An on-line monitoring and intelligent control system for high-purity cerium hydroxide precipitation reaction process

Abstract

The application discloses an online monitoring and intelligent regulation and control system for a high-purity cerium hydroxide precipitation reaction process and relates to the technical field of reaction process monitoring and regulation and control. The system comprises a precipitation process double-parameter tracing module, a local oxidation characteristic capturing module, a multi-signal decoupling identification module and a partitioned dynamic compensation execution module, and signals of the modules are connected to a PLC control unit. In the precise control of oxidation, the local oxidation characteristic capturing module is used to monitor the dissolved oxygen and ORP values of four types of high-oxidation-risk areas, such as a feeding mixing area and a vortex shear area, in real time, and in combination with a time sequence matching algorithm of the multi-signal decoupling identification module, the precipitation disturbance, the feeding disturbance and the oxidation disturbance can be accurately distinguished, the accuracy of judgment is high, the false triggering rate is low, and the hierarchical directional pulse deoxygenation strategy of the partitioned dynamic compensation execution module is matched, compared with continuous aeration of the whole tank, the consumption of inert gas is reduced, and the problem of air entrainment caused by continuous bubbling is avoided.

Description

Technical Field

[0001] This invention relates to the field of reaction process monitoring and control technology, specifically to an online monitoring and intelligent control system for a high-purity cerium hydroxide precipitation reaction process. Background Technology

[0002] Cerium hydroxide precipitation refers to the chemical reaction in which an alkali, such as sodium hydroxide or ammonia, is added to a solution containing cerium ions, causing the formation of a cerium hydroxide precipitate that is sparingly soluble in water. Cerium has unique chemical properties, existing in two stable valence states, and the precipitation and subsequent behavior of its hydroxides differ significantly. Therefore, there are two precipitation reactions for these valence states: trivalent cerium precipitation and tetravalent cerium precipitation. Trivalent cerium precipitation involves adding an alkali to a solution containing trivalent cerium, producing a white or pale yellow cerium hydroxide precipitate. Tetravalent cerium precipitation involves adding an alkali to a solution containing tetravalent cerium, directly producing a yellow to brownish-yellow cerium hydroxide precipitate. Online monitoring and control during the cerium hydroxide precipitation process can significantly reduce the difficulty of subsequent filtration and washing processes, reduce impurity adsorption residues, and ensure consistently stable reaction conditions. This reduces waste of raw materials from non-conforming batches and minimizes operational errors from manual sampling.

[0003] The patent publication number CN119247896A discloses an intelligent control method and device for metallurgical reactors based on real-time simulation and monitoring. This method combines image recognition, machine learning, and simulation techniques, utilizing geometric image recognition and parametric automatic modeling and mesh generation to obtain a large amount of simulation data. Machine learning algorithms are used to achieve real-time simulation and prediction of the metallurgical reactor, and based on this, dynamic monitoring of multiple physical fields within the reactor is completed. A deep learning optimization prediction model is established to achieve intelligent control of the metallurgical reactor. This method enables real-time simulation of multiple physical fields and intelligent control of the metallurgical reactor, solving problems such as difficulty in obtaining physical field monitoring information within the reactor during production, low simulation efficiency preventing real-time acquisition, and reliance on experience for reactor control. It is a highly efficient and intelligent control method for metallurgical reactors based on real-time simulation monitoring that accurately reflects the physical field information within the reactor.

[0004] The above and similar technical solutions are due to Ce 3+ After Ce(OH)3 is precipitated, it is rapidly oxidized to Ce(OH)4 by dissolved oxygen. The rate, extent, and local distribution of this oxidation process are difficult to detect online. Traditional offline sampling and detection have a lag and cannot capture the dynamic changes of the instantaneous oxidation front. As a result, it is impossible to accurately control the valence ratio of the product, which affects the purity, and it is also impossible to distinguish the disturbances of the system's pH and turbidity signals caused by "true precipitation formation" and "subsequent oxidation and deterioration". Summary of the Invention

[0005] The purpose of this invention is to provide an online monitoring and intelligent control system for the precipitation reaction process of high-purity cerium hydroxide, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an online monitoring and intelligent control system for the precipitation reaction process of high-purity cerium hydroxide, comprising:

[0007] The precipitation process dual-parameter tracer module, the local oxidation feature capture module, the multi-signal decoupling and identification module, and the partitioned dynamic compensation execution module are all connected to the PLC control unit.

[0008] The precipitation process dual-parameter tracer module consists of a first sensor and a second sensor deployed in the reactor. The first sensor collects the real-time conductivity change value and the real-time solid phase turbidity value of the reaction system, respectively. The instantaneous conversion rate of Ce(OH)3 precipitate is obtained by cross-correction, and the precipitation information item is obtained.

[0009] The local oxidation feature capture module consists of multiple monitoring units. It sets feature points inside the reactor to obtain the set point items, installs monitoring units based on the set point items to obtain the monitoring unit installation items, and collects the instantaneous dissolved oxygen concentration and ORP value in different areas to identify the location, rate and extent of oxidation reaction and obtain oxidation information items.

[0010] The multi-signal decoupling identification module has a built-in time-series matching algorithm model to distinguish the sources of disturbances in the system's pH, turbidity, and ORP signals, thus obtaining the first interference identification item. At the same time, it eliminates the interference from instantaneous parameter fluctuations caused by the addition of alkali / cerium salt, thus obtaining the second interference identification item.

