A method and system for controlling crystallization in potassium nitrate production

Abstract

This invention relates to the field of chemical crystallization control technology, specifically to a method and system for controlling the crystallization of potassium nitrate production. Through steps such as adaptive initialization of the feed and system, sequential establishment and precise control of vacuum and temperature gradients, coordinated regulation of supersaturation and solid content, intelligent balancing of multi-level liquid levels, and adaptive optimization based on product quality feedback, closed-loop automatic control of the entire crystallization process is achieved. Its core lies in employing multi-sensor data fusion and model predictive feedforward control to dynamically coordinate key operational variables such as feed, vacuum, and discharge. This invention enables continuous, efficient, and stable production of potassium nitrate, significantly improving the particle size and uniformity of the crystal product and increasing production efficiency.

Description

Technical Field

[0001] This invention relates to the field of chemical crystallization control technology, and in particular to a method and system for controlling crystallization in potassium nitrate production. Background Technology

[0002] Potassium nitrate, as an important chemical product, is widely used in agriculture, chemical industry, and medicine. Traditional potassium nitrate crystallization processes often employ a single-reactor batch crystallization method, where the cooling rate is controlled by cooling water, allowing the saturated solution to slowly cool and precipitate crystals in a single crystallization vessel. This process has the following significant drawbacks: each production cycle lasts several hours, making continuous production impossible and resulting in low equipment utilization; the crystals are small in size, unevenly distributed, and incomplete in shape, affecting product appearance and purity; and key parameters such as temperature and concentration are not precisely controlled, leading to unstable product quality. Therefore, there is an urgent need for a continuous, automated, and efficient crystallization process to improve the product quality and production efficiency of potassium nitrate. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for controlling crystallization in potassium nitrate production, thereby improving the product quality and production efficiency of potassium nitrate.

[0004] To achieve the above objectives, in a first aspect, the present invention provides a method for controlling crystallization in potassium nitrate production, comprising the following steps: S1. Based on the concentration and liquid level feedback, the feed rate of the saturated potassium nitrate solution into the multi-stage crystallization system is adaptively controlled to establish and stabilize the initial liquid level of each stage. S2. Based on the system initialization completion signal, the vacuum system is started sequentially, and graded pressure control is implemented to establish a temperature gradient from the first stage to the third stage crystallization device 103; S3. Based on the supersaturation data obtained from real-time calculation and the solid content data measured online, the vacuum degree and the discharge valve are adjusted in a coordinated manner through a combination of model prediction feedforward and feedback control to stabilize the supersaturation and solid content within the target range. S4. Based on the real liquid level data obtained by multi-source fusion, dynamic material balance feedforward and liquid level feedback control are used to coordinate the adjustment of the feed valve, interstage transfer pump and discharge valve to maintain the system hydraulic balance and crystal growth residence time stability. S5. Based on the crystal growth maturity and system health, determine the timing of material discharge and adaptively optimize key process settings based on the quality data feedback of the discharged crystals.

[0005] The feed rate of the saturated potassium nitrate solution into the multi-stage crystallization system is adaptively controlled based on concentration and liquid level feedback, including: The concentration of the feed liquid is monitored by an online concentration sensor. If the concentration is lower than the safety threshold, the feed rate is increased; if the concentration is higher than the safety threshold, the feed rate is decreased. Once the liquid level in each crystallization unit reaches the set value and stabilizes, the feed will be switched to normal saturated solution.

[0006] Based on the system initialization completion signal, the vacuum system is sequentially started, and graded pressure control is implemented to establish a temperature gradient from the first to the third stage crystallization device 103, including: Once the liquid level in the third-stage crystallizer 103 stabilizes at the set high liquid level, the water ring vacuum pumps connected to each stage of the crystallizer are started sequentially. The pressure of the first-stage crystallization unit 101 is monitored in real time. Once the pressure drops to the starting pressure threshold of the Roots pump, the Roots pumps are started sequentially. By controlling the Roots pump frequency with the pressure of the three-stage separator as the main loop and adjusting the vacuum fine-tuning valves with the pressures of the first and second stage separators as the secondary loops, the pressures of the first, second, and third stage separators are controlled at the first, second, and third set values, respectively, thereby enabling the material outlet temperature to reach the corresponding gradient temperature.

[0007] The method further includes: The actual temperature at each material outlet is compared with the target temperature, and the resulting temperature difference signal is used as a fine-tuning amount to dynamically correct the pressure setpoint of the corresponding separator.

[0008] The supersaturation obtained through real-time calculation includes: The temperature and online density of the circulating mother liquor in the first to third stage crystallization units 103 are collected. The actual concentration is calculated by using a preset solution property model, and the supersaturation is calculated by combining the equilibrium solubility at the current temperature. The solid content measured online was obtained using a microwave solid content analyzer installed on the discharge pipeline of the third-stage crystallization unit 103.

[0009] Among them, the vacuum level and the discharge valve are adjusted in a coordinated manner by combining model prediction feedforward and feedback control, including: The supersaturation of the first and second stage crystallization devices 102 is used as the controlled variable, and the vacuum setting value of the corresponding separator is controlled by PID; the solid content of the third stage crystallization device 103 is used as the controlled variable, and the opening of its discharge valve is controlled by PI. Meanwhile, the change in the opening of the discharge valve is used as a feedforward signal to predict the impact on supersaturation and is compensated in advance to the vacuum control loop. The change in the vacuum setpoint is used as a feedforward signal to predict the impact on solid content and is compensated in advance to the discharge valve control loop.

[0010] Among them, multi-source fusion obtains real liquid level data, including: The radar level gauge readings, differential pressure transmitter readings, and weighing module readings for each crystallization stage are obtained. An adaptive weighted fusion algorithm based on confidence assessment is then used to calculate the actual level signal for control.

