A method for detecting and controlling the production temperature of cold-soluble gelatin powder

Abstract

The present invention relates to the technical field of the production of cold-soluble gelatin powder, and proposes a method for temperature detection and control in the production of cold-soluble gelatin powder. In terms of temperature detection, the limitation of single-point detection is abandoned, and a multi-point distributed sensor network is constructed to comprehensively cover the reaction kettle, drying oven and material transmission pipeline, accurately capture the temperature at each position, and provide detailed basis for temperature control; the temperature control strategy is diverse and intelligent. The multi-model adaptive control customizes models for different production stages and automatically switches to meet the production requirements; the fuzzy logic control deals with complex non-linear temperature problems and improves the control accuracy; overall, the level of temperature detection and control is greatly improved, the stability of the production process of cold-soluble gelatin powder is guaranteed, and the product quality and production efficiency are improved; a closed-loop system is constructed through a real-time feedback adjustment mechanism, and the detection data is used for real-time regulation to avoid the accumulation of temperature deviation; the control parameters are dynamically optimized, and based on historical data and real-time feedback, the control effect is continuously optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of the production of cold-soluble gelatin powder, and specifically to a method for detecting and controlling the production temperature of cold-soluble gelatin powder. Background Art

[0002] In the field of cold-soluble gelatin powder production, temperature plays a decisive role in product quality. In traditional production processes, there are many drawbacks in temperature detection and control methods, making it difficult to meet the requirements of high-quality production.

[0003] Existing temperature detection often adopts a single-point detection method, which can only obtain the temperature at a specific position of the production equipment and cannot reflect the overall temperature distribution. For example, in a large reaction kettle, the heat of material reaction causes obvious temperature differences at different heights. Single-point detection is likely to miss key temperature changes, resulting in uneven product quality. In terms of temperature control, the strategy is single and lacks real-time feedback adjustment. Most production relies on fixed set-value control and cannot be flexibly adjusted according to changes in raw material characteristics, environmental temperature, etc. Once the production conditions fluctuate, temperature deviation will immediately occur. For example, in the drying stage, changes in external humidity affect the drying efficiency, and traditional control methods are difficult to adjust the temperature in a timely manner, easily causing the product moisture content to not meet the standard. These problems seriously restrict the improvement of the quality of cold-soluble gelatin powder and the increase of production efficiency.

[0004] Therefore, a method for detecting and controlling the production temperature of cold-soluble gelatin powder is proposed. Through multi-point detection, multi-element control strategies, and a real-time feedback optimization mechanism, accurate temperature detection and efficient control are achieved, providing reliable technical support for the production of cold-soluble gelatin powder and meeting the urgent needs of the industry for high-quality products. Summary of the Invention

[0005] Technical Problem to be Solved: In view of the deficiencies of the prior art, the present invention provides a method for detecting and controlling the production temperature of cold-soluble gelatin powder.

[0006] Technical Solution: To achieve the above-mentioned purpose, the present invention provides the following technical solution: A method for detecting and controlling the production temperature of cold-soluble gelatin powder, comprising the following steps:

[0007] Construct a multi-point distributed sensor network, and arrange temperature sensors at the top, middle, and bottom of the reaction kettle, the air inlet and outlet of the drying oven, and at specific intervals along the material transfer pipeline to collect temperature data at different positions in real time;

[0008] Apply sensor data fusion technology to perform weighted average fusion on the data of each sensor to obtain a value reflecting the overall temperature state.

[0009] Adopt multi-model adaptive control, establish corresponding temperature control models for the raw material dissolution, reaction, and drying stages of cold-soluble gelatin powder production respectively, and automatically switch models according to the production state.

[0010] Combined with fuzzy logic control, taking the temperature deviation and the rate of change of temperature deviation as input variables, the control output is determined through fuzzyfication, fuzzy rule inference, and defuzzyfication operations in sequence.

[0011] Build a closed-loop control system, and the temperature detection data is fed back to the temperature control unit in real time. The control quantity is adjusted according to the deviation between the feedback data and the set temperature and output to the actuator.

[0012] Implement dynamic optimization of control parameters, establish a relationship model between control parameters and product quality based on historical data, and use an adaptive algorithm to online adjust the control parameters according to the real-time feedback data.

