A mud self-adaptive return flow control system based on foam state

Abstract

The application belongs to the technical field of automation control and regulation technology, and relates to a mud self-adaptive reflux control system based on a foam state. The system comprises a pre-foam-removing mud generation module, a physical state feature extraction module, a foam potential index modeling module, a dynamic threshold screening module, a hierarchical cutting and stripping module, a feed-forward dosing control module and a feedback correction adjustment module. The pre-foam-removing mud generation module generates a pre-foam-removing mud flow. The physical state feature extraction module extracts mud physical feature data. The foam potential index modeling module calculates a deep foaming risk assessment model. The dynamic threshold screening module screens hierarchical control dynamic threshold parameters. The hierarchical cutting and stripping module truncates output upper-layer reflux mud and bottom-layer output mud flow. The feed-forward dosing control module generates a defoaming control signal. The feedback correction adjustment module feeds back and adjusts a preset weight product coefficient in the parameter integral operation according to the light transmittance of supernatant water. The application solves the problem that the physical state of foam-containing mud is not quantitatively perceived, so that it is difficult to implement targeted reflux control based on dynamic threshold hierarchical cutting.

Description

Technical Field

[0001] This invention belongs to the technical field of automated control and regulation technology, and relates to an adaptive reflux control system for mud based on foam state. Background Technology

[0002] In underground engineering construction, mud treatment is a critical step. The treatment process typically involves injecting chemical reagents and using solid-liquid separation mechanisms to achieve purification. However, the discharged raw foamy mud often contains a large amount of stable foam. These bubbles reduce the mud density and hinder the binding of flocculants to solid particles, affecting subsequent solid-liquid separation. Therefore, how to implement adaptive reflux control for foamy mud is a key technical challenge in this field.

[0003] To address the aforementioned needs, existing technologies have explored relevant approaches. For example, CN120215604A discloses a method and control system for treating acidic wastewater using a high-density sludge process. This scheme receives real-time pH and liquid level data from within the reaction device and inputs it into a reflux ratio calculation model. Combined with an algorithm, it calculates the dosage and sludge reflux flow rate, thereby controlling the operating parameters of relevant pump sets to maintain the pH within the reaction device within a set range, achieving automated adjustment of dosing and reflux parameters.

[0004] The aforementioned technology has limitations in dealing with foam-rich mud. The system relies solely on conventional macroscopic parameters such as pH and liquid level, lacking the ability to quantitatively perceive the physical state of the mud surface and internal depth. This processing logic struggles to combine surface images with deep-penetrating acoustic waves to analyze mud foaming risk and cannot establish a foam potential index profile model. Therefore, the system cannot dynamically filter stratified control threshold parameters, leading to confusion between high-risk foam-containing mud and the underlying high-quality mud, making it difficult to implement longitudinal profile cutting and targeted backflow control for pre-defoaming mud flows.

[0005] In summary, existing control systems suffer from a lack of multi-dimensional quantitative perception models of the physical state of the controlled object, and the control logic lacks an adaptive closed-loop mechanism for feedforward compensation and feedback correction. This results in technical problems that make it difficult for the system to implement precise targeted backflow and throttling control for dynamic thresholds. Summary of the Invention

[0006] To address the technical problem of existing control systems lacking multi-dimensional quantitative sensing and adaptive closed-loop adjustment capabilities, this invention provides a mud adaptive recirculation control system based on foam state.

[0007] An adaptive mud recirculation control system based on foam state includes: The pre-defoaming mud generation module obtains the original foamed mud, performs simultaneous surface and bottom layer chemical addition operations, and generates a pre-defoaming mud flow. The physical state feature extraction module guides the pre-defoamed mud flow into the expansion and deceleration zone, controls the image acquisition device and the ultrasonic device to acquire data synchronously, and extracts the physical feature data of the mud. The foam potential index modeling module calls the mud foaming risk quantification mapping function to parse the physical characteristic data of the mud, and performs parametric integration along the vertical depth coordinate line in combination with the preset weight product coefficient to calculate the depth foaming risk assessment model. The dynamic threshold filtering module iterates through the risk classification data contained in the deep foaming risk assessment model, compares it with the bubble concentration tolerance threshold, and filters out the stratified control dynamic threshold parameters. The layered cutting and stripping module sends the layered control dynamic threshold parameters to the dynamic interception mechanism to perform layered physical interception on the pre-defoaming mud flow, cutting off the upper layer return mud flow and the bottom layer output mud flow. The feedforward drug delivery control module monitors and records the real-time return flow load signal of the upper reflux mud, initiates the advance compensation drug delivery strategy, and generates an antifoaming control signal. The feedback correction and adjustment module guides the bottom output mud flow into the sedimentation and granulation area, adds polymer flocculant to cause solids to settle, and adjusts the preset weight product coefficient in the parameter integration calculation based on the transmittance of the supernatant.

[0008] A further embodiment of the present invention, wherein the simultaneous application of chemicals to the surface and bottom layers to generate a pre-defoamed mud flow, includes the following steps: The original foamy mud is guided into the mixing and defoaming chamber; Activate the atomizing spray valve array installed at the top of the mixing defoaming chamber to spray chemical defoamer onto the exposed bubble surface area of ​​the original foamy slurry; Turn on the circulation pump located at the bottom of the mixing and defoaming chamber, extract the bottom slurry carrying unbroken microbubbles, and inject chemical defoamer into the main pipeline and apply cutting and shearing force; The bottom mud, after being treated twice, is transported back to the mixing and defoaming chamber to be mixed and generate the pre-defoaming mud flow.

[0009] A further embodiment of the present invention, wherein the control image acquisition device and the ultrasonic device synchronously acquire and extract physical characteristic data of the mud, includes the following steps: The image acquisition device, positioned at the top of the expansion and deceleration zone, continuously captures surface foam images of the liquid surface. Based on the inter-frame pixel differences between consecutive images, the dissipation rate factor that quantifies the dynamic balance between bubble generation and rupture is extracted. Simultaneously send detection commands to the vertical ultrasonic transceiver array that is attached to the sidewall of the expansion and deceleration area, transmit acoustic pulse sequences and receive reflected echoes that penetrate the lower mud layer, and extract the energy attenuation rate factor that reflects the acoustic impedance characteristics at different immersion depths. The feature data of the surface foam image, the dissipation rate factor, and the energy decay rate factor are extracted, and then time-stamped and encapsulated to construct the physical feature data of the mud.

[0010] A further aspect of the present invention includes the following steps for extracting the energy attenuation rate factor, which reflects the acoustic impedance characteristics at different immersion depths: Obtain the effective path length of ultrasonic waves propagating in mud and the reference amplitude; Obtain the actual measured echo signal amplitude at each immersion depth; Calculate the ratio of the reference amplitude to the echo signal amplitude, and perform a base-10 logarithmic operation on the ratio to obtain the logarithmic result; Based on the signal attenuation ratio between the reference amplitude and the echo signal amplitude, and in conjunction with the effective path length, the energy attenuation rate factor corresponding to each immersion depth is calculated.

[0011] A further aspect of this invention involves calling a mud bubbling risk quantification mapping function to parse the mud physical characteristic data, performing parametric integration along the vertical depth coordinate line in combination with preset weighted product coefficients, and calculating a depth bubbling risk assessment model, including the following steps: Extract the feature data and dissipation rate factor of the surface foam image encapsulated within the physical feature data of the mud, and input them into the surface activity assessment formula to perform numerical calculations to generate an apparent risk score. The energy decay rate factor at each depth carried by the physical characteristic data of the mud is matched and extracted, and then converted into the internal enrichment density reference value at the corresponding liquid surface depth using an acoustic empirical inversion algorithm. The preset weighted product coefficients are configured for the apparent risk score and the internal enrichment density reference value, and a multiplication-addition operation is performed to obtain the fusion risk coefficient. By integrating the fusion risk coefficients at various depths and establishing a continuous analytical model for the target depth, the bubble risk assessment model for that depth is output.

[0012] A further aspect of the present invention involves the following steps: traversing and reading the risk classification data contained in the deep bubbling risk assessment model, comparing it with the bubble concentration tolerance threshold, and filtering out the stratified control dynamic threshold parameters. The connection retrieves the depth bubbling risk assessment model, which exists in the form of a continuous function. Call the storage unit to read the bubble concentration tolerance threshold that represents the allowable entry into the back-end flocculation process under the current operating conditions; A risk benchmark equation is constructed, and the corresponding depth solution is obtained by solving the risk benchmark equation within the effective water depth range through a continuous solution algorithm model. Lock the specific liquid surface depth value corresponding to the depth solution, and format and encode the specific liquid surface depth value to generate the layered control dynamic threshold parameter.

