Composite intelligent production method and system of liquid polyaluminum chloride sulfate

Abstract

The invention discloses a composite intelligent production method and system of liquid polyaluminum chloride sulphide, and belongs to the technical field of liquid polyaluminum chloride sulphide production.The method specifically comprises the steps that a sodium aluminate solution and hydrochloric acid are mixed to generate an aluminum oxyhydroxide precursor solution, and an initial pH track model is constructed; dropwise adding a sulfur source solution into the aluminum oxychloride precursor solution, obtaining a first reaction lag time according to the initial pH track model, identifying a sulfur-aluminum coordination ion structure formed in the reaction, extracting the dynamic evolution characteristics of the sulfur-aluminum coordination ion structure, obtaining a second reaction lag time, and obtaining a second reaction lag time according to the second reaction lag time; adjusting the feeding rate of the sulfur source and the alkalization degree based on the first reaction lag time and the second reaction lag time, and producing liquid polyaluminum chloride sulfate based on the adjusted feeding rate of the sulfur source and the alkalization degree; according to the invention, the generation efficiency and quality of the liquid polyaluminum chloride sulfate are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of liquid polysulfur aluminum chloride production, in particular to a composite intelligent production method and system of liquid polysulfur aluminum chloride. BACKGROUND

[0002] Liquid polysulfur aluminum chloride is a new type of efficient inorganic polymer flocculant, which is synthesized by introducing sulfur compounds into the aluminum salt reaction system, and has the molecular characteristics of high charge density and multi-tooth sulfur coordination structure. This material shows better performance than traditional flocculants in water treatment, heavy metal capture, and high-difficulty organic wastewater treatment.

[0003] Unlike ordinary liquid polyaluminum chloride, the synthesis process of liquid polysulfur aluminum chloride involves complex coordination reactions of aluminum ions and polyvalent sulfur anions, forming Al-S, Al-SO4, Al-S-OH and other bridging or chelating structures. This kind of sulfur coordination complex system gives it stronger electric neutralization and bridging adsorption capacity, but also has the following problems: strong uncertainty of sulfur-aluminum coordination reaction, sensor data lag caused by reaction time misplacement, etc. SUMMARY

[0004] In view of the deficiencies of the prior art, the application provides a composite intelligent production method and system of liquid polysulfur aluminum chloride.

[0005] To achieve the above purpose, the application provides the following technical scheme:

[0006] A composite intelligent production method of liquid polysulfur aluminum chloride, comprising:

[0007] Mixing sodium aluminate solution and hydrochloric acid to generate aluminum oxychloride precursor solution and building an initial pH trajectory model;

[0008] Adding sulfur source solution to the aluminum oxychloride precursor solution, and obtaining a first reaction lag time according to the initial pH trajectory model;

[0009] Identifying the sulfur aluminum coordination ion structure formed in the reaction and extracting its dynamic evolution characteristics to obtain a second reaction lag time;

[0010] Adjusting the sulfur source addition rate and alkalinity based on the first reaction lag time and the second reaction lag time, and producing liquid polysulfur aluminum chloride based on the adjusted sulfur source addition rate and alkalinity.

[0011] Specifically, the mixing of sodium aluminate solution and hydrochloric acid to generate aluminum oxychloride precursor solution and building an initial pH trajectory model comprises:

[0012] The hydrochloric acid is injected into the reaction tank in a reaction with the sodium aluminate solution in an intermittent wave form through an acid injection interface, and the pulse interval is adjusted according to the pH change rate of the previous cycle;

[0013] The pH signal of the reaction solution in the reaction tank is acquired through a multi-point time sequence sampling mode, and a pH response matrix is established;

[0014] Based on the pH response matrix and the evolution law under different initial proportions and temperature conditions, a plurality of trajectory curves are established, each trajectory curve corresponding to an evolution law;

[0015] The initial proportion and temperature condition are matched with the initial proportion and temperature condition corresponding to each evolution law, and based on the matching result, the trajectory curve corresponding to the current evolution law is selected from the plurality of trajectory curves, and is used as the initial pH trajectory model.

[0016] Specifically, the sulfur source solution is added dropwise to the aluminum oxychloride precursor solution to obtain a first reaction lag time, including:

[0017] A disturbance flow state window is established, a liquid phase shear response boundary is identified, and the initial addition starting point of the sulfur source solution is determined in combination with the initial pH trajectory model;

[0018] The addition is performed using a delay trigger dropwise, and is performed using a timing pulse with random noise disturbance loading, so that the addition behavior presents a non-periodic waveform;

[0019] A hydrogen sulfide gas channel is guided at the top of the reaction tank to perform gas chromatography-mass spectrometry dual mode analysis in a one-step circulation mode, and a gas concentration-time response curve is established;

[0020] In combination with a preset reference response envelope, the timestamp corresponding to the gas mutation inflection point is set as the first reaction lag time t1.

[0021] Specifically, the sulfur aluminum coordination ion structure formed in the reaction is identified and the dynamic evolution characteristics thereof are extracted to obtain a second reaction lag time, including:

[0022] A multi-section wavelength scanning sequence is excited in the reaction solution of the aluminum oxychloride precursor solution and the sulfur source solution to generate a transient response spectrum set of aluminum coordination structures, and a snapshot data packet is established after each scanning;

[0023] The snapshot data packet is converted into a two-dimensional spectrum variable trajectory graph in chronological order, a sulfur aluminum coordination peak cluster with absorption displacement behavior is extracted, and its initial appearance time, maximum intensity time and disappearance time are marked to generate a candidate coordination state time node cluster;

[0024] The candidate coordination state time node cluster is compared with a preset sulfur aluminum coordination model in spectrum shape, whether there is a cross-state mutation signal or a reconstruction signal is identified, and the evolution path of the coordination structure is derived.

[0025] The time point at which the first spectrum shape switching behavior occurs in the evolution path is defined as the second reaction lag time t2, and the coordination state category corresponding to the time point is marked.

[0026] Specifically, the adjusting the sulfur source addition rate and alkalinity based on the first reaction lag time and the second reaction lag time includes:

[0027] Collect the time series data streams of Raman spectroscopy, electrochemical impedance spectroscopy and microwave dielectric changes, and generate independent parameter sequence matrices for each;

[0028] Based on the parameter sequence matrix, a three-dimensional spatial reaction parameter matrix under a unified time axis is constructed, and each matrix unit is set to correspond to a micro-time slice and a physical property group;

[0029] Inputting the three-dimensional spatial reaction parameter matrix into a structural inversion neural network based on an attention weight mechanism, identifying the evolution path of the sulfur-aluminum coordination state, and outputting its current structural state, future evolution direction, and associated hysteresis label;

[0030] According to the first reaction lag time t1 and the second reaction lag time t2, a lag prediction window is embedded in the coordination evolution output to generate a lag intervention factor set;

[0031] Real-time monitoring of the concentration of ferrous sulfide impurities in the sulfur source raw materials, and dynamic adjustment of the reaction temperature and target alkalinity range based on its changing trend;

[0032] Based on the set of delayed intervention factors, temperature and target alkalinity range, a weight distribution table between the sulfur source addition rate and the alkalinity is constructed, the sulfur source addition rate and the alkalinity are regulated, and liquid polyaluminum sulfide chloride is produced based on the regulated sulfur source addition rate and the alkalinity.

