Powder granulation control method and system based on fuzzy control

Abstract

The invention relates to the technical field of granulation control, and discloses a powder granulation control method and system based on fuzzy control, the system comprises a data acquisition unit, a data processing unit, a data analysis unit and a data adjustment unit; by collecting key parameters such as powder humidity in real time, instant and accurate data are provided for control; through fine processing such as filtering, normalization and feature extraction, data noise is removed, the data scale is unified, and parameter dynamic change features are deeply mined; by means of a preset fuzzy rule, the parameter features are converted into fuzzy linguistic variables for reasoning, and the equipment adjustment amount can be scientifically and reasonably determined; in the execution link, a final adjustment value is determined through weighted average, and precise regulation and control of the granulation equipment are achieved; the rich and practical fuzzy rule base covers various parameter deviation combinations, so that the system can adapt to different granulation working conditions, and the granulation efficiency and the product quality are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of granulation control, and in particular to a powder granulation control method and system based on fuzzy control. Background Art

[0002] Traditional powder granulation control methods often rely on operator experience or employ relatively simple control strategies. However, the powder granulation process is highly complex, characterized by strong nonlinearity, time-varying behavior, and multivariable coupling. These methods struggle to achieve precise control of the granulation process, resulting in inconsistent product quality, low production efficiency, and high production costs. Simple control strategies are unable to adapt to rapid parameter changes and effectively handle the complexities of the granulation process, severely impacting granulation stability and product quality.

[0003] With the rapid development of industrial automation, fuzzy control technology has emerged. Fuzzy control does not rely on the precise mathematical model of the controlled object, can effectively deal with uncertainty and nonlinear problems in complex systems, and has been widely used in the field of industrial control. At present, in some industrial control scenarios, fuzzy control has demonstrated advantages that are difficult to match with traditional control methods, significantly improving the control performance and adaptability of the system. However, there are relatively few studies and practices on the systematic application of fuzzy control technology to powder granulation process control. Existing methods have obvious deficiencies in granulation process data processing, parameter adjustment strategies, etc., and cannot fully utilize the advantages of fuzzy control technology. It is difficult to meet the requirements of the powder granulation process for high-precision and high-stability control.

[0004] In this context, developing a fuzzy-based powder granulation control method and system is of great practical significance. By collecting and accurately processing key parameters in the granulation process in real time and intelligently adjusting the granulation equipment based on fuzzy rules, it is expected to achieve precise control of the powder granulation process, effectively improve granulation quality and production efficiency, reduce production costs, and promote the innovative development of powder granulation technology. Summary of the Invention

[0005] The purpose of the present invention is to provide a powder granulation control method and system based on fuzzy control, which solves the technical problems raised in the background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A powder granulation control method and system based on fuzzy control, comprising:

[0008] Data collection: real-time collection of key parameters during granulation;

[0009] Data processing: Preprocess the collected key parameters in the following ways:

[0010] Filtering: Use median processing or mean processing to obtain the filtered key parameter values;

[0011] Normalization processing: normalize the key parameters after filtering;

[0012] Normalization is done by the formula: To achieve;

[0013] Where G1 represents the normalized powder moisture, granulator speed, feed rate and particle size; G min and G max They are the normal operating lower limit and normal operating upper limit of the key parameters corresponding to powder moisture, granulator speed, feed rate and particle size under normal operation;

[0014] Feature extraction processing: extract features from the normalized key parameters, and then obtain the change rate characteristics and fluctuation amplitude characteristics of the key parameters;

[0015] Data analysis: Based on the pre-set fuzzy rule base, the feed speed adjustment amount and the pelletizer speed adjustment amount of the pelletizing equipment are determined;

[0016] Data adjustment: Based on the feed speed adjustment and pelletizer speed adjustment obtained from data analysis, the adjustment calculation and execution of pelletizing equipment are carried out.

[0017] As a further solution of the present invention: the key parameters include: powder moisture measured in real time by a moisture sensor, granulator speed measured by a speed sensor, feed rate measured by a weight sensor, and particle size measured by an optical sensor.

[0018] As a further solution of the present invention: median filtering:

[0019] Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size;

[0020] Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where g represents the acquisition timestamp corresponding to the selected key parameter, and k is an odd number;

[0021] Then, the key parameter sequence is sorted from small to large, and the middle value is taken as the key parameter value after filtering.

[0022] As a further solution of the present invention: mean filtering:

[0023] Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size;

[0024] Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where G g represents the selected key parameter, k is an odd number;

[0025] in, The preceding time of the key parameter The key parameters of the acquisition timestamp are: The subsequent time of this key parameter Key parameters of the acquisition timestamp;

[0026] Then the key parameter sequence is averaged and the average value is taken as the filtered key parameter value;

[0027] The formula for calculating the average value is as follows:

[0028] ;

[0029] Where, GL g is the key parameter value after filtering corresponding to the key parameter.