[0013] A further embodiment of the present invention, wherein the step of sending the layered control dynamic threshold parameter to the dynamic interception mechanism to perform layered physical interception on the pre-defoaming mud flow, cutting off the upper layer return mud flow and the lower layer output mud flow, includes the following steps: The dynamic threshold parameters of the layered control, which include specific liquid level depth values, are analyzed and converted into displacement control signals to drive the servo motor. The guide tube head mounted on the longitudinal slide rail is driven to move up and down until the guide tube head moves to the underwater control surface indicated by the displacement control signal; Turn on the suction negative pressure pump group that is connected to the outside of the guide pipe head to suck up and strip the upper layer of return mud located at or above the control surface; The control surface physically seals the outflow channel of the lower layer mud, and under the action of gravity static pressure, the bottom layer output mud flow located deep at the bottom end of the control surface is drained out.

[0014] A further embodiment of the present invention, wherein the monitoring and recording of the real-time return flow load signal of the upper layer of refluxed mud, initiating an advance compensation dosing strategy, and generating a defoaming control signal, includes the following steps: The instantaneous volumetric flow rate signal fed back by the electromagnetic flow meter deployed on the circulating return pipeline is scanned, and the real-time flow quality value of the upper layer return slurry is calculated and extracted within a continuous statistical period. The extracted upper layer of refluxed mud is reinjected into the mixing and defoaming chamber; The deviation and rate of change between the real-time flow quality value and the median of the preset standard return expected balance interval are calculated, and matched with the preset nonlinear agent consumption compensation matrix table. The compensation difference is obtained by analyzing the table lookup matching results, generating a pulse waveform packet containing the corrected duty cycle, and sending it to the actuator as the defoaming control signal carrying the compensation difference of the agent valve control.

[0015] A further embodiment of the present invention, wherein guiding the bottom output mud flow into the sedimentation and granulation area, adding a polymeric flocculant to cause the solid phase to settle, and adjusting the preset weighted product coefficient in the parametric integration calculation based on the transmittance of the supernatant, includes the following steps: The bottom-layer output mud flow is fed into the end flocculation and forming chamber equipped with blades; Start the variable frequency motor configured at the top of the end flocculation forming chamber to drip flocculant into the bottom layer output mud flow which is in a stirring state; The stirring action generated by the variable frequency motor driving the blades accelerates the solid phase sedimentation and separation from the liquid phase, and the supernatant liquid at the top overflows and precipitates out. The optical detection device is activated to continuously scan the transmittance of the supernatant. When the real-time value of the transmittance is less than the allowable threshold, an optimization command is sent in reverse to adaptively increase the preset weight product coefficient.

[0016] A further aspect of the present invention, wherein the adaptive increase of the preset weight product coefficient includes the following steps: Obtain the preset weight product coefficient and the preset fixed step size increment currently in use during the control cycle; The transmittance of the supernatant liquid, measured in real time during the current control cycle, and the preset allowable boundary line for compliance are obtained. Determine whether the light transmittance is less than the acceptable boundary line; If the judgment result is yes, then based on the fixed step size increment, the preset weight product coefficient currently in use is incrementally updated.

[0017] In summary, the present invention has the following beneficial technical effects: 1. The pre-defoamed mud flow is guided into the expansion and deceleration region. A multi-dimensional physical state feature set is extracted using a synchronous sampling array of surface images and deep-penetrating acoustic waves, and a foam potential index profile model is calculated. This expands the foaming risk assessment to a multi-dimensional quantitative analysis encompassing the geometric contour of the liquid surface and the acoustic characteristics of internal microbubbles, providing an objective data foundation for system control and solving the problem of the lack of quantitative perception of the physical state of foamed mud in existing technologies.

[0018] 2. The data from the foam potential index profile model is traversed and read, and the dynamic threshold parameters for stratified control are selected by comparing them with the bubble concentration tolerance threshold. These parameters are then sent to the dynamic interception mechanism to implement longitudinal profile cutting and stripping. The model is transformed into a physical control boundary, enabling the mechanism to cut off the targeted return mud components and guide the bottom layer of output mud flow out under gravity. This method achieves targeted stratified interception, overcoming the shortcomings of existing technologies in implementing targeted return based on dynamic threshold stratified cutting.

[0019] 3. A pre-compensation strategy is initiated by monitoring the real-time return flow load signal of the targeted reflux mud components, and the preset weighted product coefficient is adjusted based on the terminal's transmittance characteristics using closed-loop feedback. An adaptive control system architecture combining feedforward drug delivery and closed-loop feedback correction is constructed. The feedforward stage actively adjusts the defoaming control signal to proactively address load impacts, while the feedback stage corrects the risk assessment model in reverse. This algorithm mechanism enables the entire automated control system to adapt to the dynamic fluctuations of mud properties, improving the control system's anti-interference capability and the overall stability of closed-loop regulation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings are used to provide a further understanding of the present invention.

[0021] Figure 1 This discloses a schematic diagram of the framework in the embodiments of this application.

[0022] Figure 2 This discloses a flowchart of an embodiment of this application.

[0023] Figure 3 The present application discloses a curve showing the distribution of the depth bubbling risk assessment model in the effective water depth range in the embodiments of this application.

[0024] Figure 4 The present application discloses a dynamic compensation curve for adjusting the front-end dosage based on changes in reflux load in the embodiments of this application. Detailed Implementation

[0025] The following is in conjunction with the appendix Figure 1 - Figure 4 A preferred description of the present invention is provided below.

[0026] See attached document Figure 1 - Figure 2 This invention proposes an adaptive mud recirculation control system based on foam state, comprising the following modules: The pre-defoaming mud generation module obtains the original foamed mud, performs simultaneous surface and bottom layer chemical addition operations, and generates a pre-defoaming mud flow. The physical state feature extraction module guides the pre-defoamed mud flow into the expansion and deceleration zone, controls the image acquisition device and the ultrasonic device to acquire data synchronously, and extracts the physical feature data of the mud. The foam potential index modeling module calls the mud foaming risk quantification mapping function to parse the physical characteristic data of the mud, and performs parametric integration along the vertical depth coordinate line in combination with the preset weight product coefficient to calculate the depth foaming risk assessment model. The dynamic threshold filtering module iterates through the risk classification data contained in the deep foaming risk assessment model, compares it with the bubble concentration tolerance threshold, and filters out the stratified control dynamic threshold parameters. The layered cutting and stripping module sends the layered control dynamic threshold parameters to the dynamic interception mechanism to perform layered physical interception on the pre-defoaming mud flow, cutting off the upper layer return mud flow and the bottom layer output mud flow. The feedforward drug delivery control module monitors and records the real-time return flow load signal of the upper reflux mud, initiates the advance compensation drug delivery strategy, and generates an antifoaming control signal. The feedback correction and adjustment module guides the bottom output mud flow into the sedimentation and granulation area, adds polymer flocculant to cause solids to settle, and adjusts the preset weight product coefficient in the parameter integration calculation based on the transmittance of the supernatant.

[0027] In one embodiment of the present invention, the pre-defoaming mud generation module is used to perform the following steps: The original foamy mud is guided into the mixing and defoaming chamber; Activate the atomizing spray valve array installed at the top of the mixing defoaming chamber to spray chemical defoamer onto the exposed bubble surface area of ​​the original foamy slurry; Turn on the circulation pump located at the bottom of the mixing and defoaming chamber, extract the bottom slurry carrying unbroken microbubbles, and inject chemical defoamer into the main pipeline and apply cutting and shearing force; The bottom mud, after being treated twice, is transported back to the mixing and defoaming chamber to be mixed and generate the pre-defoaming mud flow.

[0028] The original foamed mud discharged from the tunnel boring machine (TBM) connection area is obtained. During earth pressure balance TBM construction, foaming agents are injected into the raw mud to improve the properties of the excavated soil. After excavation and screw conveying processes, the original foamed mud is a liquid-solid mixture containing soil particles, moisture, additive residues, and stabilizing foam discharged from the sand washing or screening process. A dual-path dosing operation is performed at the top and bottom to generate a pre-defoamed mud flow that has been physically and chemically intercepted.

