Efficient pretreatment method for particulate matter-containing low-calorific-value waste liquid

Abstract

The invention discloses an efficient pretreatment method for particulate matter-containing low-calorific-value waste liquid, and relates to the technical field of waste liquid pretreatment, and the method comprises the following steps: establishing a coating existence identification factor by obtaining the surface viscosity displacement track, the particle size fluctuation phase and the interface refraction characteristic of particulate matters in the waste liquid; judging whether the particulate matters in the waste liquid are coated by the low-calorific-value organic components or not according to the coating existence identification factor; and under the condition that the particulate matter is coated with the low-calorific-value organic component, based on the dynamic trajectory behavior of the coating existence identification factor, extracting the initial bubbling lag time and the particle deformation rate inhibition amplitude generated by the conditioning reaction, and generating the permeation delay expression quantity. Aiming at the problems that permeation of a conditioner is blocked and the conditioning effect is difficult to judge due to the fact that particles in the particulate matter-containing low-calorific-value waste liquid are coated by organic components, accurate identification of the conditioning failure risk and dynamic closed-loop optimization of the dosing behavior are realized, and the pretreatment efficiency and the system operation stability are improved.

Description

Technical Field

[0001] This invention relates to the field of waste liquid pretreatment technology, specifically to a high-efficiency pretreatment method for low-calorific-value waste liquid containing particulate matter. Background Technology

[0002] High-efficiency pretreatment of low-calorific-value wastewater containing particulate matter refers to a preliminary purification and conditioning method implemented before the wastewater enters its final treatment stage (such as incineration, biochemical treatment, or resource utilization) to improve treatment efficiency and ensure the stability of subsequent processes, given the complex characteristics of the wastewater containing both solid particulate impurities and low calorific value. Existing technologies typically employ a combined process of "solid-liquid separation + concentration and conditioning" to achieve this goal. The treatment process mainly includes several key steps: First, large particles and suspended solids are removed through bar filtration or sedimentation devices to reduce the load on subsequent equipment; second, fine particles are further removed using flocculation sedimentation or air flotation, often supplemented with polymeric additives to improve particle aggregation efficiency; subsequently, to address the low calorific value of the wastewater, methods such as dehydration and concentration, phase separation, or the addition of conditioning agents may be introduced to increase the concentration of combustible organic components and enhance the thermal treatment value of the wastewater; in addition, auxiliary units such as pH adjustment and oil and sand removal may be installed to prevent corrosion and scaling. Overall, existing pretreatment technologies for low-calorific-value waste liquid containing particulate matter show a trend of multi-stage coupling and precise control. These technologies aim to solve problems such as complex impurities, insufficient calorific value, and high treatment difficulty through a systematic treatment process, thereby providing a good working condition foundation for subsequent treatment.

[0003] The existing technology has the following shortcomings:

[0004] In the efficient pretreatment of low-calorific-value wastewater containing particulate matter, when the wastewater contains both fine particulate matter from complex sources and colloidal low-calorific-value organic components, the particulate matter easily adheres and aggregates with the organic components during flow and residence, forming composite particles with a surface coated by low-calorific-value colloids. Because this coating layer is typically hydrophobic and structurally inert, it significantly hinders the effective contact between the conditioner and the inorganic core of the particulate matter, causing the conditioner to act only on the particle surface and failing to adequately condition the particles. This results in the particles failing to form stable flocs, affecting subsequent sedimentation and separation. However, existing efficient pretreatment technologies for low-calorific-value wastewater containing particulate matter cannot determine the risk of ineffective conditioning based on the conditioner penetration barrier when the particulate matter is coated with low-calorific-value organic components. This makes it impossible to accurately adjust the dosing strategy, potentially leading to decreased particle treatment efficiency, conditioner waste, and ultimately, the entry of insufficiently conditioned composite particles into the incineration system, causing problems such as heat load fluctuations, abnormal carbon emissions, and even furnace coking, severely impacting the stability and operational efficiency of the treatment system.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an efficient pretreatment method for low-calorific-value waste liquid containing particulate matter, so as to solve the problems in the background art mentioned above.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for efficient pretreatment of low-calorific-value wastewater containing particulate matter, specifically comprising the following steps:

[0008] S1. By acquiring the surface viscous displacement trajectory, particle size fluctuation phase and interface refractive properties of particulate matter in waste liquid, a coating presence identification factor is established, and the coating presence identification factor is used to determine whether particulate matter in waste liquid is coated by low-calorific-value organic components.

[0009] S2. When particulate matter is coated with low-calorific-value organic components, based on the dynamic trajectory behavior of the recognition factor in the coating, the initial foaming hysteresis time and the inhibition amplitude of particle deformation rate generated by the conditioning reaction are extracted to generate the penetration delay expression level, which is used to characterize the conditioner's penetration barrier performance.

[0010] S3. Combine the delayed expression level of penetration with the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window to form a conditioning failure assessment signal, which is used to determine whether there is a risk of conditioning failure.

[0011] S4. Based on the interval intensity distribution results of the conditioning failure assessment signal, generate the dosing behavior selection path, and adjust the particle affinity polarity, action sequence segment and mixing frequency of the conditioning agent according to the dosing behavior selection path;

[0012] S5. By combining the changing trends of conditioning failure assessment signals and coating presence identification factors, a dosing response feedback channel is constructed, and the dosing behavior is dynamically adjusted based on the feedback channel to achieve closed-loop optimization of the dosing strategy.

[0013] Preferably, S1 specifically includes the following steps:

[0014] S101. By introducing oscillation disturbance into the waste liquid, the surface viscous displacement response path of particulate matter under disturbance conditions is collected, and the particle size fluctuation phase of particulate matter in the flow cycle is obtained through a particle size distribution tracking device. At the same time, the interface refractive characteristic data of particulate matter and liquid phase boundary are extracted using a multi-angle refractive scanning device.

[0015] S102. Time synchronization and unit standardization are performed on the surface viscous displacement response path, particle size fluctuation phase and interface refractive property data. Based on the dynamic response coupling relationship between the processed surface viscous displacement response path, particle size fluctuation phase and interface refractive property data, a coating existence identification factor is constructed to reflect the changing characteristics of the surface state of particles.

[0016] S103. Compare the output value of the coating presence identification factor with the preset coating judgment interval. When the output value continuously falls into the preset coating judgment interval, it is determined that the particulate matter in the waste liquid is coated by low-calorific-value organic components.

[0017] Preferably, S102 is as follows:

[0018] A unified time start point was set for the surface viscous displacement response path, particle size fluctuation phase and interface refractive characteristic data respectively, and time synchronization was achieved by data window alignment to ensure that various types of data are comparable within the same physical response period;

[0019] Physical unit conversion factors are introduced into the time-synchronized data, and normalization is performed based on the maximum and minimum values ​​of the response dimension to ensure that the data of different physical quantities have a unified calculation scale and eliminate the interference of the original units on the subsequent analysis results.

[0020] The three types of data, after time synchronization and unit standardization, are input into a nonlinear cross-convolution analysis process to extract the response offset features in the interaction effect term. Based on the change in offset amplitude, a coating existence identification factor is constructed to identify the coupled evolution characteristics of particulate matter surface state.

[0021] Preferably, S2 is as follows:

[0022] When particulate matter is coated with low-calorific-value organic components, the dynamic trajectory behavior of the coating recognition factor is continuously collected. The time point of conditioner addition is used as the starting point of analysis to identify the first moment when the coating recognition factor changes from a steady state to a fluctuating state. The time interval between this moment and the starting point of analysis is calculated to determine the initial hysteresis time.

[0023] Based on the dynamic trajectory behavior of the coating presence identification factor, the response path of particulate morphology change during conditioning reaction is continuously recorded, and the difference between the morphology change rate under the coating state and the morphology change rate under the preset benchmark state is compared to extract the particle deformation rate suppression amplitude.

[0024] The initial condensation hysteresis time and the particle deformation rate inhibition amplitude are uniformly scaled and weighted according to the temporal correlation and amplitude coupling relationship between the two within the same conditioning reaction cycle to form a penetration delay expression quantity characterizing the degree of conditioning reaction response lag, so as to reflect the penetration limitation characteristics of conditioning agent on the surface of coated particles.