[0011] The partitioned dynamic compensation execution module executes a graded oxidation compensation strategy based on the results of the first and second interference identification items to obtain the first treatment item. At the same time, it dynamically adjusts the alkaline feed rate based on the overall oxidation rate of the reactor to accelerate the precipitation process and shorten the precipitation and oxidation window period, thereby obtaining the second treatment item. This enables real-time monitoring and dynamic control of the reaction process.

[0012] Furthermore, the precipitation process dual-parameter tracer module includes:

[0013] The first sensor and the second sensor are deployed in the central liquid phase zone of the reactor. The first sensor is a four-electrode type corrosion-resistant in-situ conductivity sensor, and the second sensor is a near-infrared dynamic turbidity sensor with a pneumatic window self-cleaning function. The wavelength of the second sensor is set to obtain the wavelength setting item.

[0014] The sampling frequency of the first sensor and the purge window of the second sensor are set, and the real-time conductivity change value and real-time solid phase turbidity value of the reaction system are collected by the first sensor and the second sensor, respectively.

[0015] Furthermore, the method for obtaining the sedimented information items includes:

[0016] Pre-calibrate the conductivity baseline, turbidity baseline, and Ce content of the cerium-free background system. 3+ The theoretical final turbidity value of complete precipitation is calculated by measuring the decrease in free Ce in real time. 3+ Consumption rate;

[0017] The solid-phase formation rate of Ce(OH)3 is calculated by the real-time turbidity rise value, and the average value is taken to obtain the instantaneous conversion rate of precipitation, thereby obtaining the precipitation information item.

[0018] Furthermore, the method for obtaining the monitoring unit installation items includes:

[0019] The feature points of the set point item include at least four fixed points, including the feed mixing zone at the feed port, the radial outer vortex shear zone of the agitator, the dead corner zone at the bottom of the inner wall of the reactor, and the gas-liquid contact zone.

[0020] Each monitoring unit is coaxially integrated with a needle-type dissolved oxygen microelectrode with built-in PT100 temperature compensation, an in-situ ORP electrode, and a directional microbubble head.

[0021] The sampling frequency of the dissolved oxygen microelectrode and ORP electrode of each monitoring bubble unit is set, and the outlet position of the directional microbubble head is set, thereby obtaining the installation items of the monitoring unit.

[0022] Furthermore, the local oxidation feature capture module incorporates built-in ORP baseline automatic calibration logic, and the method for obtaining oxidation information items includes:

[0023] Set the calibration logic and the dissolved oxygen threshold. Before each batch of reaction, after the system is pre-deoxygenated to below the dissolved oxygen threshold, calibrate the ORP baseline value of that batch to obtain the batch calibration items.

[0024] The oxidation reaction judgment rules are divided into at least three levels to obtain judgment level items. Corresponding judgment logic is generated based on the judgment level items. The batch calibration items are offset based on the judgment logic. At the same time, the dissolved oxygen consumption rate at the corresponding point is adjusted based on the drift result, thereby obtaining the oxidation information item.

[0025] Furthermore, the method for obtaining the first interference identification item includes:

[0026] A time-series matching threshold is set. When the overlap of the time-series curves of the rate of decrease in conductivity and the rate of increase in turbidity exceeds the time-series matching threshold, it is determined to be a disturbance in the formation of Ce(OH)3 precipitate.

[0027] When the rate of decrease in conductivity and the rate of increase in turbidity do not exceed the time-matching threshold, but the ORP shifts positively and the dissolved oxygen consumption rate increases abnormally, it is determined to be a Ce(OH)3 oxidation and deterioration disturbance.

[0028] Furthermore, the method for obtaining the second interference identification item includes:

[0029] Set a window time value. If ORP and turbidity fluctuate within the time window when the feed port solenoid valve is in the open state, but the dissolved oxygen consumption rate does not increase abnormally, it is judged as a momentary disturbance during feeding and oxidation compensation is not triggered.

[0030] Furthermore, the first processing item includes:

[0031] For the characteristic points where oxidation is determined, inert gas is pulsed into the corresponding points for deoxygenation. At least three pulse level parameters are set, and the pulse level parameters are matched with the oxidation information items to obtain the first processing item.

[0032] Furthermore, the method for obtaining the second processing item includes:

[0033] Set a conversion rate threshold. When the instantaneous conversion rate of precipitation is lower than the conversion rate threshold, trigger the alkali rate adjustment. Set at least three oxidation levels and match the adjustment range with the oxidation information item to obtain the second treatment item.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] This online monitoring and intelligent control system for the high-purity cerium hydroxide precipitation reaction process achieves precise oxidation control. It uses a local oxidation feature capture module to monitor dissolved oxygen and ORP values ​​in real time across four high-oxidation-risk zones, including the feeding mixing zone and the eddy shear zone. Combined with a time-series matching algorithm from a multi-signal decoupling identification module, it can accurately distinguish between precipitation disturbances, feeding disturbances, and oxidation disturbances, achieving high accuracy and low false triggering rate. The accompanying zonal dynamic compensation execution module employs a graded directional pulse deoxygenation strategy, which, compared to continuous aeration throughout the reactor, reduces inert gas consumption and avoids air entrainment issues caused by continuous bubbling.

