A method for enhancing the curing performance of geopolymer in high-bentonite content slurry

Abstract

The present invention relates to the technical field of waste mud treatment, and more particularly to a method for enhancing the curing properties of geopolymers in high-bentonite-content mud. The method includes a method for adding a curing-enhancing modification agent, which comprises adding the curing-enhancing modification agent to the high-bentonite-content mud, stirring the mixture, and then allowing the mixture to cure. The curing-enhancing modification agent comprises the following raw material components in percentage by weight: 70-80% calcium chloride, 10-20% sodium chloride, 10-15% nano-silicon dioxide, 5-10% dispersant, and 0.5-3% EDTA-polymer derivative. The addition of the agent in the present invention effectively eliminates the negative impact of bentonite in the waste mud on the reaction of geopolymers, optimizes the reaction conditions of the geopolymers in complex mud environments, and improves mud utilization and success rate.

Description

Technical Field

[0001] The invention relates to the technical field of waste mud treatment, in particular to a method for enhancing and modifying the geopolymer solidification performance of mud with high bentonite content. Background Art

[0002] As a new type of cementitious material, geopolymers offer low-carbon, environmentally friendly advantages, attracting increasing research interest. They are gradually becoming an alternative to high-energy cement. Their outstanding acid and alkali resistance, high strength, and other properties also address many challenges that cement cannot address. Currently, geopolymers are widely used to recycle waste mud. By solidifying waste mud to produce functional products, waste mud can not only be consumed on-site but also transported to factories for processing into building materials, playing a significant role in environmental protection and cost savings. However, the composition and content of clay minerals in muds from different sources vary. These minerals may originate from the mud itself or from additives added during construction. For example, bentonite is commonly added to slurry shield and drilling mud to improve fluidity and facilitate construction. However, these clay minerals, with varying degrees of adsorption, can negatively impact the depolymerization and polycondensation reactions during geopolymer solidification, thereby negatively affecting the solidification of geopolymers in the mud. Among them, bentonite (montmorillonite as the main component) is the most common clay mineral in nature and on construction sites, and it also has the strongest adsorption properties.

[0003] Traditional high bentonite content mud treatment has the following problems:

[0004] (1) Traditional research has focused on enhancing the adsorption of bentonite without inhibiting its interference with geopolymer reactions. Therefore, there is currently no solution specifically for the poor effect of geopolymer solidification of slurries with high bentonite clay mineral content.

[0005] (2) There is a lack of special curing enhancement modification agent formulas and standardized operating procedures for high-bentonite mud.

[0006] (3) The current mud solidification treatment methods are simple means such as mechanical stirring and filter pressing. The process lacks real-time monitoring means, resulting in uncontrollable reaction conditions and large differences in solidification effects. Summary of the Invention

[0007] The present invention provides a method for enhancing the curing performance of geopolymer in high-bentonite-content mud, aiming to solve the shortcomings of the existing treatment of geopolymer-cured waste mud.

[0008] The present invention provides a method for enhancing the curing performance of geopolymer in high-bentonite content mud, including a method for adding a curing enhancement modification agent. The method for adding the curing enhancement modification agent comprises: adding the curing enhancement modification agent to the high-bentonite content mud, stirring evenly, and then standing to cure;

[0009] The curing enhancement and modification additive includes the following raw material components in percentage by mass:

[0010] Calcium chloride 70~80%, sodium chloride 10~20%, nano-silicon dioxide 10~15%, dispersant 5~10%, EDTA-polymer derivative 0.5~3%.

[0011] As a further improvement of the present invention, a curing enhancement and modification agent is added to the high bentonite content mud, stirred evenly and then allowed to stand for curing, specifically comprising:

[0012] A curing enhancement and modification additive is added to the slurry at a rate of 1.7-6.7% of the dry mass of bentonite, and the slurry is subjected to high-speed shear stirring. After standing, a curing agent is added to perform a curing treatment. The curing agent comprises raw material components in the following mass ratio: slag: cement: water glass = 42:38:20.

[0013] As a further improvement of the present invention, the dispersant is a polycarboxylic acid dispersant, and the EDTA-polymer derivative includes PAA-EDTA and PEG-EDTA.

[0014] As a further improvement of the present invention, the preparation method of the curing enhancement modification auxiliary agent comprises:

[0015] a1. Dissolve calcium chloride and sodium chloride in deionized water according to the composition ratio to prepare a mother liquor;

[0016] a2. Nano-silica was added to the mother liquor and ultrasonically dispersed to obtain a uniformly dispersed solution;

[0017] a3. EDTA- polymer derivative was added to the solution obtained in a2, and the mixture was stirred to obtain a uniform solution;

[0018] a4. Add a dispersant to the solution obtained in a3 and stir to mix until a homogeneous suspension is obtained.