[0025] Preferably, S3 specifically includes the following steps:

[0026] S301. Within the dosing window, continuously collect the morphology of the flocs formed during the conditioning reaction, extract the changes in the projected area of ​​the flocs, the changes in the boundary integrity, and the changes in the structural connectivity, and compare the collected results with the reference morphology under the preset intact state to obtain the floc integrity deviation value, which reflects the degree of deviation of the floc structure; and synchronously record the energy changes at the particle-liquid interface during the conditioning reaction, extract the interface energy change curve over time, and identify the peak position and amplitude changes in the energy change curve to obtain the change in the peak value of interface energy dissipation.

[0027] S302. The changes in the penetration delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window are processed in a time-series synchronization process, and a joint change sequence is constructed within the same dosing cycle. The coupling relationship of the joint change sequence is analyzed to form a conditioning failure assessment signal containing time correlation characteristics and amplitude correlation characteristics.

[0028] S303. Compare the conditioning failure assessment signal with the preset conditioning stability judgment interval. When the conditioning failure assessment signal falls into the failure judgment interval of the preset conditioning stability judgment interval within the continuous dosing cycle, it is judged that there is a risk of conditioning failure.

[0029] Preferably, S302 is as follows:

[0030] A unified time starting point was set for the changes in the osmotic delay expression level, floc integrity deviation value and interface energy dissipation peak value, and they were aligned based on the time tag of the first addition of the conditioner, and time synchronization was performed within the same dosing cycle.

[0031] Based on the data that has been synchronized over time, the changes in the penetration delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation are assembled into a joint change sequence in a progressive time order. The response offset features of adjacent segments in the sequence are extracted by window convolution processing to identify the dynamic linkage features between the parameters.

[0032] The response offset features extracted from the joint change sequence are subjected to amplitude correlation calculation and time response coupling calculation. The results are then bidirectionally superimposed to construct a conditioning failure assessment signal, forming an output index containing time correlation features and amplitude correlation features, which is used as a basis for judging the risk of subsequent conditioning failure.

[0033] Preferably, S4 is as follows:

[0034] The numerical distribution of conditioning failure assessment signals within a continuous dosing cycle was statistically analyzed in intervals. The frequency and duration of occurrence of conditioning failure assessment signals in different intensity intervals were collected and processed to obtain the interval intensity distribution results that reflect the evolution trend of conditioning status.

[0035] Based on the proportion of each intensity interval in the interval intensity distribution results, the conditioning behavior parameters are combined and rearranged to form multiple drug administration behavior parameter sequences that advance over time, and drug administration behavior selection paths are generated based on the iterative correlation between the parameter sequences.

[0036] Based on the parameter configuration corresponding to the generation path of the dosing behavior, the particle affinity polarity of the conditioner is adjusted by directional matching. At the same time, the action sequence of the conditioner is reorganized before and after, and the mixing frequency is synchronously corrected, so that the dosing behavior is updated and executed according to the generation path in subsequent dosing cycles.

[0037] Preferably, S5 is as follows:

[0038] During the continuous dosing cycle, time series data of conditioning failure assessment signals and coating presence identification factors were collected simultaneously. The direction, magnitude and rate of change of the two at the same time scale were extracted and time-aligned to form a trend combination result reflecting the correlation between conditioning state changes and coating state changes.

[0039] Based on the trend combination results, a dosing response feedback channel is constructed. The trend change of conditioning failure assessment signal is used as the conditioning risk input dimension, and the trend change of the identified factors are used as the coverage status input dimension. The correspondence between trend input and dosing behavior parameters is established in the same mapping space, forming a feedback channel structure that can be updated over time.

[0040] Based on the trend response results output in the dosing response feedback channel, the dosing behavior is dynamically regulated so that the particle affinity polarity, action sequence segment and mixing frequency point in subsequent dosing cycles are adjusted synchronously with the trend changes, thereby realizing the adaptive update of dosing behavior in continuous cycles.

[0041] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0042] 1. This invention enables precise identification of whether particulate matter in wastewater is coated with low-calorific-value organic components, and further quantifies the degree of penetration restriction of conditioners under the influence of coating. By constructing a penetration delay expression quantity with dual time and amplitude response characteristics, it can effectively capture the lag behavior of particulate matter during the conditioning process. This scheme also integrates floc structure deviation characteristics and interfacial energy dissipation characteristics to form a conditioning failure assessment signal, constructing a complete conditioning failure risk identification chain. This achieves dynamic perception of the entire process from coating identification and penetration barrier quantification to conditioning effect assessment, solving the problem of inaccurate diagnosis of conditioning failure risk in existing technologies.

[0043] 2. This invention also constructs a dosing behavior selection path based on the interval intensity distribution results of the conditioning failure assessment signal, and adjusts the particle affinity polarity, action sequence segment, and mixing frequency of the conditioning agent accordingly to achieve adaptive updating of the dosing behavior. Simultaneously, by combining the trend changes of the conditioning failure assessment signal and the coating presence identification factor, a dosing response feedback channel is constructed, enabling closed-loop control of the dosing behavior based on trend response throughout a continuous cycle. This technical approach can effectively improve the efficiency of conditioning agent action, avoid overdosing or ineffective dosing, improve floc stability and settling performance, reduce the load impact on the incineration system, and thus improve the operational stability and resource utilization of the wastewater treatment system. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0045] Figure 1 This is a schematic flowchart of the efficient pretreatment method for low-calorific-value waste liquid containing particulate matter according to the present invention. Detailed Implementation

[0046] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0047] This invention provides, for example Figure 1 The efficient pretreatment method for low-calorific-value wastewater containing particulate matter shown includes the following steps:

[0048] S1. By acquiring the surface viscous displacement trajectory, particle size fluctuation phase and interface refractive properties of particulate matter in waste liquid, a coating presence identification factor is established, and the coating presence identification factor is used to determine whether particulate matter in waste liquid is coated by low-calorific-value organic components.

[0049] In this embodiment, S1 specifically includes the following steps:

[0050] S101. By introducing oscillation disturbance into the waste liquid, the surface viscous displacement response path of particulate matter under disturbance conditions is collected, and the particle size fluctuation phase of particulate matter in the flow cycle is obtained through a particle size distribution tracking device. At the same time, the interface refractive characteristic data of particulate matter and liquid phase boundary are extracted using a multi-angle refractive scanning device.

[0051] To identify whether particulate matter in waste liquid is coated with low-calorific-value organic components, the particle response behavior in the waste liquid is first excited by introducing oscillatory disturbances. In practical applications, an electromagnetic vibrator or a controllable pulse pump can be installed in the waste liquid delivery pipeline to apply a disturbance flow field with a specific frequency and amplitude, causing the particles to undergo microscopic displacement responses under the disturbance environment. The minute displacement trajectories of the particles under the disturbance are recorded using a high-speed camera system or optical particle size analysis equipment, obtaining the surface viscous displacement response path. This path reflects the degree of viscous response of the particle surface adhesion layer and is an important feature for identifying the coating state. Simultaneously, particle size fluctuations within the disturbance period are monitored using a particle size distribution tracking device, and the particle size phase changes at different disturbance stages are extracted to construct a particle size fluctuation phase data sequence, which is used to reveal the fluctuation stability characteristics of particles under colloidal coating. Furthermore, the reflection path of the particles in the liquid environment is illuminated by a multi-angle refractive scanning device, and the refractive index change data at the particle-liquid phase boundary is collected to obtain interfacial refractive characteristic data. These three data dimensions collectively reflect the physical state of the particle surface, constructing a preliminary characterization of the particle coating state from three perspectives: perturbation response, particle size variation, and optical behavior. These three types of data were chosen because particles coated with organic components often exhibit characteristics such as sluggish response, poor particle size stability, and significant differences in optical refraction. These physical indicators can indirectly quantify the presence and influence of the coating layer.