[0030] As a further solution of the present invention: the feature extraction method is as follows:

[0031] Rate of change characteristics:

[0032] In a specified period, obtain the key parameters of each adjacent acquisition timestamp, then subtract the key parameters of the next acquisition timestamp from the key parameters of the previous acquisition timestamp, and then divide the difference by the time interval between the two adjacent acquisition timestamps to obtain the change parameters of the key parameters of each adjacent acquisition timestamp;

[0033] Then, the average value of the change parameters corresponding to the key parameters at all adjacent acquisition timestamps is calculated, and the change rate characteristics are obtained;

[0034] The formula is: ;

[0035] Where GB is the change rate characteristic of the key parameters, G1 tis the normalized key parameter of each acquisition timestamp in the specified period, t=1, 2, ..., v, where v represents the number of acquisition timestamps in the specified period, and t0 is the time interval between adjacent acquisition timestamps;

[0036] Among them, GB∈{HB, SB, FB, DB}, HB is the change rate characteristic of powder moisture, SB is the change rate characteristic of granulator speed, FB is the change rate characteristic of feed speed, and DB is the change rate characteristic of particle size;

[0037] Fluctuation amplitude characteristics:

[0038] Within a specified period, the key parameters with the largest and smallest values are extracted from the normalized key parameters at all acquisition timestamps. The key parameter with the largest value is then subtracted from the key parameter with the smallest value to obtain the fluctuation amplitude feature.

[0039] The formula is: ;

[0040] Where GF is the fluctuation amplitude characteristic of the key parameters, G1 max and G1 min are the key parameters with the maximum and minimum values after normalization;

[0041] Among them, GF∈{HF, SF, FF, DF}, HF is the fluctuation amplitude characteristic of powder humidity, SF is the fluctuation amplitude characteristic of granulator speed, FF is the fluctuation amplitude characteristic of feed rate, and DF is the fluctuation amplitude characteristic of particle size.

[0042] As a further solution of the present invention: the specific method of data analysis is as follows:

[0043] Step X1: Feature transformation:

[0044] The change rate features and fluctuation amplitude features obtained after feature extraction of key parameters are converted into fuzzy linguistic variables:

[0045] Among them, the fuzzy linguistic variables are: negative large NB, negative small NS, zero ZO, positive small PS, positive large PB;

[0046] The fuzzy linguistic variable conversion method is as follows:

[0047] For the powder moisture change rate characteristic HB;

[0048] If HB≤-HB a , then the membership degree of the fuzzy linguistic variable is NB, that is, the humidity drops significantly;

[0049] If-HB a <HB≤HB b, then the membership degree of the fuzzy linguistic variable is NS;

[0050] If −HB b <HB≤HB b , then the membership degree of the fuzzy linguistic variable is ZO;

[0051] If HB b <HB≤HB a , then the membership degree of the fuzzy linguistic variable is PS;

[0052] If HB>HBa, the membership degree of the fuzzy linguistic variable is PB;

[0053] Among them, HB a and HB b is the membership judgment threshold preset according to the powder moisture change rate characteristics, and HB a >HB b ;

[0054] The conversion method for the rate of change characteristic SB of the granulator rotation speed, the rate of change characteristic FB of the feed speed, and the rate of change characteristic DB of the particle size is similar to that for the rate of change characteristic HB of the powder moisture content;

[0055] For the fluctuation amplitude characteristic HF of powder humidity;

[0056] If HF≤HF a , then the membership degree of the fuzzy linguistic variable is NB;

[0057] If HF a <HF≤HF b , then the membership degree of the fuzzy linguistic variable is NS;

[0058] If HF b <HF≤HF c , then the membership degree of the fuzzy linguistic variable is ZO;

[0059] If HF c <HF≤HF d , then the membership degree of the fuzzy linguistic variable is PS;

[0060] If HF d >3%, then the membership degree of the fuzzy linguistic variable is PB;

[0061] Among them, HF a , HF b , HF c , HF d is the membership judgment threshold preset according to the powder humidity fluctuation amplitude characteristics, and HF a <HF b <HF c <HFd ;

[0062] The conversion method of the fluctuation amplitude characteristic SF of the granulator speed, the fluctuation amplitude characteristic FF of the feed rate and the fluctuation amplitude characteristic DF of the particle size is similar to the conversion method of the fluctuation amplitude characteristic HF of the powder moisture;

[0063] Step X2, fuzzy reasoning:

[0064] Fuzzy rules are extracted from a pre-set fuzzy rule base, and then the key parameters are converted into fuzzy linguistic variables, which are compared with the fuzzy rules in the fuzzy rule base to determine the feed speed adjustment amount and the pelletizer speed adjustment amount.

[0065] As a further solution of the present invention, the calculation and execution methods are adjusted as follows:

[0066] According to the fuzzy rules obtained by fuzzy reasoning, the feed speed adjustment amount and the pelletizer speed adjustment amount are respectively corresponding to each other; the final value of the feed speed adjustment and the final value of the pelletizer speed adjustment are determined;

[0067] The final value of the feed rate adjustment and the final value of the pelletizer speed adjustment are added to the current feed rate and pelletizer speed of the pelletizer, respectively, to obtain the adjusted feed rate and pelletizer speed of the pelletizer.

[0068] As a further solution of the present invention, the final value of the feed rate adjustment and the final value of the granulator speed adjustment are determined as follows:

[0069] The feed speed adjustment values obtained by different fuzzy rules are marked as JT j , and obtain the preset weight coefficients corresponding to different fuzzy rules and mark them as w j The granulator speed adjustment values obtained by different fuzzy rules are marked as ZT j ;

[0070] Then, the weighted average method is used to calculate the final adjustment values of the feed speed and the pelletizer speed respectively;

[0071] The formula is as follows:

[0072] ;

[0073] Where j = 1, 2, ... m, m represents the number of feed rate adjustment and pelletizer speed adjustment, JT0 and ZT0 are the final values of feed rate adjustment and pelletizer speed adjustment, respectively.

