A method and system for real-time control of sucrose crystallization based on microscopic image analysis

Abstract

The present application relates to the field of sucrose crystallization, and provides a real-time control method and system for sucrose crystallization based on microscopic image analysis, comprising the following steps: S1, obtaining a microscopic image of a crystal slurry in real time through an online image acquisition device arranged on a crystallization tank; S2, performing digital image processing on the microscopic image of the crystal slurry, extracting the number of identifiable crystals in a single frame image, the number of crystal grains per unit volume after conversion, and granularity distribution characteristic parameters, wherein the granularity distribution characteristic parameters at least include D10, D50 and D90; S3, comparing the number of crystals with a preset first threshold value, and when the number of identifiable crystals in continuous multiple frames of images reaches a seed addition window threshold value, generating a seed addition control signal to add seed crystals into the crystallization tank. The present application converts experience judgment into quantitative indicators, reduces the influence of human factors; controls fine crystals through D10 / D50 / D90, significantly improves sugar product uniformity and color value; realizes closed-loop automatic control of the crystallization process, reduces labor intensity, and stabilizes product quality.

Description

Technical Field

[0001] This invention relates to the field of sucrose crystallization technology, specifically to a real-time control method and system for sucrose crystallization based on microscopic image analysis. Background Technology

[0002] In the sugar boiling and crystallization process, the quality of the crystals directly determines the grade and selling price of the final sugar product. Traditional processes rely heavily on the experience of the operators, who determine whether to add seed crystals or adjust the operation by observing the "crystallization" and "crystallization" states of the crystal slurry in a sight glass. This method has the following problems:

[0003] Highly subjective: Inconsistent judgment standards among different workers lead to large quality fluctuations between batches;

[0004] Fine crystals are difficult to control: In the later stage of crystallization, a large number of fine crystals are generated. If they are not eliminated in time, it will lead to an increase in the sugar color value of the finished product and uneven crystal grains.

[0005] Lack of quantitative methods: Existing instruments mostly measure temperature and concentration, which cannot directly reflect the number of crystals and particle size distribution. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention aims to provide a real-time control method and system for sucrose crystallization based on microscopic image analysis. To solve these problems, this invention employs the following technical solution:

[0007] A real-time control method for sucrose crystallization based on microscopic image analysis includes the following steps:

[0008] S1, real-time acquisition of microscopic images of crystal slurry using an online image acquisition device installed on the crystallizer;

[0009] S2, perform digital image processing on the crystal slurry microscopic image, extract the number of identifiable crystals, the converted number of crystals per unit volume, and the particle size distribution characteristic parameters in a single frame image. The particle size distribution characteristic parameters include at least D10, D50, and D90.

[0010] S3, compare the number of crystals with a preset first threshold. When the number of identifiable crystals in multiple consecutive frames of images reaches the seed addition window threshold, generate a seed addition control signal to add seed crystals into the crystallization tank.

[0011] S4, determine the fine grain distribution based on D10, D50 and D90;

[0012] When D10 is less than the preset fine crystal judgment threshold or the difference between D90 and D10 is less than the preset particle size span threshold, the proportion of fine crystals is judged to exceed the second threshold. At this time, the proportion of large crystals in the particle size distribution is insufficient and the proportion of small crystals is too high. Before the crystallization discharge filtration, a process water addition control signal is generated to add process water into the crystallization tank to preferentially dissolve the excess small crystals until D10 reaches the target range and the difference between D90 and D10 reaches the set span range, then the process water addition is stopped.

[0013] S5: After the fine crystals are eliminated in S4, the system automatically closes the process water valve and increases the vacuum of the crystallizer until the amount of water evaporated is equal to the amount of process water added in the early stage, and then filters and dries.

[0014] Optionally, the first threshold is the seed addition window threshold; wherein, the seed addition window threshold is triggered after the number of identifiable crystals in 5 consecutive single-frame images reaches more than 10, and the target number of crystals is preset according to the product specifications.

[0015] Optionally, the fine grain determination threshold is 50 μm, and the preset particle size span threshold is 300 μm.

[0016] Optionally, in S4, the amount of process water added is dynamically adjusted, and after addition, D10 is increased to 60-150 μm.