[0013] Preferably, the sensors arranged at the top, middle, and bottom of the reactor are used to monitor the temperature at different height levels of the material. The sensor at the air inlet of the drying oven is used to monitor the initial temperature of the hot air, and the sensor at the air outlet of the drying oven is used to monitor the temperature of the gas discharged after the material is dried. Sensors are arranged on the material transfer pipeline at intervals of 0.5 to 2 meters to monitor the temperature change during the material transfer process.

[0014] Preferably, in the sensor data fusion technology, the weights are determined according to the importance of the sensor positions to the production process. The weight of the sensor at the bottom of the reactor is determined to be 0.4, the weight of the sensor at the top is determined to be 0.2, and the weight of the middle sensor is determined to be 0.4; the weight of the sensor at the air outlet of the drying oven is determined to be 0.6, and the weight of the other sensor is determined to be 0.4; the weights of the sensors on the material transfer pipeline near the key links are higher than those at other positions.

[0015] Preferably, the raw material dissolution stage model is , where is the temperature at time, , are coefficients obtained by fitting experimental data, represents the influence degree of heating power on temperature, represents the influence degree of stirring speed on temperature.

[0016] Preferably, the reaction stage model is , is a coefficient related to the reaction heat, determined by reaction kinetics experiments, is the initial reaction temperature.

[0017] Preferably, the drying stage model is , , are coefficients determined by experiments, represents the influence degree of heating power on temperature, represents the influence degree of ventilation volume on temperature, is the starting temperature of drying.

[0018] Preferably, in the fuzzy logic control, the temperature deviation e and the rate of change of temperature deviation ec are used as input variables, and the temperature deviation , and the rate of change of temperature deviation ec = , and a Gaussian membership function is used for fuzzification, where is the center value of the membership function, and is the standard deviation.

[0019] Preferably, in the closed-loop control system, the temperature detection data is transmitted to the temperature control unit using a programmable logic controller via the RS485 bus. The PLC calculates the control quantity adjustment value according to the control algorithm and outputs it to the heating and cooling actuators.

[0020] Preferably, in the dynamic optimization control parameters, a relationship model between the control parameters and the product quality is established using the least squares method based on historical data , where is the product quality index, and is the control parameter.

[0021] Preferably, during the online adaptive optimization, the center value and standard deviation of the membership function in the fuzzy logic control are dynamically adjusted according to the real-time ranges of the temperature deviation and the rate of change of the deviation, and the corresponding model coefficients are dynamically adjusted according to the real-time parameters of the reaction stage and the drying stage.

[0022] Advantages: Compared with the prior art, the present invention provides a method for temperature detection and control in the production of cold-soluble gelatin powder, having the following advantages:

[0023] 1. In the method for temperature detection and control in the production of cold-soluble gelatin powder, in terms of temperature detection, the limitation of single-point detection is abandoned, and a multi-point distributed sensor network is constructed to comprehensively cover the reaction kettle, drying oven, and material transfer pipeline, accurately capturing the temperature at each position and providing detailed basis for temperature control; the temperature control strategy is diverse and intelligent. The multi-model adaptive control customizes models for different production stages and automatically switches to meet the production requirements; the fuzzy logic control processes complex non-linear temperature problems and improves the control accuracy; overall, it greatly improves the temperature detection and control level, ensures the stability of the cold-soluble gelatin powder production process, and improves the product quality and production efficiency.

[0024] 2. The method for temperature detection and control in the production of cold-soluble gelatin powder constructs a closed-loop system through a real-time feedback adjustment mechanism, and adjusts in real time according to the detection data to avoid the accumulation of temperature deviation; dynamically optimizes the control parameters, and continuously optimizes the control effect based on historical data and real-time feedback. Description of the Drawings

[0025] Figure 1 Schematic diagram of the step flow of the present invention;

[0026] Figure 2 Schematic diagram of the frame structure of the present invention. Detailed implementation manners

[0027] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0028] Please refer to Figures 1 to 2 , the present invention provides a method for detecting and controlling the production temperature of cold-soluble gelatin powder, including the following content:

[0029] I. Temperature detection system

[0030] (I) Multi-point distributed sensor network

[0031] 1. Implementation of sensor layout

[0032] Reactor: As a key equipment for gelatin powder production, the internal temperature distribution in the reactor is affected by factors such as material convection and reaction heat. Installing sensors at its top (position A) can monitor the temperature in the upper layer of the reactor, which is crucial for understanding the gas evolution and heat accumulation at the top during the reaction process. For example, when the gelatin raw material reacts with additives, the temperature change at the top can reflect the impact of the volatile substances generated by the reaction on the temperature of the top space. The sensor in the middle (position B) can obtain the temperature of the main body of the middle-layer material in the reactor. The temperature at this position is relatively stable and can represent the average reaction temperature of the materials in the reactor. The sensor at the bottom (position C) is particularly important because the reaction heat usually accumulates at the bottom, and the bottom material is less affected by factors such as stirring. Its temperature change can directly reflect the intensity of the reaction. For example, in an exothermic reaction, the temperature at the bottom rises faster, which can timely monitor whether the reaction is proceeding normally.

[0033] Drying oven: The temperature at the air inlet (position D) of the drying oven determines the initial temperature of the hot air entering the drying oven, which directly affects the starting conditions of material drying. The temperature at the air outlet (position E) reflects the temperature of the gas discharged after passing through the material drying, and can reflect the degree of material drying. By comparing the temperatures at the air inlet and air outlet, the heat utilization efficiency in the drying oven and whether the material is dried sufficiently can be judged. For example, if the temperature at the air outlet is too low, it may mean that the drying is insufficient; if the temperature at the air outlet is too high, there may be a risk of over-drying the material.

[0034] Material transfer pipeline: In the transfer pipeline, the temperature of the material will change continuously due to heat exchange with the pipeline wall and the influence of the external environment. A sensor is arranged every L meters, and the value of L can be determined according to the pipeline length, material properties and production process requirements, generally between 0.5 - 2 meters. Assuming that n sensors are arranged, the positions are successively . These sensors can monitor the temperature change of the material during the transfer process in real time, so as to timely detect temperature anomalies caused by poor pipeline insulation or too long material residence time. For example, in a long-distance transfer pipeline, the sensors on the front section of the pipeline near the heating source can monitor the rising temperature of the material, and the sensors on the rear section can monitor whether the temperature drops too fast due to heat dissipation.

[0035] 2. Sensor selection implementation

[0036] Select a platinum resistance temperature sensor, whose temperature measurement principle is based on the characteristic that the resistance value of metal platinum changes with temperature. In the range of 0 - 100 °C, the resistance value of the platinum resistance has an approximate linear relationship with temperature, which can be expressed by the formula . Among them, is the resistance value of the platinum resistance at t °C, is the resistance value of the platinum resistance at 0 °C, is the temperature coefficient of the platinum resistance, and the general value is 0.00385 1 / °C. The accuracy of this sensor can reach ±0.1 °C, and the response time is short, generally within 1 - 3 seconds. It can quickly and accurately capture temperature changes, meeting the requirements of high-precision and real-time temperature detection in the production of cold-soluble gelatin powder. Its stability is high. During long-term use, the resistance value drifts little, ensuring the reliability of temperature detection.

[0037] (2) Sensor data fusion technology

[0038] Weighted average fusion algorithm implementation

[0039] Suppose there are m sensors in total, and the temperature data collected by them are respectively , and the corresponding weights are respectively . The fused temperature value T is:

[0040]

[0041] The determination of the weight needs to comprehensively consider the importance of the sensor position to the production process. In the reaction kettle, the bottom temperature has the greatest influence on the reaction process. Suppose the weight of the bottom sensor is , because the bottom temperature change directly reflects the heat release of the reaction and has a significant impact on the reaction rate and product quality. The weight of the top sensor is , and the top temperature is mainly affected by gas escape and convection, and has a relatively small impact on the overall reaction. The weight of the middle sensor is , the middle temperature can better represent the main reaction temperature of the material. In the drying oven, the outlet temperature is more crucial for judging the drying degree of the product. Set the weight of the outlet sensor , because the outlet temperature directly reflects the state of the gas discharged after the material is dried and is closely related to the water content of the product. The weight of the inlet sensor , although the inlet temperature is important, its direct impact on the quality of the final product is less than that of the outlet temperature. In the material transfer pipeline, the weight of the sensor close to the key production links (such as before entering the reaction kettle or before entering the drying oven) is relatively high. For example, the weight of the sensor in the transfer pipeline close to the reaction kettle can be set to 0.3, and the weights of the sensors in other positions decrease successively according to the distance from the key link. By reasonably setting the weights, the fused temperature value can more accurately reflect the overall temperature state and provide a reliable basis for subsequent temperature control.