[0052] Introducing oscillations into waste liquid refers to using physical devices to create periodic energy disturbances in the waste liquid, such as flow velocity oscillations, pressure pulsations, or local vibration fluctuations, to excite the fluid response characteristics of particles. Disturbance conditions refer to the combination of oscillation frequency, amplitude, waveform, and duration; different disturbance conditions will trigger different degrees of physical reactions in the particles. The surface viscous displacement response path is the tiny displacement trajectory of particles on the surface due to the attached layer under disturbance; the degree of trajectory deviation is closely related to the adhesion strength of the organic coating layer. Particle size distribution tracking devices typically use equipment such as laser particle size analyzers and dynamic image analyzers to capture the particle size changes of the particle group frame by frame during the disturbance process. Particle size fluctuation phase refers to the phase shift corresponding to the particle size change under periodic disturbance; the periodic distortion of this phase can characterize the physical buffering effect of surface attachments. Multi-angle refractive scanning equipment collects the refractive changes at the particle-liquid boundary based on the difference between the incident and reflected light paths, and captures the refractive intensity and refractive index response at different angles using a high-sensitivity receiver. Interfacial refractive properties data are comprehensive characteristics of the optical response of particle surfaces, revealing the presence of a colloidal organic coating layer with strong adsorption and a refractive index different from the background liquid. Synergistic analysis of these indicators allows for precise determination of whether particulate matter is coated with low-calorific-value organic components, providing a foundation for subsequent conditioning actions.

[0053] S102. Time synchronization and unit standardization are performed on the surface viscous displacement response path, particle size fluctuation phase and interface refractive property data. Based on the dynamic response coupling relationship between the processed surface viscous displacement response path, particle size fluctuation phase and interface refractive property data, a coating existence identification factor is constructed to reflect the changing characteristics of the surface state of particles.

[0054] S103. Compare the output value of the coating presence identification factor with the preset coating judgment interval. When the output value continuously falls into the preset coating judgment interval, it is determined that the particulate matter in the waste liquid is coated by low-calorific-value organic components.

[0055] In determining whether particulate matter is coated with low-calorific-value organic components, comparing the output value of the coating presence identification factor with a preset coating judgment interval is crucial for intelligent identification. The specific operation is as follows: First, a coating judgment interval is established based on historical data and experimental calibration results. This interval indicates that when the value of the identification factor falls within this range, its corresponding physical behavior is highly correlated with the known coating state. Then, the output value of the identification factor is calculated in real time within each processing cycle, and this output value is compared with the judgment interval one by one. If the output value of the identification factor stably falls within this interval over multiple consecutive time points, the particulate matter is considered to exhibit typical coating behavior. In implementation, a sliding window mechanism can be used to track the output value over time to avoid misjudgments caused by short-term disturbances. For example, coating confirmation can only be triggered if the identification factor falls within the judgment interval for five consecutive window periods. This judgment logic can be constructed using a conditional logic chain and combined with a dynamic threshold adjustment mechanism to adapt to different waste liquid characteristics, achieving more accurate and robust identification capabilities. For example, in the process of treating high-viscosity waste liquid, the identification factor fluctuates frequently, and the upper limit of the judgment interval can be appropriately expanded to enhance the fault tolerance.

[0056] The output value of the coating presence identification factor is a comprehensive score obtained by extracting the response offset amplitude, trend, and nonlinear coupling strength between different physical quantities through a nonlinear cross-convolution analysis process, usually expressed as a dimensionless numerical value. This output value can be continuously generated during the processing cycle, serving as a real-time indicator of particle surface state evolution. The preset coating judgment interval is an empirically set numerical range, usually determined by statistical modeling of a large amount of known coated and uncoated sample data. It represents the typical range of the identification factor output value when coating is present. The upper and lower boundaries of the interval can be adaptively adjusted according to the source of the waste liquid, particle type, or rheological properties to ensure an optimal balance between judgment sensitivity and false positive rate. The advantage of using this method is that it avoids fuzzy empirical judgments of coating status, and instead achieves standardization and automation of particle coating identification through quantifiable and repeatable numerical judgments, providing accurate decision-making basis for subsequent treatment strategies.

[0057] In this embodiment, S102 specifically refers to:

[0058] A unified time start point was set for the surface viscous displacement response path, particle size fluctuation phase and interface refractive characteristic data respectively, and time synchronization was achieved by data window alignment to ensure that various types of data are comparable within the same physical response period;

[0059] Before jointly analyzing surface viscous displacement response paths, particle size fluctuation phases, and interfacial refractive properties, a unified time start point must be established to ensure that all types of physical data are activated and recorded at the same starting moment, thus achieving temporal comparability. A unified time start point refers to using the trigger time of the waste liquid disturbance signal as the starting recording reference for all acquisition channels, simultaneously with the excitation of the disturbance signal. For example, this can be achieved by synchronously controlling the data recording actions of high-speed cameras, particle size analyzers, and refractive scanning devices using a unified trigger electrical signal. Data window alignment refers to using a sliding time window to truncate and resample data from different data sources with different sampling frequencies and acquisition periods, ensuring that various data segments have equally long sample point distributions within the same time interval. This is often achieved using linear interpolation reconstruction or time resampling algorithms. By setting a unified start point for different data and using data window alignment, the three types of physical data can be ensured to be synchronously comparable within the same physical response period, thereby capturing the true response characteristics of particles under disturbance. For example, when the oscillation disturbance is triggered at zero seconds, all response data within 10 seconds are collected. By setting the same start and end times and a uniform sampling density, surface viscous displacement, particle size fluctuation, and interface refractive intensity are mapped onto the same time axis, forming a collaborative data sequence that can be used for dynamic coupling analysis. Before alignment, the three data sources may experience time drift and response misalignment due to differences in sampling frequencies. Direct analysis would lead to coupling misjudgment. Therefore, this synchronization process is the foundation for the subsequent construction of the coating identification factor. A unified time start point ensures synchronized response start, data window alignment achieves unified sampling of the response process, and the same physical response period provides a reference for the overall behavior of particles under the same disturbance.

[0060] Physical unit conversion factors are introduced into the time-synchronized data, and normalization is performed based on the maximum and minimum values ​​of the response dimension to ensure that the data of different physical quantities have a unified calculation scale and eliminate the interference of the original units on the subsequent analysis results.

[0061] When processing three types of data—surface viscous displacement response paths, particle size fluctuation phases, and interface refractive properties—that have already undergone time synchronization, it is necessary to introduce physical unit conversion factors to unify the processing of data with different dimensions, thereby ensuring comparability and coupling. Physical unit conversion factors refer to converting the units of the original physical quantities to dimensionless parameters of a unified standard. For example, converting displacement paths measured in nanometers to standard length units, particle size change frequencies from Hertz to displacement phase differences on a unified time scale, and refractive indices to standardized optical density values. The conversion factors introduced for each type of data conversion are determined based on their original physical meaning and precisely set through equipment calibration data or physical quantity conversion coefficients. Subsequently, the converted data is normalized based on the maximum and minimum values ​​of its response dimensions. Normalization means limiting all values ​​in a data sequence to a fixed range, typically through linear scaling to convert the original data into a numerical range between zero and one. The maximum and minimum values ​​of the response dimension represent the limiting response state of this type of physical quantity within the acquisition time window. For example, within a 10-second perturbation, the maximum surface viscous displacement is 500 nanometers, and the minimum is 30 nanometers. After normalization, each data point is mapped to its relative position within this interval. The reason for normalization is that the original units of the three types of data are completely different. When directly participating in mathematical coupling analysis, the difference in dimensions can cause one variable to have a dominant influence on the overall analysis results, thereby masking the important changing trends of other variables. Through normalization, not only can the interference of the original units on the coupling analysis be eliminated, but it can also ensure that the characteristic changes of each physical variable can be considered equally when constructing the identification factor for the coating. For example, without normalization, the change amplitude of the particle size fluctuation phase may be much smaller than the change in surface displacement, and thus be ignored in subsequent analysis. After normalization, the two will jointly constitute the discrimination criterion for the particle surface state within a unified scale.

[0062] The three types of data, after time synchronization and unit standardization, are input into a nonlinear cross-convolution analysis process to extract the response offset features in the interaction effect term. Based on the change in offset amplitude, a coating existence identification factor is constructed to identify the coupled evolution characteristics of particulate matter surface state.