[0033] Specifically, the three-dimensional spatial reaction parameter matrix is ​​input into a structural inversion neural network based on an attention weight mechanism to identify the evolution path of the sulfur-aluminum coordination state, and output its current structural state, future evolution direction, and associated hysteresis label, including:

[0034] Based on the three-dimensional spatial response parameter matrix, a multi-scale data view set is constructed, and a unique identity index is set for each view;

[0035] A set of multi-scale data views is fed in parallel into an inverse deep neural network structure built on an attention weight mechanism to generate a cross-focus map. The network structure scans the weight distribution of each time segment of each view in parallel through attention heads.

[0036] Identify the key parameter mutation node in the cross-focus map, concatenate it into the candidate coordination track path in time sequence, and calculate the relative offset degree and time overlap rate between the paths to form the evolution path map cluster;

[0037] Evolution consistency test is carried out on the evolution path map cluster, the evolution trend of each path is labeled by drift, stability and mutation three-state labels, and the current coordination structure state and future evolution direction are output;

[0038] Identify the index time slice where the first state variation occurs in the future evolution direction, and define it as the associated lag label.

[0039] Specifically, the lag prediction window is embedded in the coordination evolution output according to the first reaction lag time t1 and the second reaction lag time t2, and a set of lag intervention factors is generated, including:

[0040] The first reaction lag time t1 and the second reaction lag time t2 are embedded into the coordination evolution trend vector respectively, and the interval Δt=t2-t1 is used to construct the lag interval reference index as the boundary labeling parameter of the lag prediction window;

[0041] Local attention weight scanning is performed on the coordination evolution trend vector within the lag prediction window, and a time-weighted sensitivity curve is constructed by integrating the weight of each time slice, which is used to describe the intensity of the reaction parameter on the structure evolution during the lag period;

[0042] According to the time-weighted sensitivity curve, the center position of the local gradient change rate maximum region is extracted and defined as the lag response center index, and a symmetric expansion interval is generated starting from the lag response center index to form an intervention envelope cluster;

[0043] The intervention envelope cluster is mapped to the lag factor, and a set of lag intervention factors is generated according to its distribution form, each factor in the set of lag intervention factors contains a label parameter.

[0044] Specifically, the concentration of ferrous sulfide impurities in the sulfur source raw material is monitored in real time, and the reaction temperature and target alkalinity interval are dynamically adjusted according to the change trend, including:

[0045] Collect the concentration characteristic data of ferrous sulfide in the sulfur source raw material, and compare it with the preset impurity mode to judge its fluctuation trend;

[0046] Based on the fluctuation trend, determine the reaction deviation area that appears in the future reaction process, and adjust the temperature regulation time period to construct the corresponding temperature regulation strategy;

[0047] According to the impurity fluctuation trend and the temperature regulation strategy, dynamically correct the alkalinity interval.

[0048] The application discloses a composite intelligent production system of liquid polysulfur aluminum chloride, which is used for realizing a composite intelligent production method of the liquid polysulfur aluminum chloride.

[0049] The pH trajectory construction module is used for mixing a sodium aluminate solution and hydrochloric acid to generate an aluminum oxychloride precursor solution and constructing an initial pH trajectory model.

[0050] The first time acquisition module is used for adding a sulfur source solution dropwise into the aluminum oxychloride precursor solution and acquiring a first reaction lag time according to the initial pH trajectory model.

[0051] The second time acquisition module is used for identifying a sulfur aluminum coordination ion structure formed in the reaction and extracting dynamic evolution characteristics of the sulfur aluminum coordination ion structure to acquire a second reaction lag time.

[0052] The adjusting production module adjusts a sulfur source addition rate and alkalinity based on the first reaction lag time and the second reaction lag time, and produces the liquid polysulfur aluminum chloride based on the adjusted sulfur source addition rate and alkalinity.

[0053] Specifically, the adjusting production module comprises a parameter sequence matrix construction unit, an evolution path output unit, a distribution condition adjustment unit and an adjusting production unit.

[0054] The parameter sequence matrix construction unit is used for collecting time sequence data streams of Raman spectroscopy, electrochemical impedance spectroscopy and microwave dielectric changes, and respectively generating independent parameter sequence matrices.

[0055] The evolution path output unit is used for identifying an evolution path of a sulfur aluminum coordination state, and outputting a current structure state, a future evolution direction and a related lag label of the evolution path.

[0056] The distribution condition adjustment unit is used for generating a lag intervention factor set and adjusting a reaction temperature and a target alkalinity interval.

[0057] The adjusting production unit constructs a weight distribution table between the sulfur source addition rate and the alkalinity based on the lag intervention factor set, the temperature and the target alkalinity interval, adjusts the sulfur source addition rate and the alkalinity, and produces the liquid polysulfur aluminum chloride based on the adjusted sulfur source addition rate and alkalinity.

[0058] Compared with the prior art, the application has the following beneficial effects:

[0059] The application provides a composite intelligent production method and system of liquid polysulfur aluminum chloride, which predicts the initial state of reaction by constructing a pH trajectory model, guides sulfur source dropwise addition, extracts double lag time t1 and t2 in combination with sulfur aluminum coordination structure evolution, constructs a three-dimensional reaction parameter matrix by fusing Raman spectrum, electrochemical impedance and microwave dielectric data, generates a set of lag intervention factors, and realizes the regulation and control of sulfur source dropwise addition rate and alkalinity by real-time monitoring of raw material impurity trend and dynamically adjusting the target interval of reaction temperature and alkalinity. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 A composite intelligent production method flow chart of liquid polysulfur aluminum chloride is provided in the application.

[0061] Figure 2 A two-dimensional spectrum variable trajectory of sulfur aluminum coordination structure is provided in the application.

[0062] Figure 3 A Raman spectrum in reaction is provided in the application.

[0063] Figure 4 An electrochemical impedance spectrum in reaction is provided in the application.

[0064] Figure 5 A microwave dielectric change in reaction is provided in the application.

[0065] Figure 6 A composite intelligent production system architecture diagram of liquid polysulfur aluminum chloride is provided in the application. DETAILED DESCRIPTION

[0066] The application will be described in detail below with specific embodiments. The following examples will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be noted that for those skilled in the art, without departing from the concept of the application, a number of modifications and improvements can be made. These all belong to the protection scope of the application.

[0067] In order to make the purpose, technical scheme and advantages of the application more clear and obvious, the application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the application and do not limit the application.

[0068] It should be noted that the various features of the embodiments of the present application can be combined with each other, and are within the scope of the present application. In addition, although the functional modules are divided in the device schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than the module division in the device or the order in the flowchart. In addition, the "first", "second", "third" and the like used in the present application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same function and effect.