[0017] Optionally, the online image acquisition device includes an industrial camera or a microscopic imaging system, and the digital image processing employs edge detection and particle counting algorithms.

[0018] A real-time control system for sucrose crystallization based on microscopic image analysis, used to execute the aforementioned real-time control method for sucrose crystallization based on microscopic image analysis, includes: an image acquisition module, a data processing module, a decision control module, and an execution module.

[0019] The present invention has the following beneficial effects:

[0020] Transforming empirical judgments into quantitative indicators reduces the impact of human factors; controlling fine crystals through D10 / D50 / D90 significantly improves sugar uniformity and color value; achieving closed-loop automatic control of the crystallization process reduces labor intensity and stabilizes product quality. Attached Figure Description

[0021] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the structure of a real-time control method and system for sucrose crystallization based on microscopic image analysis according to the present invention.

[0023] Figure 2 This is a control effect diagram of one embodiment of the present invention;

[0024] Figure 3 This is a control effect diagram of one embodiment of the present invention;

[0025] Figure 4 This is a control effect diagram of one embodiment of the present invention. Detailed Implementation

[0026] 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.

[0027] In the description of this invention, it should be noted that the terms "vertical," "upper," "lower," and "horizontal," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0028] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or a connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0029] During the sucrose boiling and crystallization process, the timing of seed crystal addition directly affects the subsequent uniform growth of crystals. Too many fine crystals or an insufficient proportion of large crystals result in small, unevenly sized, and unevenly colored finished sugar, making separation difficult. Existing technologies mainly suffer from the following shortcomings:

[0030] Existing sucrose refining particle size control methods mainly focus on recrystallization, sieving, and particle size control of pharmaceutical-grade sucrose excipients. For example, CN110656206A, "A Novel Method for Refining Sucrose with Controllable Particle Size," discloses steps such as dissolution and decolorization, concentration and crystallization, separation and purification, drying and sieving. It also describes adding seed crystals at a Brix value of 87–92°Bx and using ultrasonic stimulation to induce crystallization, resulting in pharmaceutical-grade sucrose crystals of 20–80 mesh. This method demonstrates that factors such as seed crystal addition time, amount, and ultrasonic frequency affect sucrose crystal particle size. However, its control targets are offline or batch pharmaceutical sucrose refining processes. It does not disclose the continuous identification of crystal quantity using online microscopic images within the sugar boiling crystallization tank, nor does it disclose using D10, D50, and D90 as closed-loop control criteria for fine crystal elimination before discharge.

[0031] Existing seed preparation methods can improve the quality of the seeds themselves, but the focus remains on seed preparation rather than the control of the sugar boiling process. For example, CN111808999B, "A Sucrose Seed and Its Preparation Method," discloses a method of concentrating a sucrose solution at -10℃ to 4℃, adding ethanol, and treating it under ultrasonic conditions to obtain sucrose seeds with uniform particle size and smooth surface, thereby improving the problems of large particle size range, numerous microcrystals, and severe secondary nucleation associated with traditional grinding seeds. This method is related to the purpose of seed addition in this case, but it does not address when to add seeds based on the number of crystals in the online image during the sugar boiling process, nor does it link the seed addition window with subsequent monitoring of D10, D50, and D90 particle size distribution and control of fine crystal removal by process water.

[0032] Existing sucrose refining production systems can equip pharmaceutical-grade sucrose production, but they do not involve online detection and control within the boiling tank. For example, CN211771375U, "A Novel Production System for Controllable Particle Size Sucrose Refining," discloses equipment such as a dissolving tank, filter, concentration tank, centrifuge, and dryer. It also includes a seed crystal dispenser and an ultrasonic generator in the concentration tank, controlling the particle size of pharmaceutical-grade sucrose crystals through a combination of seed crystal initiation and ultrasonic stimulation. While this system demonstrates the role of "seed crystal + ultrasound" in particle size control, it does not disclose online image acquisition within the crystallization tank, single-frame crystal quantity statistics, or the determination of the D90-D10 particle size span. Furthermore, it does not disclose the automatic addition of process water before filtration to preferentially dissolve fine crystals based on particle size distribution.