[0036] Meanwhile, in terms of product quality optimization, the dual-parameter tracer module for the precipitation process uses a four-electrode conductivity sensor and a 940nm near-infrared turbidity sensor for cross-calibration, resulting in small measurement errors in precipitation conversion rate. This allows for real-time control of the precipitation process, and by dynamically adjusting the alkali feed rate in conjunction with the oxidation level, free Ce can be effectively controlled. 3+ The oxidation window period has been reduced from 15 minutes in the traditional process to less than 3 minutes. The product has a stable proportion of tetravalent cerium doping and a low total content of non-cerium rare earth impurities, which fully meets the high purity requirements of energy storage anode precursors. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0038] Figure 2 This is a schematic diagram of the dual-parameter tracer module for the precipitation process of the present invention;

[0039] Figure 3 This is a schematic diagram of the local oxidation feature capture module structure of the present invention;

[0040] Figure 4 This is a schematic diagram of the multi-signal decoupling identification module structure of the present invention;

[0041] Figure 5 This is a schematic diagram of the partition dynamic compensation execution module of the present invention;

[0042] Figure 6 This is a schematic diagram of the overall implementation process of the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Cerium hydroxide, as an important rare earth compound, has broad application prospects in catalysts, polishing materials, and electronic ceramics. Preparing cerium hydroxide via controlled precipitation can effectively regulate its particle size, morphology, and crystal structure, thereby optimizing its performance. However, during the cerium hydroxide precipitation reaction, especially when generating Ce(OH)3, Ce3+ ions are readily oxidized to Ce(OH)4 by dissolved oxygen. Traditional cerium hydroxide precipitation is usually achieved by adding an alkaline precipitant to a solution containing Ce3+ ions. Ideally, Ce3+ reacts with hydroxide ions to form Ce(OH)3 precipitate. However, in actual reaction environments, dissolved oxygen is often present. Under alkaline conditions, Ce3+ is rapidly oxidized to Ce4+, subsequently forming Ce(OH)4 precipitate. The oxidation rate and extent are difficult to control precisely. Factors such as the concentration of dissolved oxygen, reaction temperature, and pH value all affect the oxidation rate of Ce3+. Even under strict control of experimental parameters, the oxidation rate may still fluctuate, resulting in a mixture of Ce3+ and Ce4+ in the final product. The proportion of dissolved oxygen is difficult to control precisely, thus affecting the purity and performance of the product. Simultaneously, the distribution of dissolved oxygen in the solution may be uneven, causing oxidation reactions to occur locally, forming an oxidation front. This localized oxidation leads to uneven composition and structure of the product, affecting its performance. The technical solution provided in this application, however, achieves precise oxidation control by using a localized oxidation feature capture module to monitor dissolved oxygen and ORP values ​​in real time in four high-oxidation-risk areas, including the feeding mixing zone and the eddy shear zone. Combined with the time-series matching algorithm of the multi-signal decoupling identification module, it can accurately distinguish between precipitation disturbances, feeding disturbances, and oxidation disturbances, achieving high accuracy and low false triggering rate. The accompanying partitioned dynamic compensation execution module employs a graded directional pulse deoxygenation strategy, which, compared to continuous full-reactor aeration, reduces inert gas consumption and avoids air entrainment problems caused by continuous bubbling. Figure 1 As shown, it includes a precipitation process dual-parameter tracer module, a local oxidation feature capture module, a multi-signal decoupling identification module, and a partitioned dynamic compensation execution module. The signals of each module are connected to the PLC control unit.

[0045] The precipitation process dual-parameter tracer module consists of a first sensor and a second sensor deployed in the reactor. They collect the real-time conductivity change value and the real-time solid phase turbidity value of the reaction system, respectively. The instantaneous conversion rate of Ce(OH)3 precipitate is obtained by cross-correction, and the precipitation information item is obtained.

[0046] It is important to note that, such as Figure 2As shown, the precipitation process dual-parameter tracer module includes: a first sensor and a second sensor deployed in the central liquid phase zone of the reactor. The first sensor is a four-electrode type corrosion-resistant in-situ conductivity sensor, and the second sensor is a near-infrared dynamic turbidity sensor with a pneumatic window self-cleaning function. The wavelength of the second sensor is set to obtain the wavelength setting item; the sampling frequency of the first sensor and the purge window of the second sensor are set. The real-time conductivity change value and real-time solid phase turbidity value of the reaction system are collected by the first sensor and the second sensor, respectively.

[0047] Specifically, a four-electrode corrosion-resistant in-situ conductivity sensor was chosen because traditional two-electrode sensors are prone to polarization and electrode scaling in high-salt, high-alkali precipitation systems, resulting in measurement errors exceeding 10%. The four-electrode structure, by applying a constant current to two electrodes and detecting the potential difference between the two electrodes, completely cancels out the polarization effect, improving measurement stability by more than five times. The 940nm near-infrared wavelength was chosen because of Ce... 3+ The characteristic absorption peak is located in the 253 nm ultraviolet region, Ce 4+ The characteristic absorption peak is located in the 320nm ultraviolet region. The 940nm near-infrared light is not absorbed by cerium ions in either of the two valence states, nor is it affected by the intrinsic fluorescence effect of rare earth elements. The detected turbidity signal is only related to the scattering effect of solid-phase precipitation, completely eliminating the interference of valence state changes on turbidity readings.