[0011] Among them, dynamic material balance feedforward and liquid level feedback control are used to coordinate the adjustment of the feed valve, interstage transfer pump and discharge valve, including: The actual liquid level of the third-stage crystallization unit 103 is used as the direct control object. The opening of its discharge valve is adjusted by a PID controller to stabilize the crystal growth residence time. A dynamic material balance model is constructed to predict the required amount of fresh material replenishment in real time based on the changes in the opening degree of the discharge valve and the total crystallization rate. The predicted amount of fresh material replenishment is then used as a feedforward signal to directly control the first-stage feed valve. The second-stage crystallization unit 102 is used as a buffer coordination stage, and its liquid level setting value is dynamically optimized. The frequency of the inter-stage transfer pump is adjusted to make it smoothly buffer between the first and third-stage crystallization units 103.

[0012] Among these, the timing of material discharge is determined based on a comprehensive assessment of crystal growth maturity and system health, and key process settings are adaptively optimized based on feedback from the quality data of the discharged crystals, including: When the crystal growth rate in the third-stage crystallization unit 103 is continuously lower than the threshold and the supersaturation is stable at a low level, offline sampling analysis is triggered; if the offline particle size analysis confirms that the crystal particle size and distribution meet the requirements and the system's control loops are operating stably, it is determined that the optimal discharge window period has been entered. Establish and update a digital twin model of the crystallization process, linking the average values ​​of key control parameters from historical batches with the final crystal quality data; when the actual product quality deviates from the target, use the digital twin model of the crystallization process to calculate the optimization adjustment amount of the key process settings and apply it to subsequent production batches.

[0013] Secondly, the present invention provides a potassium nitrate production crystallization control system, applied to a potassium nitrate production crystallization control method as provided in the first aspect, comprising: The first-stage crystallization apparatus, the second-stage crystallization apparatus, and the third-stage crystallization apparatus are connected in sequence. Vacuum system, used to provide vacuum for crystallization units at all levels, including water ring vacuum pumps, Roots pumps and their piping valves; Material conveying system, used for material conveying, including feed pipeline and regulating valve, interstage transfer pump, circulation pump and discharge pipeline and regulating valve; Sensor arrays for data acquisition include radar level gauges, pressure transmitters, temperature sensors, online density meters, concentration sensors, microwave solid content analyzers, and weighing modules installed on the support legs of the crystallization unit. The server is electrically connected to the sensor array, the vacuum system, and the material conveying system, respectively, and is configured to perform steps S1 to S5.

[0014] This invention discloses a method and system for controlling the crystallization of potassium nitrate production. Through steps such as adaptive initialization of the feeding and system, sequential establishment and precise control of vacuum and temperature gradients, coordinated regulation of supersaturation and solid content, intelligent balancing of multi-level liquid levels, and adaptive optimization based on product quality feedback, it achieves closed-loop automatic control of the entire crystallization process. Its core lies in employing multi-sensor data fusion and model predictive feedforward control to dynamically coordinate key operational variables such as feeding, vacuum, and discharging. This invention enables continuous, efficient, and stable production of potassium nitrate, significantly improving the particle size and uniformity of the crystal product and increasing production efficiency. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0016] Figure 1 This is a schematic diagram of the steps in a method for controlling crystallization during potassium nitrate production according to the first embodiment of the present invention.

[0017] Figure 2 This is a schematic flowchart of a method for controlling crystallization in potassium nitrate production provided by the present invention.

[0018] Figure 3 This is a schematic diagram of a potassium nitrate production crystallization system provided by the present invention.

[0019] Figure 4 This is a crystallization diagram provided by the present invention.

[0020] Figure 5 This is a schematic diagram of the electronic device of the present invention.

[0021] In the diagram: 101 - First-stage crystallization apparatus, 102 - Second-stage crystallization apparatus, 103 - Third-stage crystallization apparatus. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0023] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0024] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0025] The first embodiment of this application is as follows: Please see Figures 1-3 This invention provides a method for controlling crystallization during potassium nitrate production, comprising the following steps: S1. Based on the concentration and liquid level feedback, the feed rate of the saturated potassium nitrate solution into the multi-stage crystallization system is adaptively controlled to establish and stabilize the initial liquid level at each stage.

[0026] Specifically, the feeding speed is not a fixed value, but is dynamically adjusted according to the actual needs during the system initialization phase, mainly based on the following two core parameters: Baseline feed rate (V_base): An initial rate is calculated based on the design volume of the crystallization system and the expected initial filling time (e.g., main filling to be completed in 2-3 hours). This rate serves as the starting point for control.

[0027] Real-time concentration feedback: During system initialization, the first solution pumped in is a "low-concentration stock solution" (a solution that will not crystallize upon cooling). An online concentration sensor (such as a refractometer or conductivity meter) installed on the feed line monitors the concentration of the feed solution in real time.

[0028] The feed regulating valve opens at a rate of V_base to feed material into the first-stage crystallizer 101. If the feed concentration is lower than the preset "safe concentration threshold" (ensuring no crystallization during the initial cooling phase), the control system will maintain or appropriately increase the feed rate to quickly establish the liquid level and shorten the start-up time. If the feed concentration is close to or higher than the "safe concentration threshold," the control system will linearly reduce the opening of the feed regulating valve to reduce the amount of high-concentration material fed, preventing uncontrolled crystallization from occurring prematurely in the first-stage crystallizer 101 before the system temperature and vacuum are fully established, thus avoiding blockage of pipelines or pumps.

[0029] Once the liquid levels in each unit of the system are basically established, and the vacuum and temperature are close to the process setpoints, the control system will automatically switch the feed source to a 95°C saturated potassium nitrate solution. At this point, the control objective of the feed rate changes from "rapid filling" to "maintaining a steady-state liquid level and concentration," and its setpoint is transferred to the subsequent liquid level interlock control system.

[0030] The relevant data is collected using high-precision instruments and transmitted in real time to the central server (PLC / DCS): Feed velocity acquisition: An electromagnetic flow meter is installed on the main feed pipeline to measure and output the volumetric flow rate signal (m³ / s) in real time and continuously. 3 / h).

[0031] Cumulative feed volume acquisition: The central control system receives the instantaneous flow signal from the flow meter and calculates the cumulative feed volume in real time through time integration. Combined with the real-time density of the feed liquid (which can be converted from concentration and temperature or measured by an online density meter), the cumulative mass of potassium nitrate added can be accurately calculated.