[0069] Unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used in the present application are only for the purpose of describing the specific embodiments of the present application, and are not used to limit the present application. The term "and / or" used in the present application includes any and all combinations of one or more related listed items.

[0070] Embodiment 1

[0071] Please refer to Figure 1 The present application provides an embodiment: a composite intelligent production method of liquid polythioaluminate, comprising the following specific steps:

[0072] Step S1: mixing sodium aluminate solution with hydrochloric acid to generate aluminum oxychloride precursor solution and constructing initial pH trajectory model.

[0073] The specific steps of step S1 are:

[0074] Step S101: through the acid injection interface, the hydrochloric acid is injected into the reaction tank in the form of intermittent wave and reacts with the sodium aluminate solution, and the pulse interval is adjusted according to the pH change rate of the previous period solution.

[0075] In the present embodiment, during the synthesis of liquid polythioaluminate, a sodium aluminate solution with a concentration of 20% is prepared in advance, the temperature in the reaction tank is kept at 45°C, a three-way acid injection valve is used to inject 31% hydrochloric acid into the reaction tank in a pulse manner, the duration of each acid injection is set to 2 seconds, the initial interval is 15 seconds, the pH change of the reaction solution is monitored in real time at a sampling frequency of 0.5 seconds, and the pH change rate in the current period is calculated at the end of each acid injection period. When the pH change rate of two consecutive periods is less than 0.03 pH units / second, the next acid injection interval is shortened to 12 seconds; if the change rate is greater than 0.08, it is automatically extended to 18 seconds. The entire acid injection process lasts for 8 minutes, during which the pH curve decreases in steps, finally tends to a slow and stable transition interval, and forms a homogeneous aluminum oxychloride precursor solution.

[0076] Step S102: Obtain the pH signal of the reaction liquid in the reaction tank by a multi-point time sequence sampling method, and establish a pH response matrix.

[0077] In this embodiment, in the process of generating the aluminum oxychloride precursor solution by reacting sodium aluminate with hydrochloric acid, high-resolution glass pH electrodes are arranged on the wall surface of the reaction tank, respectively at the inlet, the center turbulent flow region, the vicinity of the outlet, a preset distance below the liquid level, and the bottom sedimentation zone. Each electrode collects a pH value at a preset interval and continuously records the time sequence data. All sampling values are arranged with time as the vertical axis and spatial measurement points as the horizontal axis to generate a two-dimensional pH response matrix. The matrix gradually shows the trend of weakening vertical gradient and converging horizontal distribution with the reaction process, reflecting the dynamic homogenization process of acid-base reaction in the spatial domain.

[0078] Under non-ideal mixing conditions, the pH change of acid-base reaction shows a non-uniform diffusion process with obvious local disturbance response difference. The two-dimensional matrix can not only judge the strength and phase difference of local pH fluctuation, but also analyze the coincidence degree of acid diffusion front and aluminum ion conversion zone in the reaction zone.

[0079] Step S103: Based on the pH response matrix and the evolution law under different initial proportions and temperature conditions, a plurality of trajectory curves are established, each corresponding to an evolution law.

[0080] In this embodiment, in the experiment under the initial proportion and temperature conditions, a plurality of pH matrices are obtained. Each matrix is grouped and processed according to its corresponding temperature and molar ratio label. The average pH value, maximum gradient interval position and standard deviation curve at each time point during the reaction process are extracted. Based on these characteristics, a plurality of trajectory curves are induced.

[0081] Step S104: Match the initial proportion and temperature conditions with the initial proportion and temperature conditions corresponding to each evolution law, and select the trajectory curve corresponding to the current evolution law from the plurality of trajectory curves based on the matching result, and use it as the initial pH trajectory model.

[0082] Step S2: Add a sulfur source solution to the aluminum oxychloride precursor solution, and obtain a first reaction lag time according to the initial pH trajectory model.

[0083] The specific steps of step S2 are as follows:

[0084] Step S201: Establish a disturbance flow state window, identify the liquid phase shear response boundary, and determine the initial addition starting point of the sulfur source solution in combination with the initial pH trajectory model.

[0085] In this embodiment, in the reaction tank, the stirring paddle driven by frequency conversion is used to apply periodic disturbance in the reaction stage of sodium aluminate and hydrochloric acid. By arranging the shear sensitive conductivity probe in the reaction liquid, the response time difference and amplitude of the local liquid phase conductivity value under the change of disturbance frequency are recorded, the periodic window is set, and the phase delay value of the liquid response is extracted in different stirring frequency intervals. When the disturbance frequency is greater than the critical threshold, the conductivity response tends to be synchronized, indicating that the system enters the critical zone of shear homogenization. At the same time, the pH trajectory collected in this shear window is compared with the previously constructed trajectory model to determine whether the pH of the reaction system has entered the non-buffer transition section, and the initial dosing point of the sulfur source is determined when the disturbance frequency is maintained at the critical threshold and the pH decline rate is higher than the threshold.

[0086] By periodic disturbance stirring, the macroscopic response characteristics of the liquid phase are exposed without changing the chemical composition. The delay of conductivity response caused by shear disturbance is related to the consistency of the distribution of reactants at the microscale. When the delay decreases to a critical value, the surface system crosses from the mixing lag zone to the reaction stable zone. By combining the identified inflection point section in the pH trajectory model, the physical disturbance behavior and the chemical evolution path are cross-located to determine the timing of the intervention of the sulfur source.

[0087] Step S202: Use delay trigger dropping to add a timing pulse with random noise disturbance, so that the dropping behavior presents a non-periodic waveform.

[0088] In this embodiment, during the dropping process of the sulfur source (sodium sulfide), the reference dropping period is set, and a random delay disturbance is introduced before each period. The delay value is randomly generated by Gaussian distribution. After each trigger, the pump opening time is 1.5 seconds, and the flow rate is adjusted according to the reaction pH fluctuation trend of the last period, with a maximum deviation amplitude controlled within ±10%. During the dropping process, it is monitored that the dropping flow rate does not have obvious periodic repetition within the preset time window, and the fluid dynamic behavior shows non-stable disturbance characteristics. By tracking the Raman spectrum response of aluminum-sulfur coordination structure, this non-periodic dropping method expands the dynamic adjustable interval of the reaction path.

[0089] Step S203: Direct hydrogen sulfide gas channel at the top of the reaction tank to perform gas chromatography-mass spectrometry dual mode analysis in a one-step circulation mode, and establish a gas concentration-time response curve.

[0090] In this embodiment, hydrogen sulfide is added as a byproduct during the reaction, and its gas phase release behavior is coupled with the rate of reaction of sulfur ions in the liquid phase and the coordination competition state. Due to the influence of pH and aluminum ion complexing degree on the solubility of hydrogen sulfide in the liquid phase, the escape rate shows obvious time lag characteristics; through the gas chromatography-mass spectrometry dual mode analysis of the gas channel, the separation and molecular level qualitative identification of the mixed gas components are realized, and the interference of other volatile components on the determination is avoided; by using one-step cyclic sampling method, a high-time-efficiency concentration response sequence can be constructed without interrupting the main reaction process, and the response curve reflects the rate change of hydrogen sulfide release.