[0074] As a further solution of the present invention: the fuzzy rules in the fuzzy rule base are as follows:

[0075] Rules based on humidity deviation and speed deviation

[0076] Rule 1: If the humidity deviation HF is NB and the speed deviation SF is PB, that is, the actual humidity is much lower than the target humidity and the actual speed is much higher than the target speed, then adjust the feed speed F to PB and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0077] Rule 2: If the humidity deviation HF is PB and the speed deviation SF is NB, then adjust the feed speed F to NB and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0078] Rule 3: If the humidity deviation HF is NB and the particle size deviation DF is PB, then adjust the pelletizer speed S to PB, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0079] Rule 4: If the humidity deviation HF is PB and the particle size deviation DF is NB, then adjust the pelletizer speed S to NB and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0080] Rule 5: If the speed deviation SF is NS and the particle size deviation DF is PS, that is, the actual speed is slightly lower than the target speed and the actual particle size is slightly larger than the target particle size, then adjust the feed speed F to NS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0081] Rule 6: If the speed deviation SF is PS and the particle size deviation DF is NS, then adjust the feed speed F to PS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0082] Rule 7: If the humidity deviation HF is ZO, the speed deviation SF is ZO, and the particle size deviation DF is PS, then fine-tune the feed speed F to NS, and fine-tune the pelletizer speed S to NS, and adjust the feed speed F according to the corresponding preset feed speed adjustment amount, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0083] Rule 8: If the humidity deviation HF is ZO, the speed deviation SF is ZO, and the particle size deviation DF is NS, then fine-tune the feed speed F to PS, and fine-tune the pelletizer speed S to PS, and adjust the feed speed F according to the corresponding preset feed speed adjustment amount, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0084] Rule 9: If the humidity deviation HF is NS and the humidity change rate HB is PS, and the particle size deviation DF is PB, then the pelletizer speed S is rapidly increased to PB, and the feed speed F is greatly reduced to NB. The feed speed F is adjusted according to the corresponding preset feed speed adjustment amount, and the pelletizer speed S is adjusted according to the corresponding preset pelletizer speed adjustment amount.

[0085] Rule 10: If the speed deviation SF is PS and the speed change rate SB is NS, and the particle size deviation DF is NS, then slowly reduce the feed speed F to NS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount.

[0086] A powder granulation control system based on fuzzy control is used to implement a powder granulation control method based on fuzzy control. The system includes:

[0087] Data acquisition unit, used to collect key parameters of the granulation process in real time;

[0088] The data processing unit is used to pre-process the collected key parameters and obtain the change rate characteristics and fluctuation amplitude characteristics of the key parameters;

[0089] A data analysis unit is used to determine the feed speed adjustment amount and the pelletizer speed adjustment amount of the pelletizing equipment according to a pre-set fuzzy rule library;

[0090] The data adjustment unit is used to calculate and execute the adjustment of the granulating equipment based on the feed speed adjustment amount and the granulator speed adjustment amount obtained by data analysis.

[0091] Beneficial effects of the present invention:

[0092] Comprehensive and real-time data collection: Real-time collection of key parameters such as powder moisture, granulator speed, feed rate, and particle size during the granulation process enables timely acquisition of the actual situation during the granulation process, providing an accurate and real-time data basis for subsequent control and adjustment, enabling the system to quickly respond to changes in the granulation process.

[0093] Filtering: Use median processing or mean processing to filter key parameters, effectively remove noise interference in the data, make key parameter values more accurate and stable, and provide high-quality data for subsequent data processing and analysis.

[0094] Normalization processing: Normalize the key parameters after filtering, unify the parameters in different ranges into the same interval, facilitate subsequent feature extraction and analysis, and improve the versatility and comparability of data processing.

[0095] Feature extraction processing: It can obtain the change rate characteristics and fluctuation amplitude characteristics of key parameters, analyze the parameters from different angles, and fully reflect the dynamic changes of parameters in the granulation process, providing rich information for data analysis and control decision-making.

[0096] The data analysis is scientific and reasonable: the feed speed adjustment amount of the granulating equipment and the granulator speed adjustment amount are determined based on the pre-set fuzzy rules. The characteristics of the key parameters are converted into fuzzy language variables through feature conversion, and fuzzy reasoning is performed to make the data analysis more in line with the complexity and uncertainty of the actual granulating process, and to make reasonable control decisions according to different parameter changes.

[0097] Accurate Adjustment Calculation and Execution: Final adjustment values for feed rate and pelletizer speed are determined using a weighted average method, comprehensively considering the influence of different fuzzy rules to ensure more accurate and reliable adjustment results. Adding these adjustment values to the current feed rate and pelletizer speed allows for precise adjustment of the pelletizing equipment, ensuring process stability and product quality.

[0098] The fuzzy rule base is rich and practical: the rules in the fuzzy rule base cover a variety of different parameter deviation combinations, such as humidity deviation, speed deviation, particle size deviation, etc., and can make corresponding adjustment decisions for different granulation conditions, making the system highly adaptable and flexible, able to meet the control requirements of different granulation processes, and improve the efficiency and quality of granulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] The present invention will be further described below with reference to the accompanying drawings.

[0100] Figure 1 This is a system block diagram of a powder granulation control method and system based on fuzzy control of the present invention.

