Intelligent multi-stage feeding control system for marine engineering

Abstract

The present application relates to the technical field of feeding control, and particularly relates to an intelligent multi-stage feeding control system for sea cucumber processing, which comprises a data acquisition module for obtaining initial sea cucumber elastic modulus; a multi-stage feeding analysis module for starting a first-stage feeding program based on the initial sea cucumber elastic modulus, and determining a second-stage initial liquid concentration gradient and a second-stage initial temperature gradient to start a second-stage feeding program; a multi-stage feeding control module for constructing a gentle feeding concentration gradient and predicting a first-stage concentration inside the sea cucumber, determining a second-stage process liquid concentration gradient and a second-stage process temperature gradient based on a real-time lag influence coefficient, and predicting a second-stage concentration inside the sea cucumber; and a compensation feeding analysis module for determining a compensation liquid concentration and a compensation time, determining a compensation correction time based on a real-time temperature, and predicting a third-stage concentration inside the sea cucumber. The present application improves the control precision of the multi-stage feeding process for sea cucumber processing.

Description

Technical Field

[0001] This invention relates to the field of material feeding control technology, and in particular to an intelligent multi-level material feeding control system for sea cucumber processing. Background Technology

[0002] With the development of modern processing towards intelligence, sea cucumber, as a product combining traditional tonic and industrial production, has placed higher demands on automation control in its feeding process. Traditional sea cucumber processing relies heavily on manual experience, using a one-time feeding method or simple multi-stage feeding, and simmering at high temperatures for a long time during stewing to allow the added ingredients to penetrate into the sea cucumber.

[0003] However, existing technologies generally fail to consider the unique physical characteristics of sea cucumbers as active colloidal materials during the feeding process. The sea cucumber body wall is primarily composed of collagen; when encountering sudden changes in high-concentration feed solutions, its surface proteins hinder the penetration of the feed solution into deeper layers. The mass transfer process of the added ingredients within the sea cucumber exhibits a significant time lag; that is, after an external concentration change, the internal concentration may require tens of minutes or even hours to respond. Existing technologies cannot dynamically compensate for this lag, leading to a deviation between the actual penetration effect and the expected target. This results in low utilization of effective ingredients during the feeding process and poor product quality consistency.

[0004] Chinese Patent Publication No. CN120078131A discloses a method for preparing medicinal stewed sea cucumber, belonging to the technical field of ready-to-eat sea cucumber processing, including the following steps: S1. Soaking the sea cucumber for later use; S2. Taking Chinese medicinal materials, adding water, and placing them in an ultrasonic water bath for extraction, with a frequency of 50-80kHz, a temperature of 65-70℃, and an extraction time of 25-30min, followed by filtration to obtain a medicinal soup; S3. Placing the soaked sea cucumber from step S1, the medicinal soup obtained in step S2, and seasonings in a vacuum tumbler for low-temperature vacuum flavoring, with the following parameters: total working time 140min, tumbling time 40min with 10min interval, vacuum degree 0.08MPa, frequency 10Hz, and temperature 20-25℃; S4. Filling, sterilizing, and cooling the sea cucumber and medicinal soup processed in step S3 to obtain medicinal stewed sea cucumber.

[0005] Therefore, the following problems exist in the existing technology: The problem of surface protein denaturation due to excessive concentration gradient during the initial feeding process was not considered, nor was the limitation of real-time control accuracy due to the lag in liquid penetration during feeding. Furthermore, the problem of inaccurate prediction of internal concentration in sea cucumbers due to the failure to update the diffusion coefficient through sampling detection was also not considered, resulting in low accuracy of real-time adjustment control based on predicted concentration during each feeding stage. Summary of the Invention

[0006] Therefore, the present invention provides an intelligent multi-stage feeding control system for sea cucumber processing, which overcomes the problems in the prior art that do not consider the surface protein denaturation caused by excessive concentration gradient during the initial feeding process, do not consider the characteristics of material liquid penetration lag during the feeding process which leads to limited real-time control accuracy, and do not consider updating the diffusion coefficient through sampling detection, which leads to distortion of the internal concentration prediction of sea cucumber and thus low accuracy of real-time adjustment control based on the predicted concentration during each feeding stage.