[0063] This study inputs three types of data—time-synchronized and unit-standardized surface viscous displacement response paths, particle size fluctuation phases, and interface refractive properties—into a nonlinear cross-convolution analysis pipeline. The aim is to uncover the deep coupling characteristics between different physical responses caused by potential encapsulation behavior. The nonlinear cross-convolution analysis pipeline is a processing mechanism based on time-series feature alignment and nonlinear relationship mapping. Unlike traditional linear correlation analysis, it can identify asynchronous, asymmetric, and even lagging interactive response relationships between variables at different time scales and response amplitudes. The pipeline first performs dynamic window scanning on the three types of physical data, calculating the cross-response intensity between variables at different time points to form an initial interaction term. Then, it expands the nonlinear mapping dimension through kernel function transformation to construct a high-dimensional response tensor. Within the interaction term, the focus is on extracting response offset features, i.e., the phenomenon of a lag or advance in the time or amplitude of one variable's response to another. For example, after particles form an encapsulation structure, significant changes in particle size fluctuation phase delay the feedback response to surface displacement; this time difference is a manifestation of the response offset feature. The magnitude of the offset reflects the degree of hysteresis or non-cooperative enhancement in this type of response; the larger the magnitude, the more complex the internal coupling of the system, and the more likely there is coating interference. Based on the statistical distribution and intensity variation trend of these response offset characteristics, a coating presence identification factor is constructed. This identification factor is a comprehensive quantitative index that integrates the degree of interaction and coupling between three types of physical data, used to determine whether colloidal organic matter coating exists on the surface of particulate matter. For example, if there is a continuous relative hysteresis between the change in surface viscous displacement and the phase of particle size fluctuation within multiple analysis windows, accompanied by abnormally stable interfacial refractive index in the low-frequency range, the identification factor will output a high coupling value, indicating that the coating behavior is highly suspicious; otherwise, it is considered that there is no obvious coating phenomenon. This method achieves the identification of complex evolution trends of surface states through nonlinear modeling, effectively overcoming the misjudgment problems caused by dependent variable decoupling or temporal interference in existing analyses.

[0064] S2. When particulate matter is coated with low-calorific-value organic components, based on the dynamic trajectory behavior of the recognition factor in the coating, the initial foaming hysteresis time and the inhibition amplitude of particle deformation rate generated by the conditioning reaction are extracted to generate the penetration delay expression level, which is used to characterize the conditioner's penetration barrier performance.

[0065] In this embodiment, S2 specifically refers to:

[0066] When particulate matter is coated with low-calorific-value organic components, the dynamic trajectory behavior of the coating recognition factor is continuously collected. The time point of conditioner addition is used as the starting point of analysis to identify the first moment when the coating recognition factor changes from a steady state to a fluctuating state. The time interval between this moment and the starting point of analysis is calculated to determine the initial hysteresis time.

[0067] When particulate matter is coated with low-calorific-value organic components, continuous acquisition of the dynamic trajectory behavior of the coated recognition factor is necessary to extract the initial hysteresis time. This dynamic trajectory behavior can be captured at high frequency by a real-time signal recording system, tracking its numerical changes during conditioning. The conditioner addition time is set as the starting point for analysis, at which point the system automatically initiates time tracking. Subsequently, the value of the recognition factor will fluctuate due to the chemical action of the conditioner, gradually evolving from an initial stable state to a fluctuating state. By determining the moment when the recognition factor first exhibits a significant fluctuation—that is, when its numerical curve first breaks through the stable fluctuation range—the initial reaction response triggered by the reaction can be precisely located. The time interval between this moment and the conditioner addition starting point is the initial hysteresis time. For example, in a conditioning experiment, if the recognition factor begins to abruptly change 6 seconds after the conditioner is added, the initial hysteresis time is recorded as 6 seconds.

[0068] The dynamic trajectory behavior of the coating presence identification factor refers to the time-series response of this factor to the evolution of the particulate surface state during conditioning. It typically includes a combined output of multiple physical variables to track the changing trend of the coating state. Conditioning agents are chemical additives used in pretreatment to break down the coating layer on the particulate surface and enhance its reactivity, such as inorganic flocculants or organic conditioning enhancers. The timing of conditioning agent addition is a key reference point for initiating the conditioning reaction and is used as the starting anchor point for dynamic analysis in the data record. The first moment when the coating presence identification factor transitions from a stationary state to a fluctuating state refers to the time when the identification factor value first exceeds the baseline fluctuation range, usually extracted from the time series using a statistical outlier identification algorithm. The initial hysteresis time is the time difference between this transition point and the conditioning agent addition time, used to measure the degree of delay in the effect of the conditioning agent and is an important quantitative indicator for subsequent assessment of penetration barriers.

[0069] Based on the dynamic trajectory behavior of the coating presence identification factor, the response path of particulate morphology change during conditioning reaction is continuously recorded, and the difference between the morphology change rate under the coating state and the morphology change rate under the preset benchmark state is compared to extract the particle deformation rate suppression amplitude.

[0070] During the conditioning reaction, to characterize the true response of particulate matter to the conditioning agent in the coated state, it is necessary to continuously record the morphological changes of the particulate matter based on the dynamic trajectory behavior of the coating's identifying factors. Specifically, high-speed imaging devices or dynamic image analysis equipment can be used to collect hourly data on the changes in the particle's external contour, projected area, and boundary curvature during the conditioning reaction, thus forming the particulate matter's morphological change response path. Subsequently, under the same conditioning conditions, the rate of change of particle morphology over time in the coated state is compared with pre-established baseline data, which is derived from the morphological change records of particles not coated with low-calorific-value organic components under the same conditioning conditions. By comparing the difference between the two rate curves, the degree of inhibition of particle deformation behavior by the coating state can be quantified. For example, in the baseline state, particles may show significant stretching or breakage shortly after the start of conditioning, while in the coated state, the morphological changes of similar particles are significantly delayed. The rate difference between the two is recorded and accumulated to form the magnitude of particle deformation rate inhibition.

[0071] The morphological change response path of particles during conditioning reactions refers to the continuous trajectory of particle geometry changes over time under the influence of conditioning agents, typically described by contour changes, area shrinkage, or extension trends. The morphological change rate under coating refers to the speed at which particle morphological changes progress over time under the influence of low-calorific-value organic component coating; this rate is usually slow, delayed, or discontinuous. The morphological change rate under a preset baseline state refers to the reference rate of particle morphological changes obtained from historical experiments or standard samples under conditions without coating interference, used as a comparison benchmark. Difference comparison refers to calculating the difference between the coating state rate and the baseline rate point-by-point within the same time scale to identify the inhibitory effect of coating behavior on conditioning reactions. The particle deformation rate inhibition magnitude is a quantitative result based on this difference, used to describe the degree to which the coating layer weakens the particle morphological response ability, and is an important input parameter for subsequently constructing penetration delay expression levels.

[0072] The initial condensation hysteresis time and the particle deformation rate inhibition amplitude are uniformly scaled and weighted according to the temporal correlation and amplitude coupling relationship between the two within the same conditioning reaction cycle to form a penetration delay expression quantity characterizing the degree of conditioning reaction response lag, so as to reflect the penetration limitation characteristics of conditioning agent on the surface of coated particles.

[0073] To achieve the fusion of initial hysteresis lag time and particle deformation rate inhibition magnitude, these two physical quantities first need to be mapped to a unified computational scale. Time and rate magnitude values ​​can be converted to standard percentages using normalization functions, ensuring comparability within the same coordinate dimension. Subsequently, using the start of the conditioning reaction as a common starting point, a time series over the entire reaction cycle is constructed, and the two normalized parameters are projected onto this timeline, establishing time correlation matrices and amplitude coupling matrices at corresponding time points. Combining the synchronicity and intensity ratio of these two factors during the reaction process, a weighted function is used to combine them, outputting a hysteresis index curve that dynamically changes over time. The integral value of this curve represents the penetration delay expression level. For example, if the hysteresis lag is significantly prolonged while the particle deformation rate decreases significantly, the expression level will rise rapidly, indicating that the conditioning agent's penetration effect is significantly limited, suggesting the need for drug adjustment intervention.