[0101] Figure 2 It is a flow chart of a powder granulation control method and system based on fuzzy control of the present invention. DETAILED DESCRIPTION

[0102] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0103] Example 1

[0104] See also Figure 1 and Figure 2As shown, the present invention is a powder granulation control method based on fuzzy control, comprising the following steps:

[0105] Data collection: real-time collection of key parameters during granulation;

[0106] Key parameters include:

[0107] Powder humidity: measured in real time by a humidity sensor, in percentage (%);

[0108] Pellet mill speed: measured by a speed sensor, in revolutions per minute (r / min);

[0109] Feed rate: measured by weight sensor, unit is grams per minute (g / min);

[0110] Particle size: measured by optical sensor, in millimeters (mm);

[0111] In this embodiment, the key parameters are collected once per second to ensure real-time data.

[0112] Data processing: Preprocess the collected key parameters in the following ways:

[0113] Normalization processing:

[0114] Normalize key parameters;

[0115] Normalization is done by the formula: To achieve;

[0116] Where G1 represents the normalized powder moisture, granulator speed, feed rate and particle size; G min and G max They are the normal operating lower limit and normal operating upper limit of the key parameters corresponding to powder moisture, granulator speed, feed rate and particle size under normal operation;

[0117] In this embodiment, the normalization process is used to eliminate the dimensional differences between different key parameters so that the different key parameters can be compared and calculated on the same scale;

[0118] Feature extraction processing:

[0119] The normalized key parameters are subjected to feature extraction. The feature extraction method is as follows:

[0120] Rate of change characteristics:

[0121] In a specified period, obtain the key parameters of each adjacent acquisition timestamp, then subtract the key parameters of the next acquisition timestamp from the key parameters of the previous acquisition timestamp, and then divide the difference by the time interval between the two adjacent acquisition timestamps to obtain the change parameters of the key parameters of each adjacent acquisition timestamp;

[0122] Then, the average value of the change parameters corresponding to the key parameters at all adjacent acquisition timestamps is calculated, and the change rate characteristics are obtained;

[0123] The formula is: ;

[0124] Where GB is the change rate characteristic of the key parameters, G1 t is the normalized key parameter of each acquisition timestamp in the specified period, t=1, 2, ..., v, where v represents the number of acquisition timestamps in the specified period, and t0 is the time interval between adjacent acquisition timestamps;

[0125] Among them, GB∈{HB, SB, FB, DB}, HB is the change rate characteristic of powder moisture, SB is the change rate characteristic of granulator speed, FB is the change rate characteristic of feed speed, and DB is the change rate characteristic of particle size;

[0126] Fluctuation amplitude characteristics:

[0127] Within a specified period, the key parameters with the largest and smallest values are extracted from the normalized key parameters at all acquisition timestamps. The key parameter with the largest value is then subtracted from the key parameter with the smallest value to obtain the fluctuation amplitude feature.

[0128] The formula is: ;

[0129] Where GF is the fluctuation amplitude characteristic of the key parameters, G1 max and G1 min are the key parameters with the maximum and minimum values after normalization;

[0130] Among them, GF∈{HF, SF, FF, DF}, HF is the fluctuation amplitude characteristic of powder moisture, SF is the fluctuation amplitude characteristic of granulator speed, FF is the fluctuation amplitude characteristic of feed rate, and DF is the fluctuation amplitude characteristic of particle size;

[0131] Data analysis: Based on pre-set fuzzy rules, the feed rate adjustment amount and pelletizer speed adjustment amount of the pelletizing equipment are determined;

[0132] Step X1: Feature transformation:

[0133] The change rate features and fluctuation amplitude features obtained after feature extraction of key parameters are converted into fuzzy linguistic variables:

[0134] Among them, the fuzzy linguistic variables are: negative large NB, negative small NS, zero ZO, positive small PS, positive large PB;

[0135] The fuzzy linguistic variable conversion method is as follows:

[0136] For the powder moisture change rate characteristic HB;

[0137] If HB≤−0.1% / s, the membership degree of the fuzzy linguistic variable is NB, which means that the humidity drops significantly;

[0138] If −0.1% / s<HB≤−0.05% / s, the membership degree of the fuzzy linguistic variable is NS;

[0139] If −0.05% / s<HB≤0.05% / s, the membership degree of the fuzzy linguistic variable is ZO;

[0140] If 0.05% / s<HB≤0.1% / s, the membership degree of the fuzzy linguistic variable is PS;

[0141] If HB>0.1% / s, the membership degree of the fuzzy linguistic variable is PB;

[0142] For the particle size change rate characteristic DB;

[0143] If DB≤−0.05 mm / s, the membership degree of the fuzzy linguistic variable is NB;

[0144] If −0.05 mm / s<DB≤−0.02 mm / s, the membership degree of the fuzzy linguistic variable is NS;

[0145] If −0.02 mm / s<DB≤0.02 mm / s, the membership degree of the fuzzy linguistic variable is ZO;

[0146] If 0.02mm / s<DB≤0.05mm / s, the membership degree of the fuzzy linguistic variable is PS;

[0147] If DB>0.05mm / s, the membership degree of the fuzzy linguistic variable is PB;

[0148] The conversion method of the granulator speed change rate characteristic SB and the feed rate change rate characteristic FB is similar to the conversion method of the powder moisture change rate characteristic HB and the particle size change rate characteristic DB;

[0149] For the fluctuation amplitude characteristic HF of powder humidity;

[0150] If HF≤0.5%, the membership degree of the fuzzy linguistic variable is NB;