[0042] II. Temperature control strategy

[0043] (I) Multi-model adaptive control

[0044] 1. Model establishment and implementation

[0045] Model for the raw material dissolution stage: The main goal of this stage is to dissolve the gelatin raw material quickly and evenly. Let the heating power be , the stirring speed be , and the temperature change rate be , and establish the model:

[0046] Among them, is the temperature at time , , are the model coefficients obtained by fitting experimental data. represents the influence coefficient of heating power on temperature. For example, in multiple experiments, when the heating power increases by 100 W, the temperature rises by 0.5 °C, then k1 = 0.005 °C / W. represents the influence coefficient of stirring speed on temperature. If the stirring speed increases by 100 r / min and the temperature drops by 0.2 °C, then . is the model error term, which reflects the interference of other unconsidered factors on temperature, such as environmental temperature fluctuations and differences in the initial temperature of the raw materials. This model considers the combined influence of heating power and stirring speed on temperature. By adjusting these two parameters, the temperature in the raw material dissolution stage can be effectively controlled, and the dissolution efficiency and uniformity can be improved.

[0047] Model for the reaction stage: The temperature needs to be precisely controlled during the reaction stage to ensure the progress of the reaction. Let the reaction time be t, the reaction rate be r, and the temperature be , and establish the model:

[0048]

[0049] The initial reaction temperature is set by the production process and is generally determined according to the variety of gelatin powder and reaction requirements. For example, the initial reaction temperature of a specific gelatin powder is set at 50 °C. is a coefficient related to the heat of reaction and is determined by reaction kinetics experiments. For example, through experimental determination, when the reaction rate increases by a certain amount, the temperature rises by 0.3 °C, then . is the model error term for this stage, reflecting the influence of other factors on temperature during the reaction, such as changes in catalyst activity and heat dissipation of the reaction vessel. This model takes the reaction rate as the key variable and precisely controls the temperature in the reaction stage by controlling the reaction rate to ensure that the reaction proceeds as expected and improve product quality and yield.

[0050] Drying stage model: The temperature needs to be controlled within a suitable range during the drying stage to ensure that the product moisture content meets the standard. Let the heating power be , the ventilation volume be , the drying time be , and the temperature be . The model is established as follows:

[0051] is the initial drying temperature, which is determined according to product requirements and drying process. For example, the initial drying temperature of a certain cold-soluble gelatin powder is 60 °C. , is a model coefficient and is determined through experiments. represents the influence coefficient of heating power on temperature. For example, when the heating power increases by 100 W, the temperature rises by 0.4 °C, then . represents the influence coefficient of ventilation volume on temperature. If the ventilation volume increases by a certain amount, the temperature drops by 0.1 °C, then . is the model error term for the drying stage, reflecting the influence of other factors on temperature during drying, such as the sealing of the drying oven and the loading amount of materials. This model comprehensively considers the influence of heating power and ventilation volume on temperature, and by adjusting these two parameters, the temperature in the drying stage can be effectively controlled to ensure that the product moisture content meets the standard.

[0052] 2. Model switching implementation

[0053] During the production process, the production stage is judged by monitoring the production status parameters, and the control model is automatically switched.

[0054] During the raw material dissolution stage, the completion of dissolution is judged by detecting the dissolution degree of the material. An on-line concentration detector, such as a near-infrared spectroscopy analyzer, can be used to monitor the material dissolution rate in real time according to the relationship between the characteristic spectrum and concentration of the gelatin solution. When the material dissolution rate reaches over 95%, it is determined that the raw material dissolution stage ends, and the system automatically switches to the reaction stage model for temperature control.