[0033] Existing integrated monitoring platforms for sugar boiling can collect various process parameters and study soft measurement and control strategies for supersaturation, but their focus is primarily on the experimental platform and multi-parameter monitoring. For example, CN102505048A, "Open Intelligent Monitoring Integrated Experimental Platform for Sugarcane Boiling Process," discloses a boiling tank module, an auxiliary working module, a sensor data acquisition module, and a valve control module. It uses data such as temperature, density, conductivity, vacuum degree, liquid level, and valve opening degree to perform soft measurement of supersaturation and dynamic parameter optimization control. This scheme can serve as a background for monitoring and optimizing the sugar boiling process, but it does not disclose the use of the number of identifiable crystals obtained from online microscopic images as a seed crystal in the window threshold, nor does it disclose the use of D10, D50, D90, and D90-D10 thresholds to directly determine the proportion of fine crystals and large crystals.

[0034] Existing automated control schemes for sugar boiling in sugar factories can achieve modular automated operation of multiple processes, but they still mainly rely on process parameters such as lauryl permeability, viscosity, temperature, vacuum, and liquid level. For example, CN111235325A, "Automatic Control System and Method for Sugar Boiling in Sugar Factories," discloses automated control of processes such as vacuuming, bottom material feeding, bottom material concentration, powder or seed extraction, crystal formation, crystal solidification, crystal growth, sugar paste finishing, sugar paste concentration, sugar dispensing, and tank washing. The control is achieved through the use of level gauges, vacuum pressure gauges, material thermometers, steam drum thermometers, steam drum pressure gauges, sugar paste resistance viscometers, and optical lauryl permeability meters. This scheme can reduce reliance on manual operation, but it does not disclose the use of microscopic images to obtain real-time crystal quantity and particle size distribution parameters, nor does it disclose the use of "determining the seed crystal addition window based on crystal quantity" and "determining the fine crystals by D10 / D50 / D90" to form the same closed-loop control process.

[0035] Therefore, there is an urgent need for a sucrose crystallization control method and system that can quantify the number of crystal grains in real time to guide the addition of seed crystals and actively eliminate fine crystals before discharge based on online particle size distribution parameters.

[0036] The present invention aims to solve the following problems: (1) the timing and amount of seed crystal addition during the crystallization process depend on human experience and cannot be quantified; (2) when the proportion of fine crystals is too high or the proportion of large crystals is insufficient, there is a lack of active elimination methods based on real-time particle size distribution data, which leads to fluctuations in the quality of finished sugar.

[0037] The purpose of this invention is to solve the following problems:

[0038] How to scientifically determine the seeding window based on the number of grains;

[0039] How to automatically eliminate fine crystals using particle size distribution data before filtration to improve the quality of finished sugar?

[0040] The specific technical solution is as follows:

[0041] like Figure 1As shown, a real-time control method for sucrose crystallization based on microscopic image analysis includes the following steps:

[0042] S1, real-time acquisition of microscopic images of crystal slurry using an online image acquisition device installed on the crystallizer;

[0043] S2, perform digital image processing on the crystal slurry microscopic image, extract the number of identifiable crystals, the converted number of crystals per unit volume, and the particle size distribution characteristic parameters in a single frame image. The particle size distribution characteristic parameters include at least D10, D50, and D90.

[0044] S3, compare the number of crystals with a preset first threshold. When the number of identifiable crystals in multiple consecutive frames of images reaches the seed addition window threshold, generate a seed addition control signal to add seed crystals into the crystallization tank.

[0045] S4, determine the fine grain distribution based on D10, D50 and D90;

[0046] When D10 is less than the preset fine crystal judgment threshold or the difference between D90 and D10 is less than the preset particle size span threshold, the proportion of fine crystals is judged to exceed the second threshold. At this time, the proportion of large crystals in the particle size distribution is insufficient and the proportion of small crystals is too high. Before the crystallization discharge filtration, a process water addition control signal is generated to add process water into the crystallization tank to preferentially dissolve the excess small crystals until D10 reaches the target range and the difference between D90 and D10 reaches the set span range, then the process water addition is stopped.

[0047] S5: After the fine crystals are eliminated in S4, the system automatically closes the process water valve and increases the vacuum of the crystallizer until the amount of water evaporated is equal to the amount of process water added in the early stage, and then filters and dries.