[0032] Liquid level acquisition: High-precision radar level gauges are installed on the first-stage crystallization unit 101, the second-stage crystallization unit 103, and the third-stage crystallization unit 103 to continuously measure and output the liquid level height signal (mm). This signal is the most direct feedback for liquid level control.

[0033] Liquid level stability is achieved through a multi-stage linkage and feedforward compensation liquid level control loop, ensuring that the liquid level remains constant during dynamic material transfer in each crystallization unit.

[0034] Liquid level setting and initialization: The liquid level setting value (L1_sp) for the first-stage crystallization device 101 is 2000 mm. The liquid level setting value (L2_sp) of the second-stage crystallization device 102 is 50-100 mm from the top cover; The liquid level setting value (L3_sp) of the third-stage crystallization device 103 is 50-100 mm from the top cover; After feeding begins, the control system gradually raises the liquid level at each stage to its respective set value as the initial target.

[0035] During continuous operation, level control is no longer a simple single-point control, but a coupled system: Main control loop (downstream drive): The starting control point is the liquid level (L3) at the final discharge point (third-stage crystallization unit 103).

[0036] When L3 decreases, the PLC first increases the frequency of the three-stage discharge pump in an attempt to maintain L3. If L3 continues to decrease (indicating that internal material consumption is faster than replenishment), a feedforward signal is triggered.

[0037] Inter-stage interlocked material replenishment: Based on the decreasing trend and rate of L3, the PLC calculates the amount of material that needs to be replenished and proportionally increases the frequency of the transfer pumps in the second- to third-stage crystallization units 103. Similarly, the decrease in L2 will trigger a chain reaction that increases the frequency of the transfer pumps in the first- to second-stage crystallization units 102.

[0038] Source feed compensation (feedforward-feedback composite): Feedback regulation: The liquid level (L1) of the first-stage crystallizer 101 serves as the final buffer for the material balance of the entire system. When L1 decreases due to increased downstream flow, the PID controller adjusts the opening of the feed regulating valve to increase the feed rate of fresh saturated solution, causing L1 to return to the set value.

[0039] Feedforward regulation: Based on the magnitude and speed of the frequency increase of the secondary and tertiary transfer pumps, the control system calculates an estimated additional material demand in advance and directly adds a feedforward signal to the feed regulating valve. This is equivalent to increasing the feed before L1 decreases significantly, which significantly reduces liquid level fluctuations and improves the system's response speed and stability.

[0040] When the liquid level in any crystallization unit reaches the high-high alarm value, all upstream transfer pumps and feed regulating valves will immediately interlock and stop. When the liquid level in any crystallization unit reaches the low-low alarm value, downstream discharge or transfer pumps will interlock and reduce frequency or stop to prevent pump cavitation and damage. S2. Based on the system initialization completion signal, the vacuum system is started sequentially, and graded pressure control is implemented to establish a temperature gradient from the first stage to the third stage crystallization device 103.

[0041] Specifically, the vacuum system is not simply started after feeding is completed, but is intelligently triggered by the system initialization completion status confirmed in step 1. When the control system confirms that its liquid level (L3) has reached and stabilized at the set high liquid level value (50-100mm from the top cover) for a preset stabilization time (e.g., 5 minutes), this signifies that the system "container is ready." Simultaneously, the liquid levels (L1) of the first-stage crystallizer 101 and (L2) of the second-stage crystallizer 102 must also be confirmed to be within their respective set ranges. Furthermore, the concentration sensor on the feed line must confirm that the material has been switched to a normal saturated solution. All relevant circulation pumps are operating normally, and the bottom discharge valves of each crystallizer are in a preset closed or low-flow state.

[0042] All the aforementioned liquid level, concentration, and equipment status signals are acquired through their respective transmitters (liquid level, concentration, and pump operation feedback contacts) and transmitted in real time to the field PLC control cabinet via 4-20mA or industrial Ethernet signals. The control program within the PLC continuously scans these signals. When all the Boolean conditions—"L3 stable high liquid level," "L1 / L2 liquid level normal," "concentration meets standard," and "pump operation normal"—are true, the "vacuum system start permission" logic flag within the PLC is set. Subsequently, this permission signal is displayed as "vacuum system start permitted" on the field control cabinet's human-machine interface (HMI) and is also transmitted to the central server as the enable condition for a remote start command.

[0043] Then, after the program automatically issues a start command, the PLC first starts the water ring vacuum pumps corresponding to the first, second, and third stage crystallizers 103. The water ring pumps, acting as primary vacuum devices, begin evacuating the chambers of their respective crystallizers. The pressure transmitter at the top of each crystallizer measures and transmits the absolute pressure value (unit: kPa) to the PLC in real time. The PLC continuously compares the pressure values ​​of each separator. The direct control objective at this point is to first bring the absolute pressure of the entire system (usually using the highest-pressure first-stage separator as a reference) below the safe start-up threshold pressure of the Roots pump (10 kPa).

[0044] When all water ring pumps are operating normally and the pressure of the first-stage crystallization unit 101 (a critical reference point) drops to ≤10 kPa, the "start-up preparation" conditions for starting the Roots pump are met. To reduce the impact on the high-power Roots pump and ensure vacuum stability, the control system employs a delayed sequential start-up. After the conditions are met, the PLC first issues a command to open the isolation valve of the vacuum buffer tank connected before the Roots pump inlet. After the valve is fully open signal is returned, there is a delay of several seconds before the PLC sends a start command to the Roots pump frequency converter. After the Roots pump starts, it operates in series with the water ring pump, forming a higher pumping capacity, thereby achieving the lower process vacuum level more quickly.

[0045] The pressure control of the third-stage crystallization unit 103 is implemented through "graded setpoints and dynamic correlation balance," rather than independent control of each stage, with graded settings: Pressure setpoint (P1_sp) for the first-stage crystallization unit 101: 12 kPa (absolute pressure); Pressure setpoint (P2_sp) for the second-stage crystallizer 102: 5 kPa; Pressure setting value (P3_sp) of the third-stage crystallization unit 103: 5 kPa.