[0055] During the reaction stage, the reaction process is judged by monitoring the concentration of the reaction progress indicator. A high-performance liquid chromatograph is used to detect the concentration of a specific reaction intermediate in the reaction system. When the concentration of this intermediate reaches 90% of the expected reaction progress, it is determined that the reaction is nearly completed and preparation is made to switch to the drying stage model.

[0056] During the drying stage, the water content of the product is detected by an on-line moisture detector. When the water content of the product reaches the target range (for example, the target water content is 5% - 8%), it is determined that the drying is completed. Through this model switching mechanism based on real-time production status parameters, it can be ensured that the temperature control always matches the production stage, improving production efficiency and product quality.

[0057] (2) Fuzzy logic control

[0058] 1. Implementation of input quantity determination

[0059] The temperature deviation e and the temperature deviation change rate ec are used as input quantities. The temperature deviation , is the current actual temperature, is the set temperature. For example, during the reaction stage, the set temperature is 60°C, and the current actual temperature T is 60.5°C, then the temperature deviation e = 0.5°C. The temperature deviation change rate ec = , is the temperature at the previous moment, is the time interval. Assuming that the temperature at the previous moment is 60.3°C and the time interval is 1 minute, then the temperature deviation change rate ec = . By accurately calculating these two input quantities, the deviation between the current temperature and the set temperature and the temperature change trend can be reflected, providing basic data for subsequent fuzzy logic control.

[0060] 2. Implementation of fuzzy processing

[0061] Both e and ec are divided into multiple fuzzy subsets, such as negative big (NB), negative medium (NM), negative small (NS), zero (Z), positive small (PS), positive medium (PM), and positive big (PB). Taking the membership function of the temperature deviation e as an example, assume a Gaussian membership function:

[0062]

[0063] Among them, is the membership degree of \(e\) belonging to the fuzzy subset \(A\), is the central value of the membership function, is the standard deviation, and different fuzzy subsets correspond to different and values. For example, for the negative big (NB) fuzzy subset, let , , when the temperature deviation , substituting it into the membership function gives , indicating that the degree to which the temperature deviation of -6°C belongs to the negative big fuzzy subset is 0.88. By performing fuzzy processing on \(e\) and \(ec\), the precise temperature deviation and deviation change rate are transformed into fuzzy language variables, facilitating subsequent fuzzy rule reasoning.

[0064] 3. Implementation of Fuzzy Rule Reasoning

[0065] Determine the fuzzy rules based on experience and experiments. For example, if \(e\) is NB and \(ec\) is NB, then the control output (such as the heating power adjustment ) is PB; if \(e\) is Z and \(ec\) is PS, then is NS, etc. These rules form the fuzzy rule base. When constructing the fuzzy rule base, various situations in the production process need to be fully considered. For example, in the reaction stage, when the temperature deviation is negative big and the deviation change rate is also negative big, it indicates that the current temperature is far lower than the set temperature and is still dropping rapidly. At this time, it is necessary to significantly increase the heating power, so the control output is positive big. In practical applications, the fuzzy rule base can be verified and optimized through a large number of experiments and production data to ensure its rationality and effectiveness.

[0066] 4. Implementation of Defuzzification Operation

[0067] The centroid method is used for defuzzification to calculate the final control output. Let the fuzzy output be , and its calculation formula is:

[0068]

[0069] Among them, is the element in the fuzzy output set, is its corresponding membership degree. For example, after fuzzy rule reasoning, there are three elements in the fuzzy output set, and the corresponding membership degrees are , , , then the final control output . Through the defuzzification operation, the fuzzy control output is transformed into an accurate control quantity, which is used to adjust actuators such as heating and cooling to achieve precise temperature control.

