of the grouting pressure of this hole section is increased by about 10% to 30%, the pump speed is increased or the valve opening is adjusted through the frequency converter to make the actual pressure stable near the new upper limit;

[0252] The grouting duration of this hole section is appropriately extended, for example, to 150% of the original set time, and the grouting process enters the stable pressure compensation mode, that is, the borehole pressure is kept stable within a small fluctuation range through closed-loop control, so as to ensure sufficient penetration and diffusion of the slurry.

[0253] 2. Over-grouting / slurry escape correction strategy, when the state recognition is "over-grouting" or "slurry escape", the system automatically executes the pressure reduction and speed reduction logic, including:

[0254] The grouting pump speed is reduced by about 20% to 50% based on the current value, so as to reduce the instantaneous flow and pressure increase rate;

[0255] By adjusting the water-cement ratio through the grouting device controller, the water-cement ratio is reduced (i.e. the slurry is increased in cohesiveness and viscosity), thereby enhancing the sealing capacity of the slurry and inhibiting long-distance escape;

[0256] At the same time, a safe single-section cumulative grouting amount threshold V is set for the current hole section m When the cumulative grouting amount approaches or reaches V m , automatic grouting is limited, and a warning is issued.

[0257] 3. Blockage dredging strategy: when the state is identified as "blockage", the system starts the "back pressure-unloading-pulsation dredging program", which can be executed in turn: short-time reverse unloading: by quickly closing the main valve and opening the bypass valve, the high-pressure slurry in the hole is released, so that the pressure drops sharply to relieve the stress at the blockage site; bypass pressure relief: keep the bypass valve at a certain opening, so that part of the slurry flows back, reducing the pressure difference in the pipeline; pulsation excitation: periodically switch the main pump output state (such as between high pressure / low pressure or slurry supply / unloading mode) at a frequency of 1-3 Hz to form periodic pressure pulses to mechanically vibrate and dredge the blockage site.

[0258] If the blockage cannot be eliminated by the above program, the system automatically stops the pump and issues a manual maintenance prompt to avoid equipment damage and overpressure accidents.

[0259] 1. Working condition matching regulation table and data record: all the above-mentioned deviation correction actions are driven by the embedded working condition matching regulation table. The regulation table takes the grouting state label, real-time monitoring parameters, and geological response information as input, and outputs the corresponding optimal control scheme. The trigger condition, execution parameters, and result evaluation of each correction process are automatically recorded by the system for subsequent engineering analysis and retraining and optimization of the rule base and adaptive threshold model.

[0260] 2. Normal working condition control: when the evaluation result is "normal", the system continues to grout according to the optimized process parameters until the designed grouting amount is reached or the termination criteria are met, and then automatically switches to the next hole or hole section construction. The operator can take over the system for manual control at any time to ensure that the system has a safety redundancy for manual intervention.