[0046] The pressure control loop of the three-stage separator is set as the main loop. Its pressure transmitter signal (P3) is compared with the set value (P3_sp = 5 kPa), and after PID calculation, the output control signal directly adjusts the frequency of the Roots pump's inverter. The Roots pump, as the main power source for vacuuming the entire system, determines the system's ultimate vacuum and pumping rate.

[0047] The feedforward-feedback composite regulation works as follows: the pressure control of the first and second stage separators serves as a secondary loop. Their respective pressure transmitter signals (P1, P2) are compared with the setpoints. Their PID outputs do not directly control the vacuum pump, but are used to dynamically adjust the opening of the "vacuum fine-tuning valve" installed on the exhaust pipe of their respective separators. By changing the flow resistance of this branch, the pressure of this stage is finely adjusted. To avoid excessive "backflow" pressure differences between the three stages due to different pressure settings interfering with the flow field, when the actual pressure difference between P1 and P2 exceeds a certain safety value, the vacuum fine-tuning valve of P2 is finely adjusted so that its pressure fluctuates slightly while meeting the process requirements (5 kPa), maintaining a stable pressure difference with P1 and ensuring smooth material flow.

[0048] Temperature control and pressure control are deeply coupled to achieve precise management of supersaturation. Platinum resistance thermometers (PT100) are installed on the material pipelines at each key node (the outlet of the first-stage flash evaporator, the outlet of the second-stage crystallizer 102, and the outlet of the third-stage crystallizer 103). In conjunction with temperature transmitters, the material temperatures (T1, T2, T3) are collected in real time and the signals are transmitted to the PLC / DCS.

[0049] In vacuum crystallization, the temperature of the material is primarily determined by the boiling point of the solution at that pressure. Therefore, controlling the pressure is fundamental to controlling the temperature. This scheme adopts a strategy of "pressure as the primary factor, with temperature control as an auxiliary factor." As mentioned earlier, by controlling P1, P2, and P3, the theoretical values ​​of T1, T2, and T3 (corresponding boiling point temperatures at the given pressure) are essentially determined. Due to variations in material concentration and flow rate, the actual temperature may deviate slightly from the boiling point. Therefore, the system compares the measured temperatures (T1, T2, T3) with their respective temperature setpoints (T1_sp = 70°C, T2_sp = 50°C, T3_sp = 30°C).

[0050] This temperature difference signal is used as a correction input to a secondary controller in a cascade control loop. The output of this secondary controller, as a fine-tuning signal, is superimposed on the output of the corresponding pressure control loop (primary controller). For example, if the measured temperature of T2 is 52°C, which is higher than the setpoint of 50°C, the correction loop will output a signal to temporarily and slightly lower the pressure setpoint of P2 from 5 kPa (e.g., to 4.8 kPa), thereby further reducing the boiling point and causing T2 to drop. This process is performed dynamically and in small increments to ensure accurate temperature control.

[0051] S3. Based on the supersaturation data obtained from real-time calculation and the solid content data measured online, the vacuum degree and the discharge valve are adjusted in a coordinated manner through a combination of model prediction feedforward and feedback control to stabilize the supersaturation and solid content within the target range.

[0052] Specifically, since supersaturation and solid content are both indirect parameters, this scheme uses a combination of multi-sensor data fusion and online calculation using a thermodynamic model to obtain them in real time.

[0053] Collection of supersaturation: Temperature: High-precision platinum resistance thermometers are installed on the circulating mother liquor outlet pipes of each crystallization unit to measure the temperature T1, T2, and T3 (unit: °C) in real time.

[0054] Solution density: At the same measurement point, an online vibrating densitometer was installed to continuously measure the density ρ_soln of the circulating mother liquor (unit: g / cm³). 3 ).

[0055] Calculation process: Concentration inversion: The control system has a built-in ternary model of the temperature-density-concentration relationship of potassium nitrate solution. By substituting the real-time collected T and ρ_soln into the model, the actual mass concentration of potassium nitrate in the solution, C_actual (unit: g / 100g water), can be calculated.

[0056] Saturation concentration lookup: The system also includes a built-in database of potassium nitrate solubility curves. Based on the real-time temperature T, the equilibrium saturation concentration C_sat(T) of potassium nitrate at that temperature is obtained through interpolation.

[0057] Supersaturation calculation: Absolute supersaturation ΔC = C_actual - C_sat(T) (unit: g / 100g water) Supersaturation σ = (C_actual - C_sat(T)) / C_sat(T) (dimensionless) This scheme uses supersaturation σ as the core control index because it better reflects the strength of the driving force for nucleation and growth.

[0058] Solid content collection: An online microwave solids content analyzer is installed on the bottom slurry discharge pipe of the third-stage crystallization unit 103. Based on the microwave penetration method, this instrument can directly and non-contactly measure the mass percentage of solid crystals in the slurry, i.e., the solids content M (unit: %). Simultaneously, an online densitometer is installed in parallel on the slurry pipe to measure the apparent density ρ_slurry of the slurry. By comparing the theoretical slurry density calculated from the solids content M with the measured density, online calibration and fault diagnosis of the solids content measurements can be performed.

[0059] All raw measurement signals are processed by adaptive Kalman filtering to eliminate noise and measurement fluctuations. The system continuously calculates the moving averages (σ_avg, M_avg) and standard deviations (σ_std, M_std) of σ and M over the past 15 minutes. The averages reflect the current level, and the standard deviations reflect process stability; both serve as the basis for control decisions.

[0060] A dual-closed-loop-cross-feedforward collaborative control system was constructed, with supersaturation control as the dominant factor and solid content control as the subordinate factor. The two are dynamically coupled through material and energy balance.

[0061] The supersaturation (σ1, σ2) in the first and second stage crystallization units 102 is stabilized within the metastable region of 0.05-0.15 and 0.02-0.08, respectively, to suppress explosive nucleation and promote orderly crystal growth. The solid content M at the outlet of the third stage crystallization unit 103 is stabilized within the range of 20-25% to ensure sufficient growth residence time for the crystals while maintaining the flowability of the output material.

[0062] Oversaturation control loop (main loop): Controller: A PID controller with anti-integral saturation function is adopted.

[0063] Controlled variables: supersaturation (σ1, σ2) of the first-stage and second-stage crystallization devices 102.