[0070] III. Real-time Feedback Adjustment Mechanism

[0071] (I) Establishment of Closed-loop Control System

[0072] 1. Implementation of System Structure

[0073] The temperature data collected by the temperature detection system is transmitted to the temperature control unit through a data transmission line (such as the RS485 bus) after data fusion processing. The temperature control unit generally uses a programmable logic controller (PLC), which has a built-in temperature control algorithm. The PLC calculates the control quantity adjustment value according to the deviation between the feedback data and the set temperature. For example, in the reaction stage, if the set temperature is 60°C and the actual temperature after data fusion is 60.5°C, the PLC calculates that the heating power needs to be reduced according to the control algorithm. Then, the PLC outputs the control signal to the actuators such as heating and cooling. The heating actuator can use an electric heating element to control the heating power by adjusting the current magnitude; the cooling actuator can use a cold water circulation device to control the cooling water volume by adjusting the valve opening. The actuator adjusts the temperature of the production equipment according to the control signal output by the PLC to form a closed-loop control. In the entire closed-loop control system, the real-time and accuracy of data transmission are crucial. The RS485 bus has advantages such as long transmission distance and strong anti-interference ability, which can ensure the rapid and accurate transmission of temperature data to the temperature control unit.

[0074] 2. Implementation of Control Algorithm

[0075] The proportional-integral-derivative (PID) control algorithm is adopted. Let the deviation , and the control output be:

[0076]

[0077] Among them, is the proportional coefficient, is the integral coefficient, is the differential coefficient. In practical applications, by adjusting these three coefficients, the system can quickly and stably adjust the temperature to the set value. For example, in the drying stage, when the temperature deviation is large, the proportional term plays a major role in quickly reducing the deviation. If , the temperature deviation , then the output of the proportional term is 4. The integral term is used to eliminate the steady-state error. As time accumulates, the integral term continuously accumulates the deviation, prompting the system to gradually eliminate the steady-state error. The differential term Adjust the control quantity in advance according to the deviation change rate to prevent temperature overshoot. For example, when the temperature deviation change rate is 0.5 °C / min and Kd = 1, the output of the differential term is 0.5. By reasonably adjusting the PID parameters, the closed-loop control system can achieve efficient and stable temperature control at different production stages.

[0078] (2) Dynamically optimize control parameters

[0079] 1. Optimization implementation based on historical data

[0080] The system records the historical temperature data, control parameters, and product quality data during the production process. Through data analysis algorithms such as the least squares method, a relationship model between control parameters and product quality is established. Let the product quality index be Q and the control parameter be , and establish the model:

[0081] Among them, is the functional relationship between control parameters and product quality, and is the error term. For example, in the drying stage, the control parameter is the heating power, is the ventilation volume, and the product quality index Q is the product moisture content. By collecting a large amount of historical production data and using the least squares method for fitting, the functional relationship is obtained. By continuously updating the historical data and optimizing the model, the optimal control parameter combination is obtained. For example, when the target product moisture content is 6%, the optimal heating power is calculated to be 300 W and the ventilation volume is 15 m³ / min through the model.

[0082] 2. Online adaptive optimization implementation

[0083] During the production process, according to the real-time feedback data, the control parameters are adjusted using an adaptive algorithm. For example, in fuzzy logic control, according to the real-time ranges of temperature deviation and deviation change rate, the center value and standard deviation of the membership function are dynamically adjusted to make the fuzzy control rules more suitable for the actual production situation. Specifically, let the real-time range of temperature deviation be , and the real-time range of temperature deviation change rate be . When is relatively small, it indicates that the temperature is relatively stable. At this time, the standard deviation of the membership function can be appropriately reduced to make the division of fuzzy subsets more refined and improve the sensitivity of control. For example, originally, the center value of the membership function of the negative small (NS) fuzzy subset and the standard deviation . If the real-time temperature deviation range is reduced to , then can be adjusted to , so that when the temperature deviation is small, the control quantity can be adjusted more accurately according to the fuzzy rules.

[0084] For the reaction stage, if it is real-time monitored that the reaction rate fluctuates greatly, the coefficients related to the reaction heat can be dynamically adjusted according to the reaction stage model . By online monitoring the key parameters in the reaction system, such as the change in reactant concentration, the reaction heat release rate, etc., the adjustment value of is calculated using an adaptive algorithm. Suppose through a series of calculations and analyses, it is found that when the reaction rate increases, the actual temperature rise is larger than expected, indicating that the original value is too large, and it can be appropriately reduced to ensure the accuracy of temperature control. For example, the original , after adaptive adjustment, it can be reduced to .