[0074] Unified scale mapping refers to converting the initial hysteresis time and particle deformation rate inhibition amplitude of physical quantities to the same dimensions and numerical range to facilitate subsequent correlation analysis. The conditioning reaction cycle refers to the entire time process from the conditioner's initial entry into the reaction system to the basic completion of the particulate response, serving as the time benchmark for calculating all variables. Temporal correlation refers to the synchronicity of the changing trends of two indicators over time, while amplitude coupling reflects the consistency of response between their numerical changes. Weighted combination is used to fuse these two different types of response signals according to a proportional relationship, constructing a single indicator to reflect the overall hysteresis degree. The penetration delay expression level, characterizing the degree of conditioning reaction response hysteresis, is a comprehensive parameter integrating time and amplitude information, used to quantify the degree of obstruction during conditioner penetration. A higher penetration delay expression level means that the conditioner has more difficulty penetrating the coating layer to reach the particle core, reflecting a more significant limitation of conditioner penetration on the coated particle surface, and is an important basis for assessing conditioning risk.

[0075] S3. Combine the delayed expression level of penetration with the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window to form a conditioning failure assessment signal, which is used to determine whether there is a risk of conditioning failure.

[0076] In this embodiment, S3 specifically includes the following steps:

[0077] S301. Within the dosing window, continuously collect the morphology of the flocs formed during the conditioning reaction, extract the changes in the projected area of ​​the flocs, the changes in the boundary integrity, and the changes in the structural connectivity, and compare the collected results with the reference morphology under the preset intact state to obtain the floc integrity deviation value, which reflects the degree of deviation of the floc structure; and synchronously record the energy changes at the particle-liquid interface during the conditioning reaction, extract the interface energy change curve over time, and identify the peak position and amplitude changes in the energy change curve to obtain the change in the peak value of interface energy dissipation.

[0078] During the designated dosing window, a high-frequency image acquisition device was used to dynamically capture the flocs formed during the conditioning reaction, recording their morphological image sequences at different time points. Image processing algorithms were used to extract the area change trend of the flocs in a two-dimensional projection, the degree of closure of the contour boundary lines, and the number and density of connected regions within the flocs to quantify their overall structural characteristics. Subsequently, the structural parameters extracted at each time point were compared and analyzed with pre-defined complete floc morphology samples, and a weighted comprehensive score of three types of deviations was calculated as the floc integrity deviation value. Simultaneously, a micro-energy probe device was used to monitor the energy exchange behavior between the particles and the surrounding liquid phase interface during the conditioning reaction, generating a dynamic curve of energy change over time. Key peak regions were located from the curve, and their amplitude variation range and frequency of occurrence were extracted to form an interface energy dissipation peak change index. These two parameters, as microscopic response quantities reflecting the conditioning effect, will be used together with the penetration delay expression quantity to construct a conditioning failure assessment signal.

[0079] The dosing window refers to the time interval from the initial injection of the conditioner into the waste liquid to the end of the preliminary reaction stage, typically corresponding to the critical period for the conditioner and particulate matter to fully interact. The flocs formed during the conditioning reaction are structured particle clusters formed by the aggregation of particulate matter under the action of the conditioner; their morphology directly affects sedimentation efficiency and treatment effect. Continuous morphology acquisition uses high-frequency image scanning to acquire the morphological evolution of flocs in real time. Changes in the projected area of ​​the flocs reflect fluctuations in the degree of aggregation, changes in boundary integrity describe the continuity and closure of the floc edges, and changes in structural connectivity measure the compactness of its internal structure. The reference morphology under the preset intact state is a morphological model standard established based on the ideal reaction effect, serving as a comparison template for deviation assessment. The floc integrity deviation value is a quantitative result obtained by comparing the three structural parameters with the reference model, used to characterize whether there are any anomalies in the current structure. The energy change at the particle-liquid interface represents the microscale reaction intensity. By capturing the fluctuations in interface energy over time, a change curve is established. The change in the peak value of interface energy dissipation is analyzed by identifying significant fluctuation points in these curves to determine the intensity of the reaction and the energy consumption distribution, serving as an important assessment signal for the dynamics of interface-level penetration.

[0080] S302. The changes in the penetration delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window are processed in a time-series synchronization process, and a joint change sequence is constructed within the same dosing cycle. The coupling relationship of the joint change sequence is analyzed to form a conditioning failure assessment signal containing time correlation characteristics and amplitude correlation characteristics.

[0081] S303. Compare the conditioning failure assessment signal with the preset conditioning stability judgment interval. When the conditioning failure assessment signal falls into the failure judgment interval of the preset conditioning stability judgment interval within the continuous dosing cycle, it is judged that there is a risk of conditioning failure.

[0082] The key to comparing treatment failure assessment signals with preset treatment stability judgment intervals lies in establishing a numerical reference system with discriminative capabilities. First, a large amount of historical treatment data needs to be analyzed to statistically determine the numerical range, fluctuation frequency, and trend of the treatment failure assessment signal under stable treatment conditions. This constructs a numerical distribution model, which is then used to define the treatment stability judgment interval. In practical applications, the treatment failure assessment signal for the current cycle is generated in real time after each dosing cycle, and its corresponding value is matched point-by-point with the judgment interval. If the treatment failure assessment signal consistently falls within the failure judgment interval of the preset treatment stability judgment interval across multiple consecutive dosing cycles, it indicates that the signal is numerically abnormal and has failed to return to the normal fluctuation range. This suggests that the treatment behavior has deviated from the expected response path, thus indicating a risk of treatment ineffectiveness. The significance of this approach is that it captures non-linear trends through quantitative indicators, providing a concrete decision-making basis for early warning and dosing adjustments.

[0083] The preset conditioning stability judgment interval is a numerical band constructed using statistical methods, covering the entire range from normal conditioning state to failure state. It typically consists of multiple segmented intervals, including a standard judgment interval for identifying the conditioning stability range and a failure judgment interval for judging conditioning anomalies. The failure judgment interval is the set of numerical values ​​corresponding to typical abnormal characteristics such as strong fluctuations, imbalance trends, and periodic shifts in the assessment signal during the conditioning process. This interval is usually located at the upper and lower boundaries of the judgment interval or within a specific cluster, exhibiting strong anomalous signal aggregation. By setting these intervals, stratified identification of conditioning anomalies of different intensities and modes can be achieved, enhancing the accuracy and sensitivity of conditioning state judgment. When the conditioning failure assessment signal remains within the failure judgment interval, it means that key indicators in the conditioning response have shown irreversible or out-of-control characteristics, thus providing an early warning signal to the system and guiding it into a dynamic intervention process for medication strategies.

[0084] In this embodiment, S302 specifically refers to:

[0085] A unified time starting point was set for the changes in the osmotic delay expression level, the floc integrity deviation value, and the peak value of interfacial energy dissipation. The values ​​were aligned based on the time tag of the initial addition of the conditioner and time-series synchronization was performed within the same dosing cycle to ensure that the three values ​​are time-comparable within the same physical response range.

[0086] To ensure comparability of changes in osmotic delay expression levels, floc integrity deviation values, and peak interfacial energy dissipation during data analysis, a unified time starting point needs to be established for these three parameters, which have different sources and response dimensions. Specifically, the initial addition time of the conditioner can be selected as a unified time reference label, marked as the zero point of each parameter's response curve. Next, the time series recorded in their respective acquisition channels are relocated so that their starting nodes correspond to the conditioner addition event, thus completing initial alignment. Based on this, a dosing cycle range corresponding to the conditioning reaction cycle is set, for example, a 10-minute interval from complete conditioner mixing to stable reaction. Within this cycle, the data of the three parameters at the same time step are window-aligned, and paired point-by-point using a data frame synchronization algorithm to form a corresponding synchronized data point sequence. After time-series synchronization, the three types of parameters will have temporal consistency within the same physical response interval, ensuring comparability of their trends in a unified time domain, thereby providing an accurate temporal basis for subsequent coupling relationship analysis and conditioning failure risk assessment. The expression level of penetration delay reflects the degree of obstruction of conditioner penetration, the floc integrity deviation value characterizes the degree of deviation of microstructure, and the change of interfacial energy dissipation peak measures the intensity of the reaction. Without a unified time reference, the dynamic response relationship between the three cannot be revealed. Therefore, a unified time starting point and time-series synchronization processing are key means to ensure the effectiveness of data correlation.

[0087] Based on the data that has been synchronized over time, the changes in the penetration delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation are assembled into a joint change sequence in a progressive time order. The response offset features of adjacent segments in the sequence are extracted by window convolution processing to identify the dynamic linkage features between the parameters.