[0019] As a further improvement of the present invention, the method for enhancing the curing performance of geopolymers in high-bentonite-content slurries further includes constructing an intelligent real-time detection and control system, the construction process of which includes:

[0020] b1. Bentonite characteristic response sensor is set on the pumping path of the mud mixing tank to detect the bentonite content in the mud;

[0021] b2. A multi-modal element sensor is provided at the input and output ports of the mud mixing tank to detect the concentration of a specified ion;

[0022] b3 set curing agent feeding point, additive feeding point, the curing agent feeding point, additive feeding point are docking mud mixing bin, the curing agent feeding point for storing the curing agent, the additive feeding point for storing curing enhancement modification additives;

[0023] b4. Use feedback control software to obtain parameter information of bentonite characteristic response sensors and multimodal element sensors, make real-time predictions of bentonite content and ion concentration, make decisions on the optimal amount of material addition for curing agent and curing enhancement and modification additives, and control the material addition at the curing agent addition point and additive addition point.

[0024] As a further improvement of the present invention, the bentonite characteristic response sensor includes a CEC rapid measurement unit, which has a built-in ammonium ion saturation column and a conductivity detection cavity. The ammonium ion saturation column is a microfluidic exchange column filled with a displacement fluid, and the ammonium ion replacement of the cations between the montmorillonite layers is completed when the mud passes through at a fixed low flow rate; the conductivity detection cavity monitors the changes in the conductivity of the displacement fluid in real time through a multi-electrode array, and calculates the CEC value in combination with the Langmuir adsorption model.

[0025] As a further improvement of the present invention, the bentonite characteristic response sensor includes a NIR auxiliary recognition unit for auxiliary detection of bentonite content, and the NIR auxiliary recognition unit carries a 510~530 cm -1 interval and 1070~1090 cm -1 A near-infrared spectrometer with a characteristic wavelength range of 1 / 40 nm was used to establish a quantitative relationship between spectral absorbance and bentonite content through regression algorithm.

[0026] As a further improvement of the present invention, the process of establishing a quantitative relationship between spectral absorbance and bentonite content by a regression algorithm includes:

[0027] c1. Prepare standard samples with known bentonite content to cover the detection range, and measure the absorbance of each standard sample at the characteristic peak wavelength range using a near-infrared spectrometer;

[0028] c2. Based on the characteristic peak wavelength range of each sample tested, calculate the correlation coefficient between the corresponding wavelength of each sample and the characteristic peak wavelength of the bentonite standard, retain the wavelengths with a correlation coefficient greater than 0.9, and automatically select and optimize the optimal wavelength combination through an intelligent algorithm;

[0029] c3. Use partial least squares regression to build a regression model, decompose the spectral data and content data simultaneously, extract the common latent variables, and use the latent variables to construct a linear equation: predicted content = a × absorbance 1 + b × absorbance 2 + constant, where a and b are 520 cm -1 and 1080cm -1 The contribution weights of the two characteristic wavelengths, absorbance 1 and absorbance 2 correspond to 520cm -1 and 1080cm -1 The sum or average absorbance of the two characteristic wavelength ranges; when the predicted content is in the low content region of the detection range (<25%), the predicted value = the standard PLSR result; when the predicted content is in the high content region of the detection range (>25%), the predicted value = the standard PLSR result + correction coefficient × ln (absorbance); coefficients a and b are obtained by fitting the standard sample set through the partial least squares method, and the correction coefficient is determined by nonlinear fitting of high-concentration samples and is dynamically adjusted with environmental parameters.

[0030] As a further improvement of the present invention, the multimodal element sensor includes

[0031] ISE unit: using all-solid-state Ca 2+ / Na + Dual-channel ion selective electrode, using perovskite / graphene composite membrane as electrode membrane material, Ca 2+ Ion selective electrodes and Na + The ion selective electrodes are covered with electrode membranes to form Ca 2+ Sensitive membrane and Na + The sensitive membrane is separated by microchannels into two parts: 2+ Sensitive membrane and Na + Two flow channels of the sensitive membrane for dual ion simultaneous detection;

[0032] Fluorescence unit: an excitation light source of an integrated dual-wavelength excitation fiber optic sensor, an Al fluorescent probe based on luminogallion organic matter, and a Si fluorescent probe based on ammonium molybdate complex. The excitation light source wavelengths for the Al fluorescent probe and the Si fluorescent probe are 350-380 nm and 500-510 nm, respectively.

[0033] As a further improvement of the present invention, the feedback control software includes a concentration prediction model of a BP neural network and a multimodal dynamic compensation algorithm;

[0034] BP neural network concentration prediction model: Based on a large database, it obtains sensor parameters or manually input parameters to make real-time predictions of bentonite content and ion concentration, and make decisions on the optimal amount of material to be added. Parameters include ISE, fluorescence intensity, pH, CEC, and flow rate;

[0035] Multimodal dynamic compensation algorithm: Through sensor drift compensation and inter-ion interference compensation, sensor drift compensation includes ISE potential compensation and fluorescence signal attenuation correction, to correct the concentration prediction model decision.