[0151] If 0.5%<HF≤1%, the membership degree of the fuzzy linguistic variable is NS;

[0152] If 1%<HF≤2%, the membership degree of the fuzzy linguistic variable is ZO;

[0153] If 2%<HF≤3%, the membership degree of the fuzzy linguistic variable is PS;

[0154] If HF>3%, the membership degree of the fuzzy linguistic variable is PB;

[0155] For the particle size fluctuation amplitude characteristic DF;

[0156] If DF≤0.02mm, the membership degree of the fuzzy linguistic variable is NB, indicating that the particle size fluctuation is very small;

[0157] If 0.02mm<DF≤0.05mm, the membership degree of the fuzzy linguistic variable is NS;

[0158] If 0.05mm<DF≤0.1mm, the membership degree of the fuzzy linguistic variable is ZO;

[0159] If 0.1mm<DF≤0.15mm, the membership degree of the fuzzy linguistic variable is PS;

[0160] If DF>0.15 mm, the membership degree of the fuzzy linguistic variable is PB, which means that the particle size fluctuates greatly;

[0161] The conversion method of the fluctuation amplitude characteristic SF of the granulator speed and the fluctuation amplitude characteristic FF of the feed rate is similar to the conversion method of the fluctuation amplitude characteristic HF of the powder moisture and the fluctuation amplitude characteristic DF of the particle size;

[0162] Step X2, fuzzy reasoning:

[0163] Extracting fuzzy rules from a pre-set fuzzy rule base, then comparing the converted fuzzy language variables corresponding to the key parameters with the fuzzy rules in the fuzzy rule base to determine the feed speed adjustment amount and the pelletizer speed adjustment amount;

[0164] Data adjustment: adjust the operating parameters of the granulation equipment;

[0165] Step K1, determining the feed speed adjustment amount and the pelletizer speed adjustment amount corresponding to each fuzzy rule obtained by fuzzy reasoning; and determining the final feed speed adjustment value and the final pelletizer speed adjustment value;

[0166] Specifically:

[0167] The feed speed adjustment values obtained by different fuzzy rules are marked as JT j, and obtain the preset weight coefficients corresponding to different fuzzy rules and mark them as w j The granulator speed adjustment values obtained by different fuzzy rules are marked as ZT j ;

[0168] Then, the weighted average method is used to calculate the final adjustment values of the feed speed and the pelletizer speed respectively;

[0169] The formula is as follows:

[0170] ;

[0171] Where j = 1, 2, ... m, m represents the number of feed rate adjustment and pelletizer speed adjustment, JT0 and ZT0 are the final values of feed rate adjustment and pelletizer speed adjustment, respectively;

[0172] Step K2, adding the final value of the feed rate adjustment and the final value of the granulator speed adjustment to the current feed rate and granulator speed of the granulator, respectively, to obtain the adjusted feed rate and granulator speed of the granulator;

[0173] This embodiment effectively ensures real-time data by collecting key parameters at a high frequency, allowing the system to respond promptly to changes in granulation. Normalization eliminates dimensional differences between different parameters, laying a unified foundation for subsequent analysis. The extraction of change rate and fluctuation amplitude characteristics depicts the dynamic changes of parameters in multiple dimensions, providing rich information for fuzzy rule analysis. Based on the fuzzy rules, the adjustment amount is determined and the final value is obtained through weighted averaging. This makes the adjustment of feed rate and granulator speed scientific and reasonable, improves the automation and intelligence level of granulation equipment, and ensures stable granulation quality.

[0174] Example 2

[0175] As the second embodiment of the present invention, when the present application is specifically implemented, compared with the first embodiment, the technical solution of this embodiment is different from that of the first embodiment only in that in this embodiment, the preprocessing method further includes: filtering processing:

[0176] The filtering process uses median processing or mean processing;

[0177] Median filter:

[0178] Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size;

[0179] Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where g represents the acquisition timestamp corresponding to the selected key parameter, and k is an odd number;

[0180] Then, the key parameter sequence is sorted from small to large, and the middle value is taken as the key parameter value after filtering;

[0181] Mean filtering:

[0182] Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size;

[0183] Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where G g represents the selected key parameter, k is an odd number;

[0184] In this embodiment, The preceding time of the key parameter The key parameters of the acquisition timestamp are: The subsequent time of this key parameter Key parameters of the acquisition timestamp;

[0185] Then the key parameter sequence is averaged and the average value is taken as the filtered key parameter value;

[0186] The formula for calculating the average value is as follows:

[0187] ;

[0188] Where, GL g is the key parameter value after filtering corresponding to the key parameter;

[0189] This embodiment, based on Example 1, adds a median or mean filtering method, effectively removing noise interference from the collected data and significantly improving the accuracy and stability of the data. The filtered data is used for subsequent processing, avoiding erroneous judgments and adjustments caused by noise, and further enhancing the reliability of the granulation process control.