[0088] Based on the synchronized data, the changes in penetration delay expression, floc integrity deviation, and interface energy dissipation peak value are combined in a progressively advancing time sequence. This means that the data corresponding to the three parameters at each time point need to be linearly arranged into a unified sequence along the time axis. During this process, it is necessary to ensure that the three parameters are aligned at each moment and that the sampling time intervals are consistent. The constructed joint change sequence is essentially a multivariate time series, where each time node contains a three-dimensional data vector. To further analyze the change characteristics in this sequence, a window convolution processing method can be used. A fixed-width time window slides along the sequence, and the numerical trend of changes in penetration delay expression, floc integrity deviation, and interface energy dissipation peak value is extracted within each window. Response offset features refer to the magnitude, direction, and persistence of numerical changes between the same parameters in adjacent segments before and after the sliding window, reflecting whether there is a certain regular linkage between the parameters. For example, when penetration delay expression increases, floc integrity deviation also increases significantly, while interface energy dissipation peak value fluctuates dramatically. This coordinated trend can be identified and quantified by the convolution kernel in window convolution. By analyzing the correlation and coupling of these response offset features, the dynamic linkage characteristics among the three types of parameters during the time evolution process can be revealed, providing a mathematical basis and response mechanism basis for constructing an accurate conditioning failure assessment model. This process not only enables multi-dimensional data feature extraction but also effectively avoids the risks caused by single parameter errors, improving the robustness and reliability of subsequent conditioning failure risk assessment.

[0089] The response offset features extracted from the joint change sequence are subjected to amplitude correlation calculation and time response coupling calculation. The results are then bidirectionally superimposed to construct a conditioning failure assessment signal, forming an output index containing time correlation features and amplitude correlation features, which is used as a basis for judging the risk of subsequent conditioning failure.

[0090] The magnitude correlation and time response coupling calculations are performed on the response shift features extracted from the joint change sequence to identify the degree of correlation between the three types of parameters in terms of numerical change intensity and temporal change path. The magnitude correlation calculation analyzes the correlation between the change amplitudes of different parameters within the same time window to determine whether their fluctuations are synchronized and whether they have a consistent upward or downward trend; for example, when the expression level of osmotic delay increases sharply, the floc integrity deviation value also increases significantly, indicating a positive correlation in magnitude. The time response coupling calculation compares the differences in the response start time, response duration, and change rate of each parameter to explore whether their linkage rhythm in the time series is consistent; for example, whether a shift in one parameter will trigger a response in the other two parameters after a fixed delay. Bidirectional overlay refers to the weighted fusion of the magnitude correlation and time response coupling calculation results, considering both the value change trend and the temporal pattern of the response to generate a comprehensive evaluation result. This result is the conditioning failure assessment signal, an index sequence output in digital form, used to express the degree of potential anomalies in the conditioning process. The evaluation signal not only has time-related characteristics, such as the order of response shifts, but also amplitude-related characteristics, such as the coupling relationship of response intensity. This dual characteristic can more accurately reflect the abnormal performance of the opsonizing agent's mechanism of action, help the system determine whether there is a risk of opsonizing ineffectiveness, and thus provide a scientific basis and response basis for subsequent drug administration strategy adjustments.

[0091] S4. Based on the interval intensity distribution results of the conditioning failure assessment signal, generate the dosing behavior selection path, and adjust the particle affinity polarity, action sequence segment and mixing frequency of the conditioning agent according to the dosing behavior selection path;

[0092] In this embodiment, S4 specifically refers to:

[0093] The numerical distribution of conditioning failure assessment signals within a continuous dosing cycle was statistically analyzed in intervals. The frequency and duration of occurrence of conditioning failure assessment signals in different intensity intervals were collected and processed to obtain the interval intensity distribution results that reflect the evolution trend of conditioning status.

[0094] During conditioning, to assess the evolution of conditioning failure signals as dosing behavior changes, it is first necessary to collect time-series data on the values ​​of conditioning failure assessment signals over consecutive dosing cycles and construct signal intensity variation curves. By setting fixed intensity interval thresholds, different numerical segments are divided into multiple intensity intervals. Based on this, the frequency and duration of conditioning failure assessment signals within each intensity interval are statistically analyzed, and their repeatability and persistence characteristics in different cycles are recorded. For example, if the signal frequently occurs in a high-intensity interval and lasts for a long time during a certain dosing cycle, it indicates a significant abnormal tendency in the conditioning state during that cycle. By matrix-collecting the frequency and duration data of all intensity intervals over multiple consecutive cycles, an interval intensity distribution result reflecting the changing trend of conditioning response can be formed, thus providing a stable data foundation for subsequent generation path construction.

[0095] The numerical distribution of conditioning failure assessment signals over continuous dosing cycles refers to the set of signal values ​​at preset time points within multiple consecutive time windows, reflecting the continuous change in the conditioning system's response state. Interval statistics is the process of classifying and calculating signal values ​​according to pre-defined intensity ranges to identify the occurrence patterns of specific signal value ranges. The frequency of conditioning failure assessment signals within different intensity ranges reflects their probability of occurrence in various conditioning states, while the duration reflects the temporal stability of that state. Aggregation processing integrates the data from both frequency and duration dimensions to form a multi-cycle statistical feature vector. The resulting interval intensity distribution not only exhibits temporal evolution patterns but also reveals trends in conditioning states as a result of drug behavior, providing a quantitative basis for the fine-tuning of dosing parameters.

[0096] Based on the proportion of each intensity interval in the interval intensity distribution results, the conditioning behavior parameters are combined and rearranged to form multiple drug administration behavior parameter sequences that advance over time, and drug administration behavior selection paths are generated based on the iterative correlation between the parameter sequences.

[0097] To construct a dosing behavior adjustment strategy with closed-loop response characteristics, it is first necessary to identify the dynamic trend of conditioning state over time based on the proportion changes of each intensity interval in the interval intensity distribution results. Specifically, this can be achieved by calculating the proportion of different intensity intervals in each dosing cycle and constructing an intensity evolution trajectory based on the rate of change. Subsequently, the key nodes of conditioning instability are located based on the time periods exhibiting abnormal deviations or concentrated fluctuations in these trajectories. On this basis, conditioning behavior parameters such as particle affinity polarity, action sequence segments, and mixing frequency points are combined and rearranged to generate multiple sets of parameter combinations covering different conditioning response stages, and these combinations are organized into parameter sequences according to the time progression. Furthermore, by analyzing the synergistic improvement relationship between adjacent parameter sequences in response indicators, the optimal evolution direction is identified, and an iterative dosing behavior selection path is constructed to support the dynamic optimization of subsequent conditioning agent behavior.

[0098] The proportion of each intensity interval in the interval intensity distribution results refers to the changing pattern of the proportion of different signal intensity levels in the overall conditioning state over multiple dosing cycles, reflecting the evolution trend of conditioning response. Conditioning behavior parameters are three-dimensional regulatory variables of conditioning agents during actual action, including particle affinity polarity, action sequence segment, and mixing frequency. Combination rearrangement involves rearranging these parameters in different value ranges to form a set of conditioning behaviors covering a wider range of response possibilities. A time-progressive sequence of multiple dosing behavior parameters arranges these combinations sequentially according to the predicted conditioning response evolution trend, reflecting the phased evolution logic of future dosing strategies. The iterative correlation between parameter sequences reflects the optimized continuity of the feedback changes in conditioning failure signals after parameter adjustments. The dosing behavior selection path is a conditioning behavior adjustment trajectory ultimately determined based on the logical dependencies between the above sequences, used to dynamically guide the dosing strategy to iteratively converge towards a stable target.

[0099] Based on the parameter configuration corresponding to the generation path of the dosing behavior, the particle affinity polarity of the conditioner is adjusted by directional matching. At the same time, the action sequence of the conditioner is reorganized before and after, and the mixing frequency is synchronously corrected, so that the dosing behavior is updated and executed according to the generation path in subsequent dosing cycles.