[0029] The original foamy slurry, continuously discharged during tunnel boring machine (TBM) operations and rich in foam, is stably transported through pipelines and injected into a pre-designed mixing and defoaming chamber. The mixing and defoaming chamber is a cylindrical or square steel tank with a volume determined according to the processing flow rate. The mixing and defoaming chamber is equipped with a variable frequency driven mechanical stirring structure, such as a multi-layer variable diameter impeller stirrer whose speed can be continuously adjusted from 30 r / min to 120 r / min. The reason for setting this speed range is that when the speed is below 30 r / min, it is not possible to effectively break the macroscopic bubble skeleton inside the original foamy slurry and to evenly disperse the chemical defoamer. When the speed is above 120 r / min, excessive shearing will cause excessive conversion of mechanical energy into heat energy, and may even cause secondary cavitation and foaming of the original foamy slurry.

[0030] The control system activates the atomizing spray valve array installed at the top of the mixing and defoaming chamber. A water-based diluted chemical defoamer, typically a blend of mineral oil-based and silicone-based defoamers, is uniformly sprayed from the top onto the exposed bubble surface of the original foamy slurry where it comes into contact with air. This process eliminates macroscopic surface foam. The reason for setting this mass fraction range is that when the mass fraction is below 0.2%, the surface tension of the bubbles on the liquid surface cannot be effectively destroyed, resulting in a slow defoaming rate; when the mass fraction is above 1.0%, the defoaming effect shows diminishing marginal returns and can easily lead to excessive chemical residues in the slurry, severely affecting the effluent quality of subsequent flocculation processes.

[0031] Start the circulation pump located at the bottom of the mixing and defoaming chamber, for example, select a shear rate of not less than 1000 s. -1 The high-shear pump is designed with a shear rate threshold that ensures the high-concentration emulsion is instantly torn into micron-sized droplets, rapidly penetrating and destroying the polar liquid film of the microbubbles. Below this threshold, emulsification and dispersion will be uneven. An external circulation pump draws highly fluid bottom mud carrying unbroken microbubbles from the bottom of the mixing and defoaming chamber. A metering pump injects a high-concentration chemical defoamer into the main circulation pipeline before the external circulation pump inlet. The typical mass fraction of the high-concentration emulsion is 5% to 15%. This typical concentration is set because the bottom mud has high viscosity and strong microbubble stability, requiring a higher concentration of agent to ensure the osmotic pressure needed for local bubble rupture. A concentration below 5% is insufficient to break through the microbubble barrier, while a concentration above 15% can easily cause local agglomeration of the agent, leading to waste. The high-speed rotation of the impeller inside the external circulation pump applies a cutting and shearing force to the dosed mud.

[0032] The bottom layer mud, which has undergone secondary chemical dosing and endured physical shear forces, is transported back to the upper middle part of the mixing and defoaming chamber at a predetermined angle and high flow rate. The kinetic energy of the bottom layer mud impacts and mixes into the overall mud within the mixing and defoaming chamber, breaking down the weak bubble skeleton deeply embedded within the original foamy mud. After dual intervention of surface chemical suppression and internal physical shearing, it is continuously discharged from the overflow port of the mixing and defoaming chamber, forming a pre-defoamed mud flow that has been cut off by physical and chemical intervention. At this point, macroscopic foam is eliminated, the internal microbubble structure is destroyed, and the overall gas content is reduced, creating conditions for subsequent identification and separation. The simultaneous surface and bottom layer chemical dosing operation includes two dosing paths: surface atomized spraying and bottom layer pipeline injection; through combined physical and chemical intervention, the stability of the foam liquid film is disrupted.

[0033] In a specific embodiment, assuming a tunnel boring machine (TBM) operation site, the continuous output flow rate is 50 m³ / s. 3 The original foamed mud, with a solids content of 40% and an air content of 30%, was introduced into a 12 m³ / h container. 3 Inside the mixing and defoaming chamber, a three-layer impeller agitator operates at a speed of 80 r / min. The control system then activates the atomizing spray valve array to atomize and spray a 0.5% (by mass) silicone composite defoamer emulsion at a pressure of 0.4 MPa and a flow rate of 10 L / h.

[0034] Simultaneously, the circulation pump installed at the bottom of the mixing and defoaming chamber is activated, and the circulation flow rate of the external circulation pump is set to 15 m³ / s. 3 At a flow rate of 3 L / h, a metering pump injects a 10% mineral oil-based defoamer into the external circulation pump. The mixture of mud and defoamer, after being processed by the external circulation pump, is transported back to the upper part of the mixing and defoaming chamber tangentially along the side wall. After approximately 10 minutes of stable operation, sampling and testing of the pre-defoamed mud stream discharged from the mixing and defoaming chamber showed that the air content of the pre-defoamed mud stream decreased to the range of 4% to 6%, and no visually visible stable foam layer appeared on the surface. The physical and chemical intervention cutoff operation achieved the expected defoaming and degassing effects.

[0035] In one embodiment of the present invention, the physical state feature extraction module is configured to perform the following steps: The image acquisition device, positioned at the top of the expansion and deceleration zone, continuously captures surface foam images of the liquid surface. Based on the inter-frame pixel differences between consecutive images, the dissipation rate factor that quantifies the dynamic balance between bubble generation and rupture is extracted. Simultaneously send detection commands to the vertical ultrasonic transceiver array that is attached to the sidewall of the expansion and deceleration area, transmit acoustic pulse sequences and receive reflected echoes that penetrate the lower mud layer, and extract the energy attenuation rate factor that reflects the acoustic impedance characteristics at different immersion depths. The feature data of the surface foam image, the dissipation rate factor, and the energy decay rate factor are extracted, and then time-stamped and encapsulated to construct the physical feature data of the mud.

[0036] The pre-defoaming mud flow is guided into the expansion and deceleration zone. Simultaneous acquisition of images and ultrasonic equipment is controlled to extract physical characteristic data of the mud, mapping the transient foaming activity level. The pre-defoaming mud flow from the mixing and defoaming chamber is introduced through a guide plate into an expansion buffer separation chamber with an enlarged cross-sectional area. This is typically a horizontally positioned cuboid channel. The width-to-depth ratio of the expansion buffer separation chamber is set according to process requirements; for example, the width is 5 to 10 times the diameter of the inlet pipe. This ratio ensures that the cross-sectional area rapidly expands after the pre-defoaming mud flow enters the expansion buffer separation chamber, thereby drastically reducing the horizontal flow velocity to laminar flow. A ratio below 5 times will result in insufficient velocity reduction, leaving localized turbulence within the expansion buffer separation chamber. A ratio above 10 times will lead to an excessively large expansion buffer separation chamber volume, wasting equipment space and causing mud deposition in stagnant water areas.

[0037] The geometric design of the expanded buffer separation chamber forces the horizontal flow velocity of the pre-defoamed mud flow to be lower than the inlet velocity value in the turbulent or transitional flow region. Reduce to the horizontal velocity value corresponding to laminar flow. That is, the Reynolds number of the fluid. Below the critical value of 2000, the Reynolds number is calculated using mud density, flow velocity, hydraulic equivalent diameter, and dynamic viscosity. The rationale for setting this critical Reynolds number is that, in fluid mechanics, a Reynolds number below 2000 represents the classic laminar flow boundary. In this laminar flow state, there is no macroscopic mixing between flow layers, effectively eliminating the binding of internal bubbles by turbulent kinetic energy, allowing tiny bubbles to float stably solely on their own buoyancy. At this point, fluid particles flow smoothly and regularly in a linear fashion parallel to the wall, with no macroscopic mixing between flow layers. The formulas for calculating the inlet and outlet velocities are:

[0038]

[0039] in, This is the inlet flow velocity value, in m / s; This represents the horizontal velocity value corresponding to laminar flow, in m / s. The volumetric flow rate of the pre-defoamed mud flow is expressed in cubic meters per second (m³). 3 / h; The cross-sectional area of ​​the inlet pipe, in meters. 2 ; The increased cross-sectional area of ​​the buffer separation chamber is measured in m². 2 .

[0040] The reduction in horizontal flow velocity weakens the turbulent shearing and mixing effects of the fluid, creating stable conditions for the manifestation of gravity settling and hydrostatic equilibrium effects. Under these conditions, the remaining microbubbles, previously bound by turbulent kinetic energy within the mud matrix, break free due to their own buoyancy and continue to float upwards. They collide and merge near the liquid surface, forming a larger, more easily identifiable enriched foam layer. The control system simultaneously triggers sensor arrays of two different modes.