[0190] Example 3

[0191] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments. The technical solution of this embodiment is different from the first and second embodiments only in that this embodiment further provides a fuzzy rule in a fuzzy rule base, which is as follows:

[0192] Rules based on humidity deviation and speed deviation

[0193] Rule 1: If the humidity deviation HF is NB and the speed deviation SF is PB, that is, the actual humidity is much lower than the target humidity and the actual speed is much higher than the target speed, then adjust the feed speed F to PB and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0194] Explanation: When the humidity is very low, the powder has better fluidity and the particles are less likely to agglomerate. At this time, the rotation speed is also higher, which means that the equipment has a stronger processing capacity;

[0195] In order to improve the granulation efficiency under this state, increasing the feed rate can make full use of the processing capacity of the equipment and enable more materials to be granulated under suitable conditions;

[0196] For example, if the original feeding speed is 20g / min, it can be increased to about 30g / min according to this rule (the specific value is fine-tuned according to the actual situation);

[0197] Rule 2: If the humidity deviation HF is PB and the speed deviation SF is NB, then adjust the feed speed F to NB and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0198] Explanation: When the humidity is high, the powder is sticky and easily agglomerated, while the rotation speed is low, and the equipment's ability to handle the material is limited. At this time, if the feed speed is too fast, the material will accumulate in the equipment and cannot be fully processed, affecting the granulation quality. Therefore, it is necessary to significantly reduce the feed speed, for example, from 25g / min to 15g / min, so that the equipment can better handle high-humidity materials.

[0199] Rules based on humidity deviation and particle size deviation

[0200] Rule 3: If the humidity deviation HF is NB and the particle size deviation DF is PB, then adjust the pelletizer speed S to PB, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0201] Explanation: Low humidity makes the powder relatively dry and loose, and the particle size is relatively large, which may be because the particles are not fully sheared and mixed during the granulation process;

[0202] Increasing the granulator speed can increase the friction, collision and mixing between particles and between particles and equipment components, so that large particles can be further refined and shaped, so that the particle size is closer to the target value;

[0203] For example, the original speed is 280r / min, which can be increased to 320r / min;

[0204] Rule 4: If the humidity deviation HF is PB and the particle size deviation DF is NB, then adjust the pelletizer speed S to NB and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0205] Explanation: Under high humidity, the powder is sticky but the particle size is small. This may be because the rotation speed is too high, resulting in excessive particle crushing. Reducing the rotation speed can reduce the degree of particle crushing, allowing the particles to agglomerate and form better under the condition of high stickiness, thereby increasing the particle size. For example, reducing the rotation speed from 350r / min to 300r / min.

[0206] Rules based on speed deviation and particle size deviation

[0207] Rule 5: If the speed deviation SF is NS and the particle size deviation DF is PS, that is, the actual speed is slightly lower than the target speed and the actual particle size is slightly larger than the target particle size, then adjust the feed speed F to NS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0208] Explanation: When the speed is slightly lower, the equipment's processing capacity decreases slightly, while the particle size increases slightly. If the feed rate remains unchanged or is too fast, the problem of large particle size may be exacerbated. Properly reducing the feed rate will allow the equipment more time and energy to process the material, gradually restoring the particle size to normal; for example, reducing the feed rate from 22g / min to 20g / min.

[0209] Rule 6: If the speed deviation SF is PS and the particle size deviation DF is NS, then adjust the feed speed F to PS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0210] Explanation: A slightly higher speed means that the equipment has a slightly stronger processing capacity, while a slightly smaller particle size may be due to a relatively insufficient material supply. Properly increasing the feed rate can allow the equipment to granulate more material under a stronger processing capacity, which helps to achieve the target particle size. For example, increase the feed rate from 18g / min to 20g / min.

[0211] Integrating multi-parameter rules

[0212] Rule 7: If the humidity deviation HF is ZO, the speed deviation SF is ZO, and the particle size deviation DF is PS, then fine-tune the feed speed F to NS, and fine-tune the pelletizer speed S to NS, and adjust the feed speed F according to the corresponding preset feed speed adjustment amount, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0213] Explanation: When the humidity and rotation speed are appropriate, the particle size is slightly larger, indicating that the current feed rate and rotation speed combination may cause the material to stay in the equipment for too short a time, or the processing intensity may be slightly higher;

[0214] By simultaneously fine-tuning and reducing the feed rate and rotation speed, the material can have a more appropriate processing time and intensity in the equipment, making the particle size closer to the standard. For example, the feed rate can be reduced by 1-2g / min and the rotation speed can be reduced by 5-10r / min;

[0215] Rule 8: If the humidity deviation HF is ZO, the speed deviation SF is ZO, and the particle size deviation DF is NS, then fine-tune the feed speed F to PS, and fine-tune the pelletizer speed S to PS, and adjust the feed speed F according to the corresponding preset feed speed adjustment amount, and adjust the pelletizer speed S according to the corresponding preset pelletizer speed adjustment amount;

[0216] Explanation: When the humidity and rotation speed are normal and the particle size is slightly small, it means that the current feed rate and rotation speed combination may cause the material to stay in the equipment for too long, or the processing intensity may be slightly low;

[0217] Appropriate fine-tuning of the feed rate and rotation speed can improve the equipment's material processing efficiency and increase the particle size to the target range. For example, increasing the feed rate by 1-2 g / min and the rotation speed by 5-10 r / min;

[0218] Rules that take into account the rate of change of parameters

[0219] Rule 9: If the humidity deviation HF is NS and the humidity change rate HB is PS, and the particle size deviation DF is PB, then the pelletizer speed S is rapidly increased to PB, and the feed speed F is greatly reduced to NB. The feed speed F is adjusted according to the corresponding preset feed speed adjustment amount, and the pelletizer speed S is adjusted according to the corresponding preset pelletizer speed adjustment amount.