[0261] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A visualized grouting quality online evaluation and self-correction system, characterized in that, The data acquisition module, the data processing module, the state analysis module, the geological information module and the control execution module are included. The data acquisition module integrates multiple sensors for real-time acquisition of pressure, flow, cumulative grouting volume, back grouting volume, slurry conductivity and formation vibration and pore pressure parameters during grouting. The data processing module calibrates and filters the data of each sensor and synchronizes the time to generate standardized grouting process data. The state analysis module uses a preset atlas analysis algorithm and an expert rule base to evaluate the standardized data in real time to identify the grouting state and quality. The geological information module associates historical geological survey data with real-time grouting data to generate a geological response atlas for assisting evaluation and control decision-making. The geological information module mainly includes a geological profile database unit, a spatial index and hole section matching unit, a geological-grouting data fusion and geological response graph generation unit, and a formation classification and threshold correction unit. The geological profile database unit is used to store and manage various types of geological information obtained during the engineering survey stage. The geological-grouting data fusion and geological response graph generation unit is used to associate real-time grouting data with static geological information, which is the static parameters of lithology, k value, N value and fracture index in the formation unit. During grouting, each real-time grouting data frame includes hole number, hole section number or current grouting depth information. The geological information module locates the corresponding geological profile of the grouting hole according to the hole number, finds the corresponding formation unit record according to the current grouting depth or hole section number, and binds the grouting parameters at that moment with the static parameters in the formation unit to form a "geological-grouting" fusion record. At the end of grouting or during construction, the system draws the distribution of each formation along the hole section direction with depth as the vertical axis, and superimposes grouting pressure, cumulative grouting volume, back grouting ratio and pore pressure change parameters on each hole section to form a "depth-formation-grouting effect in-hole longitudinal geological response graph" with different colors or symbols. Along the curtain axis direction, the formation distribution of multiple grouting holes is spliced into a 2D profile with distance as the horizontal axis and depth as the vertical axis. The grouting parameter statistical values are superimposed at the corresponding hole section positions to generate a curtain line profile geological response graph for evaluating the overall continuity and compactness of the curtain. The average pressure-flow curve, average unit grouting volume and typical pore pressure response curve of each grouting hole section in the formation category are generated to form a formation category response graph. The formation classification and threshold correction unit divides the formation into several standard categories according to the lithology, permeability coefficient k, and fracture index information in the geological profile database; when the state analysis module identifies the state of a certain hole section, the geological information module provides the formation category and the corresponding threshold correction coefficient of the hole section, and the threshold correction can be realized in the form of multiplicative coefficient or additive offset: ; wherein is the correction coefficient determined by the formation category; The control execution module automatically adjusts the operating parameters of the grouting equipment, including valve opening and closing, slurry flow rate and grouting sequence, based on the evaluation results output by the state analysis module to achieve adaptive correction control of the grouting process.

2. The visualized grouting quality online evaluation and self-correction system according to claim 1, characterized in that, The data acquisition module includes pressure sensors, flow meters, slurry concentration or conductivity sensors, back grouting volume measuring devices installed on the grouting pipeline and drill holes, and vibration acoustic sensors and pore water pressure sensors laid in the grouting area for acquiring multi-source data of grouting dynamics and formation response.

3. The visualized grouting quality online evaluation and self-correction system according to claim 1, characterized in that, The data processing module comprises a signal conditioning circuit, an A / D converter and a data synchronization unit, and is configured to amplify, filter, and convert the original signals of different sensors into digital signals, and to perform time alignment processing on the digital signals, and output the processed data in a predetermined format.

4. The visualized grouting quality online evaluation and self-correction system according to any one of claims 1-3, characterized in that, The state analysis module comprises a built-in grouting process expert rule base, which comprises threshold conditions and logical rules for identifying under-grouting, over-grouting, blockage and escaped grouting states; the state analysis module compares the real-time monitored pressure-flow curve and grouting amount change curve with the rule base, and outputs a corresponding grouting state signal and an alarm prompt when the conditions of a certain rule are met.

5. The visualized grouting quality online evaluation and self-correction system according to any one of claims 1-3, characterized in that, The control execution module comprises an electric valve, a variable-speed grouting pump, a grout mixing device controller and a sequence control unit. When the state analysis module determines that the grouting state is normal, the control execution module operates according to the preset process parameters. When an abnormal state is detected, the control execution module automatically executes a correction measure, including starting or stopping the grouting pump, opening or closing the valve to change the flow direction of the grout, adjusting the grout ratio or switching the grouting hole, to correct the abnormality and restore normal grouting.