[0048] Specifically, the four-electrode type corrosion-resistant in-situ conductivity sensor is coated with a 200μm thick polytetrafluoroethylene anti-corrosion layer, suitable for reaction systems with pH 0-14 and total salinity ≤5mol / L. It has a built-in PT100 temperature compensation unit, a sampling frequency set to 5Hz, and a measurement error ≤0.5%FS. The near-infrared dynamic turbidity sensor uses a 940nm laser light source. The pneumatic window self-cleaning logic is set to automatically purge the sensor window with 0.1MPa high-purity argon every 8s for 0.5s, which can completely remove attached precipitate particles and avoid turbidity reading drift. The turbidity measurement range covers 0-10000NTU with an error ≤1%FS.

[0049] It is important to note that the methods for obtaining precipitation information items include: pre-calibrating the conductivity baseline, turbidity baseline, and Ce content of the cerium-free background system. 3+ The theoretical final turbidity value of complete precipitation is calculated by measuring the decrease in free Ce in real time. 3+ Consumption rate; the solid phase formation rate of Ce(OH)3 is calculated by the real-time turbidity rise value, and the average value is taken to obtain the instantaneous conversion rate of precipitation, and then the precipitation information item is obtained.

[0050] Specifically, three baseline parameters are pre-calibrated, the first being the conductivity baseline. The conductivity value measured at the reaction temperature, after preparing a cerium-free solution with the same impurity ion concentration and alkaline anion concentration as the reaction system, is used to exclude CeCl₂. − CeNO 3− CeNH 4+ The first is the conductivity contribution of non-cerium ions, and the second is the turbidity baseline. The turbidity value measured after adding trivalent cerium salt solution to the reactor and before adding alkali is used to exclude the turbidity contribution of tiny impurity particles in the raw materials. The third is Ce. 3+ Theoretical final turbidity value of complete precipitation The turbidity value measured under pilot-scale conditions is the result of complete precipitation of trivalent cerium salt at this concentration, without oxidation or agglomeration. Recalibration is performed for each batch of raw materials. The conversion rate is calculated using the following formula:

[0051] ;

[0052] in The conversion rate is determined by the conductivity method. The conductivity of the trivalent cerium salt solution before adding alkali is reduced by The obtained pure Ce 3+ Contribution of conductivity The conductivity value is detected in real time.

[0053] ;

[0054] in The conversion rate is determined by the turbidimetric method. The turbidity value is taken as the real-time detection value. As the final instantaneous conversion rate of precipitation, the corrected error is ≤2%; if and If the difference exceeds 5%, it is determined to be a localized uneven mixing, and the stirring rate is automatically adjusted by 10% to eliminate localized interference from excessive alkalinity or acidity.

[0055] The local oxidation feature capture module consists of multiple monitoring units. It sets feature points inside the reactor to obtain the set point items, installs monitoring units based on the set point items to obtain the monitoring unit installation items, and collects the instantaneous dissolved oxygen concentration and ORP value in different areas to identify the location, rate and extent of oxidation reaction and obtain oxidation information items.

[0056] It is important to note that, such as Figure 3As shown, the method for obtaining the monitoring unit installation items includes: setting the characteristic points of the point items to include at least four fixed points, including the feed mixing zone at the feed port, the radial outer vortex shear zone of the agitator, the dead corner zone at the bottom of the inner wall of the reactor, and the gas-liquid contact zone; each monitoring unit is coaxially integrated with a needle-type dissolved oxygen microelectrode with built-in PT100 temperature compensation, an in-situ ORP electrode, and a directional micro-bubbling head, setting the sampling frequency of the dissolved oxygen microelectrode and ORP electrode of each monitoring bubble unit, and setting the outlet position of the directional micro-bubbling head, thereby obtaining the monitoring unit installation items.

[0057] Specifically, the oxidation reaction does not occur uniformly throughout the entire reactor; over 95% of the oxidation begins in four high-risk areas: uneven mixing at the feed inlet leads to the release of free Ce. 3+ Prolonged exposure, high shear force in the eddy zone breaking up sediment and exposing a fresh reaction interface, insufficient stirring in the bottom dead corner leading to dissolved oxygen accumulation, and direct contact between the gas and liquid contact zone and the air leaking in from above the liquid surface—covering these four points can capture almost all oxidation initiation signals. The monitoring and bubbling unit are coaxially integrated to achieve precise deoxygenation by "blowing where oxidation occurs." At the same time, the bubbling airflow can sweep away sediment on the electrode surface, achieving electrode self-cleaning.

[0058] Specifically, the installation parameters for the four fixed points are as follows: Feed inlet mixing zone: located 10cm directly below the cerium salt / alkali solution feed inlet, representing the initial uneven mixing area after feeding; Radial outer vortex shear zone of the stirring paddle: located 2cm radially outward of the stirring paddle blades, representing the area with the greatest shear force in the system; Dead corner zone at the bottom of the reactor wall: located 5cm from the bottom of the reactor and 5cm from the inner baffle, representing the weakest stirring dead zone; Gas-liquid contact zone: located 3cm below the liquid surface, preventing exposure to the liquid surface due to surface fluctuations. The coaxial integration parameters for each monitoring unit are as follows: the needle-type dissolved oxygen microelectrode and the in-situ ORP electrode are installed side-by-side along the central axis, with the directional micro-bubbling head fixed 1cm directly below the two electrodes. The air outlet faces the electrode probe surface, allowing the airflow during bubbling to simultaneously remove dissolved oxygen around the electrodes and blow away deposits adhering to the electrode surface, eliminating the need for manual cleaning and extending electrode lifespan by more than 3 times; the sampling frequency of the dissolved oxygen microelectrode and the ORP electrode is set to 2Hz, capable of capturing instantaneous oxidation signals at the 1s level.