[0064] Manipulated variables: Vacuum degree setting values ​​(P1_sp, P2_sp) of the corresponding crystallization device.

[0065] Control logic: When the supersaturation σ exceeds the upper limit of the set range, the controller outputs a signal to increase the vacuum setting value of the corresponding separator (e.g., fine-tuning P1_sp from 12 kPa to 11.5 kPa). Stronger evaporative cooling causes the solution temperature to drop faster, while solvent evaporation increases the concentration. However, the decrease in saturated concentration due to temperature drop is usually more significant, and the combined effect causes the supersaturation σ to fall back. Conversely, when σ is too low, the vacuum setting value is slightly reduced to slow down cooling and evaporation, causing σ to rise again.

[0066] Solid content control loop (slave loop): Controller: A proportional-integral (PI) controller is used.

[0067] Controlled variable: Solid content M at the outlet of the third-stage crystallization unit 103.

[0068] Manipulated variable: The opening degree of the slurry discharge regulating valve at the bottom of the third-stage crystallization unit 103.

[0069] Control logic: When the solid content M is higher than the set upper limit, the opening of the discharge valve is increased to accelerate the discharge of crystal products and reduce the solid content in the crystallization device. When M is lower than the set lower limit, the opening of the discharge valve is decreased to prolong the residence time of the crystals in the system, allowing them to continue growing and thus increasing the solid content.

[0070] The two control loops mentioned above are interconnected and interfere with each other. For example, adjusting the discharge valve (changing the solid content) will affect the total amount of crystals in the system and the surface area available for growth, thus affecting the supersaturation; conversely, adjusting the vacuum level (changing the supersaturation) will affect the crystallization rate, thus changing the solid content. Therefore, this scheme introduces a model prediction feedforward (MPFF) and dynamic decoupling algorithm: Feedforward compensation: When the solids content controller adjusts the discharge valve opening, the change and its rate of change are input into a dynamic material balance model in real time. This model quickly predicts the impact (Δσ1_ff, Δσ2_ff) of this discharge change on the supersaturation of the first and second stage crystallizers 102 within the next 1-3 minutes. Subsequently, this predicted impact is used as a feedforward signal and superimposed on the output of the supersaturation controller in advance, allowing for immediate fine-tuning of P1_sp and P2_sp, thereby counteracting the supersaturation disturbance caused by the discharge change.

[0071] Similarly, when the supersaturation controller adjusts the vacuum setting value, this change is also input into an energy and crystallization kinetics coupling model to predict its potential impact (ΔM_ff) on the solid content M of the third-stage crystallizer 103, and is used as a feedforward signal to compensate the feed valve controller in advance.

[0072] Decoupling control: In the control algorithm, the aforementioned feedforward compensation quantity and the deviation feedback quantity of the controlled variable (σ, M) are weighted and synthesized to form the final control command. This effectively reduces the coupling oscillation between the two control loops, enabling the system to smoothly and cooperatively achieve dual-objective stability.

[0073] All data acquisition, calculation, and control command output are synchronized with a 1-second cycle. The control system has a built-in expert rule base. When the system detects a continuous increase in the standard deviation of oversaturation σ_std (process instability), it will automatically and appropriately reduce the gain of the PID controller and enhance filtering to make the control action smoother; conversely, when the process is stable, it will appropriately increase the response speed.

[0074] If the supersaturation σ1 of the first-stage crystallization unit 101 continuously exceeds the safety upper limit (e.g., 0.25) and reaches the set time, the system determines that there is a risk of explosive nucleation and will automatically trigger the "oversaturation" alarm, and execute interlocking actions: suspend feeding, gradually increase the vacuum level of this stage to forcibly and rapidly reduce the supersaturation, and increase the frequency of the circulating pump in this stage to enhance mixing. If the solid content M continuously falls below the lower limit (e.g., 15%), and the supersaturation σ3 is also at a low level, the system determines that crystal growth is almost stagnant, will trigger the "growth abnormality" alarm, and will automatically execute diagnostic procedures to check the vacuum system, temperature sensor, and concentration model.

[0075] S4. Based on the real liquid level data obtained by multi-source fusion, the feed valve, interstage transfer pump and discharge valve are adjusted in a coordinated manner through dynamic material balance feedforward and liquid level feedback control to maintain the hydraulic balance of the system and the stability of the crystal growth residence time.

[0076] Specifically, to overcome the limitations of single liquid level measurement in complex environments of crystallization devices (such as foam, boiling, and crystal adhesion), this solution adopts a multi-source data acquisition method that combines main and auxiliary measurements with soft measurement integration.

[0077] Main measurement: High-frequency modulated continuous wave (FMCW) radar level gauges are installed vertically downwards at the top of the gas phase space of the first-stage crystallization unit 101, the second-stage crystallization unit 103, and the third-stage crystallization unit. This non-contact measurement is unaffected by changes in medium density, temperature, or vacuum level, and has strong penetrating power against foam and vapor, providing a direct distance signal (in mm, converted to liquid level height L_radar).

[0078] The radar level gauge has a built-in echo processing algorithm that can effectively distinguish between real liquid surface echoes and false echoes caused by agitators and internal components.

[0079] Auxiliary Measurement and Verification: Differential pressure transmitter: A capillary remote-transmission dual-flange differential pressure transmitter is installed between the bottom liquid phase zone and the top gas phase zone of each crystallization unit. It indirectly calculates the liquid level height L_dp = ΔP / (ρ * g) by measuring the static pressure difference ΔP, where ρ is the medium density updated based on real-time data from the online density meter, and g is the acceleration due to gravity.

[0080] Weighing Module: High-precision weighing sensors are installed on the support legs of the secondary and tertiary crystallization devices 103 to monitor the total mass m_total of the entire crystallization device (including the shell, internal components, and materials) in real time. By subtracting the tare weight and combining it with real-time solid content M and density ρ data, the total mass of the liquid-solid mixture inside the device can be calculated, and then the equivalent average liquid level L_weight based on mass can be derived. This method is unaffected by bubbles, boiling, or internal components, and is a reliable benchmark for verifying the volumetric liquid level.