[0100] To achieve efficient closed-loop execution of the dosing strategy under dynamic feedback, the core behavioral characteristics of the conditioning agent need to be adjusted item by item according to the parameter configuration corresponding to the dosing behavior replacement path. Specifically, this can be achieved by analyzing the feedback performance of each set of parameters in the replacement path in the conditioning failure assessment signal of the previous dosing cycle, identifying their key roles in affinity matching, segmented response, and frequency synergy. In adjusting the particle affinity polarity, conditioning agent components with corresponding affinity potential energy are selected based on the surface charge distribution of particles and the polarity characteristics of the organic coating layer to achieve directional matching adjustment. Subsequently, for the action sequence segment of the conditioning agent, the timing of different components during the addition process is adjusted to complete the recombination, thereby optimizing the adsorption and penetration synergy of the conditioning agent on the surface of the composite particles. Finally, based on the response feedback of the conditioning reaction system to the mixing disturbance frequency, the mechanical shear frequency or ultrasonic frequency during conditioning agent mixing is synchronously corrected to improve the dynamic contact efficiency between the conditioning agent and the composite particles, thereby ensuring that the dosing behavior is strictly updated and executed along the replacement path in subsequent cycles, achieving continuous optimization.

[0101] The parameter configuration in the dosing behavior selection path refers to the specific combination of conditioning agent behavior parameters at each step of the evolution of the selection path, including three dimensions: particle affinity polarity, action sequence segment, and mixing frequency. Particle affinity polarity is a measure of the degree of polarity matching between conditioning agent particles and the surface of the treated particles, determining their adsorption capacity at the interface. Directional matching adjustment involves selecting conditioning agent structures with complementary or compatible polarities based on the polarity characteristics of the target particles to improve their interfacial binding efficiency. Action sequence segment refers to the order in which different conditioning agent components are added in the dosing process; rearrangement involves reordering this sequence to adapt to the layered structure and penetration logic of the composite particle coating. The mixing frequency is the perturbation frequency parameter of the conditioning agent during the mixing process, affecting its contact dynamics with the particles; synchronous correction involves timely adjustment of the frequency based on feedback to maintain phase synchronization with the particle response frequency, ensuring a closed-loop connection in the structural, temporal, and frequency dimensions of the dosing process.

[0102] S5. By combining the changing trends of conditioning failure assessment signals and coating presence identification factors, a dosing response feedback channel is constructed, and the dosing behavior is dynamically adjusted based on the feedback channel to achieve closed-loop optimization of the dosing strategy.

[0103] In this embodiment, S5 specifically refers to:

[0104] During the continuous dosing cycle, time series data of conditioning failure assessment signals and coating presence identification factors were collected simultaneously. The direction, magnitude and rate of change of the two at the same time scale were extracted and time-aligned to form a trend combination result reflecting the correlation between conditioning state changes and coating state changes.

[0105] To achieve dynamic and coordinated analysis between changes in conditioning and coating states during continuous dosing cycles, time-series data of conditioning failure assessment signals and coating presence identification factors can be acquired using a high-frequency sensor array. During acquisition, the dosing trigger signal is used as a unified starting point marker, and the two data sequences are updated synchronously on a unified time scale. In the acquired time series, the instantaneous numerical change direction at each time point is extracted, the change amplitude within adjacent time periods is calculated, and the change rate parameter is obtained by combining the change within a unit time span. Subsequently, the two data sequences are time-aligned, and a window pairing method is used to identify synchronous response patterns within the same time period, thereby constructing a set of trend combination results corresponding to a set of time points. This trend combination result can capture whether the physicochemical response generated by conditioning behavior is synchronous with or deviates from the changes in particulate surface coating state, thus providing basic data correlation for subsequent feedback channel construction.

[0106] Time-series data of conditioning failure assessment signals and coating presence identification factors reflect the evolution of the conditioning agent's influence on floc structure and the degree of organic coating on particulate matter surface, respectively. The direction of change indicates the positive or negative reversal of the signal trend, the magnitude of change indicates the absolute order of magnitude of signal increase or decrease, and the rate of change indicates the speed of change per unit time. Time alignment processing refers to logically pairing the two signals at each time node based on a unified time label, ensuring comparability at any given time. The trend combination result involves vector concatenation or fusion calculation of the three change characteristics of the conditioning and coating signals, ultimately forming a composite dynamic information unit that simultaneously describes the evolution of both states. This unit reveals the logical relationship between the source of response deviation and the feedback target in actual operation of the dosing behavior.

[0107] Based on the trend combination results, a dosing response feedback channel is constructed. The trend change of conditioning failure assessment signal is used as the conditioning risk input dimension, and the trend change of the identified factors are used as the coverage status input dimension. The correspondence between trend input and dosing behavior parameters is established in the same mapping space, forming a feedback channel structure that can be updated over time.

[0108] In the process of drug dosing control, to achieve closed-loop optimization of the conditioning strategy, a drug dosing response feedback channel can be constructed based on trend combination results. Specifically, the trend change index of the conditioning failure assessment signal is first used as the conditioning risk input dimension, and the trend change index encompassing the identification factors is used as the encompassing state input dimension. A continuous mapping model is then performed on the historical trend data of these two dimensions. By establishing a dual-input mapping space with time as the evolution axis, a matching relationship is constructed between the trend input state at each moment and the response results of historical drug dosing behavior parameters. In this mapping space, multiple state blocks are formed using clustering or fitting learning, with each block associated with a set of drug dosing behavior parameters. As the input trend state migrates in the space, automatic switching between different drug dosing behavior parameters is achieved, thereby constructing a feedback channel structure that can dynamically respond over time, thus providing a logical driving basis for subsequent drug dosing adjustments.

[0109] The dosing response feedback channel is a control structure built upon the mapping relationship between trend inputs and output responses. The trend change of the conditioning failure assessment signal refers to its continuous increase, decrease, or oscillation over time, serving as a sensitive reflection of system stability. The conditioning risk input dimension is a risk judgment parameter formed by standardizing this trend value. The trend change of the coating presence identification factor reflects the stability evolution of the particle surface state and is transformed into a coating state input dimension to participate in the mapping construction. The correspondence between trend inputs and dosing behavior parameters is reflected in the mapping of a certain combination of input trends to a set of conditioning agent behavior configurations, including particle affinity polarity, action sequence segments, and mixing frequencies. The time-updable feedback channel structure means that this mapping relationship has real-time evolution capabilities, enabling adaptive adjustment of parameter recommendation results after data updates, thereby supporting the dynamic optimization decision-making of the dosing system based on state changes during actual operation.

[0110] Based on the trend response results output in the dosing response feedback channel, the dosing behavior is dynamically regulated so that the particle affinity polarity, action sequence segment and mixing frequency point in subsequent dosing cycles are adjusted synchronously with the trend changes, thereby realizing the adaptive update of dosing behavior in continuous cycles.

[0111] In the adaptive control of conditioning behavior, the conditioning agent dosing behavior can be dynamically adjusted based on the trend response results output from the dosing response feedback channel. Specifically, in the feedback channel, the trend response result at each moment is determined by the joint trend input of the conditioning failure assessment signal and the coating presence identification factor. After mapping transformation, a set of dosing behavior adjustment parameters is output. This set of parameters includes the particle affinity polarity matching level, the order of action sequences, and the correction value for the mixing frequency. The system updates the dosing strategy for the next dosing cycle based on the current feedback results. For example, when the trend response results show an increasing risk trend and an enhanced coating trend, the system automatically increases the polarity matching level of the conditioning agent, adds components affecting interfacial energy release in the action sequence segments earlier, and lowers the mixing frequency to prolong the reaction contact time. This process enables real-time updates of dosing behavior and self-integration of parameters during continuous operation.

[0112] The trend response results output from the dosing response feedback channel are a comprehensive expression of the changes in conditioning and coating states over continuous cycles. The output parameter configuration is typically determined by the position of the trend combination result in the mapping space. Particle affinity polarity refers to the interfacial affinity matching ability between conditioning agent particles and treated particles, which is achieved by adjusting the composition ratio of polar groups in the conditioning agent. The action sequence segment refers to the specific order in which the functional components of the conditioning agent are added; different sequences affect the interfacial guidance and structure formation mechanism of the conditioning reaction. The mixing frequency is the oscillation frequency setting when the conditioning agent and waste liquid undergo shear mixing after addition, affecting particle contact and surface penetration processes. Synchronous adjustment of these three factors ensures a high degree of adaptability between the dosing strategy and the actual conditioning state in each cycle, thereby improving the conditioning efficiency and response stability of the entire system.