[0059] It is important to note that the local oxidation feature capture module has built-in ORP baseline automatic calibration logic. The method for obtaining oxidation information items includes: setting calibration logic, setting dissolved oxygen threshold, and before each batch reaction, after the system is pre-deoxygenated to below the dissolved oxygen threshold, calibrating the ORP baseline value of that batch to obtain batch calibration items; dividing the oxidation reaction judgment rules into at least three levels to obtain judgment level items, generating corresponding judgment logic based on the judgment level items, shifting the batch calibration items based on the judgment logic, and adjusting the dissolved oxygen consumption rate at the corresponding points based on the drift results to obtain oxidation information items.

[0060] Specifically, the impurity content, ionic strength, and temperature vary between different batches of raw materials, and there is no unified standard for determining the absolute value of ORP. Using the relative offset of the baseline calibrated for each batch can eliminate batch differences. The dual parameter determination of ORP and dissolved oxygen consumption rate can avoid the influence of variable valence impurities such as CeFe on the single ORP. 3+ CeMn 2+ Misjudgments caused by interference: When oxidation occurs, Ce(OH)3 consumes dissolved oxygen, while ORP shifts caused by impurities do not increase with the rate of dissolved oxygen consumption. The accuracy of dual-parameter determination is ≥99%.

[0061] Specifically, the automatic ORP baseline calibration logic is as follows: Before each batch of reaction, the entire reactor is pre-deoxygenated by purging with inert gas. When the dissolved oxygen at all points is ≤0.05mg / L, the ORP value is continuously collected for 30 seconds, and the average value is taken as the ORP baseline for that batch. The oxidation reaction judgment rules are divided into three levels. The three-level oxidation judgment logic is as follows: Mild oxidation: The ORP at a single point shifts positively by 20-50mV compared to the ORP baseline, and the dissolved oxygen consumption rate at that point in the past 10 seconds increases by 30%-60% compared to the baseline value. The oxidation rate is 100% - 100% - 2%, corresponding to approximately 0.1% - 0.5% of Ce(OH)3 being oxidized; moderate oxidation: the single-site ORP shifts positively by 50-100mV compared to the ORP baseline, and the dissolved oxygen consumption rate increases by 60% - 100% compared to the baseline value, corresponding to approximately 0.5% - 2% of Ce(OH)3 being oxidized; severe oxidation: the single-site ORP shifts positively by ≥100mV compared to the ORP baseline, and the dissolved oxygen consumption rate increases by ≥100% compared to the baseline value, corresponding to ≥2% of Ce(OH)3 being oxidized.

[0062] The multi-signal decoupling identification module has a built-in time-series matching algorithm model to distinguish the sources of disturbance in the system's pH, turbidity, and ORP signals, thus obtaining the first interference identification item. At the same time, it eliminates the interference from instantaneous parameter fluctuations caused by the addition of alkali / cerium salt, thus obtaining the second interference identification item.

[0063] It is important to note that, such as Figure 4 As shown, the method for obtaining the first interference identification item includes: setting a time-series matching degree threshold; when the overlap of the time-series curves of the conductivity decrease rate and the turbidity increase rate exceeds the time-series matching degree threshold, it is determined to be a Ce(OH)3 precipitation generation disturbance; when the conductivity decrease rate and the turbidity increase rate do not exceed the time-series matching degree threshold, but the ORP shifts positively and the dissolved oxygen consumption rate increases abnormally, it is determined to be a Ce(OH)3 oxidation and deterioration disturbance.

[0064] Specifically, the Ce(OH)3 precipitation reaction is a process in which the consumption of free ions and the formation of solid phase occur simultaneously. Therefore, the decrease in conductivity and the increase in turbidity are completely synchronized. On the other hand, the Ce(OH)3 oxidation reaction is a process of solid phase valence state change. There is no consumption of free ions and the change in the amount of solid phase is minimal. Therefore, there is no significant change in conductivity and turbidity. It is only accompanied by an increase in ORP and the consumption of dissolved oxygen. The two types of disturbances can be completely distinguished by the degree of time matching.

[0065] Specifically, the timing matching threshold is set to 90%. The overlap between the conductivity decrease rate time series curve and the turbidity increase rate time series curve in the past 5 seconds is calculated using the Pearson correlation coefficient. If the correlation coefficient is ≥0.9, i.e. the matching degree is ≥90%, the signal disturbance is determined to be a Ce(OH)3 precipitation formation disturbance, and oxidation compensation is not triggered. If the correlation coefficient is <0.9, and the two conditions of ORP positive offset ≥20mV and dissolved oxygen consumption rate increase ≥30% are met simultaneously, the signal disturbance is determined to be a Ce(OH)3 oxidation and deterioration disturbance, and oxidation compensation logic is triggered.

[0066] It should be noted that the method for obtaining the second interference identification item includes: setting a window time value. When ORP and turbidity fluctuations occur within the time window value when the feed port solenoid valve is in the open state, but the dissolved oxygen consumption rate does not increase abnormally, it is determined to be a momentary disturbance during feeding and will not trigger the oxidation compensation action.