[0091] Step S204: In combination with the preset reference response envelope, the time stamp corresponding to the gas mutation inflection point is set as the first reaction lag time t1.

[0092] In this embodiment, after the sulfur source is added and the hydrogen sulfide gas is collected, the obtained gas concentration-time response curve is compared with the standard reference envelope established in advance. The envelope is the upper and lower envelope formed by the hydrogen sulfide response trajectories collected under n different drop adding rhythms and temperature conditions, which contains the fluctuation range and inflection point interval of the expected reaction window. The envelope matching method is used to extract the change segment in the test curve where the instantaneous rising rate is greater than 30% of the reference average rising rate for the first time and lasts for three time points, and the starting time of the segment is marked as the first reaction lag time t1.

[0093] Since the reaction lag characteristic has a working condition dependence, a single peak or rate is not enough as a universal index. By time mapping the current response curve with the envelope, the abnormal interval is identified without relying on explicit function fitting, and then the first reaction lag time t1 is calibrated by matching the change rate with the structural response critical point.

[0094] Step S3: Identify the sulfur-aluminum coordination ion structure formed in the reaction and extract its dynamic evolution characteristics to obtain the second reaction lag time.

[0095] The specific steps of step S3 are:

[0096] Step S301: In the reaction solution of aluminum hydroxide chloride precursor solution and sulfur source solution, a multi-section wavelength scanning sequence is excited to generate a set of instantaneous response spectra of aluminum coordination structure, and a snapshot data package is established after each scanning.

[0097] In this embodiment, after each group of scanning is completed, the corresponding scanning time, waveband interval, peak intensity distribution, baseline offset parameter and sampling position information are automatically packaged to form a set of structured snapshot data packages. The spectrum set generated during the whole reaction process is used to track the configuration evolution of sulfur-aluminum coordination structure at different time points.

[0098] During the reaction of sulfur source and aluminum hydroxyl chloride, the coordination behavior of aluminum ion and sulfide anion will form dynamic Al-S, Al-SO3 bridge structure, and its vibration characteristics in Raman spectrum is a series of specific instantaneous peak clusters in waveband. Due to the short life cycle and reversibility of these coordination structures, it is necessary to capture the whole process of formation, stability and disappearance through multi-wavelength and high time density scanning mode. The spectrum set obtained by scanning is essentially the joint projection of reaction state in frequency domain and time domain. In order to retain the original structure response information, each set of scanning results is packaged as a snapshot data packet, and then spliced in time sequence, mode extracted and spectrum shape classified, finally the dynamic evolution of coordination structure is reconstructed.

[0099] As shown in Figure 2 , step S302: converting the snapshot data packet into a two-dimensional spectrum variable trajectory graph in time sequence, extracting the sulfur aluminum coordination peak cluster with specific absorption displacement behavior, and marking the initial appearance time, maximum intensity time and disappearance time to generate a candidate coordination state time node cluster.

[0100] In this embodiment, as shown in Figure 2 , the snapshot data set is integrated in the order of scanning time axis to form a two-dimensional spectrum variable trajectory graph, with wave number as horizontal axis, reaction time as vertical axis and Raman intensity as color scale. The wave number unit is cm -1 , the reaction time unit is min and the Raman intensity unit is a.u., which is used for relative comparison of coordination structure change trend in the reaction process. Two-dimensional peak domain extraction is performed on the image to identify three peak cluster regions with obvious absorption evolution characteristics, which are concentrated in wave number about 500cm -1 , 650-700cm -1 and 1100-1130cm -1 . The time nodes of first appearance, peak intensity corresponding time and complete disappearance of each peak cluster along the time axis direction are extracted. On this basis, the candidate coordination state time node cluster is constructed, and the candidate label is given according to waveband classification.

[0101] The snapshot data is presented in the form of two-dimensional image, which can significantly enhance the visual recognition ability of spectral peak position drift, intensity change and overlapping interference. The absorption behavior of specific coordination structure in the process of formation, evolution and dissociation often shows peak value movement, energy bandwidth expansion and resonance decay characteristics. By time domain labeling of these behaviors, the mapping relationship between spectral behavior and reaction stage can be established, and the key turning points in the reaction path can be extracted.

[0102] Step S303: comparing the candidate coordination state time node cluster with the preset sulfur aluminum coordination model in spectrum shape, identifying whether there is a cross-state mutation signal or a reconstruction signal, and deducing the evolution path of coordination structure.

[0103] In this embodiment, the candidate coordination state time node cluster obtained by spectrum deconstruction is compared with a preset sulfur-aluminum coordination model database in a one-to-one spectrum shape comparison. The database contains reference Raman profile templates of seven typical aluminum-sulfur complex structures, including Al-S single bridge bond, Al2(μ-S)2 ring structure, Al-SO3 internal ligand, etc. Peak position interval, bandwidth expansion, intensity ratio, and multi-peak overlap phase are used as characteristic dimensions during comparison. Weight scoring algorithm is used to calculate the fitting similarity of each snapshot peak cluster.

[0104] The coordination evolution of the sulfur-aluminum system has high polymorphism. A single sulfur source forms multiple complex configurations in different reaction windows. The spectrum at a single time point cannot reflect the global path. By comparing the time sequence snapshots with the multi-element reference templates, the occurrence process of structure reconstruction is identified from the relative intensity change, spectral line drift trend, and bandwidth expansion behavior. In particular, the cross-state mutation often shows spectral peak center shift and energy band splitting, forming identifiable features in the two-dimensional spectrum variable map.

[0105] Step S304: defining the time point at which the first spectrum shape switching behavior occurs in the evolution path as a second reaction lag time t2, and marking the coordination state category corresponding to the time point.

[0106] In this embodiment, in the two-dimensional spectrum variable trajectory graph, the coordination structure evolution path is identified based on spectrum shape comparison. The peak value drift interval corresponding to each state switching in the path is mapped on the time axis. The first occurring behavior and its corresponding time period are determined. The time point at which the switching behavior occurs is determined by fitting the cut point of the maximum slope of the spectrum intensity change in the conversion interval. The time point is defined as the second reaction lag time t2.

[0107] The second reaction lag time t2 is essentially the first dominant coordination transition time after the liquid phase sulfur source enters the main complex reaction mechanism. There is a structural response delay between the dropping behavior. The spectrum shape switching, as a direct external signal of structural change, has stable characteristic transfer performance, including main peak wave number mutation, characteristic peak disappearance or new birth, bandwidth narrowing, etc. By analyzing the spectrum intensity derivative of each state node on the evolution path, the significant turning point of the peak cluster can be captured, and the relative starting position of the reaction kinetics can be quantified from the spectrum level.