[0081] The data fusion module within the PLC receives three signals per second: L_radar, L_dp, and L_weight. The system evaluates the confidence level of each signal in real time. For example, when the radar signal strength is high and the echo is clear, the confidence level is high; when the differential pressure measurement changes significantly due to large fluctuations in density ρ, its confidence level temporarily decreases; the weighing signal is usually used as a high-confidence benchmark, but its response speed is slightly slower. Using an existing adaptive weighted fusion algorithm, weights are dynamically allocated based on the real-time confidence level to calculate an optimal estimate, namely the "true liquid level" L_true. This value serves as the sole authoritative liquid level signal (L1_true, L2_true, L3_true) for subsequent control and monitoring.

[0082] The system continuously calculates the standard deviation of L_true at each level within a short period (e.g., 1 minute) as the "level fluctuation index". A high fluctuation index may indicate uneven feeding, severe vacuum fluctuations, or unstable operation of the circulating pump. Linear fitting is performed on L_true to determine whether it rises slowly, falls slowly, or remains stable over a long period (e.g., 10 minutes), providing early warning of potential material imbalances.

[0083] When L_radar and L_dp show a continuous and regular deviation and L_dp>L_radar, the system determines that there may be a large amount of stable foam layer on the liquid surface, triggering a "foam layer alarm" and suggesting that the operator check the defoamer addition system or adjust the vacuum level.

[0084] When L_true remains stable, but the weighing module signal L_weight continues to rise slowly and the discharge solid content M does not increase accordingly, the system judges that there may be a large amount of crystals adhering and accumulating on the wall or bottom of the container, triggering the "anti-accumulation warning", and automatically starting the backup circulation pump or increasing the frequency of the current circulation pump for flushing.

[0085] The goal of liquid level control is to maintain a stable total material inventory in the system while ensuring that each level has a certain safe buffer volume, and to precisely control the growth residence time of crystals at each level.

[0086] Hierarchical feedforward-feedback control is adopted: Primary control: Maintaining the overall material balance of the entire system, which is ultimately reflected in the stability of the liquid level L1_true in the first-stage crystallization unit 101. L1 is the inlet for fresh material and also the "reservoir" for the material inventory of the entire system.

[0087] Secondary control: Stabilize the liquid level L3_true in the third-stage crystallization unit 103. L3 directly determines the growth residence time of the final product crystals and is the most critical parameter affecting crystal particle size.

[0088] Control variables: Main controlled variable: Fresh saturated solution feed regulating valve entering the first-stage crystallization unit 101 (controls total material input).

[0089] Secondary manipulated variable: the frequency of the interstage transfer pump connecting the crystallization units at each stage (controlling the material distribution between stages).

[0090] Auxiliary manipulated variable: Crystal slurry discharge regulating valve at the bottom of the third-stage crystallization unit 103 (controls the final product output).

[0091] The deviation between L3_true and the setpoint L3_sp is calculated using a PID controller and then directly and quickly controls the discharge regulating valve at the bottom of the third-stage crystallization unit 103. This is a fast-response loop that ensures a constant residence time. The system has a built-in dynamic material balance model. This model receives in real time the change in the opening of the discharge regulating valve ΔV_out, and the total crystallization growth rate R_crystal estimated based on the supersaturation model.

[0092] To compensate for the current discharge rate ΔV_out and the amount of material "solidified" due to crystal growth R_crystal, and to maintain the stability of L1 and L2, the total amount of fresh material that needs to be replenished in the future is ΔF_in. This predicted value ΔF_in serves as a strong feedforward signal and is directly superimposed on the control output of the first-stage feed regulating valve. This makes the adjustment of the feed rate almost synchronized with the discharge and crystallization consumption, greatly reducing the disturbance to L1.

[0093] The control of the liquid level L2_true in the second-stage crystallization unit 102 adopts a "floating setpoint" strategy. Its setpoint L2_sp is not completely fixed, but is dynamically optimized within a range based on the actual conditions of L1 and L3.

[0094] S5. Based on the crystal growth maturity and system health, determine the timing of material discharge and adaptively optimize key process settings based on the quality data feedback of the discharged crystals.

[0095] Specifically, the discharge decision is not based solely on a single parameter (such as time or liquid level), but rather dynamically determined by a "discharge readiness index" that comprehensively reflects the maturity of crystal growth and the health of system operation. The system continuously calculates the increase in solid content M per unit time (dM / dt) within the third-stage crystallizer. When dM / dt remains consistently below a very low threshold (e.g., <0.1% / min) for a period of time, it indicates that crystal growth has entered a plateau phase, and the main growth phase is essentially complete. The supersaturation σ3 of the third-stage crystallizer 103 is monitored. When σ3 is stably maintained near its set lower limit (e.g., stable around 0.02), it indicates that most of the solute in the solution has precipitated, the growth driving force has weakened, and the crystals are maturing.

[0096] When online indicators suggest that discharge conditions may be met, the system will not immediately discharge the material but will automatically trigger an offline sampling procedure. A sample of the crystal slurry from the bottom of the third-stage crystallizer 103 is obtained via an automatic sampling valve and sent to a connected online particle size analyzer (e.g., based on dynamic image analysis) for rapid analysis. The analyzer provides the crystal grain size distribution (PSD) within 2-3 minutes, specifically the median grain size D50 and span ((D90-D10) / D50). Only when D50 ≥ 0.3 mm and the span is below a set value (e.g., <1.0) is the crystal growth and uniformity physically confirmed.

[0097] While determining crystal maturity, the system checks whether each control loop is in a stable state: Whether the fluctuations of each liquid level (L1, L2, L3) are within the allowable range.

[0098] Whether the critical supersaturation (σ1, σ2) is within its "metastable region" setting range.

[0099] Are all vacuum pumps, circulation pumps, and valves operating smoothly without frequent adjustments?

[0100] Only when the crystallization maturity meets the standard and the system is in good health will the system be determined to be in the "optimal discharge window".

[0101] When the "discharge readiness index" (which integrates growth rate, supersaturation, estimated particle size, and system stability) exceeds the set threshold, the central control system automatically generates a discharge command.