[0067] Specifically, the sudden change in local ion concentration during the addition of alkali or cerium salt can cause instantaneous fluctuations in ORP, conductivity, and turbidity. These fluctuations are unrelated to the oxidation reaction, and if not addressed, they can lead to false triggering of oxidation compensation, wasting inert gas and even interfering with the normal precipitation process. The feeding window is set to a 1.5-second interval, from 0.5 seconds before the feeding solenoid valve opens to 1 second after it closes. This interval represents the range within which instantaneous fluctuations during feeding occur. If ORP and turbidity fluctuations occur within this time window, but the dissolved oxygen consumption rate does not increase by ≥30%, it is considered an instantaneous disturbance during feeding, and no oxidation compensation action is triggered. If the fluctuations continue after the window ends, and the dissolved oxygen consumption rate meets the oxidation determination criteria, it is re-determined as an oxidation disturbance. This logic can reduce the false trigger rate of oxidation compensation to below 1%.

[0068] The partitioned dynamic compensation execution module executes a graded oxidation compensation strategy based on the results of the first and second interference identification items to obtain the first treatment item. At the same time, it dynamically adjusts the alkaline feed rate based on the overall oxidation rate of the reactor to accelerate the precipitation process and shorten the precipitation and oxidation window period, thereby obtaining the second treatment item. This enables real-time monitoring and dynamic control of the reaction process.

[0069] It is important to note that, such as Figure 5As shown, the first processing item includes: for the characteristic point where oxidation is determined to have occurred, an inert gas is pulsed into the corresponding point for deoxygenation, and at least three pulse level parameters are set. The pulse level parameters are matched with the oxidation information item to obtain the first processing item.

[0070] Specifically, compared to traditional continuous aeration throughout the reactor, directional pulsed bubbling targets only the small area where oxidation occurs for deoxygenation, significantly reducing inert gas consumption. Compared to continuous aeration, pulsed aeration avoids the problems of violent liquid level fluctuations and air entrainment caused by continuous bubbling. Furthermore, the pulsed airflow has stronger turbulence and higher deoxygenation efficiency. The pulse level parameters are matched one-to-one with the oxidation level. For mild oxidation, the corresponding pulse parameters are: 0.1 MPa high-purity CEN2 or CEAr, 1 second pulse duration, and 5 second pulse interval. For moderate oxidation, the corresponding pulse parameters are: High-purity ceN2 or ceAr with an inlet pressure of 0.15 MPa, a single pulse duration of 2 s, and a pulse interval of 3 s. For severe oxidation, the corresponding pulse parameters are: high-purity ceN2 or ceAr with an inlet pressure of 0.2 MPa, a single pulse duration of 3 s, and a pulse interval of 2 s. The stopping condition is: the ORP at the corresponding point returns to the range of ±10 mV from the ORP baseline, and the dissolved oxygen consumption rate falls back to within ±10% of the baseline value. Bubbling is stopped immediately. Compared with the continuous aeration process of the whole reactor, this scheme reduces the inert gas consumption by ≥55% and significantly reduces the operating cost.

[0071] It should be noted that the method for obtaining the second treatment item includes: setting a conversion rate threshold, triggering alkali rate adjustment when the instantaneous conversion rate of precipitation is lower than the conversion rate threshold, setting at least three oxidation levels, and matching the adjustment range with the oxidation information item to obtain the second treatment item.

[0072] Specifically, free Ce 3+ The oxidation rate of precipitated Ce is 3+ More than 8 times, accelerating the precipitation process can quickly remove free Ce 3+ The conversion to a precipitate reduces the time that easily oxidized substrates remain, thereby reducing the degree of oxidation at its source. The 80% conversion threshold is set because when the precipitation conversion exceeds 80%, the remaining free Ce... 3+ If the proportion is less than 20%, continuing to increase the alkali feed rate will cause the system pH to rise rapidly to over 7.8, triggering the co-precipitation of non-cerium trivalent rare earth impurities and affecting product purity.

[0073] Specifically, the conversion rate threshold is set at 80%. If the instantaneous conversion rate of precipitation is <80%, the adjustment range of the alkali feed rate is matched with the oxidation level: 5% for mild oxidation, 10% for moderate oxidation, and 15% for severe oxidation. During the adjustment process, the pH of the system is monitored in real time to maintain the pH at 7.2-7.8 (Ce). 3+In the optimal precipitation range, if the pH reaches 7.8, the alkali rate is stopped. If the instantaneous conversion rate of precipitation is ≥80%, the alkali feed rate is no longer adjusted regardless of the oxidation level. Oxidation is controlled only by directional pulse bubbling. This adjustment logic can compress the precipitation oxidation window from 15 minutes in the traditional process to less than 3 minutes. The tetravalent cerium doping ratio of the product is stable at ≤0.2%, and the total content of non-cerium rare earth impurities is ≤10ppm.