[0036] The beneficial effects of the present invention are as follows: (1) The addition of additives effectively eliminates the negative impact of bentonite in waste mud on the reaction of geopolymers, optimizes the reaction conditions of geopolymers in complex mud environments, and improves mud utilization and success rate. (2) The use of ISE-fluorescence dual sensing equipment detects ion concentration to reflect the mud solidification environment. (3) Through feedback control software, based on a big data model, material placement decisions are made according to input data, achieving real-time and efficient improvement and solidification synchronization. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of the method for enhancing the curing performance of geopolymer of high bentonite content slurry of the present invention;

[0038] Figure 2 This is a working principle diagram of the bentonite characteristic response sensor of the present invention;

[0039] Figure 3 It is a working principle diagram of the multimodal element sensor of the present invention;

[0040] Figure 4 This is a comparison chart of the unconfined compressive strength of the solidified bodies of shield mud and drilling mud tested under standard curing conditions at 3, 7, 14 and 28 days in Example 1 of the present invention;

[0041] Figure 5 This is a comparison chart of the unconfined compressive strength of the solidified bodies of shield mud and drilling mud in Example 2 of the present invention tested under standard curing conditions at 3, 7, 14 and 28 days;

[0042] Figure 6 This is a comparison chart of the unconfined compressive strength of the solidified bodies of shield mud and drilling mud in Example 3 of the present invention tested under standard curing conditions at 3, 7, 14 and 28 days. DETAILED DESCRIPTION

[0043] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0044] The present invention is a patent for a method for treating Ca 2+ and Na +A method for improving the solidification strength of waste mud based on the ion selective adsorption exchange mechanism and the geopolymer reaction principle, including a solidification modification additive formula and an intelligent real-time detection-control system, can be applied to the resource utilization of waste mud, especially for waste mud with high bentonite content. It is a method worthy of promotion and application for improving geopolymer-based solidified waste mud.

[0045] Specifically, the method for enhancing the curing performance of geopolymer in high-bentonite content mud includes a method for adding a curing enhancement modification agent, wherein the method comprises: adding the curing enhancement modification agent to the high-bentonite content mud, stirring evenly, and then standing to cure;

[0046] The curing enhancement and modification additive includes the following raw material components in percentage by mass:

[0047] Calcium chloride 70~80%, sodium chloride 10~20%, nano-silicon dioxide 10~15%, dispersant 5~10%, EDTA-polymer derivative 0.5~3%.

[0048] Calcium chloride is used to provide Ca 2+ source, selectively adsorbing ions in the mud and optimizing ion balance; sodium chloride is used to provide Na + source to optimize ion balance; nano-silica is used to promote gel cross-linking; the dispersant is a polycarboxylic acid dispersant to prevent the agglomeration of the additives; EDTA-polymer derivatives include PAA-EDTA and PEG-EDTA, which serve as polymers to block the bentonite adsorption sites and form a dense structure layer.

[0049] The preparation method of the curing enhancement modification auxiliary agent comprises:

[0050] a1. Dissolve calcium chloride and sodium chloride in deionized water in proportion to prepare a 3~5mol / L mother liquor;

[0051] a2. Add nano-silica to the mother liquor and ultrasonically disperse for 30 min at an ultrasonic frequency of 20 to 40 kHz and an ultrasonic power of 300 W.

[0052] a3. Add EDTA-polymer derivative (PAA-EDTA, PEG-EDTA) to the solution obtained in a2 and stir to mix at a stirring speed of 300 rpm for 1 min.

[0053] a4. Finally, add the polycarboxylic acid dispersant and stir until a homogeneous suspension is obtained. The stirring speed is 500 rpm and the stirring time is 2 h.

[0054] Dosage and usage of curing enhancement modifier:

[0055] Add curing enhancement modification additives according to 1.7~6.7% of the dry mass of bentonite in the waste mud, stir at high speed shearing speed of 1200rpm for 15 minutes, let it stand for 10 minutes, and then add curing agent for curing treatment.

[0056] The curing agent components include slag, cement and water glass. The specific mass ratio of each raw material component is as follows: S95 slag: PO42.5 cement: water glass (modulus 1.2) = 42:38:20.

[0057] Additives, that is, additives used to improve the effect of geopolymer curing agents, are selected based on the clay mineral adsorption mechanism, geopolymer reaction mechanism and experimental results. The experimental results are: CaCl2>NaCl.

[0058] Clay minerals in waste mud first cause agglomeration during the solidification process, which is usually visible in the early stage (a few hours). These clay mineral aggregates will adsorb Si, Al, Na, Ca and the hydrated gel products of the curing agent, affecting the micro-environment of the solidification. Due to their high cation exchange capacity and surface area, the adsorbed Ca 2+ Tends to form CaCO3. In geopolymer solidification, the three-dimensional network polymerization gelation and hydrated gel of silica and aluminum are the main sources of solidified body strength. However, the adsorption of clay minerals reduces the formation of silica-alumina tetrahedrons and Na + / Ca 2+ , resulting in a decrease in strength structure. At the same time, even if a hydrated gel and three-dimensional network structure are formed, their distribution may be uneven, weakening the connection of the strength structure. Therefore, mud with a high clay mineral content usually exhibits lower solidification strength.