[0085] In the drying stage, if the real-time feedback data of the product moisture content shows a large deviation from the target moisture content, the heating power influence coefficient and the ventilation volume influence coefficient can be adjusted simultaneously according to the drying stage model . For example, when the product moisture content is higher than the target value and it is real-time monitored that the temperature in the drying oven is low, can be appropriately increased to enhance the temperature-raising effect of the heating power, and at the same time, according to the influence of the ventilation volume on the temperature, is reasonably adjusted to optimize the drying process. Through this online adaptive optimization mechanism, the system can automatically and quickly adjust the control parameters according to the real-time changes in the production process, ensure that the temperature control is always in the best state, and effectively improve the stability of product quality and production efficiency.

[0086] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.​

Claims

1. A production temperature detection and control method for cold-soluble gelatin powder, characterized in that: The following steps are involved: Build a multi-point distributed sensor network, arrange temperature sensors at the top, middle and bottom of the reactor, the air inlet and outlet of the drying box, and at specific intervals on the material transmission pipeline to collect temperature data at different locations in real time; Using sensor data fusion technology, the data of each sensor is weighted averaged and fused to obtain a value reflecting the overall temperature state; Multi-model adaptive control is adopted to establish corresponding temperature control models for the raw material dissolution, reaction and drying stages of cold-soluble gelatin powder production, and the models are automatically switched according to the production status; the raw material dissolution stage model is ,in is the temperature at time t, , are the coefficients obtained by fitting the experimental data, Indicates the influence of heating power on temperature. Indicates the influence of stirring speed on temperature; the reaction stage model is , is a coefficient related to the heat of reaction, determined by reaction kinetics experiments, is the initial temperature of the reaction; the drying stage model is , , is the coefficient determined by experiment, Indicates the influence of heating power on temperature. Indicates the influence of ventilation volume on temperature. is the drying starting temperature; Combined with fuzzy logic control, the temperature deviation and the temperature deviation change rate are used as inputs, and the control output is determined by fuzzification, fuzzy rule reasoning and defuzzification operations in sequence; Build a closed-loop control system to feed back temperature detection data to the temperature control unit in real time, adjust the control amount according to the deviation between the feedback data and the set temperature, and output it to the actuator; Implement dynamic optimization of control parameters, establish a control parameter and product quality relationship model based on historical data, and use adaptive algorithms to adjust control parameters online according to real-time feedback data.

2. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: The sensors arranged at the top, middle and bottom of the reactor are used to monitor the temperature of the materials at different height levels. The air inlet sensor of the drying box is used to monitor the initial temperature of the hot air. The air outlet sensor of the drying box is used to monitor the temperature of the exhaust gas after the materials are dried. The sensors on the material transmission pipeline are arranged every 0.5 to 2 meters to monitor the temperature changes during the material transmission process.

3. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: In the sensor data fusion technology, the weight is determined according to the importance of the sensor position to the production process. The weight of the sensor at the bottom of the reactor is determined to be 0.4, the weight of the top sensor is determined to be 0.2, and the sensor weight is determined to be 0.4; the weight of the sensor at the air outlet of the drying box is determined to be 0.6, and the sensor weight is determined to be 0.4; the weight of the sensor near the key link of the material transmission pipeline is higher than that of other positions.

4. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: In the fuzzy logic control, the temperature deviation e and the temperature deviation change rate ec are used as input quantities. , temperature deviation change rate , using Gaussian membership function Fuzzy processing is performed, where is the central value of the membership function, is the standard deviation.

5. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: In the closed-loop control system, the temperature detection data is transmitted to the temperature control unit using a programmable logic controller via the RS485 bus, and the PLC calculates the control amount adjustment value according to the control algorithm and outputs it to the heating and cooling actuators.

6. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: In the dynamic optimization control parameters, the least square method is used based on historical data to establish the relationship model between the control parameters and product quality. ,in As product quality indicators, is the control parameter.

7. The method for detecting and controlling the production temperature of cold-soluble gelatin powder according to claim 1, characterized in that: During the online adaptive optimization, the central value and standard deviation of the membership function in the fuzzy logic control are dynamically adjusted according to the real-time range of the temperature deviation and the deviation change rate, and the corresponding model coefficients are dynamically adjusted according to the real-time parameters of the reaction stage and the drying stage.

Images