[0220] Explanation: Humidity is slightly low but has an upward trend, and particle size is large, indicating that changes in humidity may have further adverse effects on particle size;

[0221] Rapidly increasing the rotation speed can enhance the material processing capacity to cope with problems such as viscosity changes caused by increased humidity, while significantly reducing the feed rate to prevent material accumulation in the equipment and deterioration of granulation quality due to increased humidity and large particle size;

[0222] Rule 10: If the speed deviation SF is PS and the speed change rate SB is NS, and the particle size deviation DF is NS, then slowly reduce the feed speed F to NS and adjust the feed speed F according to the corresponding preset feed speed adjustment amount;

[0223] Explanation: The speed is slightly high but has a downward trend, and the particle size is slightly smaller. Slowly reducing the feed speed can enable the equipment to maintain a relatively stable granulation effect during the process of gradually decreasing speed, avoiding insufficient material processing due to unchanged feed speed and further reducing the particle size.

[0224] This example presents a specific fuzzy rule library. Each rule is formulated based on the relationships between different parameters and is accompanied by detailed explanations and examples, providing clear guidance for adjusting granulation equipment parameters. The fuzzy rules comprehensively consider the combination of multiple parameters, including humidity deviation, rotational speed deviation, particle size deviation, and parameter change rate. By coordinating the feed rate and granulator speed, the interactions between these multiple parameters are fully utilized, achieving comprehensive optimization of the granulation process, thereby effectively improving granulation efficiency and quality.

[0225] Example 4

[0226] As the fourth embodiment of the present invention, when this application is specifically implemented, compared with the first, second and third embodiments, the technical solution of this embodiment is to combine and implement the solutions of the above-mentioned first, second and third embodiments.

[0227] This embodiment combines the solutions of Examples 1, 2, and 3, ensuring not only real-time and accurate data but also a rich set of feature extraction methods and scientific data analysis methods, along with a detailed fuzzy rule library. This comprehensive integration fully leverages the advantages of the first three embodiments, providing a comprehensive, efficient, and intelligent control strategy for the granulation process, maximizing the performance of the granulation equipment and ensuring high-quality granulation production.

[0228] A powder granulation control system based on fuzzy control is used to implement a powder granulation control method based on fuzzy control. The system includes:

[0229] Data acquisition unit, used to collect key parameters of the granulation process in real time;

[0230] The data processing unit is used to pre-process the collected key parameters and obtain the change rate characteristics and fluctuation amplitude characteristics of the key parameters;

[0231] A data analysis unit is used to determine the feed speed adjustment amount and the pelletizer speed adjustment amount of the pelletizing equipment according to a pre-set fuzzy rule library;

[0232] The data adjustment unit is used to calculate and execute the adjustment of the granulating equipment based on the feed speed adjustment amount and the granulator speed adjustment amount obtained by data analysis.

[0233] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0234] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A powder granulation control method based on fuzzy control, characterized in that: The following steps are involved: Data collection: real-time collection of key parameters during granulation; Data processing: pre-process the collected key parameters by using median processing or mean processing, and obtain the filtered key parameter values; Then, the filtered key parameters are normalized; then, the normalized key parameters are feature extracted, and the change rate characteristics and fluctuation amplitude characteristics of the key parameters are obtained; Data analysis: Based on the pre-set fuzzy rule base, the feed speed adjustment amount and the pelletizer speed adjustment amount of the pelletizing equipment are determined; Data adjustment: Based on the feed speed adjustment and pelletizer speed adjustment obtained from data analysis, the adjustment calculation and execution of pelletizing equipment are carried out.

2. A powder granulation control method based on fuzzy control according to claim 1, characterized in that: Key parameters include: powder moisture measured in real time by a moisture sensor, granulator speed measured by a speed sensor, feed rate measured by a weight sensor, and granule size measured by an optical sensor.

3. A powder granulation control method based on fuzzy control according to claim 2, characterized in that: Median filter: Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size; Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where g represents the acquisition timestamp corresponding to the selected key parameter, and k is an odd number; Then, the key parameter sequence is sorted from small to large, and the middle value is taken as the key parameter value after filtering.

4. The powder granulation control method based on fuzzy control according to claim 2, characterized in that: Mean filtering: Select a key parameter and mark it as G, where G∈{H, S, F, D}, H is the powder moisture, S is the granulator speed, F is the feed rate, and D is the particle size; Then, taking the acquisition timestamp corresponding to the key parameter as the base point, the key parameters on the k consecutive acquisition timestamps before and after the acquisition timestamp are obtained and grouped into a key parameter sequence [ ], where G g represents the selected key parameter, k is an odd number; Then the key parameter sequence is averaged and the average value is taken as the filtered key parameter value.

5. The powder granulation control method based on fuzzy control according to claim 2, characterized in that: Normalization is done by the formula: To achieve; Where G1 represents the normalized powder moisture, granulator speed, feed rate and particle size; G min and G max They are the normal operating lower limit and normal operating upper limit of the key parameters corresponding to powder moisture, granulator speed, feed rate and particle size during normal operation.

6. The powder granulation control method based on fuzzy control according to claim 2, characterized in that: The feature extraction method is as follows: Rate of change feature extraction: In a specified period, obtain the key parameters of each adjacent acquisition timestamp, then subtract the key parameters of the next acquisition timestamp from the key parameters of the previous acquisition timestamp, and then divide the difference by the time interval between the two adjacent acquisition timestamps to obtain the change parameters of the key parameters of each adjacent acquisition timestamp; Then, the average value of the change parameters corresponding to the key parameters at all adjacent acquisition timestamps is calculated, and the change rate characteristics are obtained; Fluctuation amplitude feature extraction: Within a specified period, the key parameters with the largest and smallest values are extracted from the normalized key parameters at all acquisition timestamps. The key parameter with the largest value is then subtracted from the key parameter with the smallest value to obtain the fluctuation amplitude feature.