[0059] Therefore, this method is based on Na + / Ca 2+ Salt is used as an admixture to pre-treat the waste mud to eliminate / cover the effect of bentonite adsorption on geopolymer curing: Na added by the admixture + / Ca 2+ It will be adsorbed by bentonite first, the adsorption of bentonite is offset, and bentonite tends to adsorb Na + / Ca 2+ ions, which depends on the fact that bentonite itself has a large specific surface area structure, + / Ca 2+ Ions have a greater affinity, these ions can effectively fill the interlayer space of clay minerals and maintain the stability of their layered structure. Therefore, when the geopolymer curing agent is added later, the probability of the ions dissolved from the raw materials and the hydration products generated by the reaction being adsorbed by bentonite will be greatly reduced. In addition, Na + / Ca 2+Chloride is an ion required for geopolymer reaction to form hydrated gel and three-dimensional network structure. The addition of admixtures improves the ion concentration in the mud and also promotes the geopolymer reaction. The reason for choosing chloride salt is that the geopolymer reaction is more sensitive to the concentration of other ions, including but not limited to the influence of other anion concentration on pH, competition with cations such as aluminum and silicon, and thus affecting the polymerization reaction of geopolymer. - Although ions may also participate in the geopolymer reaction, the chlorine-containing and insoluble Friedel and Kuzel salts generated by the reaction can provide strength for the solidified mud and further improve the porosity of the solidified mud.

[0060] like Figure 1 As shown, the method for enhancing the curing performance of geopolymer of high bentonite content slurry also includes constructing an intelligent real-time detection and control system, and the construction process includes:

[0061] b1. Bentonite characteristic response sensor is set on the pumping path of the mud mixing tank to detect the bentonite content in the mud;

[0062] b2. A multi-modal element sensor is provided at the input and output ports of the mud mixing tank to detect the concentration of a specified ion;

[0063] b3 set curing agent feeding point, additive feeding point, the curing agent feeding point, additive feeding point are docking mud mixing bin, the curing agent feeding point for storing the curing agent, the additive feeding point for storing curing enhancement modification additives;

[0064] b4. Use feedback control software to obtain parameter information of bentonite characteristic response sensors and multimodal element sensors, make real-time predictions of bentonite content and ion concentration, make decisions on the optimal amount of material addition for curing agent and curing enhancement and modification additives, and control the material addition at the curing agent addition point and additive addition point.

[0065] The intelligent real-time detection and control system is regulated based on the additive's function, enabling rapid, high-volume, and time-saving slurry processing. It consists of a hardware identification and detection module and feedback-based control software. The hardware includes a bentonite characteristic response sensor for bentonite content detection and a multimodal ion sensor for real-time ion concentration monitoring. The feedback-based control software analyzes input data based on a large model to adjust additive / curing agent dosage.

[0066] like Figure 2As shown in the figure, the bentonite characteristic response sensor is used to identify and detect the bentonite content in the mud during the mud pumping process, providing data reference for subsequent additive / curing agent adjustment. The bentonite characteristic response sensor includes a CEC rapid measurement unit and an NIR auxiliary identification unit.

[0067] The CEC rapid determination unit has a built-in ammonium ion saturation column and a conductivity detection chamber. The ammonium ion saturation column is filled with a microfluidic exchange column (5mm in diameter, 10cm in length) containing NH4Cl solution (1mol / L). When the mud passes through at a fixed low flow rate, the ammonium ion replacement of the cations between the montmorillonite layers is completed; the conductivity detection chamber uses a multi-electrode array (spacing 2mm) to monitor the conductivity changes of the replacement fluid in real time, and combines the Langmuir adsorption model to calculate the CEC value, which can complete the preliminary detection of bentonite content within minutes.

[0068] NIR auxiliary recognition unit is equipped with 510~530 cm -1 interval and 1070~1090 cm -1 A near-infrared spectrometer with a characteristic wavelength range of 10 nm was used to analyze the Si-O-Al bending vibration peak and Si-O-Si stretching vibration peak of montmorillonite, and a quantitative relationship between the spectral absorbance and the bentonite content was established through a regression algorithm, which was used for auxiliary detection of bentonite content.

[0069] The calculation process of establishing the quantitative relationship between spectral absorbance and bentonite content through regression algorithm includes: data acquisition → spectral preprocessing → characteristic wavelength screening → modeling → prediction.

[0070] Prepare standard samples with known bentonite content, covering the detection range (0-100%). Then use a near-infrared spectrometer to measure the absorbance of each sample within the characteristic peak wavelength range.