[0102] The instruction does not simply open the discharge valve, but executes a smooth discharge procedure: first, it finely adjusts the discharge regulating valve at the bottom of the third-stage crystallization device 103 to a preset "batch discharge opening", and at the same time, it interlocks to slightly increase the feed rate to compensate for the material output, maintain the overall material balance of the system, and avoid drastic fluctuations in the liquid level.

[0103] The system incorporates a simplified crystallization mechanism model that links key control variables (such as the first-stage supersaturation setpoint σ1_sp, the second-stage growth temperature T2_sp, and the third-stage residence time (determined by L3_sp)) with predicted product quality (particle size D50 and span). After each batch is discharged and offline particle size analysis is completed, the actual D50 and span are entered into the system database and stored in association with the average values ​​of control variables (such as σ1_avg and T2_avg) during the production process of that batch of products.

[0104] Compare the actual quality of the current batch with the target quality (e.g., D50 target = 0.35mm, span target < 0.8). If the actual D50 is smaller, the optimization engine will initiate diagnostics: Check if σ1_avg is too high. If it is too high, it may lead to excessive nucleation, which will crowd out the growth space.

[0105] Check if T2_avg is too low? Growth may be too slow at low temperatures.

[0106] Check if the actual average dwell time calculated based on L3_avg is insufficient. Using accumulated historical data models, a small-scale optimization algorithm (such as response surface methodology or gradient descent) is run to calculate the direction and recommended adjustment amount for each key setting value to reduce quality deviation. For example: "To improve D50, it is recommended to decrease the σ1_sp setting value by 5% in the next batch, while increasing T2_sp by 1°C." These recommendations are presented to operators for confirmation in the form of "process optimization proposals," or, in high-level automation mode, are automatically, slightly, and gradually loaded into the control settings of the next production batch after confirmation by safety logic.

[0107] The system uses the quality data of the final output crystals (mainly particle size distribution and purity) as the highest feedback signal to periodically optimize the control setpoints for steps 1 to 4. For example: Step 1 (Feed): Based on the new σ1_sp and T2_sp, the feeding system may receive new fine-tuning requirements for feed concentration or temperature to ensure that the solution state entering the first-stage crystallization unit 101 matches the new process.

[0108] Step 2 (Vacuum and Temperature): Adjusting T2_sp directly translates to recalculating and setting the pressure setpoint P2_sp of the secondary separator.

[0109] Step 3 (Supersaturation and Solid Content): Direct adjustment of σ1_sp changes the target of the supersaturation control loop. Simultaneously, to align with the new growth target, the solid content setpoint M_sp of the third-stage crystallizer 103 may also be optimized to balance yield and particle size.

[0110] Step 4 (Liquid Level): To achieve the new target residence time, the liquid level setpoint L3_sp of the third-stage crystallization unit 103 may be recalculated and adjusted.

[0111] All optimization adjustments must be made within a predefined safe operating space, and the magnitude of each adjustment is limited (e.g., each adjustment does not exceed 2% of the original value) to ensure production safety and a smooth transition.

[0112] This system modifies the OLSO crystallization unit, resulting in crystals with a particle size of over 0.3mm after passing through the second-stage crystallization unit 102, and with uniform particle size. The continuous feeding and discharging system has lower energy consumption and higher production efficiency than the old system. Precise control of temperature, pressure, and flow rate throughout the entire process ensures higher and more stable product quality. Figure 4 The crystallization image shown appears as white spherical crystals, with a particle size of approximately 250-350 μm, and a maximum size of 450 μm. The potassium oxide content is 46.2%, and the chloride content is 0.05%. This scheme can achieve the following: 1. Continuous crystallization: Feeding, crystallization and discharging are carried out simultaneously, with a large production capacity (over 10 tons per unit).

[0113] 2. Low energy consumption, utilizing vacuum flash cooling, resulting in significant energy savings.

[0114] 3. The crystals are of high quality, capable of producing crystal particles of about 2.5mm with a purity of over 99%.

[0115] The second embodiment of this application is as follows: This invention provides a potassium nitrate production crystallization control system, applied to a potassium nitrate production crystallization control method as provided in the first embodiment, comprising: The first-stage crystallization device 101, the second-stage crystallization device 102, and the third-stage crystallization device are connected in sequence. Vacuum system, used to provide vacuum for crystallization units at all levels, including water ring vacuum pumps, Roots pumps and their piping valves; Material conveying system, used for material conveying, including feed pipeline and regulating valve, interstage transfer pump, circulation pump and discharge pipeline and regulating valve; Sensor arrays for data acquisition include radar level gauges, pressure transmitters, temperature sensors, online density meters, concentration sensors, microwave solid content analyzers, and weighing modules installed on the support legs of the crystallization unit. The server is electrically connected to the sensor array, the vacuum system, and the material conveying system, respectively, and is configured to perform steps S1 to S5.

[0116] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0117] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0118] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and, when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the potassium nitrate production crystallization control method as described above. Figure 5 The diagram shown is a hardware structure diagram of any device with data processing capabilities within a potassium nitrate production crystallization control system provided in an embodiment of the present invention, except... Figure 5 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0119] Accordingly, this application also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the potassium nitrate production crystallization control method described above. The computer-readable storage medium can be an internal storage unit of any data-processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data-processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data-processing device, and can also be used to temporarily store data that has been output or will be output.

[0120] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0121] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for controlling crystallization during potassium nitrate production, characterized in that, Includes the following steps: S1. Based on the concentration and liquid level feedback, the feed rate of the saturated potassium nitrate solution into the multi-stage crystallization system is adaptively controlled to establish and stabilize the initial liquid level of each stage. S2. Based on the system initialization completion signal, the vacuum system is started sequentially, and graded pressure control is implemented to establish a temperature gradient from the first stage to the third stage crystallization device 103; S3. Based on the supersaturation data obtained from real-time calculation and the solid content data measured online, the vacuum degree and the discharge valve are adjusted in a coordinated manner through a combination of model prediction feedforward and feedback control to stabilize the supersaturation and solid content within the target range. S4. Based on the real liquid level data obtained by multi-source fusion, dynamic material balance feedforward and liquid level feedback control are used to coordinate the adjustment of the feed valve, interstage transfer pump and discharge valve to maintain the hydraulic balance of the system and the stability of crystal growth residence time. S5. Based on the crystal growth maturity and system health, determine the timing of material discharge and adaptively optimize key process settings based on the quality data feedback of the discharged crystals.