[0041] One is an industrial camera suspended at the center of the top wall of the expanded buffer separation chamber. It is usually an industrial area scan camera with a megapixel level or higher. The frame rate is set in the range of 10 to 30 frames per second. The reason for setting this frame rate range is that if the frame rate is lower than 10 frames per second, it is impossible to accurately capture the transient dynamic changes of bubble rupture and generation on the surface of the mud. If the frame rate is higher than 30 frames per second, it will result in too much image data, which will increase the computing load of the processor and have limited improvement on the accuracy of subsequent dissipation rate factor calculation. The industrial camera continuously shoots the liquid surface at the preset frame rate to generate a set of time-series images. The processor analyzes the time-series images.

[0042] On the one hand, surface foam images that can characterize the foam coverage area and morphology of a single frame are extracted. In actual processing, surface foam images are usually further processed into numerical features such as foam coverage percentage, maximum bubble diameter, and bubble diameter distribution histogram. On the other hand, through the difference operation of consecutive image frames in the time series image, the dissipation rate factor that quantifies the dynamic balance between bubble generation and rupture is calculated. The formula for calculating the dissipation rate factor is as follows:

[0043] in, It is the dissipation rate factor; The coordinates of the current frame The normalized grayscale value of the pixel at that location; This is the normalized grayscale value of the pixel corresponding to the coordinates in the previous frame; This represents the total number of pixels in the image.

[0044] Secondly, there is a vertical ultrasonic transceiver array closely fitted to the vertical wall of one side of the expansion buffer separation chamber. It consists of multiple piezoelectric ceramic transducers with an operating frequency between 1 MHz and 5 MHz. The vertical ultrasonic transceiver array is composed of multiple sensor units located at different immersion depths. After receiving the synchronous trigger command, each sensor unit simultaneously transmits a sequence of acoustic pulses at a preset frequency to the opposite side wall of the expansion buffer separation chamber and receives the reflected or transmitted echo signals after penetrating the lower layer of mud.

[0045] The processor calculates the energy attenuation rate factor for each depth level based on the amplitude difference between the transmitted and received signals. This energy attenuation rate factor reflects the acoustic impedance characteristics of the mud medium at different immersion depths, and these acoustic impedance characteristics are related to the enrichment density of microbubbles. The formula for calculating the energy attenuation rate factor is as follows:

[0046] in, In depth Energy decay rate factor at a given location, in dB / m; The effective path length of ultrasonic waves propagating in mud, in meters, is usually equal to the width of the expanded buffer separation chamber. The reference amplitude, in units of V, represents the length of the same path that an ultrasonic wave travels in a bubble-free reference liquid. The amplitude of the signal received later; In depth The actual measured amplitude of the echo signal, in V.

[0047] The extracted surface foam image feature data, dissipation rate factor, and multiple energy decay rate factors representing depth profiles are combined with uniform acquisition timestamp alignment and structured encapsulation to construct mud physical feature data that can dynamically map the transient foaming activity level of the mud. This data package object contains timestamps, foam coverage, dissipation rate factors, and an array of energy decay rate factors corresponding to the ultrasonic probe depth. The mud physical feature data provides the physical characteristics of the mud surface and interior, offering data support for subsequent foaming risk quantification.

[0048] In a specific embodiment, the volumetric flow rate is... 50 m 3 / h of pre-defoamed mud flow is introduced into a width An expanded buffer separation chamber with a diameter of 1.5m and an effective water depth of 0.8m is included. The mud inlet pipe has a diameter of 0.15m and a cross-sectional area of... It is 0.01767 m 2 Substituting into the formula, the inlet velocity is calculated. It is approximately 0.785 m / s.

[0049] After entering the expanded buffer separation chamber, the cross-sectional area of ​​the water passage is increased. Expanded to 1.2 m 2 Substituting into the formula, the horizontal flow velocity of the mud is calculated. The speed was reduced to approximately 0.0116 m / s, and the mud density under this condition was 1100 kg / m³. 3 Parameters such as dynamic viscosity of 0.025 Pa·s indicate a Reynolds number of approximately 519, below 2000, confirming laminar flow. At this point, an industrial area-array camera with a resolution of 1920x1080 pixels captures images of the liquid surface at 15 frames per second. The image analysis module calculates the current foam surface coverage to be 15%, and based on the differential calculation formula of consecutive image frames, obtains the dissipation rate factor. It is 0.35.

[0050] Simultaneously, a vertical ultrasonic transceiver array comprising four transducers was triggered, with the transducers located at depths of 0.2 m, 0.4 m, 0.6 m, and 0.8 m below the water surface, respectively. The calibrated reference amplitude... The voltage is 5 V. In actual measurements, the signal amplitude received at each depth... The values ​​are 0.5 V, 1.5 V, 3.0 V, and 4.0 V, respectively. According to the formula... The energy decay rate factors at each depth were calculated as follows: ; ; ; At the timestamp 2023-10-27T10:30:05.100Z, the above data was encapsulated into mud physical characteristic data. The content of the mud physical characteristic data is {timestamp: 2023-10-27T10:30:05.100Z, surface foam coverage: 0.15, dissipation rate factor: 0.35, energy decay rate profile: [13.33, 6.97, 2.96, 1.29]}. The mud physical characteristic data depicts the surface and internal foaming risk status of the mud at that moment.

[0051] In one embodiment of the present invention, the bubble potential index modeling module is used to perform the following steps: Extract the feature data and dissipation rate factor of the surface foam image encapsulated within the physical feature data of the mud, and input them into the surface activity assessment formula to perform numerical calculations to generate an apparent risk score. The energy decay rate factor at each depth carried by the physical characteristic data of the mud is matched and extracted, and then converted into the internal enrichment density reference value at the corresponding liquid surface depth using an acoustic empirical inversion algorithm. The preset weighted product coefficients are configured for the apparent risk score and the internal enrichment density reference value, and a multiplication-addition operation is performed to obtain the fusion risk coefficient. By integrating the fusion risk coefficients at various depths and establishing a continuous analytical model for the target depth, the bubble risk assessment model for that depth is output.

[0052] The mud foaming risk quantification mapping function is invoked to parse the mud physical characteristic data. Parametric integration is performed along the vertical depth coordinate line using preset weighted product coefficients to calculate a depth foaming risk assessment model characterizing the potential regeneration foaming capacity of each liquid surface depth level. The central processing unit extracts feature data related to the surface foam image of the liquid surface state from the mud physical characteristic data constructed in the previous step; specifically, this includes foam coverage and dissipation rate factor. The foam coverage and dissipation rate factor are then substituted into a preset surface activity assessment formula for weighted calculation. The surface activity assessment formula is as follows:

[0053] in, This represents the apparent risk score; a higher apparent risk score indicates a more severe problem with surface foaming. The foam coverage rate ranges from 0 to 1. It is the dissipation rate factor; and Empirical weighting coefficients for surface features are used to balance the importance of coverage and dynamic stability in the evaluation, such as based on historical data statistics. Typical values ​​range from 0.5 to 0.8. Typical values ​​range from 0.2 to 0.5. The basis for setting the above empirical weighting coefficient range is that, since the macroscopic foam coverage rate is the most intuitive and important indicator for assessing the foaming state, it is given a high weight of 0.5 to 0.8; while the dissipation rate factor, as a dynamic auxiliary judgment indicator, is given a relatively low weight of 0.2 to 0.5, so as to ensure that the assessment results can reflect both the static status quo and dynamic changes. This generates a dimensionless scalar that comprehensively reflects the macroscopic performance and dynamic stability of the liquid surface foam, namely the apparent risk score.

[0054] Simultaneously, the processor matches and extracts an energy decay rate factor profile array reflecting depth information from the mud physical characteristic data, obtained from measurements using a vertical ultrasonic transceiver array. It then invokes an acoustic empirical inversion algorithm, based on laboratory calibration, which reflects the mapping relationship between the energy decay rate factor and the bubble volume concentration. Each energy decay rate factor in the energy decay rate factor profile array is converted, one by one, into a reference value for the internal enrichment density of microbubbles embedded within the mud at the corresponding depth. The core conversion formula is:

[0055] in, For depth The internal enrichment density reference value is expressed as an equivalent volume percentage. A higher internal enrichment density reference value indicates greater depth. The higher the concentration of bubbles at a given location; For depth Energy decay rate factor at the location; The acoustic inversion coefficient is typically set to 0.1% / (dB / m) through experimental calibration. The basis for setting this typical value is that the inversion coefficient is based on the empirical fitting value of linear regression between acoustic attenuation and bubble volume ratio in high solids content mud media. This value can be used to map acoustic attenuation to bubble physical concentration.