[0048] Preferably, the first threshold is the seed addition window threshold; wherein, the seed addition window threshold is triggered when the number of identifiable crystals in 5 consecutive single-frame images reaches more than 10, and the target number of crystals is preset according to the product specifications.

[0049] Preferably, the fine grain determination threshold is 50 μm, and the preset particle size span threshold is 300 μm.

[0050] Preferably, in S4, the amount of process water added is dynamically adjusted, and after addition, D10 increases to 60-150 μm.

[0051] Preferably, the online image acquisition device includes an industrial camera or a microscopic imaging system, and the digital image processing employs edge detection and particle counting algorithms.

[0052] A real-time control system for sucrose crystallization based on microscopic image analysis, used to execute the aforementioned real-time control method for sucrose crystallization based on microscopic image analysis, comprising:

[0053] The image acquisition module is used to acquire images of the crystal slurry inside the crystallizer in real time.

[0054] The data processing module is used to process the crystal slurry image to obtain the number of crystals and the grain size distribution parameters D10, D50, and D90.

[0055] The decision control module is used to execute the real-time control method and output the seed crystal addition signal and the process water addition signal;

[0056] The execution module is used to adjust the opening degree of the seed crystal dosing valve and the process water dosing valve according to the control signal.

[0057] In this invention:

[0058] Image-based seed control strategy: Using online microscopy or industrial cameras, images of the crystal slurry inside the crystallization tank are captured in real time, and the number of crystals per unit volume is identified and calculated through image processing algorithms.

[0059] Once the required number of crystal grains is detected, the system automatically triggers the seed crystal addition program to ensure a sufficient number of crystal nuclei.

[0060] Fine grain elimination strategy based on D10 / D50 / D90:

[0061] Real-time acquisition of granularity distribution feature values ​​through online image granularity analysis:

[0062] D10: Cumulative 10% of particle size;

[0063] D50: median diameter;

[0064] D90: Cumulative particle size distribution;

[0065] When D10 is too small, or the difference between D90 and D10 is too large, indicating that the proportion of fine crystals is too high, the system automatically adds a certain amount of process water to the crystallizer before discharge to dissolve some of the fine crystals, thereby optimizing the particle size distribution.

[0066] Beneficial effects:

[0067] Transform experience-based judgments into quantitative indicators to reduce the influence of human factors;

[0068] By controlling the fine crystals using D10 / D50 / D90, the uniformity and color value of sugar products are significantly improved.

[0069] Achieving closed-loop automatic control of the crystallization process reduces labor intensity and stabilizes product quality.

[0070] Example:

[0071] A sugar boiling and crystallization system is controlled using the method of this invention:

[0072] Step 1: Image Acquisition and Processing

[0073] An online industrial camera is installed on the side wall of the crystallization tank, capturing an image of the crystal slurry every second. The image processing system automatically identifies the crystal outlines, counts the number of identifiable crystals in a single frame image, and calculates the number of crystals per milliliter of crystal slurry, as well as D10, D50, and D90, based on image calibration coefficients. The image calibration coefficients include at least the conversion ratio between pixel length and actual length, and the conversion relationship between field of view area.

[0074] Step 2: Control of Seed Crystal Addition

[0075] The seed addition window threshold is set at a minimum of 10 identifiable crystals in each of five consecutive single-frame images. When five consecutive single-frame images reach the seed addition window threshold, and the online particle size distribution shows that the crystals have entered a stage where uniform growth can be induced, the system determines that the current situation is not a shortage of crystal nuclei, but rather that it has entered a suitable window period for adding seed crystals to promote uniform growth. At this time, the system automatically opens the seed valve and adds seed crystals at a dosage of 0.5 kg / t, allowing the crystals to continue to grow uniformly based on the existing crystal nuclei and obtain uniform large particles.

[0076] Step 3: Fine grain elimination control

[0077] The allowable range for the process is set as follows:

[0078] D10≥60μm;

[0079] D90−D10≤300μm;

[0080] When any indicator exceeds the standard, the system automatically adds 1–3% (w / w) of process water and stirs continuously for 5 minutes to dissolve the fine crystals until the particle size distribution returns to the set range.

[0081] Step 4: Evaporation to increase yield

[0082] After the fine crystals are eliminated, start the heating system of the crystallizer or the external circulation evaporator to concentrate and evaporate the crystal slurry. Evaporate the corresponding proportion of water according to the feed amount to raise the concentration of the mother liquor back to the target supersaturation range, thereby improving the sucrose yield. Then discharge and filter.