6. A method for online evaluation and self-correction of grouting quality, characterized in that, The method comprises the following steps: Multi-parameter data acquisition: during the grouting construction process, real-time grouting data are collected by pressure, flow, back grouting amount, concentration, conductivity, acoustic vibration and hole pressure sensors arranged; Real-time data processing: the collected multi-source data are processed synchronously and standardized, abnormal interference is filtered out, and the data are converted into a unified time sequence grouting parameter set; State recognition analysis: the processed data are input into the expert rule base for analysis and comparison, and the current grouting state is identified in combination with the grouting atlas characteristics, such as whether under-grouting, over-grouting, blockage or escaped grouting occurs, and an evaluation result and a prompt information are generated; Geological information fusion: the real-time evaluation result is associated with the pre-stored geological model and parameters, the response characteristics of the current hole section stratum to grouting are analyzed, the conclusion of state recognition is corrected or verified accordingly, and the evaluation threshold is adjusted if necessary; The geological information module fuses the geological profile and parameters of the position where the borehole is located with the above evaluation result: If the current hole section is a high-permeability sandy gravel layer, the system allows a larger cumulative grouting amount and a higher flow upper limit, and accordingly relaxes the discrimination threshold of "over-grouting / escaped grouting"; If it is a low-permeability clay layer or a weakly weathered rock layer, the threshold of "over-grouting / escaped grouting" is appropriately tightened, and the sensitivity to pressure surge is increased; If it is a hard rock layer with developed fissures, more attention is paid to vibration energy, acoustic emission count and hole pressure response to judge the fissure filling condition and curtain integrity; Automatic correction control: the grouting construction parameters are adjusted in real time according to the evaluation result, and corresponding correction control measures are automatically executed when an abnormality is determined, including pressure reduction, pump stop, grout formula adjustment or grouting sequence adjustment; If the evaluation is normal, the grouting is continuously performed according to the optimized parameters until completion; The grouting state recognition analysis comprises: A pressure-flow-time-concentration four-dimensional atlas matrix is generated using the processed grouting parameters, and the matrix is generated by setting a sliding window accumulation with a sampling frequency of ≥10 Hz per second, serving as the basis for time sequence analysis; The system compares the atlas characteristic curve with the built-in expert rule base, and the rule base comprises a complex logical criterion established based on experience: If the flow rate ΔQ / Δt decreases over time and the pressure ΔP / Δt increases over time, then the system is determined to be blocked. t t If the flow rate ΔQ / Δt decreases over time and the pressure ΔP / Δt increases over time, then the system is determined to be blocked.​ If the pressure remains low for a long time and the cumulative grouting volume is abnormally higher than the expected reference value, it is determined to be over-grouting or slurry escaping state; If the pressure rises rapidly and the grouting volume is far lower than the standard segment value, it is determined to be under-grouting or poor sealing; An adaptive dynamic threshold identification algorithm can be optionally introduced, which is based on statistical history of similar strata or hole segment grouting data, automatically fitting a reasonable range of key identification variables, and improving the adaptability and accuracy of the model to different geological conditions.

7. The method for grouting quality online evaluation and self-correction of claim 6, characterized in that, The real-time data processing specifically includes: Using a bidirectional time window interpolation algorithm to align the time stamps of the collected pressure, flow rate, back slurry volume, and conductivity multi-source data, ensuring that each parameter has a corresponding value at the same time; Based on the sliding average filtering algorithm or the median filtering algorithm, the original signal is denoised, the abnormal fluctuations are smoothed, and the high-frequency interference is eliminated; The jump data or invalid data appearing in the collection process are processed by three-sigma rule outlier rejection; A linear drift correction model is used to automatically correct the sensor data with slow drift, maintaining measurement stability; Different physical quantities are normalized and transformed according to their normal working intervals, and the data dimensions are unified to facilitate subsequent atlas construction and state identification.

8. The method for grouting quality online evaluation and self-correction of any one of claims 6-7, characterized in that, The automatic deviation correction control includes the following implementable strategies: If the state identification module outputs the "under-grouting" label, the system will increase the upper limit of the grouting pressure by 10-30% or extend the grouting duration to 150% of the original design value based on the corresponding hole segment stratum permeability and design target, while enabling the pressure stabilization compensation mode; If the state identification is "over-grouting" or "slurry escaping", the system automatically executes the pressure reduction and speed reduction logic, including: Adjusting the grouting pump speed by 20-50%, Reducing the slurry water-cement ratio to improve the adhesion, And limit the single-stage grouting cumulative amount does not exceed the safety threshold V m ; If the state identification is "blockage", the system starts the back pressure-unloading-pulsating dredging program, which sequentially performs short-term reverse unloading, opens the bypass valve to reduce the pressure difference, and periodically switches the main pump output at a frequency of 1-3 Hz for pulsating excitation; All deviation correction actions are driven by the embedded working condition matching control rule table, which selects the optimal control scheme based on real-time monitoring parameters and pre-stored geological response information, and records the deviation correction process in real time for subsequent analysis and model retraining.

Images