[0071] c2. Based on the characteristic peak wavelength range of each sample tested, calculate the correlation coefficient between the corresponding wavelength of each sample and the characteristic peak wavelength of the bentonite standard. Retain wavelengths with a correlation coefficient greater than 0.9, and automatically select and optimize the optimal wavelength combination through intelligent algorithms (such as genetic algorithms).

[0072] c3. Use partial least squares regression (PLSR) to build a regression model, decompose the spectral data (X) and content data (Y) simultaneously, extract the common principal components (latent variables), and use the latent variables to construct a linear equation: predicted content = a × absorbance 1 + b × absorbance 2 + constant, where a and b are 520 cm -1 and 1080cm -1 The contribution weights of the two characteristic wavelengths, absorbance 1 and absorbance 2 correspond to 520cm -1 and 1080cm -1The sum or average of the absorbance values within two characteristic wavelength ranges. To address absorbance saturation in high-concentration bentonite samples in mud, a nonlinear compensation term is added: For the low-concentration range (<25%), the predicted value = the standard PLSR result; for the high-concentration range (>25%), the predicted value = the standard PLSR result + correction factor × ln(absorbance). Coefficients a and b are obtained by fitting the standard sample set using the partial least squares method; the correction factor is determined by nonlinear fitting of high-concentration samples and dynamically adjusted based on environmental parameters.

[0073] like Figure 1 As shown, modal element sensors are placed at the input and output ports of the mud mixing chamber to detect the concentration of specific ions. Combining and optimizing traditional fluorescence sensing and ion-selective electrode (ISE) technologies, the modal element sensors utilize an ISE-fluorescence dual-mode probe structure to enable simultaneous, rapid, and efficient detection of multiple elements (here, sodium, calcium, silicon, and aluminum). Free sodium and calcium are primarily present in mud as ions, so ISE technology is used for detection. Free silicon and aluminum are primarily present in mud as monomers bound to oxygen atoms, so fluorescence sensing technology is used for detection. By measuring the concentrations of calcium, sodium, silicon, and aluminum, the levels of these elements can be used to reflect the effectiveness of additive control and whether the geopolymer reaction environment has reached optimal conditions.

[0074] like Figure 3 As shown, the modal element sensor includes:

[0075] ISE unit: using all-solid-state Ca 2+ / Na + Dual-channel ion-selective electrode, the electrode membrane material is a perovskite / graphene composite membrane (thickness 50μm), and dual-ion synchronous detection is achieved through microchannel separation.

[0076] Fluorescence unit: integrated dual-wavelength excitation fiber optic sensor laser light source, embedded Al fluorescent probe (based on Lumogallion organic matter) and Si-responsive complex fluorescent probe (based on ammonium molybdate-(NH4)6Mo7O 24 ), the excitation light source wavelengths are 350~380nm and 500~510nm respectively.

[0077] Working principle of ISE unit: slurry is injected from the inlet and divided into two paths through the microchannel, flowing through the Ca 2+ and Na + Sensitive membrane (i.e. electrode membrane) area, each selectively corresponds to Na + and Ca 2+ , and generate membrane potential, and output signals through the conductive layer.

[0078] Ca 2+ Sensitive film: Eu-doped3+ Fluoride perovskite (KCaF3) and graphene composite film, with Ca 2+ Response; Na⁺ sensitive film: doped with Al 3+ NASICON-type perovskite (Na3Zr 1.8 Al 0.2 Si2PO 12 ) and graphene composite membrane, and Na + response.

[0079] The present invention combines fluorescence sensor and ion selective electrode (ISE) technology to monitor the ion concentration in mud in real time, aiming to measure the sodium (Na) in mud efficiently and accurately. + ), calcium (Ca 2+ ), silicon (Si) and aluminum (Al 3+ ) ion concentration, thereby optimizing the geopolymer solidification process and achieving large-scale, standardized solidification treatment of waste mud. The working principle of this method is explained as follows:

[0080] ① Na + and Ca 2+ Ion concentration is measured using a dedicated ion-selective electrode, leveraging the membrane's selective permeability and potential difference. When the electrode is immersed in the sample, the ions interact with the membrane surface, generating a potential difference. This potential difference is linearly related to the ion concentration in the sample, and the ion concentration can be calculated. These two ions exhibit good conductivity and strong ion exchange capacity in aqueous solution, making them suitable for precise measurement using ion-selective electrodes.

[0081] ② Regarding the detection of silicon and aluminum ions, silicon typically exists in a polymerized state, forming silicates, and its concentration as a monomer in water is relatively low, making it difficult to measure directly using traditional ion-selective electrodes. The measurement of aluminum ions is affected by its complex behavior in acidic water and is also unsuitable for standard ion-selective electrodes. Therefore, fluorescence sensor technology is used. Fluorescent probes bind to Si and Al ions in the sample, generating a fluorescence signal of a specific wavelength. By irradiating the sample with an excitation light source, the fluorescent probe is excited, and the intensity of the fluorescence signal is proportional to the ion concentration. The detector measures the fluorescence intensity in real time and converts it to the corresponding ion concentration using a standard curve.