[0056] This generates an array of internal enrichment density reference values. The previously calculated apparent risk score and the array of internal enrichment density reference values ​​are multiplied by their respective independently configured preset weighted product coefficients, and a multiplication-addition operation is performed for each depth level. The fusion calculation formula is as follows:

[0057] in, For depth The risk factor of integration at the location; and These are the preset weighted product coefficients assigned to the apparent risk score and the internal enrichment density reference value, respectively. The preset weighted product coefficients are adaptively adjusted based on the priority considerations of the current operating condition; for example, when emphasizing internal risk, they can be set to... and The basis for setting the relative magnitude of this preset weighted product coefficient is that when the mud enters the deep treatment stage, the large surface foam has been basically eliminated, and the potential hazards mainly come from the tiny air bubbles embedded inside. Therefore, the weighting coefficient of the internal enrichment density reference value is set accordingly. Set as the weighting coefficient for apparent risk score Two to three times that of internal microbubbles to amplify their contribution to total risk.

[0058] A set of fused risk coefficients, integrating surface and internal risk information, is obtained and distributed along the vertical coordinate. Using these discrete fused risk coefficients, correlated with their corresponding depths, as input nodes, a continuous analytical model for the target depth is established based on the least squares polynomial fitting principle. This outputs a continuous function curve describing the potential regenerative foaming capacity at each liquid depth level—the depth-based foaming risk assessment model. .

[0059] The mud foaming risk quantification mapping function is a collection of formulas and algorithms used in this step, including surface assessment, acoustic inversion, and fusion calculation. The output depth foaming risk assessment model reflects the changing trend of regeneration foaming risk with depth.

[0060] In a specific embodiment, the processor retrieves the mud physical characteristic data with a timestamp of 2023-10-27T10:30:05.100Z to extract the surface foam coverage rate. The dissipation rate factor is 0.15. The weighting coefficient is 0.35. It is 0.6. The value is 0.4. Substitute this value into the formula to calculate the apparent risk score. Simultaneously, the processor extracted an energy attenuation rate profile of [13.33, 6.97, 2.96, 1.29] dB / m. This was achieved using acoustic inversion coefficients. % / (dB / m), the calculated internal enrichment density reference value profile is as follows %, %, %, %.

[0061] Use the pre-set weighted product coefficients and Calculate the fusion risk coefficient at each depth. ; ; ; The aforementioned four fused risk coefficient data points were input into the numerical fitting module. Through least squares quadratic polynomial fitting, the final deep bubbling risk assessment model was generated. The function expression of the deep bubbling risk assessment model is as follows: The depth foaming risk assessment model depicts the continuous trend of decreasing potential foaming risk of mud as depth increases from 0.2 m to 0.8 m below the liquid surface.

[0062] In one embodiment of the present invention, the dynamic threshold filtering module is configured to perform the following steps: The connection retrieves the depth bubbling risk assessment model, which exists in the form of a continuous function. Call the storage unit to read the bubble concentration tolerance threshold that represents the allowable entry into the back-end flocculation process under the current operating conditions; A risk benchmark equation is constructed, and the corresponding depth solution is obtained by solving the risk benchmark equation within the effective water depth range through a continuous solution algorithm model. Lock the specific liquid surface depth value corresponding to the depth solution, and format and encode the specific liquid surface depth value to generate the layered control dynamic threshold parameter.

[0063] The risk classification data carried by the deep foaming risk assessment model is traversed and read, compared with the bubble concentration tolerance threshold of the safety line, and the dynamic threshold parameters for layered control used to cut off unqualified isolation zones are selected. The basis for calculating the dynamic threshold parameters is that physical cutting is based on an accurate quantitative model. By finding the roots of the continuous function equation, the abstract risk index can be transformed into the boundary of physical depth, ensuring that high-risk mud does not enter the next stage. The control microprocessor connects to and retrieves the deep foaming risk assessment model generated by the previous steps. The deep foaming risk assessment model uses a continuous function... It exists in the form of.

[0064] Based on the currently set operating mode, the processor sends a read command to its internal non-volatile memory unit, namely the safety redundancy register, a high-reliability memory with power-loss protection to ensure that critical settings are not lost after an unexpected power outage. The processor retrieves and stores a pre-calibrated bubble concentration tolerance threshold that represents the allowable bubble concentration to enter the downstream flocculation process under the current operating conditions. The bubble concentration tolerance threshold is experimentally determined based on the downstream process requirements for the quality of the influent slurry, with a typical value range of 1.5 to 3.0. The basis for setting this reference constant range is that this reference constant represents the maximum tolerance of the subsequent flocculation process for microbubbles in the slurry. A value below 1.5 will cause frequent backflow actions at the front end, severely reducing the overall processing capacity of the system; a value above 3.0 will cause too much slurry carrying microbubbles to enter the terminal flocculation stage, destroying the polymer cross-linking structure and causing solid-liquid separation failure.

[0065] A risk baseline equation is established, and the risk baseline equation is solved within the effective water depth range using a continuous solution algorithm model to obtain the cutoff depth where the risk curve intersects the threshold line. The key analytical logic involved in this step can be expressed by the following equation and solution conditions:

[0066] in, To calculate the specific liquid surface depth value, in meters, the specific liquid surface depth value must meet the requirement of a real root of the effective physical depth range; For the deep bubbling risk assessment model, it is a continuous function; The bubble concentration tolerance threshold is read from the safety redundancy register.

[0067] Once the system solves and locks in the depth solution of the risk baseline equation, it records the specific liquid surface depth value corresponding to the depth solution and formats and encodes this value, for example, as a 32-bit floating-point signal representing underwater depth. This generates and outputs dynamic threshold parameters for layered control, used to drive subsequent physical actuators and to separate unqualified mud from qualified mud. By continuously solving the risk baseline equation, the system transforms the continuous function reflecting foaming risk into a physical boundary indicating the execution depth, completing the transition to physical intervention commands.

[0068] In a specific embodiment, continuing from the previous embodiments, the processor retrieves the generated deep bubbling risk assessment model. The processor reads the bubble concentration tolerance threshold for the current flocculant type from the safety redundancy register. The value is set to 2.5. The system constructs and solves equations based on the risk baseline equation. ,Right now Simplify to The real-valued solution at the shallow cutoff location is obtained through calculation using the formula. m.

[0069] The calculation results of the real number solution at the shallow cutoff position were locked, and it was determined that 0.533 m below the liquid surface is the dividing line between high-risk mud and low-risk qualified mud. Therefore, the system encodes the specific liquid surface depth value of 0.533 m into a digital instruction packet, which serves as a dynamic threshold parameter for layered control and is sent to the actuator in the next step to guide the physical cutting and stripping operation.

[0070] See attached document Figure 3 The solid line in the figure represents the parabolic trend of the depth foaming risk assessment model function. It can be observed that the foaming risk decreases with increasing water depth. The horizontal dashed line represents the preset bubble concentration tolerance threshold of 2.5. The vertical dotted line represents the intersection of the solid line and the horizontal dashed line. This intersection point corresponds to a specific liquid surface depth of 0.533 m on the horizontal axis. This depth is the final selection output for the stratified control dynamic threshold parameter used to implement physical interception. The figure above illustrates the process of transforming the abstract algorithm model into a physical control boundary solution.

[0071] In one embodiment of the present invention, the layered cutting and peeling module is used to perform the following steps: The dynamic threshold parameters of the layered control, which include specific liquid level depth values, are analyzed and converted into displacement control signals to drive the servo motor. The guide tube head mounted on the longitudinal slide rail is driven to move up and down until the guide tube head moves to the underwater control surface indicated by the displacement control signal; Turn on the suction negative pressure pump group that is connected to the outside of the guide pipe head to suck up and strip the upper layer of return mud located at or above the control surface; The control surface physically seals the outflow channel of the lower layer mud, and under the action of gravity static pressure, the bottom layer output mud flow located deep at the bottom end of the control surface is drained out.

[0072] The dynamic threshold parameters of the layered control are sent to the dynamic interception mechanism, which is mainly an electromechanical system consisting of longitudinal slide rails, servo motors and guide pipe heads. The interception height of the dynamic interception mechanism can be adjusted in real time. Layered physical interception is implemented for the pre-defoaming mud flow, cutting off the upper return mud flow in the high-risk area on the surface and the bottom output mud flow in the safe area at the bottom.