[0113] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0114] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0115] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0117] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0119] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for efficient pretreatment of low-calorific-value waste liquid containing particulate matter, characterized in that, Specifically, the following steps are included: S1. By acquiring the surface viscous displacement trajectory, particle size fluctuation phase and interface refractive properties of particulate matter in waste liquid, a coating presence identification factor is established, and the coating presence identification factor is used to determine whether particulate matter in waste liquid is coated by low-calorific-value organic components. S2. When particulate matter is coated with low-calorific-value organic components, based on the dynamic trajectory behavior of the recognition factor in the coating, the initial foaming hysteresis time and the inhibition amplitude of particle deformation rate generated by the conditioning reaction are extracted to generate the penetration delay expression level, which is used to characterize the conditioner's penetration barrier performance. S3. Combine the delayed expression level of penetration with the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window to form a conditioning failure assessment signal, which is used to determine whether there is a risk of conditioning failure. S4. Based on the interval intensity distribution results of the conditioning failure assessment signal, generate the dosing behavior selection path, and adjust the particle affinity polarity, action sequence segment and mixing frequency of the conditioning agent according to the dosing behavior selection path; S5. By combining the changing trends of conditioning failure assessment signals and coating presence identification factors, a dosing response feedback channel is constructed, and the dosing behavior is dynamically adjusted based on the feedback channel to achieve closed-loop optimization of the dosing strategy.

2. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 1, characterized in that, S1 specifically includes the following steps: S101. By introducing oscillation disturbance into the waste liquid, the surface viscous displacement response path of particulate matter under disturbance conditions is collected, and the particle size fluctuation phase of particulate matter in the flow cycle is obtained through a particle size distribution tracking device. At the same time, the interface refractive characteristic data of particulate matter and liquid phase boundary are extracted using a multi-angle refractive scanning device. S102. Time synchronization and unit standardization are performed on the surface viscous displacement response path, particle size fluctuation phase and interface refractive property data. Based on the dynamic response coupling relationship between the processed surface viscous displacement response path, particle size fluctuation phase and interface refractive property data, a coating existence identification factor is constructed to reflect the changing characteristics of the surface state of particles. S103. Compare the output value of the coating presence identification factor with the preset coating judgment interval. When the output value continuously falls into the preset coating judgment interval, it is determined that the particulate matter in the waste liquid is coated by low-calorific-value organic components.

3. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 2, characterized in that, S102 specifically refers to: A unified time start point was set for the surface viscous displacement response path, particle size fluctuation phase and interface refractive characteristic data respectively, and time synchronization was achieved by data window alignment to ensure that various types of data are comparable within the same physical response period; Physical unit conversion factors are introduced into the time-synchronized data, and normalization is performed based on the maximum and minimum values ​​of the response dimension to ensure that the data of different physical quantities have a unified calculation scale and eliminate the interference of the original units on the subsequent analysis results. The three types of data, after time synchronization and unit standardization, are input into a nonlinear cross-convolution analysis process to extract the response offset features in the interaction effect term. Based on the change in offset amplitude, a coating existence identification factor is constructed to identify the coupled evolution characteristics of particulate matter surface state.

4. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 1, characterized in that, S2 specifically refers to: When particulate matter is coated with low-calorific-value organic components, the dynamic trajectory behavior of the coating recognition factor is continuously collected. The time point of conditioner addition is used as the starting point of analysis to identify the first moment when the coating recognition factor changes from a steady state to a fluctuating state. The time interval between this moment and the starting point of analysis is calculated to determine the initial hysteresis time. Based on the dynamic trajectory behavior of the coating presence identification factor, the response path of particulate morphology change during conditioning reaction is continuously recorded, and the difference between the morphology change rate under the coating state and the morphology change rate under the preset benchmark state is compared to extract the particle deformation rate suppression amplitude. The initial hysteresis time and the particle deformation rate inhibition amplitude are uniformly scaled and weighted according to the temporal correlation and amplitude coupling relationship between the two within the same conditioning reaction cycle to form a penetration delay expression quantity characterizing the degree of conditioning reaction response lag, so as to reflect the penetration limitation characteristics of conditioning agent on the surface of coated particles.

5. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 1, characterized in that, S3 specifically includes the following steps: S301. Within the dosing window, continuously collect the morphology of the flocs formed during the conditioning reaction, extract the changes in the projected area of ​​the flocs, the changes in the boundary integrity, and the changes in the structural connectivity, and compare the collected results with the reference morphology under the preset intact state to obtain the floc integrity deviation value, which reflects the degree of deviation of the floc structure; and synchronously record the energy changes at the particle-liquid interface during the conditioning reaction, extract the interface energy change curve over time, and identify the peak position and amplitude changes in the energy change curve to obtain the change in the peak value of interface energy dissipation. S302. The changes in the penetration delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation within the dosing window are processed in a time-series synchronization process, and a joint change sequence is constructed within the same dosing cycle. The coupling relationship of the joint change sequence is analyzed to form a conditioning failure assessment signal containing time correlation characteristics and amplitude correlation characteristics. S303. Compare the conditioning failure assessment signal with the preset conditioning stability judgment interval. When the conditioning failure assessment signal falls into the failure judgment interval of the preset conditioning stability judgment interval within the continuous dosing cycle, it is judged that there is a risk of conditioning failure.

6. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 5, characterized in that, S302 specifically refers to: A unified time starting point was set for the changes in the osmotic delay expression level, floc integrity deviation value and the peak value of interfacial energy dissipation. The results were aligned based on the time tag of the first addition of the conditioner and time-series synchronization was performed within the same dosing cycle. Based on the data that has been synchronized over time, the changes in the permeation delay expression level, the floc integrity deviation value and the peak value of interface energy dissipation are assembled into a joint change sequence in a progressive time order. The response offset features of adjacent segments in the sequence are extracted by window convolution processing to identify the dynamic linkage features between the parameters. The response offset features extracted from the joint change sequence are subjected to amplitude correlation calculation and time response coupling calculation. The results are then bidirectionally superimposed to construct a conditioning failure assessment signal, forming an output index containing time correlation features and amplitude correlation features, which is used as a basis for judging the risk of subsequent conditioning failure.

7. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 1, characterized in that, S4 specifically refers to: The numerical distribution of conditioning failure assessment signals within a continuous dosing cycle was statistically analyzed in intervals. The frequency and duration of occurrence of conditioning failure assessment signals in different intensity intervals were collected and processed to obtain the interval intensity distribution results that reflect the evolution trend of conditioning status. Based on the proportion of each intensity interval in the interval intensity distribution results, the conditioning behavior parameters are combined and rearranged to form multiple drug administration behavior parameter sequences that advance over time, and drug administration behavior selection paths are generated based on the iterative correlation between the parameter sequences. Based on the parameter configuration corresponding to the generation path of the dosing behavior, the particle affinity polarity of the conditioner is adjusted by directional matching. At the same time, the action sequence of the conditioner is reorganized before and after, and the mixing frequency is synchronously corrected, so that the dosing behavior is updated and executed according to the generation path in subsequent dosing cycles.

8. The efficient pretreatment method for low-calorific-value wastewater containing particulate matter according to claim 1, characterized in that, S5 specifically refers to: During the continuous dosing cycle, time series data of conditioning failure assessment signals and coating presence identification factors were collected simultaneously. The direction, magnitude and rate of change of the two at the same time scale were extracted and time-aligned to form a trend combination result reflecting the correlation between the changes in conditioning state and the changes in coating state. Based on the trend combination results, a dosing response feedback channel is constructed. The trend change of conditioning failure assessment signal is used as the conditioning risk input dimension, and the trend change of the identified factors is used as the coverage status input dimension. The correspondence between trend input and dosing behavior parameters is established in the same mapping space, forming a feedback channel structure that can be updated over time. Based on the trend response results output in the dosing response feedback channel, the dosing behavior is dynamically regulated so that the particle affinity polarity, action sequence segment and mixing frequency in subsequent dosing cycles are adjusted synchronously with the trend changes, thereby achieving adaptive updating of dosing behavior in continuous cycles.

Images