[0074] In the specific implementation process, such as Figure 6 As shown, a rare earth separation company's 500L pilot line aims to produce trivalent cerium hydroxide with a purity of 99.8%. Previously, the company used a traditional process to produce trivalent cerium hydroxide: continuous argon gas purging throughout the reactor, offline sampling to detect valence state, resulting in a tetravalent cerium doping ratio generally between 0.8% and 1.2%, a valence state qualification rate of only 62%, and a argon gas consumption of 12 m³ / batch during continuous argon purging. This process was costly, and the purity fluctuated significantly between batches, failing to meet the supply requirements for energy storage anode precursors. This time, the monitoring and control system proposed in this scheme was used for modification. The specific implementation process is as follows: Pre-preparation and baseline calibration: 500L of cerium chloride feed solution after impurity removal, concentration 0.3 mol / L, total non-cerium rare earth impurities 32 ppm, total Fe / Mn variable valence impurities 7 ppm, reaction temperature set to 25℃, a four-electrode conductivity sensor and a 940nm near-infrared turbidity sensor installed in the central liquid phase zone of the reactor, conductivity sampling frequency 5Hz, turbidity... The sensor purges the window every 8 seconds. Monitoring is conducted at four key points: 10cm below the feed inlet, 2cm outside the agitator, in the dead zone at the bottom of the vessel, and 3cm below the liquid surface. The DO / ORP sampling frequency is 2Hz. After 12 minutes of pre-deoxygenation with argon gas throughout the vessel, the DO at all points drops to 0.03mg / L, below the threshold of 0.05mg / L. The automatic batch baseline is established: cerium-free background conductivity baseline is 2.3, with an initial pure Ce³⁺ conductivity contribution of 12.1mS / cm; turbidity baseline is 12NTU, with a theoretical final turbidity value of 740NTU; and ORP baseline is −120mV. The precipitation reaction then begins, with stirring at 350rpm. A 1.5mol / L ammonia solution is added at an initial rate of 1L / min. The system calculates the instantaneous conversion rate of the precipitation in real time: after 5 minutes of alkali addition, the real-time conductivity becomes 10.2mS / cm, and the real-time turbidity becomes 320NTU. The conductivity conversion rate is... =12.1−(10.2−2.3) / 12.1×100%≈41.3%, Turbidity method conversion rate =320−12740−12×100%≈42.3%, Instantaneous conversion rate after cross-correction ≈41.8%, matching degree 97%, judged as a normal precipitation process. At 7 minutes after adding alkali, the monitoring unit in the dead zone at the bottom of the vessel transmitted data: ORP value -70mV, a positive shift of 50mV from the ORP baseline. The DO consumption rate over the past 10 seconds increased from the baseline of 0.01mg / (L·s) to 0.018mg / (L·s), an increase of 80%, meeting the criteria for moderate oxidation. At this point, the temporal matching degree between the rate of decrease in conductivity and the rate of increase in turbidity was 81%, lower than... At 90% threshold, and with no feeding activity currently in progress, feeding disturbances were ruled out. The final determination was Ce(OH)3 oxidation and deterioration disturbance, triggering a compensation strategy. At this point, only the micro-bubbling head in the dead zone at the bottom of the vessel was activated, executing the pulse parameters corresponding to moderate oxidation: 0.15 MPa high-purity argon gas, single pulse duration 2 s, interval 3 s. After bubbling for 30 s, the ORP at this point dropped to -124 mV, and the DO consumption rate returned to 0.011 mg / (L·s). Bubbling was then stopped, and the current instantaneous conversion rate of the precipitate was 62%. The pH was below the 80% threshold, so the ammonia feed rate was increased by 10% to 1.1 L / min. Throughout the process, the pH of the system was monitored and remained stable between 7.4 and 7.6, not exceeding the impurity co-precipitation threshold of 7.8. At the 10-minute mark of alkali addition, ammonia was added for 1 second to fine-tune the pH. The ORP at the feed inlet momentarily dropped to -93 mV, but the DO consumption rate did not increase abnormally. The fluctuation occurred within the window of 0.5 seconds before the feed solenoid valve opened and 1 second after it closed. The system determined this to be a momentary disturbance during feed. The process was initiated without triggering compensation, avoiding unnecessary gas release and waste. At the 12-minute mark of alkali addition, the real-time cross-corrected conversion rate was 99.3%. Alkali addition was stopped, and stirring was maintained for 2 minutes. Afterward, the ORP at all points stabilized within the ORP baseline ± 5mV range, and DO ≤ 0.04 mg / L. The product was then filtered and washed with anhydrous ethanol. The final product test results were: 99.82% purity of cerium hydroxide in its trivalent state, and 8 ppm of non-cerium rare earth impurities. The total argon consumption for this batch was 4.9 m³ / min. 3 Compared with the traditional full-reactor continuous aeration process, it reduces energy consumption by 59%, meeting the energy reduction target of ≥55%. After 12 batches of continuous operation, the valence qualification rate was 100%, and the fluctuation of tetravalent cerium content between batches was ≤0.05%, which is far lower than the 0.7% fluctuation of the traditional process.