[0073] The motion controller within the system receives and analyzes the layered control dynamic threshold parameters containing specific liquid level depth values ​​issued by the microprocessor. The conversion module inside the motion controller translates the specific liquid level depth data, such as a depth value in meters, into a sequence of target displacement level commands for driving the servo motor, based on a preset electromechanical calibration relationship. This sequence could be represented as a set of drive pulses or a corresponding voltage value. The electromechanical calibration conversion relationship can be expressed as:

[0074] in, The control command value output to the servo driver, in units such as V or pulse count; The electromechanical calibration proportional coefficient of the system, such as V / m, is used to correlate physical displacement with control signals; The specific liquid level depth value obtained from the analysis, in meters (m); The zero-point offset of the control signal is used to calibrate the deviation between the physical zero point and the electrical zero point.

[0075] The target displacement level command sequence is sent to the servo driver, which drives the guide tube head, which is suspended on the longitudinal slide rail on one side of the expansion buffer separation chamber, to move vertically up and down until the upper edge of the guide tube head's suction inlet is aligned with the control surface located below the water surface, as indicated by the target displacement level command sequence. The control surface is used to distinguish the virtual horizontal plane that separates high-risk and qualified mud.

[0076] Once the position alignment is confirmed by encoder feedback, the control system activates the suction negative pressure pump unit connected to the external pipeline of the guide pipe head. Typically, a wear-resistant positive displacement pump, such as a diaphragm pump or screw pump, is selected to handle slurries with high solids content. The suction generated by the suction negative pressure pump unit draws in and strips away slurry components rich in secondary foaming potential located at or above the control surface. The suctioned slurry is defined as the upper return slurry. During this process, the system only selectively sends this high-risk upper slurry back to the front end for reprocessing, thus achieving targeted return control. The basis for defining and suctioning the upper return slurry is that the liquid layer above the control surface is defined in the algorithm model as a high-risk area where the foaming risk index exceeds the bubble concentration tolerance threshold. Only by forcibly and physically stripping the slurry in this area and sending it back to the front end for reprocessing can the impact of residual foaming agent on downstream processes be effectively avoided.

[0077] As the high-risk upper layer of mud is removed, an invisible liquid-liquid boundary is formed on the control surface, which physically seals off the outflow channel of the lower layer of mud. Under the gravity-driven hydrostatic pressure effect of the liquid level difference between the inside and outside of the expanded buffer separation chamber, the mud located deep at the bottom of the control surface and already in the defoaming safety zone is drained out through a pre-set drain pipe, forming the bottom layer output mud flow that can enter the subsequent process, i.e. qualified lower layer mud with a risk index below the threshold.

[0078] In a specific embodiment, continuing from the previous embodiment, the motion controller receives a value containing the specific liquid level depth. The dynamic threshold parameter for layered control is 0.533 m. The total stroke of the longitudinal slide rail of the dynamic interception mechanism is 1.0 m. The servo system uses analog voltage control from 0 V to 10 V, with 0 V corresponding to the top position and 10 V corresponding to the bottom position. Therefore, the electromechanical calibration proportional coefficient is... 10 V / m, zero offset It is 0 V.

[0079] The motion controller calculates the control command value accordingly. The output 5.33 V voltage control command signal is applied to the servo driver, driving the servo motor to lower the guide tube head to a position 0.533 m below the liquid surface. After the position sensor confirms the position, a 3 kW diaphragm pump is started as a suction negative pressure pump group, beginning to pump the mud located in the depth range of 0 m to 0.533 m. The upper return mud, with a flow rate of approximately one-third of the total flow rate, is guided to the inlet of the mixing and defoaming chamber.

[0080] At the same time, the drain pipe located at a depth of 0.9 m at the bottom of the expansion buffer separation chamber begins to stably discharge the bottom output mud flow located below a depth of 0.533 m under the action of static pressure. The discharge bottom output mud flow is about two-thirds of the total flow and is transported to the subsequent flocculation unit.

[0081] In one embodiment of the present invention, the feedforward drug delivery control module is configured to perform the following steps: The instantaneous volumetric flow rate signal fed back by the electromagnetic flow meter deployed on the circulating return pipeline is scanned, and the real-time flow quality value of the upper layer return slurry is calculated and extracted within a continuous statistical period. The extracted upper layer of refluxed mud is reinjected into the mixing and defoaming chamber; The deviation and rate of change between the real-time flow quality value and the median of the preset standard return expected balance interval are calculated, and matched with the preset nonlinear agent consumption compensation matrix table. The compensation difference is obtained by analyzing the table lookup matching results, generating a pulse waveform packet containing the corrected duty cycle, and sending it to the actuator as the defoaming control signal carrying the compensation difference of the agent valve control.

[0082] The system monitors and records the real-time return flow load signal of the upper layer of reflux slurry, initiates an advance compensation dosing strategy based on a front-end intervention closed-loop correction mechanism, and generates a defoaming control signal to compensate for the difference in the control of the chemical valve. The system continuously scans and collects the instantaneous volumetric flow rate signal from electromagnetic flow meters on the circulating return pipelines deployed in the upper layer of reflux slurry, such as the pipeline connecting the upper outlet of the expansion buffer separation chamber to the inlet of the mixing and defoaming chamber. After processing by the processor within a preset continuous statistical period, the system extracts the real-time flow quality value of the upper layer of reflux slurry per unit time. The integral formula for the real-time flow quality value is expressed as:

[0083] in, For a moment Real-time flow quality values, in t / h; This is a preset reference value for mud density, in t / m³. 3 ; The instantaneous volumetric flow rate fed back by the flow meter in the recirculation pipeline, in cubic meters (m³). 3 / h; For integration time; The set continuous statistical period, in hours (h).

[0084] Simultaneously, the metered upper-layer recirculated mud flow is redirected through the recirculation pipe and injected into the inlet of the mixing and defoaming chamber, merging with the newly entering original foamed mud to participate in a new round of mixing and defoaming intervention, thus forming a physically closed-loop recirculation path. The processor continuously obtains real-time flow quality values ​​and compares them in real-time with the median of the standard recirculation expected balance interval (which represents the system's target operating state and has upper and lower limits, aiming to ensure that the recirculation effectively treats the mud without causing excessive impact on the front end), performing deviation and rate of change calculations. The formulas for deviation and rate of change are as follows:

[0085]

[0086] in, For a moment The absolute deviation of the flow rate; This represents the absolute deviation of the flow rate at the previous discrete sampling time. The midpoint of the expected equilibrium interval for standard backflow; This is the rate of change of the deviation (i.e., the first derivative). This represents the discrete sampling time interval.

[0087] The system uses the calculated absolute deviation value of the flow rate and the deviation change rate as dual indexes to match and query a preset nonlinear reagent consumption compensation matrix table in memory. The two-dimensional lookup table is generated by experience or simulation, with the row index being the deviation level and the column index being the change rate level. The values ​​in the table are the corresponding reagent compensation percentages.

[0088] The processor analyzes the lookup and matching results of the nonlinear agent consumption compensation matrix table, i.e., the compensation difference, and converts the compensation difference into specific control parameters. Specifically, the agent compensation percentage is directly applied to the agent valve opening control. At the same time, according to a preset coupling reduction ratio, such as 1:0.5, the speed and frequency gain of the external circulation pump are synchronously calculated and generated, and a pulse waveform packet containing the corrected duty cycle is generated. This pulse waveform packet containing the corrected duty cycle is used as the defoaming control signal. The compensation coefficient is converted into a composite instruction set of specific actuators such as valve opening percentage and motor frequency, and sent to the solenoid valve driver that controls the opening of the atomizing spray valve at the top of the mixing defoaming chamber, and the frequency converter that controls the speed of the external circulation pump, respectively. This completes the feedforward closed-loop correction of the front-end defoaming intensity.

[0089] Among them, through the feedforward closed-loop monitoring mechanism of flow load, the system can detect and actively adjust the front-end processing load in advance when the back end faces fluctuations in mud properties, thus maintaining the dynamic balance of the entire pipeline network operation.

[0090] In a specific embodiment, the system monitors the flow rate of the upper recirculated slurry. The standard expected balance range for the upper recirculated slurry is set to 8.5 t / h to 10.5 t / h, and the median of the standard expected balance range is... It is 9.5 t / h. At a certain moment... The volumetric flow rate is measured by an electromagnetic flow meter deployed on the circulating return pipeline, combined with the set mud density reference value. 1.1 t / m 3 The processor calculates the real-time flow quality value using a formula. It is 12.1 t / h.