[0082] ③Finally, the intelligent control system combines real-time data from the ion-selective electrode and fluorescence sensor for comprehensive analysis and feedback. Based on the set optimal reaction conditions, the system dynamically adjusts the ion concentration in the mud to ensure optimal results during the geopolymer curing process.

[0083] like Figure 1As shown, the feedback control software primarily consists of an improved BP neural network concentration prediction model and a multimodal dynamic compensation algorithm. The improved BP neural network concentration prediction model uses a large database to predict bentonite content and ion concentration in real time using various parameters (ISE, fluorescence intensity, pH, CEC, flow rate, etc.) input by sensors or manual input, thereby making decisions on the optimal material dosage. The multimodal dynamic compensation algorithm primarily compensates for sensor drift (ISE potential compensation, fluorescence signal attenuation correction) and inter-ion interference, correcting the concentration prediction model's decisions.

[0084] The large database is a summary of data obtained through tests on a large number of experimental groups, including bentonite content (characterization methods include X-ray diffraction (XRD), X-ray fluorescence (XRF), and near-infrared spectroscopy (NIR), calcium, sodium, silicon, and aluminum ion concentrations, pH value, CEC, curing agent dosage, unconfined compressive strength (UCS), and mud moisture content.

[0085] Decisions on the optimal amount of material to be delivered are made by feeding real-time monitoring data into a trained model. The model analyzes the reaction environment based on the training data and ultimately provides feedback and control. Each data detected in the model is assigned an influence weight.

[0086] Sensor drift compensation: Based on the law of sensor drift, mathematical models (such as linear regression, curve fitting, etc.) are used to estimate and correct drift.

[0087] Inter-ion interference compensation: Through extensive testing, the rules of mutual interference are derived, forming a linear regression model, which is then corrected. The microchannel separation and specialized membrane structure within the ISE unit also serve as anti-interference measures.

[0088] The improved BP neural network concentration prediction model builds on the traditional BP framework by integrating deep learning optimization techniques with domain knowledge embedded training. The improvements include: 1) Network structure optimization, employing a dual-hidden layer design. The first hidden layer captures global features, while the second layer extracts high-order nonlinear relationships, suppressing noise through gradual dimensionality reduction. 2) Regularization strategy upgrades, randomly discarding entire neuron connection paths with a probability of p=0.3 to force the network to learn redundant features. 3) Dynamic training mechanism, collecting new data every 30 minutes and calculating the distribution difference between the new and old data. If the threshold θ=0.1 is reached, parameter fine-tuning is triggered, freezing the first two layers and updating only the last two layers.

[0089] The calculation process of the multimodal dynamic compensation algorithm is: simultaneous extraction of multimodal features → cross-interference modeling → online weight allocation → dynamic compensation correction → confidence feedback update.

[0090] The following examples are given to illustrate the application of this method.

[0091] Example 1:

[0092] Curing experiments were conducted using waste slurry from slurry shield tunneling (13% bentonite dry matter content) and drilling mud (25% bentonite dry matter content) from engineering projects. The curing agent dosage was 20% of the dry mass of the mud in each case. A curing enhancement modifier was added before curing, with a gradient of 1.7% to 6.7% addition. The CP group was the group without the curing enhancement modifier. The curing enhancement modifier used consisted of the following raw materials in the following mass percentages: calcium chloride 70%, sodium chloride 14.5%, nano-silica 10%, dispersant 5%, and EDTA-polymer derivative 0.5%.

[0093] Under standard curing conditions, the unconfined compressive strength of the solidified body was tested at 3, 7, 14 and 28 days. The results are as follows: Figure 4 As shown, Figure 4 The left picture in the middle corresponds to shield mud, and the right picture corresponds to drilling mud. Figure 4 The dosage of the medium additive is 1.7%, 3.4%, 5% and 6.7% respectively.

[0094] Shield slurry group: The best effect was achieved when the modifier content was 3.4% of the dry mass of bentonite, and the 28-day curing strength increased to 315.1% of the original strength (164.5kPa → 518.3kPa).

[0095] Drilling mud group: The best effect was achieved when the modifier dosage was 3.4% of the dry mass of bentonite, and the 28-day curing strength increased to 183.1% of the original (160.1kPa→293.2kPa).

[0096] Example 2:

[0097] Curing experiments were conducted using waste slurry from slurry shield tunneling (13% bentonite dry matter content) and drilling mud (25% bentonite dry matter content) from engineering projects. The curing agent dosage was 20% of the dry mass of the mud in each case. A curing enhancement modifier was added before curing, with a gradient of 1.7% to 6.7% addition. The CP group was the group without the curing enhancement modifier. The curing enhancement modifier used consisted of the following raw materials in the following mass percentages: calcium chloride 74.5%, sodium chloride 10%, nano-silica 10%, dispersant 5%, and EDTA-polymer derivative 0.5%.