[0075] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. An online monitoring and intelligent control system for a high-purity cerium hydroxide precipitation reaction process, comprising: The precipitation process dual-parameter tracer module, the local oxidation feature capture module, the multi-signal decoupling and identification module, and the partitioned dynamic compensation execution module are all connected to the PLC control unit. Its features are: The precipitation process dual-parameter tracer module consists of a first sensor and a second sensor deployed in the reactor. The first sensor collects the real-time conductivity change value and the real-time solid phase turbidity value of the reaction system, respectively. The instantaneous conversion rate of Ce(OH)3 precipitate is obtained by cross-correction, and the precipitation information item is obtained. The local oxidation feature capture module consists of multiple monitoring units. It sets feature points inside the reactor to obtain the set point items, installs monitoring units based on the set point items to obtain the monitoring unit installation items, and collects the instantaneous dissolved oxygen concentration and ORP value in different areas to identify the location, rate and extent of oxidation reaction and obtain oxidation information items. The multi-signal decoupling identification module has a built-in time-series matching algorithm model to distinguish the sources of disturbances in the system's pH, turbidity, and ORP signals, thus obtaining the first interference identification item. At the same time, it eliminates the interference from instantaneous parameter fluctuations caused by the addition of alkali / cerium salt, thus obtaining the second interference identification item. The partitioned dynamic compensation execution module executes a graded oxidation compensation strategy based on the results of the first interference identification item and the second interference identification item to obtain the first processing item. At the same time, it dynamically adjusts the alkaline feed rate based on the overall oxidation rate of the reactor to accelerate the precipitation process and shorten the precipitation and oxidation window period to obtain the second processing item, thereby realizing real-time monitoring and dynamic control of the reaction process. The method for obtaining the first interference identification item includes: A time-series matching threshold is set. When the overlap of the time-series curves of the rate of decrease in conductivity and the rate of increase in turbidity exceeds the time-series matching threshold, it is determined to be a disturbance in the formation of Ce(OH)3 precipitate. When the rate of decrease in conductivity and the rate of increase in turbidity do not exceed the time-matching threshold, but the ORP shifts positively and the dissolved oxygen consumption rate increases abnormally, it is determined to be a Ce(OH)3 oxidation and deterioration disturbance. The method for obtaining the second interference identification item includes: Set a window time value. When ORP and turbidity fluctuate within the time window when the feed port solenoid valve is in the open state, but the dissolved oxygen consumption rate does not increase abnormally, it is judged as a momentary disturbance of feed and oxidation compensation action is not triggered. The first processing item includes: For the characteristic points where oxidation is determined, inert gas is pulsed into the corresponding points for deoxygenation. At least three pulse level parameters are set, and the pulse level parameters are matched with the oxidation information items to obtain the first processing item. The method for obtaining the second processing item includes: Set a conversion rate threshold. When the instantaneous conversion rate of precipitation is lower than the conversion rate threshold, trigger the alkali rate adjustment. Set at least three oxidation levels and match the adjustment range with the oxidation information item to obtain the second treatment item.

2. The online monitoring and intelligent control system for the high-purity cerium hydroxide precipitation reaction process according to claim 1, characterized in that: The precipitation process dual-parameter tracking module includes: The first sensor and the second sensor are deployed in the central liquid phase zone of the reactor. The first sensor is a four-electrode type corrosion-resistant in-situ conductivity sensor, and the second sensor is a near-infrared dynamic turbidity sensor with a pneumatic window self-cleaning function. The wavelength of the second sensor is set to obtain the wavelength setting item. The sampling frequency of the first sensor and the purge window of the second sensor are set, and the real-time conductivity change value and real-time solid phase turbidity value of the reaction system are collected by the first sensor and the second sensor, respectively.

3. The online monitoring and intelligent control system for the high-purity cerium hydroxide precipitation reaction process according to claim 1, characterized in that: The method for obtaining the sedimentation information items includes: Pre-calibrate the conductivity baseline, turbidity baseline, and Ce content of the cerium-free background system. 3+ The theoretical final turbidity value of complete precipitation is calculated by measuring the decrease in free Ce in real time. 3+ Consumption rate; The solid-phase formation rate of Ce(OH)3 is calculated by the real-time turbidity rise value, and the average value is taken to obtain the instantaneous conversion rate of precipitation, thereby obtaining the precipitation information item.

4. The online monitoring and intelligent control system for the high-purity cerium hydroxide precipitation reaction process according to claim 1, characterized in that: The method for obtaining the monitoring unit installation items includes: The feature points of the set point item include at least four fixed points, including the feed mixing zone at the feed port, the radial outer vortex shear zone of the agitator, the dead corner zone at the bottom of the inner wall of the reactor, and the gas-liquid contact zone. Each monitoring unit is coaxially integrated with a needle-type dissolved oxygen microelectrode with built-in PT100 temperature compensation, an in-situ ORP electrode, and a directional microbubble head. The sampling frequency of the dissolved oxygen microelectrode and ORP electrode of each monitoring bubble unit is set, and the outlet position of the directional microbubble head is set, thereby obtaining the installation items of the monitoring unit.

5. The online monitoring and intelligent control system for the high-purity cerium hydroxide precipitation reaction process according to claim 1, characterized in that: The local oxidation feature capture module incorporates ORP baseline automatic calibration logic, and the methods for obtaining oxidation information items include: Set the calibration logic and the dissolved oxygen threshold. Before each batch of reaction, after the system is pre-deoxygenated to below the dissolved oxygen threshold, calibrate the ORP baseline value of that batch to obtain the batch calibration items. The oxidation reaction judgment rules are divided into at least three levels to obtain judgment level items. Corresponding judgment logic is generated based on the judgment level items. The batch calibration items are offset based on the judgment logic. At the same time, the dissolved oxygen consumption rate at the corresponding point is adjusted based on the drift result, thereby obtaining the oxidation information item.

Images