[0091] The system calculates the absolute deviation of the flow rate. t / h represents a positive deviation. Simultaneously, the rate of change of deviation is calculated by combining the deviation data from the previous sampling period. A positive value indicates that the return load is still increasing. The system uses the absolute flow deviation value "+2.6t / h" and the deviation change rate of "positive" as dual indexes to query the nonlinear reagent consumption compensation matrix table, and the table match yields a compensation difference value of "+2.6t / h".

[0092] The processor generates a defoaming control signal. Assuming the basic flow rate of the atomizing spray valve is 10 L / h and the basic frequency of the external circulation pump is 40 Hz, after compensation calculation, the new control command is to increase the flow rate of the atomizing spray valve by 20% to 12 L / h; and increase the operating frequency of the external circulation pump by 10% to 44 Hz, thereby adjusting the defoaming capacity at the front end to cope with the increased backflow load.

[0093] See attached document Figure 4 The solid black line in the figure represents the dynamic curve of real-time flow quality values. Its starting point coincides with the bottom black horizontal dashed line, which represents the median of the standard return flow expected balance range of 9.5 t / h. It then rises to a maximum of 12.1 t / h before declining. The gray dotted line represents the flow rate action curve of the atomizing spray valves calculated based on the deviation. It follows the trend of the solid line, synchronously and adaptively increasing from a basic flow rate of 10.0 L / h to a maximum of 12.0 L / h. These curves verify that the feedforward compensation mechanism can actively sense the deviation of the system's return flow load and implement nonlinear dynamic correction and adjustment of the dosage to ensure stable system operation.

[0094] In one embodiment of the present invention, the feedback correction adjustment module is configured to perform the following steps: The bottom-layer output mud flow is fed into the end flocculation and forming chamber equipped with blades; Start the variable frequency motor configured at the top of the end flocculation forming chamber to drip flocculant into the bottom layer output mud flow which is in a stirring state; The stirring action generated by the variable frequency motor driving the blades accelerates the solid phase sedimentation and separation from the liquid phase, and the supernatant liquid at the top overflows and precipitates out. The optical detection device is activated to continuously scan the transmittance of the supernatant. When the real-time value of the transmittance is less than the allowable threshold, an optimization command is sent in reverse to adaptively increase the preset weight product coefficient.

[0095] The process involves guiding the bottom-layer output mud flow into a non-turbulent sedimentation and granulation zone, adding a polymeric flocculant to cause solids to settle, and adjusting a preset weighted product coefficient based on the transmittance of the supernatant. The bottom-layer output mud flow, having undergone risk screening and physical isolation through the aforementioned steps, is then smoothly connected to an end-flocculation chamber equipped with multiple layers of wide blades.

[0096] The control system activates the variable frequency motor located at the top of the end flocculation chamber, causing the blades to agitate the slurry at a preset speed, such as 5 to 15 r / min. This low speed range is chosen because it provides sufficient micro-hybrid power to facilitate collisions between the chemicals and slurry particles, while avoiding strong macro-shear forces that could break up the already formed fragile flocculent network. Speeds below 5 r / min would result in uneven chemical diffusion, while speeds above 15 r / min would cause uneven chemical diffusion. The r / min speed causes the flocs to break up and, along the radial position of the end flocculation chamber, outputs the diluted long-chain polymeric flocculant into the mud flow through multiple burettes into the stirred bottom layer. The diluted flocculant is typically an anionic polyacrylamide aqueous solution with a mass fraction of approximately 0.05% to 0.2%. This mass fraction range is set because this concentration of anionic polyacrylamide aqueous solution has a moderate viscosity, which can ensure that the polymer chains are fully extended and bridged in the mud. If the concentration is too low, the bridging ability will be insufficient and the sedimentation will be slow. If the concentration is too high, the agent itself will encapsulate the mud particles, resulting in colloidal protection, which will hinder solid-liquid separation.

[0097] The blade shearing and stirring mechanism within the end-flocculation chamber ensures sufficient contact between the reagent and the particles while protecting the newly formed network floc structure on the surface of the mud particles from being dispersed by the water flow, thus accelerating solid-phase sedimentation and separation from the liquid phase. In this stratified structure, the uppermost supernatant overflows and precipitates. An optical detection device, such as an online turbidimeter, installed on the supernatant overflow outlet pipe, is activated to continuously scan and measure the transmittance of the flowing liquid.

[0098] When the system measures the real-time transmittance value and it is lower than the preset acceptable threshold (the minimum acceptable transmittance threshold) due to changes in mud properties caused by upstream geological changes, the system determines that the back-end processing effect has deteriorated and sends an optimization instruction to the central processing unit node. The basis for setting this acceptable threshold is that 90% transmittance is generally considered the bottom line for water quality that allows the supernatant to meet recycling or discharge standards in engineering. A value below 90% means that suspended particles in the mud have not settled effectively and the water quality does not meet the standards. The adjustment mechanism of the front-end risk assessment model must be triggered immediately to improve the purity of the final separation.

[0099] The optimization instruction triggers the processor to adaptively adjust a key parameter used in the bubble potential index modeling module to calculate the fusion risk coefficient. Specifically, it increases the preset weighted product coefficient, which is assigned an internal enrichment density reference value in the bubble potential index modeling module. The feedback adjustment control logic of the preset weighted product coefficient can be expressed as follows:

[0100] in, After feedback adjustment, in the next control cycle The new values ​​of the internal preset weight product coefficients to be used; For the current control cycle The current value of the internally preset weighted product coefficients currently in use; This is a preset fixed step size increment; Transmittance measured in real time; To meet the acceptable boundary line, a feedback compensation and adjustment technology chain is constructed to correct the upstream risk assessment model from the perspective of end-processing effects.

[0101] Among them, the aforementioned internally preset weighted product coefficients The increase in sensitivity means that the system is more sensitive to the acoustic detection results of internal microbubbles when assessing overall risk. This increased sensitivity will shift the risk curve upwards in subsequent calculations, leading to a higher risk threshold, i.e., the specific liquid level depth. The value increases under the same acoustic signal, thereby physically stripping out a thicker upper layer of mud for recirculation, dynamically improving the purity of the mud flow entering the final flocculation and forming stage.

[0102] In a specific embodiment, the bottom-layer output slurry flows into the end-flocculation chamber, where an agitator operates at 10 r / min, while a 0.1% polyacrylamide solution is added dropwise at a flow rate of 5 L / h. Initially, the system operates stably, and the supernatant discharged from the overflow outlet is measured by an online turbidity meter, showing a transmittance of... It stabilizes at around 95%. This is the acceptable boundary line set by the system. The initial transmittance was 90%. After a period of time, as the tunnel boring machine advanced into new strata, the clay particle content of the slurry increased, leading to a decrease in flocculation effect and a drop in transmittance measured by the online turbidity meter. The value slowly decreased to 88%. Since 88% < 90%, the system immediately triggered a feedback compensation mechanism.

[0103] The processor reads the current value of the internal preset weight product coefficient in the bubble potential index modeling module currently in use. The value is 4.8. This is the system's set fixed step increment. The value is 0.2. The processor calculates the new coefficient value: The calculated new value of 5.0 for the internally preset weighted product coefficient is immediately updated in the risk calculation model of the bubble potential index modeling module.

[0104] New value of the internally preset weighted product coefficient Under the influence of this, the system's hierarchical control dynamic threshold parameter cutting depth As the volume increases, the amount of upper-layer return mud also increases. After several cycles of feedback and compensation mechanisms self-adjustment, the quality of the bottom-layer output mud flow entering the terminal flocculation stage improves accordingly, increasing light transmittance. The system's performance improved to 93%, restoring it to stable operation.

[0105] Each of the modules can be implemented in whole or in part through software, hardware, or a combination thereof. It supports hardware embedded in or independent of the processor in the computer device, and also supports software stored in the memory of the computer device, so that the processor can call and execute the operations corresponding to each of the above modules.