[0098] Under standard curing conditions, the unconfined compressive strength of the solidified body was tested at 3, 7, 14 and 28 days. The results are as follows: Figure 5 As shown, Figure 5 The left picture in the middle corresponds to shield mud, and the right picture corresponds to drilling mud. Figure 5 The dosage of the medium additive is 1.7%, 3.4%, 5% and 6.7% respectively.

[0099] Shield slurry group: The best effect was achieved when the modifier content was 3.4% of the dry mass of bentonite, and the 28-day curing strength increased to 346.5% of the original (164.5kPa→570kPa).

[0100] Drilling mud group: The best effect was achieved when the modifier dosage was 5.0% of the dry mass of bentonite, and the 28-day curing strength increased to 284.3% of the original (160.1kPa→455.1kPa).

[0101] Example 3:

[0102] Curing experiments were conducted using waste slurry from slurry shield tunneling (13% bentonite dry matter content) and drilling mud (25% bentonite dry matter content) from engineering projects. The curing agent dosage was 20% of the dry mass of the mud in each case. A curing enhancement modifier was added before curing, with a gradient of 1.7% to 6.7% addition. The CP group was the group without the curing enhancement modifier. The curing enhancement modifier used consisted of the following raw materials in the following mass percentages: 80% calcium chloride, 10% sodium chloride, 15% nano-silica, 7% dispersant, and 3% EDTA-polymer derivative.

[0103] Under standard curing conditions, the unconfined compressive strength of the solidified body was tested at 3, 7, 14 and 28 days. The results are as follows: Figure 6 As shown, Figure 6 The left picture in the middle corresponds to shield mud, and the right picture corresponds to drilling mud. Figure 6 The dosage of the medium additive is 1.7%, 3.4%, 5% and 6.7% respectively.

[0104] Shield slurry group: The best effect was achieved when the modifier content was 3.4% of the dry mass of bentonite, and the 28-day curing strength increased to 201.0% of the original strength (164.5kPa→330.6kPa).

[0105] Drilling mud group: The best effect was achieved when the modifier dosage was 3.4% of the dry mass of bentonite, and the 28-day curing strength increased to 117.6% of the original (160.1kPa→188.2kPa).

[0106] In the above embodiments, the dispersant and EDTA-polymer derivative should be used in small and appropriate amounts, otherwise the effect of the curing agent will be affected.

[0107] This method can effectively eliminate the negative effects of bentonite in geopolymer reactions, increase the strength of solidified slurries, and improve the efficiency of solid waste resource recovery. Within the optimal dosage range, the maximum strength can be increased by 250% to 320%. Although the solidification effect of different slurries may vary significantly, the strength improvement is indeed significant and is also very effective in eliminating the effect of bentonite in geopolymer reactions.

[0108] On the basis of using additives to solidify and modify the waste mud, combined with the intelligent real-time detection and control system, real-time and efficient control of solidification is achieved, effectively solving the problem of different solidification effects of multi-source mud. The specific comparison with traditional methods is shown in the following table:

[0109]

[0110] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art of the present invention, without departing from the concept of the present invention, several simple deductions or substitutions can be made, which should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for improving the curing performance of geopolymers in high bentonite content slurries, characterized in that: The invention comprises a method for adding a curing enhancement modification auxiliary agent, wherein the method comprises: adding the curing enhancement modification auxiliary agent into a slurry with a high bentonite content, stirring the slurry evenly, and then standing the slurry to cure the slurry; The curing enhancement and modification additive includes the following raw material components in percentage by mass: Calcium chloride 70-80%, sodium chloride 10-20%, nano-silicon dioxide 10-15%, dispersant 5-10%, EDTA-polymer derivative 0.5-3%; Add curing enhancement and modification additives to the high bentonite content mud, stir evenly and let it stand for curing, specifically including: Add a curing enhancement and modification additive according to 1.7-6.7% of the dry mass of bentonite in the mud, stir at high speed, let it stand, and then add a curing agent for curing treatment; the curing agent includes the following raw material components in the following mass ratio: slag: cement: water glass = 42:38:20; The dispersant is a polycarboxylic acid dispersant, and the EDTA-polymer derivative includes PAA-EDTA and PEG-EDTA; It also includes building an intelligent real-time detection and control system. The construction process includes: b1. Bentonite characteristic response sensor is set on the pumping path of the mud mixing tank to detect the bentonite content in the mud; b2. A multi-modal element sensor is provided at the input and output ports of the mud mixing tank to detect the concentration of a specified ion; b3 set curing agent feeding point, additive feeding point, the curing agent feeding point, additive feeding point are docking mud mixing bin, the curing agent feeding point for storing the curing agent, the additive feeding point for storing curing enhancement modification additives; b4. By obtaining parameter information of bentonite characteristic response sensors and multimodal element sensors through feedback control software, real-time prediction of bentonite content and ion concentration is performed, and the optimal amount of curing agent and curing enhancement modifier material feeding is decided, and the curing agent feeding point and the additive feeding point are controlled; The multimodal element sensor includes an ISE unit and a fluorescence unit.

2. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 1, characterized in that: The preparation method of the curing enhancement modification auxiliary agent comprises: a1. Dissolve calcium chloride and sodium chloride in deionized water according to the composition ratio to prepare a mother liquor; a2. Nano-silica was added to the mother liquor and ultrasonically dispersed to obtain a uniformly dispersed solution; a3. EDTA- polymer derivative was added to the solution obtained in a2, and the mixture was stirred to obtain a uniform solution; a4. Add a dispersant to the solution obtained in a3 and stir to mix until a homogeneous suspension is obtained.

3. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 1, characterized in that: The bentonite characteristic response sensor includes a CEC rapid measurement unit, which has a built-in ammonium ion saturation column and a conductivity detection cavity. The ammonium ion saturation column is a microfluidic exchange column filled with a displacement fluid, and completes the ammonium ion replacement of cations between montmorillonite layers when mud passes through at a fixed low flow rate; the conductivity detection cavity monitors the change in the conductivity of the displacement fluid in real time through a multi-electrode array, and calculates the CEC value in combination with the Langmuir adsorption model.

4. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 1, characterized in that: The bentonite characteristic response sensor includes an NIR auxiliary recognition unit for auxiliary detection of bentonite content, and the NIR auxiliary recognition unit is equipped with a 510-530 cm -1 interval and 1070~1090 cm -1 A near-infrared spectrometer with a characteristic wavelength range of 1 / 40 nm was used to establish a quantitative relationship between spectral absorbance and bentonite content through regression algorithm.

5. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 4, characterized in that: The process of establishing a quantitative relationship between spectral absorbance and bentonite content through regression algorithm includes: c1. Prepare standard samples with known bentonite content to cover the detection range, and measure the absorbance of each standard sample at the characteristic peak wavelength range using a near-infrared spectrometer; c2. Based on the characteristic peak wavelength range of each sample tested, calculate the correlation coefficient between the corresponding wavelength of each sample and the characteristic peak wavelength of the bentonite standard, retain the wavelengths with a correlation coefficient greater than 0.9, and automatically select and optimize the optimal wavelength combination through an intelligent algorithm; c3. Use partial least squares regression to build a regression model, decompose the spectral data and content data simultaneously, extract the common latent variables, and use the latent variables to construct a linear equation: predicted content = a × absorbance 1 + b × absorbance 2 + constant, where a and b are 520 cm -1 and 1080cm -1 The contribution weights of the two characteristic wavelengths, absorbance 1 and absorbance 2 correspond to 520cm -1 and 1080cm -1 The sum or average absorbance of the two characteristic wavelength ranges; when the predicted content is in the low content region of the detection range (<25%), the predicted value = the standard PLSR result; when the predicted content is in the high content region of the detection range (>25%), the predicted value = the standard PLSR result + correction coefficient × ln (absorbance); coefficients a and b are obtained by fitting the standard sample set through the partial least squares method, and the correction coefficient is determined by nonlinear fitting of high-concentration samples and is dynamically adjusted with environmental parameters.

6. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 1, characterized in that: ISE unit: using all-solid-state Ca 2+ / Na + Dual-channel ion selective electrode, using perovskite / graphene composite membrane as electrode membrane material, Ca 2+ Ion selective electrodes and Na + The ion selective electrodes are covered with electrode membranes to form Ca 2+ Sensitive membrane and Na + The sensitive membrane is separated by microchannels into two parts: 2+ Sensitive membrane and Na + Two flow channels of the sensitive membrane for dual ion simultaneous detection; Fluorescence unit: an excitation light source of an integrated dual-wavelength excitation fiber optic sensor, an Al fluorescent probe based on luminogallion organic matter, and a Si fluorescent probe based on ammonium molybdate complex. The excitation light source wavelengths for the Al fluorescent probe and the Si fluorescent probe are 350-380 nm and 500-510 nm, respectively.

7. The method for enhancing the curing performance of geopolymer of high bentonite content slurry according to claim 1, characterized in that: The feedback control software includes a concentration prediction model of a BP neural network and a multi-modal dynamic compensation algorithm; BP neural network concentration prediction model: Based on a large database, it obtains sensor parameters or manually input parameters to make real-time predictions of bentonite content and ion concentration, and make decisions on the optimal amount of material to be added. Parameters include ISE, fluorescence intensity, pH, CEC, and flow rate; Multimodal dynamic compensation algorithm: Through sensor drift compensation and inter-ion interference compensation, sensor drift compensation includes ISE potential compensation and fluorescence signal attenuation correction, to correct the concentration prediction model decision.

Images