2. The method for controlling crystallization in potassium nitrate production as described in claim 1, characterized in that, Based on concentration and liquid level feedback, the feed rate of saturated potassium nitrate solution into the multi-stage crystallization system is adaptively controlled, including: The concentration of the feed liquid is monitored by an online concentration sensor. If the concentration is lower than the safety threshold, the feed rate is increased; if the concentration is higher than the safety threshold, the feed rate is decreased. Once the liquid level in each crystallization unit reaches the set value and stabilizes, the feed will be switched to normal saturated solution.

3. The method for controlling crystallization in potassium nitrate production as described in claim 1, characterized in that, Based on the system initialization completion signal, the vacuum system is sequentially started, and graded pressure control is implemented to establish a temperature gradient from the first to the third stage of the crystallization device 103, including: Once the liquid level in the third-stage crystallizer 103 stabilizes at the set high liquid level, the water ring vacuum pumps connected to each stage of the crystallizer are started sequentially. The pressure of the first-stage crystallization unit 101 is monitored in real time. Once the pressure drops to the starting pressure threshold of the Roots pump, the Roots pumps are started sequentially. By controlling the Roots pump frequency with the pressure of the three-stage separator as the main loop and adjusting the vacuum fine-tuning valves with the pressures of the first and second-stage separators as the secondary loops, the pressures of the first, second, and third-stage separators are controlled at the first, second, and third set values, respectively, thereby enabling the material outlet temperature to reach the corresponding gradient temperature.

4. The method for controlling crystallization in potassium nitrate production as described in claim 3, characterized in that, The method further includes: The actual temperature at each material outlet is compared with the target temperature, and the resulting temperature difference signal is used as a fine-tuning amount to dynamically correct the pressure setpoint of the corresponding separator.

5. The method for controlling crystallization in potassium nitrate production as described in claim 1, characterized in that, The supersaturation obtained through real-time calculation includes: The temperature and online density of the circulating mother liquor in the first to third stage crystallization unit 103 are collected. The actual concentration is calculated by using a preset solution property model, and the supersaturation is calculated by combining the equilibrium solubility at the current temperature. The solid content measured online was obtained using a microwave solid content analyzer installed on the discharge pipeline of the third-stage crystallization unit 103.

6. The method for controlling crystallization in potassium nitrate production as described in claim 5, characterized in that, By combining model prediction feedforward and feedback control, the vacuum level and discharge valve are adjusted in a coordinated manner, including: The supersaturation of the first and second stage crystallization devices 102 is used as the controlled variable, and the vacuum setting value of the corresponding separator is controlled by PID; the solid content of the third stage crystallization device 103 is used as the controlled variable, and the opening of its discharge valve is controlled by PI. Meanwhile, the change in the opening of the discharge valve is used as a feedforward signal to predict the impact on supersaturation and is compensated in advance to the vacuum control loop. The change in the vacuum setpoint is used as a feedforward signal to predict the impact on solid content and is compensated in advance to the discharge valve control loop.

7. The method for controlling crystallization in potassium nitrate production as described in claim 1, characterized in that, Multi-source fusion obtains real liquid level data, including: The radar level gauge readings, differential pressure transmitter readings, and weighing module readings for each crystallization stage are obtained. An adaptive weighted fusion algorithm based on confidence assessment is then used to calculate the actual level signal for control.

8. The method for controlling crystallization in potassium nitrate production as described in claim 7, characterized in that, Through dynamic material balance feedforward and liquid level feedback control, the feed valve, interstage transfer pump, and discharge valve are coordinated and adjusted, including: The actual liquid level of the third-stage crystallization unit 103 is used as the direct control object. The opening of its discharge valve is adjusted by a PID controller to stabilize the crystal growth residence time. A dynamic material balance model is constructed to predict the required amount of fresh material replenishment in real time based on the changes in the opening degree of the discharge valve and the total crystallization rate. The predicted amount of fresh material replenishment is then used as a feedforward signal to directly control the first-stage feed valve. The second-stage crystallization unit 102 is used as a buffer coordination stage, and its liquid level setting value is dynamically optimized. The frequency of the inter-stage transfer pump is adjusted to make it smoothly buffer between the first and third-stage crystallization units 103.

9. The method for controlling crystallization in potassium nitrate production as described in claim 1, characterized in that, The timing of material discharge is determined based on a comprehensive assessment of crystal growth maturity and system health. Furthermore, key process settings are adaptively optimized based on feedback from the quality data of the discharged crystals, including: When the crystal growth rate in the third-stage crystallization unit 103 is continuously lower than the threshold and the supersaturation is stable at a low level, offline sampling analysis is triggered; if the offline particle size analysis confirms that the crystal particle size and distribution meet the requirements and the system's control loops are operating stably, it is determined that the optimal discharge window period has been entered. Establish and update a digital twin model of the crystallization process, linking the average values ​​of key control parameters from historical batches with the final crystal quality data; when the actual product quality deviates from the target, use the digital twin model of the crystallization process to calculate the optimization adjustment amount of the key process settings and apply it to subsequent production batches.

10. A potassium nitrate production crystallization control system, applied to the potassium nitrate production crystallization control method as described in claim 1, characterized in that, include: The first-stage crystallization apparatus, the second-stage crystallization apparatus, and the third-stage crystallization apparatus are connected in sequence. Vacuum system, used to provide vacuum for crystallization units at all levels, including water ring vacuum pumps, Roots pumps and their piping valves; Material conveying system, used for material conveying, including feed pipeline and regulating valve, interstage transfer pump, circulating pump and discharge pipeline and regulating valve; The sensor array, used for data acquisition, includes radar level gauges, pressure transmitters, temperature sensors, online density meters, concentration sensors, microwave solid content analyzers, and weighing modules installed on the support legs of the crystallization device, all installed on various levels of crystallization devices or pipelines. The server is electrically connected to the sensor array, the vacuum system, and the material conveying system, respectively, and is configured to perform steps S1 to S5.

Images