[0106] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A mud adaptive recirculation control system based on foam state, characterized in that, include: The pre-defoaming mud generation module obtains the original foamed mud, performs simultaneous surface and bottom layer chemical addition operations, and generates a pre-defoaming mud flow. The physical state feature extraction module guides the pre-defoamed mud flow into the expansion and deceleration zone, controls the image acquisition device and the ultrasonic device to acquire data synchronously, and extracts the physical feature data of the mud. The foam potential index modeling module calls the mud foaming risk quantification mapping function to parse the physical characteristic data of the mud, and performs parametric integration along the vertical depth coordinate line in combination with the preset weight product coefficient to calculate the depth foaming risk assessment model. The dynamic threshold filtering module iterates through the risk classification data contained in the deep bubbling risk assessment model, compares it with the bubble concentration tolerance threshold, and filters out the stratified control dynamic threshold parameters. The layered cutting and stripping module sends the layered control dynamic threshold parameters to the dynamic interception mechanism to perform layered physical interception on the pre-defoaming mud flow, cutting off the upper layer return mud flow and the bottom layer output mud flow. The feedforward drug delivery control module monitors and records the real-time return flow load signal of the upper reflux mud, initiates the advance compensation drug delivery strategy, and generates an antifoaming control signal. The feedback correction and adjustment module guides the bottom output mud flow into the sedimentation and granulation area, adds polymer flocculant to cause solids to settle, and adjusts the preset weight product coefficient in the parameter integration calculation based on the transmittance of the supernatant.

2. The mud adaptive recirculation control system based on foam state according to claim 1, characterized in that, The simultaneous application of chemicals to the surface and bottom layers to generate a pre-defoamed mud flow includes the following steps: The original foamy mud is guided into the mixing and defoaming chamber; Activate the atomizing spray valve array installed at the top of the mixing defoaming chamber to spray chemical defoamer onto the exposed bubble surface area of ​​the original foamy slurry; Turn on the circulation pump located at the bottom of the mixing and defoaming chamber, extract the bottom slurry carrying unbroken microbubbles, and inject chemical defoamer into the main pipeline and apply cutting and shearing force; The bottom mud, after being treated twice, is transported back to the mixing and defoaming chamber to be mixed and generate the pre-defoaming mud flow.

3. The mud adaptive recirculation control system based on foam state according to claim 1, characterized in that, The control image acquisition device and the ultrasonic device synchronously acquire and extract physical characteristic data of the mud, including the following steps: The image acquisition device, positioned at the top of the expansion and deceleration zone, continuously captures surface foam images of the liquid surface. Based on the inter-frame pixel differences between consecutive images, the dissipation rate factor that quantifies the dynamic balance between bubble generation and rupture is extracted. Simultaneously send detection commands to the vertical ultrasonic transceiver array that is attached to the sidewall of the expansion and deceleration area, transmit acoustic pulse sequences and receive reflected echoes that penetrate the lower mud layer, and extract the energy attenuation rate factor that reflects the acoustic impedance characteristics at different immersion depths. The feature data of the surface foam image, the dissipation rate factor, and the energy decay rate factor are extracted, and then time-stamped and encapsulated to construct the physical feature data of the mud.

4. The mud adaptive recirculation control system based on foam state according to claim 3, characterized in that, The extraction of the energy attenuation rate factor, which reflects the acoustic impedance characteristics at different immersion depths, includes the following steps: Obtain the effective path length of ultrasonic waves propagating in mud and the reference amplitude; Obtain the actual measured echo signal amplitude at each immersion depth; Calculate the ratio of the reference amplitude to the echo signal amplitude, and perform a base-10 logarithmic operation on the ratio to obtain the logarithmic result; Based on the signal attenuation ratio between the reference amplitude and the echo signal amplitude, and in conjunction with the effective path length, the energy attenuation rate factor corresponding to each immersion depth is calculated.

5. The mud adaptive recirculation control system based on foam state according to claim 3, characterized in that, The process of calling the mud foaming risk quantification mapping function to parse the mud physical characteristic data, performing parametric integration along the vertical depth coordinate line and combining preset weighted product coefficients to calculate the depth foaming risk assessment model includes the following steps: Extract the feature data and dissipation rate factor of the surface foam image encapsulated within the physical feature data of the mud, and input them into the surface activity assessment formula to perform numerical calculations to generate an apparent risk score. The energy decay rate factor at each depth carried by the physical characteristic data of the mud is matched and extracted, and then converted into the internal enrichment density reference value at the corresponding liquid surface depth using an acoustic empirical inversion algorithm. The preset weighted product coefficients are configured for the apparent risk score and the internal enrichment density reference value, and a multiplication-addition operation is performed to obtain the fusion risk coefficient. By integrating the fusion risk coefficients at various depths and establishing a continuous analytical model for the target depth, the bubble risk assessment model for that depth is output.

6. The mud adaptive recirculation control system based on foam state according to claim 1, characterized in that, The process of traversing and reading the risk classification data contained in the deep bubbling risk assessment model, comparing it with the bubble concentration tolerance threshold, and filtering out the stratified control dynamic threshold parameters includes the following steps: The connection retrieves the depth bubbling risk assessment model, which exists in the form of a continuous function. Call the storage unit to read the bubble concentration tolerance threshold that represents the allowable entry into the back-end flocculation process under the current operating conditions; A risk benchmark equation is constructed, and the corresponding depth solution is obtained by solving the risk benchmark equation within the effective water depth range through a continuous solution algorithm model. Lock the specific liquid surface depth value corresponding to the depth solution, and format and encode the specific liquid surface depth value to generate the layered control dynamic threshold parameter.

7. The mud adaptive recirculation control system based on foam state according to claim 6, characterized in that, The step of sending the layered control dynamic threshold parameters to the dynamic interception mechanism to implement layered physical interception of the pre-defoaming mud flow, cutting off the upper layer return mud flow and the lower layer output mud flow, includes the following steps: The dynamic threshold parameters of the layered control, which include specific liquid level depth values, are analyzed and converted into displacement control signals to drive the servo motor. The guide tube head mounted on the longitudinal slide rail is driven to move up and down until the guide tube head moves to the underwater control surface indicated by the displacement control signal; Turn on the suction negative pressure pump group that is connected to the outside of the guide pipe head to suck up and strip the upper layer of return mud located at or above the control surface; The control surface physically seals the outflow channel of the lower layer mud, and under the action of gravity static pressure, the bottom layer output mud flow located deep at the bottom end of the control surface is drained out.

8. The mud adaptive recirculation control system based on foam state according to claim 2, characterized in that, The monitoring and recording of the real-time return flow load signal of the upper layer of refluxed mud, initiating the advance compensation dosing strategy, and generating a defoaming control signal include the following steps: The instantaneous volumetric flow rate signal fed back by the electromagnetic flow meter deployed on the circulating return pipeline is scanned, and the real-time flow quality value of the upper layer return slurry is calculated and extracted within a continuous statistical period. The extracted upper layer of refluxed mud is reinjected into the mixing and defoaming chamber; The deviation and rate of change between the real-time flow quality value and the median of the preset standard return expected balance interval are calculated, and matched with the preset nonlinear agent consumption compensation matrix table. The compensation difference is obtained by analyzing the table lookup matching results, generating a pulse waveform packet containing the corrected duty cycle, and sending it to the actuator as the defoaming control signal carrying the compensation difference of the agent valve control.

9. The mud adaptive recirculation control system based on foam state according to claim 1, characterized in that, The process of guiding the bottom-layer output mud flow into the sedimentation and granulation area, adding a polymeric flocculant to cause the solid phase to settle, and adjusting the preset weighted product coefficient in the parametric integration calculation based on the transmittance of the supernatant includes the following steps: The bottom-layer output mud flow is fed into the end flocculation and forming chamber equipped with blades; Start the variable frequency motor configured at the top of the end flocculation forming chamber to drip flocculant into the bottom layer output mud flow which is in a stirring state; The stirring action generated by the variable frequency motor driving the blades accelerates the solid phase sedimentation and separation from the liquid phase, and the supernatant liquid at the top overflows and precipitates out. The optical detection device is activated to continuously scan the transmittance of the supernatant. When the real-time value of the transmittance is less than the allowable threshold, an optimization command is sent in reverse to adaptively increase the preset weight product coefficient.

10. A mud adaptive recirculation control system based on foam state according to claim 9, characterized in that, The adaptive adjustment of the preset weight product coefficient includes the following steps: Obtain the preset weight product coefficient and the preset fixed step size increment currently in use during the control cycle; The transmittance of the supernatant liquid, measured in real time during the current control cycle, and the preset allowable boundary line for compliance are obtained. Determine whether the light transmittance is less than the acceptable boundary line; If the judgment result is yes, then based on the fixed step size increment, the preset weight product coefficient currently in use is incrementally updated.

Images