[0108] Step S4: adjusting the sulfur source addition rate and alkalinity based on the first reaction lag time and the second reaction lag time, and producing liquid poly-sulfur aluminum chloride based on the adjusted sulfur source addition rate and alkalinity.

[0109] The specific steps of step S4 are:

[0110] As Figure 3 , Figure 4 and Figure 5As shown, step S401: Collect the time series data streams of Raman spectrum, electrochemical impedance spectrum and microwave dielectric change, and generate independent parameter sequence matrix respectively.

[0111] In this embodiment, during the reaction of liquid polythioaluminum chloride, three sensing collection channels are connected respectively: as shown in the figure Figure 3 As shown, the Raman spectrum covers the 400-900 cm -1 interval, and the characteristic peak intensity array is generated by scanning at a period of 30 seconds, and in Figure 3 , the abscissa represents the wave number, and the unit is cm -1 , the ordinate is the Raman intensity, and the unit is a.u., the blue line is the initial stage of the reaction, the yellow line is the middle stage of the reaction, and the red line is the late stage of the reaction; as shown in Figure 4 , the electrochemical impedance measurement adopts equal-interval Nyquist test, extracts the impedance nodes of the real part and the imaginary part, and forms the impedance sequence matrix according to the equivalent circuit parameters after each test fitting, and in Figure 4 , the abscissa represents the real part impedance, and the unit is Ω, and the ordinate is the imaginary part impedance, and the unit is Ω, the blue line is the initial stage of the reaction, the impedance is higher, the interface charge transfer is slower, and it is represented as a large radius arc, the yellow line is the middle stage of the reaction, the charge transfer is accelerated, the impedance is reduced, and the arc radius is reduced, and the red line is the late stage of the reaction, and the system tends to be stable; as shown in Figure 5 , the microwave dielectric monitoring samples the dielectric constant and loss angle at a frequency of 2.45 GHz, and records the nonlinear attenuation trend within 5 minutes, and generates a complex dielectric parameter matrix indexed by time, and in Figure 5 , the abscissa represents the frequency, and the unit is GHz, and the ordinate is the dielectric constant, and the unit is F / m, the blue line is the initial stage of the reaction, the ion concentration is low, the polarization degree is moderate, the yellow line is the middle stage of the reaction, the polarization is enhanced after dropping, the overall dielectric constant is increased, and the red line is the late stage of the reaction, and the coordination structure is stabilized, and the polarization degree falls back.

[0112] Raman spectrum, electrochemical impedance spectrum and microwave dielectric parameter respectively map the structure configuration, charge transfer characteristics and molecular polarization state of the reaction system, and three of them together constitute the physical quantitative channel of the reaction process at the bond energy-interface-medium level. Due to the different collection mechanisms, each type of data presents the characteristics of asynchronous sampling, different sampling frequencies and parameter dimensions. By aligning the time stamps of the three, and respectively constructing independent parameter sequence matrix, the original change trend can be preserved to the greatest extent.

[0113] Step S402: Based on the parameter sequence matrix, a three-dimensional space reaction parameter matrix under a unified time axis is constructed, and each matrix unit corresponds to a micro time slice and a physical attribute group.

[0114] Step S403: inputting the three-dimensional space reaction parameter matrix into a structure inversion neural network based on an attention weight mechanism, identifying an evolution path of a sulfur-aluminum coordination state, and outputting a current structure state, a future evolution direction, and a related lag label.

[0115] The specific steps of step S403 are as follows:

[0116] Step S4031: based on the three-dimensional space reaction parameter matrix, a multi-scale data view set is constructed, and a unique identity index is set for each view.

[0117] In this embodiment, based on the three-dimensional space reaction parameter matrix that has been constructed, a multi-scale view set is generated according to a time axis sliding window mechanism, different granularity units and window overlap rates are set, local sub-matrices of different time spans are extracted from the overall matrix as independent views, each view retains all physical property sequences of three modalities of Raman, electrochemistry and dielectric, a series of tensor subsets under time scale transformation are generated through matrix slicing operation, and after the view is generated, an identity index containing window start and end time, modality code and sequence number is assigned to the view.

[0118] Step S4032: inputting the multi-scale data view set into an inversion type deep neural network structure constructed based on an attention weight mechanism in parallel, generating a cross-focus map, and the network structure scans the time segment weight distribution of each view in parallel through an attention head.

[0119] In this embodiment, the constructed multi-scale data view set is input into an inversion type deep neural network, the network adopts a multi-head attention mechanism as the core structure, each attention head scans a sub-view under a specific time scale, and focuses on the physical property weight distribution of different time segments in the view; the network performs position embedding on the three types of modality data in the input layer, combines the modality attribute code and the time index to form a sequence tensor, and generates a query, a key, a value and a vector group through linear transformation; each attention head independently calculates the attention weight, and weights and aggregates the input sequence based on the attention weight, and finally generates a cross-focus map in the middle layer.

[0120] Step S4033: identifying a key parameter mutation node in the cross-focus map, concatenating the candidate coordination track path in time sequence, calculating the relative offset degree and time overlap rate between the paths, and forming an evolution path map cluster.

[0121] In this embodiment, in the cross-focus map output by the inversion neural network, the attention value mutation area within each time scale is extracted, and the high-intensity weight superposition nodes that appear in the Raman, electrochemical and dielectric modes at different time periods are identified as key parameter mutation points; these nodes are concatenated in chronological order to construct multiple candidate coordination trajectory paths, each path contains a set of modal mutation point sequences arranged in chronological order, and the physical property type and view source corresponding to each node are marked; then, the relative offset index is calculated based on the synchronization of the nodes between the paths, which is used to measure the response phase difference of different paths in the same structural transition event; and the time overlap rate between the paths is further calculated to analyze the consensus area in the evolution process; finally, a set of coordination evolution path map clusters with internal hierarchical differences and time interval intersections are generated as the core graphical expression unit reflecting dynamic structural reconstruction.

[0122] Step S4034: performing an evolution consistency check on the evolution path map cluster, annotating the evolution trend of each path with three-state labels of drift, stability and mutation, and outputting the current coordination structure state and future evolution direction.

[0123] In this embodiment, for the multiple candidate coordination trajectory paths generated in the previous step, evolutionary consistency tests are performed in the path map cluster respectively. The test logic is based on the stability of the relative offset between each path in a continuous time segment, combined with the synchronous weight distribution characteristics of Raman, electrochemical and dielectric modes, to evaluate whether the path is in a convergence, drift or mutation state; if the adjacent paths have a high overlap rate on the time axis and the difference in physical property response is within the tolerance range, they are marked as stable; if there is a continuous displacement trend between the paths and the offset direction is consistent, it is marked as drift; if there is a sudden view switching in a short time and the focus of attention is reconstructed, it is marked as mutation; after all path labels are labeled, the coordination structure state category of the current reaction and its predicted evolution direction are output according to the label ratio and path aggregation trend in the current time window.