[0083] refer to Figure 2-4According to the data, during the crystallization stage of 0–0.5 h, the number of crystals rapidly increased from 0 to over 200, and then fluctuated within the range of approximately 300–360. In the length direction, D10, D50, and D90 were approximately 60–70 μm, 190–200 μm, and 350–370 μm respectively after about 3 h; in the width direction, D10, D50, and D90 were approximately 45–55 μm, 130–140 μm, and 245–255 μm respectively after about 3 h. Therefore, when D10 is below 50 μm, it indicates a high proportion of small crystals; when D90-D10 is less than 300 μm, it indicates insufficient particle size distribution, a low proportion of large crystals, and that small crystals still dominate overall. Therefore, D10 below 50μm or D90-D10 less than 300μm can be used as the criteria for determining whether there is too much fine crystal or too little large crystal; the preferred target after eliminating fine crystal is D10 reaching 60-150μm and D90-D10 reaching or exceeding 300μm within a set range.

[0084] Compared with traditional manual operation, the following benefits are achieved by using this method:

[0085] The average particle size deviation of the finished sugar was reduced by approximately 25%.

[0086] Worker interventions decreased by approximately 70%;

[0087] The caramel color value of the finished product shows a downward trend.

[0088] The components, modules, mechanisms, and devices in this invention that are not described in detail are all general standard parts or components known to those skilled in the art. Their structures and principles can be learned by those skilled in the art through technical manuals or conventional experimental methods.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A real-time control method for sucrose crystallization based on microscopic image analysis, characterized in that, Includes the following steps: S1, real-time acquisition of microscopic images of crystal slurry using an online image acquisition device installed on the crystallizer; S2, perform digital image processing on the crystal slurry microscopic image, extract the number of identifiable crystals, the converted number of crystals per unit volume, and the particle size distribution characteristic parameters in a single frame image. The particle size distribution characteristic parameters include at least D10, D50, and D90. S3, compare the number of crystals with a preset first threshold. When the number of identifiable crystals in multiple consecutive frames of images reaches the seed addition window threshold, generate a seed addition control signal to add seed crystals into the crystallization tank. S4, determine the fine grain distribution based on D10, D50 and D90; When D10 is less than the preset fine crystal judgment threshold or the difference between D90 and D10 is less than the preset particle size span threshold, the proportion of fine crystals is judged to exceed the second threshold. At this time, the proportion of large crystals in the particle size distribution is insufficient and the proportion of small crystals is too high. Before the crystallization discharge filtration, a process water addition control signal is generated to add process water into the crystallization tank to preferentially dissolve the excess small crystals until D10 reaches the target range and the difference between D90 and D10 reaches the set span range, then the process water addition is stopped. S5: After the fine crystals are eliminated in S4, the system automatically closes the process water valve and increases the vacuum of the crystallizer until the amount of water evaporated is equal to the amount of process water added in the early stage, and then filters and dries.

2. The real-time control method for sucrose crystallization based on microscopic image analysis according to claim 1, characterized in that, The first threshold is the seed addition window threshold; the seed addition window threshold is triggered when the number of identifiable crystals in 5 consecutive single-frame images reaches more than 10, and the target number of crystals is preset according to the product specifications.

3. The real-time control method for sucrose crystallization based on microscopic image analysis according to claim 1, characterized in that, The fine grain determination threshold is 50μm, and the preset particle size span threshold is 300μm.

4. The real-time control method for sucrose crystallization based on microscopic image analysis according to claim 1, characterized in that, In S4, the amount of process water added is dynamically adjusted, and after addition, D10 increases to 60-150 μm.

5. The real-time control method for sucrose crystallization based on microscopic image analysis according to claim 1, characterized in that, Online image acquisition devices include industrial cameras or microscopic imaging systems, and digital image processing employs edge detection and particle counting algorithms.

6. A real-time control system for sucrose crystallization based on microscopic image analysis, characterized in that, A method for real-time control of sucrose crystallization based on microscopic image analysis, as described in any one of claims 1-5, comprises: an image acquisition module, a data processing module, a decision control module, and an execution module.

Images