7. The powder granulation control method based on fuzzy control according to claim 6, characterized in that: The specific methods of data analysis are as follows: Step X1: Feature transformation: The change rate features and fluctuation amplitude features obtained after feature extraction of key parameters are converted into fuzzy linguistic variables: Among them, the fuzzy linguistic variables are: negative large NB, negative small NS, zero ZO, positive small PS, positive large PB; The rate of change characteristic of powder moisture is recorded as HB, the rate of change characteristic of granulator speed is recorded as SB, the rate of change characteristic of feed rate is recorded as FB, and the rate of change characteristic of particle size is recorded as DB; At the same time, the fluctuation amplitude characteristic of powder humidity is recorded as HF, the fluctuation amplitude characteristic of granulator speed is recorded as SF, the fluctuation amplitude characteristic of feed rate is recorded as FF, and the fluctuation amplitude characteristic of particle size is recorded as DF; The fuzzy linguistic variable conversion method is as follows: For the powder moisture change rate characteristic HB; If HB≤-HB a , then the membership degree of the fuzzy linguistic variable is NB, that is, the humidity drops significantly; If-HB a <HB≤HB b , then the membership degree of the fuzzy linguistic variable is NS; If −HB b <HB≤HB b , then the membership degree of the fuzzy linguistic variable is ZO; If HB b <HB≤HB a , then the membership degree of the fuzzy linguistic variable is PS; If HB>HBa, the membership degree of the fuzzy linguistic variable is PB; Among them, HB a and HB b is the membership judgment threshold preset according to the powder moisture change rate characteristics, and HB a >HB b ; The conversion method for the rate of change characteristic SB of the granulator rotation speed, the rate of change characteristic FB of the feed speed, and the rate of change characteristic DB of the particle size is similar to that for the rate of change characteristic HB of the powder moisture content; For the fluctuation amplitude characteristic HF of powder humidity; If HF≤HF a , then the membership degree of the fuzzy linguistic variable is NB; If HF a <HF≤HF b , then the membership degree of the fuzzy linguistic variable is NS; If HF b <HF≤HF c , then the membership degree of the fuzzy linguistic variable is ZO; If HF c <HF≤HF d , then the membership degree of the fuzzy linguistic variable is PS; If HF d >3%, then the membership degree of the fuzzy linguistic variable is PB; Among them, HF a , HF b , HF c , HF d is the membership judgment threshold preset according to the powder humidity fluctuation amplitude characteristics, and HF a <HF b <HF c <HF d ; The conversion method of the fluctuation amplitude characteristic SF of the granulator speed, the fluctuation amplitude characteristic FF of the feed rate and the fluctuation amplitude characteristic DF of the particle size is similar to the conversion method of the fluctuation amplitude characteristic HF of the powder moisture; Step X2, fuzzy reasoning: Fuzzy rules are extracted from a pre-set fuzzy rule base, and then the key parameters are converted into fuzzy linguistic variables, which are compared with the fuzzy rules in the fuzzy rule base to determine the feed speed adjustment amount and the pelletizer speed adjustment amount.

8. The powder granulation control method based on fuzzy control according to claim 7, characterized in that: The adjustment calculation and execution method are as follows: According to the fuzzy rules obtained by fuzzy reasoning, the feed speed adjustment amount and the pelletizer speed adjustment amount are respectively corresponding to each other; the final value of the feed speed adjustment and the final value of the pelletizer speed adjustment are determined; The final value of the feed rate adjustment and the final value of the pelletizer speed adjustment are added to the current feed rate and pelletizer speed of the pelletizer, respectively, to obtain the adjusted feed rate and pelletizer speed of the pelletizer.

9. The powder granulation control method based on fuzzy control according to claim 8, characterized in that: The final adjustment value of the feed rate and the final adjustment value of the pelletizer speed are determined as follows: The feed speed adjustment values obtained by different fuzzy rules are marked as JT j , and obtain the preset weight coefficients corresponding to different fuzzy rules and mark them as w j The granulator speed adjustment values obtained by different fuzzy rules are marked as ZT j ; Then, the weighted average method is used to calculate the final adjustment values of the feed speed and the pelletizer speed respectively; The formula is as follows: ； Where j = 1, 2, ... m, m represents the number of feed rate adjustment and pelletizer speed adjustment, JT0 and ZT0 are the final values of feed rate adjustment and pelletizer speed adjustment, respectively.

10. A powder granulation control system based on fuzzy control, the system being used to implement a powder granulation control method based on fuzzy control according to any one of claims 1 to 9, characterized in that: The system includes: Data acquisition unit, used to collect key parameters of the granulation process in real time; The data processing unit is used to pre-process the collected key parameters and obtain the change rate characteristics and fluctuation amplitude characteristics of the key parameters; A data analysis unit is used to determine the feed speed adjustment amount and the pelletizer speed adjustment amount of the pelletizing equipment according to a pre-set fuzzy rule library; The data adjustment unit is used to calculate and execute the adjustment of the granulating equipment based on the feed speed adjustment amount and the granulator speed adjustment amount obtained by data analysis.

Images