[0124] The essence of evolutionary consistency testing is to identify the phase synergy of these path behaviors, and to determine their evolutionary properties in physical space by clustering and trend-classifying the local dynamic characteristics of multiple trajectories in a path map cluster in a specific time window.

[0125] Step S4035: Identify the index time slice where the first state variation occurs in the future evolution direction, and define it as the associated lag label.

[0126] In the present embodiment, after obtaining the labeled drift-stable-mutation tri-state coordination trajectory path, the path atlas cluster is scanned frame by frame along the time axis corresponding to the future evolution trend; the time slice index at which the stable or drift state first transitions to the mutation state is recorded, that is, the starting node at which the path set first shows a structural reorganization tendency in the predicted evolution is determined, and the time index corresponding to this node is defined as the associated lag label.

[0127] Step S404: embedding a lag prediction window in the coordination evolution output according to the first reaction lag time t1 and the second reaction lag time t2, and generating a set of lag intervention factors.

[0128] The specific steps of step S404 are:

[0129] Step S4041: embedding the first reaction lag time t1 and the second reaction lag time t2 into the coordination evolution trend vector, respectively, to construct a lag interval reference index with the interval Δt=t2-t1 as the boundary parameter of the lag prediction window.

[0130] In the present embodiment, after identifying the first reaction lag time t1 and the second reaction lag time t2, the two time points are embedded into the current coordination evolution trend vector; the trend vector is a multidimensional vector group composed of a series of marked state sequences and feature change rates, and its time dimension has been standardized. The window boundary is set with t1 and t2 as anchor points, the interval Δt=t2-t1 is calculated, and the interval is used as a reference index range for constructing the lag interval.

[0131] The lag behavior is essentially a time misalignment between input disturbance and structural response, and its boundary is difficult to clearly define by single-point labeling. By labeling t1 and t2 as the disturbance starting and response appearing nodes of the reaction chain, respectively, and taking the interval Δt between them as the window quantization basis, the time domain range of the lag behavior has clear boundary semantics in the vector space. The lag interval does not depend on the absolute time value, but has structural significance through its association with the evolution trend vector.

[0132] Step S4042: performing local attention weight scanning on the coordination evolution trend vector within the lag prediction window, constructing a time-weighted sensitivity curve through time slice weight integration, and the time-weighted sensitivity curve is used to depict the strength of the reaction parameter on the structural evolution during the lag period.

[0133] In the embodiment, a corresponding section of the coordination evolution trend vector sequence is intercepted within a hysteresis prediction window Δt defined by the first reaction hysteresis time t1 and the second reaction hysteresis time t2, a local attention weight scanning is performed, a sliding mechanism with a time slice as a unit is used to reduce the weight of the attribute sub-dimension under different physical modalities in each time slice, then the attention response strength of each time slice is integrated and superimposed to construct a weighted sensitivity curve that changes over time.

[0134] Step S4043: According to the time-weighted sensitivity curve, the center position of the local gradient change rate maximum region is extracted, defined as the hysteresis response center index, and a symmetric expansion interval is generated from the hysteresis response center index to form an intervention envelope cluster.

[0135] In the embodiment, the time-weighted sensitivity curve is subjected to first-order derivative processing, the region with the maximum local gradient change rate is identified, and the gradient change curvature center of the section is calculated, and the index position thereof on the time axis is extracted, defined as the hysteresis response center index; then, taking the center point as the symmetric axis, a dynamic window with time symmetry is constructed by expanding a fixed proportion of time length in the positive and negative time directions according to the decay trend of the sensitivity curve, forming an intervention envelope cluster.

[0136] Step S4044: The intervention envelope cluster is mapped with a hysteresis factor, and a hysteresis intervention factor set is generated according to its distribution form, each factor in the hysteresis intervention factor set contains a label parameter.

[0137] In the embodiment, the generated intervention envelope cluster is discretized according to the time dimension, and the entire envelope interval is divided into multiple equal-width sub-sections. Each sub-section is mapped to a hysteresis intervention factor unit according to multiple parameter indicators such as the gradient value of the sensitivity curve covered by the sub-section, the hysteresis response center distance, and the modal activation intensity. Each factor unit is assigned a set of label parameters, and the label content includes: the corresponding modal weight dominant type, such as electrochemical dominant type; the response position state, such as the post-peak falling segment; the intervention recommendation intensity level, such as medium; and the time index interval. All sub-section factor sets constitute the final hysteresis intervention factor set.

[0138] Step S405: The concentration of ferrous sulfide impurities in the sulfur source raw material is monitored in real time, and the reaction temperature and target alkalinity interval are dynamically adjusted according to the change trend.

[0139] The specific steps of step S405 are:

[0140] Step S4051: Collect the concentration characteristic data of ferrous sulfide in the sulfur source raw material, and compare it with the preset impurity mode to judge its fluctuation trend.

[0141] In this embodiment, the instantaneous concentration of ferrous sulfide is sampled in real time by using spectral absorption method or electrochemical detection technology, a sampling period is set, the collected results are normalized, and a concentration sequence changing with time is generated; then the concentration sequence is compared with a preset reference model of impurity behavior, the reference model is obtained by training the fluctuation characteristics of FeS (ferrous sulfide) in a stable state in historical batches, including short-period oscillation, burst peak, background drift and other mode characteristics, the similarity between the change trend of the current sequence and each characteristic curve in the model is compared, and whether the fluctuation trend of the current ferrous sulfide is abnormal or tends to be unstable is judged.

[0142] Step S4052: Based on the fluctuation trend, a reaction deviation region occurring in a future reaction process is determined, a temperature regulation time period is adjusted, and a corresponding temperature regulation strategy is constructed.

[0143] In this embodiment, according to the fluctuation trend of the ferrous sulfide concentration identified in the previous step, the fluctuation trend is fitted into the reaction evolution state atlas constructed in the sulfur-aluminum system, and the coordination deviation behavior caused by the fluctuation trend in the future time axis is evaluated; according to the fluctuation amplitude and the change direction, a plurality of potential structure disturbance nodes are set in the reaction chain timeline, and it is judged which time periods have the risk of reaction state deviation in combination with the existing evolution path model, a heat-sensitive intervention priority window is set for these predicted reaction deviation regions, the original temperature control curve is adjusted, the heat response action is activated or delayed in advance, and a dynamic temperature regulation time period is formed.

[0144] Further, the temperature regulation strategy is oriented to dynamic feedback control, and under the driving of the fluctuation trend, the temperature control time node is locally displaced, and the heating or cooling rate is adjusted to avoid the reaction lag or reaction interruption induced by impurities.

[0145] Step S4053: According to the impurity fluctuation trend and the temperature regulation strategy, the basicity interval is dynamically corrected.

[0146] In this embodiment, the influence of the reaction parameter drift caused by the impurity change on the final basicity formation mechanism is further analyzed, the threshold change of the acid or alkali amount required under different impurity disturbance backgrounds is quantified through the acid-base titration curve evolution model in the reaction process; the temperature regulation time period and the fluctuation trend are projected into the acid-base equilibrium evolution space to identify the displacement trend of the basicity inflection point and the compression degree of the reaction window, and accordingly the target basicity interval is adaptively contracted or expanded.

[0147] Step S406: Based on the set of lag intervention factors, the temperature and the target basicity interval, a weight distribution table between the sulfur source addition rate and the basicity is constructed, the sulfur source addition rate and the basicity are regulated, and based on the regulated sulfur source addition rate and the basicity, liquid polyaluminum sulfide is produced.

[0148] In the embodiment, the weight distribution table takes the hysteresis response strength, the modal dominant label, the temperature control trigger period, and the alkalization window sensitivity as input indicators, maps the dynamic contribution coefficients of each control unit to the sulfur source rate and the alkalization parameter in a two-dimensional matrix structure, and each weight node represents a resource priority configuration direction under a certain response state.

[0149] On one hand, the label of the hysteresis intervention factor is used to set the response slope of the rate change; on the other hand, the correction interval of the target alkalinity degree and the temperature regulation strategy form an overlapping discrimination area on the time axis, which is used to determine the adjustment frequency of the alkalization step; through the dynamic reference of the weight table, when the coordination reaction enters different evolution stages, the adjustment emphasis of the sulfur source and the alkalinity degree can be automatically switched, so that the structure construction is kept in a continuous and stable evolution situation.

[0150] Further, the weight table is a real-time adjustment reference mapping table, which can be embedded into a neural network or a fuzzy control engine to perform priority sorting and adaptive correction; the actual drop rate of the sulfur source solution and the alkalinity correction function take the table as the core input variable, and through real-time sensing of the change trend of the evolution state, a closed-loop control is formed in the reaction system.

[0151] It should be noted that the finally produced liquid polyaluminum sulfide chloride realizes stable structure construction on the basis of the above dynamic regulation, and the coordination integrity and the hydroxyl distribution reach a predictable range, thereby improving the structure controllability and stability of the product.

[0152] Embodiment 2

[0153] Please refer to Figure 6 The application provides another embodiment of a composite intelligent production system of liquid polyaluminum sulfide chloride, which comprises a pH trajectory construction module, a first time acquisition module, a second time acquisition module, and an adjustment production module.

[0154] The pH trajectory construction module is used to mix a sodium aluminate solution and hydrochloric acid to generate an aluminum hydroxide chloride precursor solution and construct an initial pH trajectory model.

[0155] The first time acquisition module is used to drop a sulfur source solution into the aluminum hydroxide chloride precursor solution and acquire a first reaction hysteresis time according to the initial pH trajectory model.

[0156] The second time acquisition module is used to identify a sulfur aluminum coordination ion structure formed in the reaction and extract its dynamic evolution characteristics to acquire a second reaction hysteresis time.

[0157] The adjustment production module adjusts the sulfur source drop rate and the alkalinity degree based on the first reaction hysteresis time and the second reaction hysteresis time, and produces the liquid polyaluminum sulfide chloride based on the adjusted sulfur source drop rate and the alkalinity degree.

[0158] The adjustment production module comprises a parameter sequence matrix construction unit, an evolution path output unit, an allocation condition adjustment unit and an adjustment production unit;

[0159] The parameter sequence matrix construction unit is configured to collect time series data streams of Raman spectroscopy, electrochemical impedance spectroscopy and microwave dielectric changes, and generate independent parameter sequence matrices respectively;

[0160] The evolution path output unit is configured to identify an evolution path of a sulfur-aluminum coordination state, and output a current structure state, a future evolution direction and a related hysteresis label of the evolution path;

[0161] The allocation condition adjustment unit is configured to generate a set of hysteresis intervention factors, and adjust a reaction temperature and a target alkalinity interval;

[0162] The adjustment production unit is configured to construct a weight distribution table between a sulfur source addition rate and an alkalinity based on the set of hysteresis intervention factors, the temperature and the target alkalinity interval, to regulate the sulfur source addition rate and the alkalinity, and to produce liquid polysulfur aluminum chloride based on the adjusted sulfur source addition rate and the alkalinity.

[0163] In addition, the parts of the above technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive repetition.

[0164] The specific embodiments described above further illustrate the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A composite intelligent production method for liquid polyaluminum sulfide chloride, characterized in that: include: Sodium aluminate solution is mixed with hydrochloric acid to generate aluminum oxychloride precursor solution, and an initial pH trajectory model is constructed; adding a sulfur source solution dropwise to the aluminum oxychloride precursor solution, and obtaining a first reaction lag time according to the initial pH trajectory model; Identify the structure of the sulfur-aluminum coordination ion formed in the reaction and extract its dynamic evolution characteristics to obtain the second reaction lag time; The sulfur source addition rate and the alkalinity are adjusted based on the first reaction lag time and the second reaction lag time, and liquid polyaluminum sulfide chloride is produced based on the adjusted sulfur source addition rate and alkalinity.

2. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 1, characterized in that: The sodium aluminate solution is mixed with hydrochloric acid to generate an aluminum oxychloride precursor solution, and an initial pH trajectory model is constructed, including: Through the acid injection interface, hydrochloric acid is injected in an intermittent waveform manner to react with the sodium aluminate solution in the reaction tank. The pulse interval is adjusted according to the pH change rate of the solution in the previous cycle. The pH signal of the reaction solution in the reaction tank is obtained by multi-point time series sampling, and a pH response matrix is ​​established; Based on the pH response matrix and the evolution law under different initial ratios and temperature conditions, multiple trajectory curves are established, each trajectory curve corresponds to an evolution law; The initial ratio and temperature conditions are matched with the initial ratio and temperature conditions corresponding to each evolution law, and based on the matching result, a trajectory curve corresponding to the current evolution law is selected from the multiple trajectory curves and used as the initial pH trajectory model.

3. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 2, characterized in that: Adding a sulfur source solution dropwise to the aluminum oxychloride precursor solution to obtain a first reaction lag time comprises: Establish a disturbed flow state window, identify the liquid phase shear response boundary, and determine the initial addition starting point of the sulfur source solution in combination with the initial pH trajectory model; Use delayed trigger dripping and random noise perturbation to load the timing pulse to make the dripping behavior present a non-periodic waveform; A hydrogen sulfide gas channel is guided at the top of the reaction tank to perform gas chromatography-mass spectrometry dual-mode analysis in a one-step circulation manner, and a gas concentration-time response curve is established; Combined with the preset reference response envelope, the timestamp corresponding to the gas mutation inflection point is set as the first reaction lag time t1.

4. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 3, characterized in that: The identification reaction forms a sulfur-aluminum coordination ion structure and extracts its dynamic evolution characteristics to obtain the second reaction lag time, including: A multi-segment wavelength scanning sequence is excited in a reaction solution of an aluminum hydroxide chloride precursor solution and a sulfur source solution to generate a collection of instantaneous response spectra of the aluminum coordination structure, and a snapshot data package is created after each scan; The snapshot data packets are converted into a two-dimensional spectrum trajectory diagram in chronological order, the sulfur-aluminum coordination peak cluster with absorption shift behavior is extracted, and its initial appearance time, maximum intensity moment and disappearance moment are marked to generate a candidate coordination state time node cluster; Compare the candidate coordination state time node clusters with the preset sulfur-aluminum coordination model to identify whether there are cross-state mutation signals or reconstruction signals, and deduce the evolution path of the coordination structure; The time point at which the first spectrum shape switching behavior occurs in the evolution path is defined as the second reaction lag time t2, and the coordination state category corresponding to the time point is marked.

5. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 4, characterized in that: The adjusting of the sulfur source addition rate and the alkalinity based on the first reaction lag time and the second reaction lag time comprises: Collect the time series data streams of Raman spectroscopy, electrochemical impedance spectroscopy and microwave dielectric changes, and generate independent parameter sequence matrices for each; Based on the parameter sequence matrix, a three-dimensional spatial reaction parameter matrix under a unified time axis is constructed, and each matrix unit is set to correspond to a micro-time slice and a physical property group; Inputting the three-dimensional spatial reaction parameter matrix into a structural inversion neural network based on an attention weight mechanism, identifying the evolution path of the sulfur-aluminum coordination state, and outputting its current structural state, future evolution direction, and associated hysteresis label; According to the first reaction lag time t1 and the second reaction lag time t2, a lag prediction window is embedded in the coordination evolution output to generate a lag intervention factor set; Real-time monitoring of the concentration of ferrous sulfide impurities in the sulfur source raw materials, and dynamic adjustment of the reaction temperature and target alkalinity range based on its changing trend; Based on the set of delayed intervention factors, temperature and target alkalinity range, a weight distribution table between the sulfur source addition rate and the alkalinity is constructed, the sulfur source addition rate and the alkalinity are regulated, and liquid polyaluminum sulfide chloride is produced based on the regulated sulfur source addition rate and the alkalinity.

6. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 5, characterized in that: The three-dimensional spatial reaction parameter matrix is ​​input into a structural inversion neural network based on an attention weight mechanism to identify the evolution path of the sulfur-aluminum coordination state and output its current structural state, future evolution direction and associated hysteresis label, including: Based on the three-dimensional spatial response parameter matrix, a multi-scale data view set is constructed, and a unique identity index is set for each view; A set of multi-scale data views is fed in parallel into an inverse deep neural network structure built on an attention weight mechanism to generate a cross-focus map. The network structure scans the weight distribution of each time segment of each view in parallel through attention heads. Identify key parameter mutation nodes in the cross-focus map, concatenate them into candidate coordination trajectory paths in chronological order, and calculate the relative offset and time overlap between the paths to form an evolutionary path map cluster; The evolutionary path map cluster is subjected to evolutionary consistency test, and the evolutionary trend of each path is marked by three-state labels: drift, stability and mutation, and the current coordination structure state and future evolution direction are output; The index time slice where the first state variation occurs in the future evolution direction is identified and defined as the associated lag label.

7. A composite intelligent production method for liquid polyaluminum sulfide chloride according to claim 6, characterized in that: According to the first reaction lag time t1 and the second reaction lag time t2, a lag prediction window is embedded in the coordination evolution output to generate a lag intervention factor set, including: The first reaction lag time t1 and the second reaction lag time t2 are respectively embedded into the coordination evolution trend vector, and the lag interval reference index is constructed with the interval Δt=t2-t1 as the boundary calibration parameter of the lag prediction window; Performing a local attention weight scan on the coordination evolution trend vector within the hysteresis prediction window, and constructing a time-weighted sensitivity curve by time-slice weight integration, wherein the time-weighted sensitivity curve is used to characterize the intensity of the reaction parameters on the structural evolution during the hysteresis period; According to the time-weighted sensitivity curve, the center position of the area with the maximum local gradient change rate is extracted and defined as the hysteresis response center index, and a symmetrical expansion interval is generated with the hysteresis response center index as the starting point to form an intervention envelope cluster; The intervention envelope cluster is subjected to hysteresis factor mapping, and a hysteresis intervention factor set is generated according to its distribution form, wherein each factor in the hysteresis intervention factor set includes a label parameter.

8. The composite intelligent production method of liquid polyaluminum sulfide chloride according to claim 7, characterized in that: The real-time monitoring of the concentration of ferrous sulfide impurities in the sulfur source raw material and the dynamic adjustment of the reaction temperature and the target alkalinity range according to the change trend thereof include: Collect characteristic data on the concentration of ferrous sulfide in the sulfur source raw material and compare it with the preset impurity pattern to determine its fluctuation trend; Based on the fluctuation trend, the reaction deviation area that will appear in the future reaction process is determined, and the temperature control time period is adjusted to construct a corresponding temperature control strategy; Dynamically correct the alkalinity range based on impurity fluctuation trends and temperature adjustment strategies.

9. A composite intelligent production system for liquid polyaluminum sulfide chloride, used to implement a composite intelligent production method for liquid polyaluminum sulfide chloride according to any one of claims 1 to 8, characterized in that: include: pH trajectory construction module, first time acquisition module, second time acquisition module and regulation production module; The pH trajectory building module is used to mix the sodium aluminate solution with hydrochloric acid to generate an aluminum oxychloride precursor solution and build an initial pH trajectory model; The first time acquisition module is used to add a sulfur source solution to the aluminum oxychloride precursor solution and obtain a first reaction lag time according to the initial pH trajectory model; The second time acquisition module is used to identify the sulfur-aluminum coordination ion structure formed in the reaction and extract its dynamic evolution characteristics to obtain the second reaction lag time; The regulating production module regulates the sulfur source addition rate and the alkalinity based on the first reaction lag time and the second reaction lag time, and produces liquid polyaluminum sulfide chloride based on the regulated sulfur source addition rate and alkalinity.

10. A composite intelligent production system for liquid polyaluminum sulfide chloride according to claim 9, characterized in that: The regulation production module includes: a parameter sequence matrix construction unit, an evolution path output unit, an allocation condition adjustment unit and a regulation production unit; The parameter sequence matrix construction unit is used to collect the time series data streams of Raman spectroscopy, electrochemical impedance spectroscopy and microwave dielectric change, and generate independent parameter sequence matrices for each; The evolution path output unit is used to identify the evolution path of the sulfur-aluminum coordination state and output its current structural state, future evolution direction and associated hysteresis label; The distribution condition adjustment unit is used to generate a set of hysteresis intervention factors and adjust the reaction temperature and target alkalinity range; The regulating production unit constructs a weight distribution table between the sulfur source addition rate and the alkalinity based on the set of delayed intervention factors, temperature and target alkalinity range, regulates the sulfur source addition rate and alkalinity, and produces liquid polyaluminum sulfide chloride based on the adjusted sulfur source addition rate and alkalinity.

Images