[0007] To achieve the above objectives, the present invention provides an intelligent multi-stage feeding control system for sea cucumber processing, comprising: The data acquisition module is used to obtain the initial elastic modulus of sea cucumbers, the real-time liquid concentration sequence in the first feeding process, and the surface impedance of each sea cucumber per unit time in the second feeding process. The multi-level feeding analysis module is used to determine the surface porosity activation rate of sea cucumbers based on the initial elastic modulus of sea cucumbers, to determine whether to start the first-level feeding program based on the surface porosity activation rate of sea cucumbers, and to start the second-level feeding program based on the internal first-level concentration of sea cucumbers determined by the deviation rate between the measured first-level concentration of the feed liquid and the estimated feed liquid concentration. A multi-stage feeding control module is used to construct a mild feeding concentration gradient, predict the primary concentration inside the sea cucumber based on the mild feeding concentration gradient and the primary diffusion coefficient, determine the secondary process liquid concentration gradient and the secondary process temperature gradient based on the real-time lag effect coefficient, and predict the secondary concentration inside the sea cucumber based on the secondary process liquid concentration gradient and the secondary diffusion coefficient. The compensation feeding analysis module is used to determine the compensation liquid concentration and compensation time based on the difference between the secondary concentration and the target tertiary concentration inside the sea cucumber, to correct the compensation time based on the real-time temperature in the compensation feeding program, and to predict the tertiary concentration inside the sea cucumber based on the compensation liquid concentration, the compensation correction time and the tertiary diffusion coefficient.

[0008] Furthermore, the multi-level feeding analysis module determines to start the first-level feeding procedure based on the fact that the proportion of sea cucumbers whose surface pore activation rate is greater than the preset sea cucumber surface pore activation rate threshold meets the first preset condition. The first preset condition is that the proportion of the ocean parameter quantity is greater than a preset proportion threshold.

[0009] Furthermore, the multi-stage feeding control module determines a preset first-stage target concentration sequence based on the target first-stage concentration change rate; wherein, the target first-stage concentration change rate is determined based on the average value of the sea cucumber surface pore activation rate that is greater than the preset sea cucumber surface pore activation rate threshold.

[0010] Furthermore, the multi-stage feeding control module constructs a gentle feeding concentration gradient based on the real-time first-stage concentration change rate; The real-time first-level concentration change rate is determined based on the target first-level concentration change rate and the real-time rate adjustment coefficient. The real-time rate adjustment coefficient is determined based on the ratio of the real-time difference between the real-time feed concentration sequence and the preset first-level target concentration sequence in the first-level feeding program to the preset difference threshold.

[0011] Furthermore, the multi-stage feeding analysis module determines the first-stage correction concentration inside the sea cucumber based on the product of the deviation rate between the measured first-stage concentration and the estimated first-stage concentration and the first-stage concentration inside the sea cucumber. Based on the difference between the first-stage correction concentration inside the sea cucumber and the second-stage target concentration inside the sea cucumber, the module determines the second-stage initial concentration gradient and the second-stage initial temperature gradient to initiate the second-stage feeding procedure.

[0012] Furthermore, the multi-level feeding control module determines the real-time lag influence coefficient based on the ratio of the maximum to the minimum surface resistivity of each sea cucumber per unit time in the secondary feeding program.

[0013] Furthermore, the multi-stage feeding control module determines the secondary process liquid concentration gradient and the secondary process temperature gradient based on the product of the real-time lag effect coefficient and the secondary initial liquid concentration gradient and the secondary initial temperature gradient.

[0014] Furthermore, the compensation feeding analysis module determines the compensation liquid concentration based on the difference between the secondary concentration inside the sea cucumber and the target tertiary concentration inside the sea cucumber, combined with the historical difference distribution, and determines the compensation time based on the compensation liquid concentration and the deviation between the historical target tertiary concentration inside the sea cucumber and the historical tertiary concentration inside the sea cucumber.

[0015] Furthermore, the compensation feeding analysis module corrects the compensation time based on the difference between the real-time temperature and the preset temperature to determine the compensation correction time.

[0016] Furthermore, it also includes a diffusion coefficient update module, which is used to adjust the diffusion coefficients at each level based on the residuals between the measured concentration inside the sea cucumber and the predicted concentrations at each level inside the sea cucumber; The measured concentration inside the sea cucumber is obtained by sampling and testing the sea cucumbers after each feeding procedure is completed using the data acquisition module. The concentration levels inside the sea cucumber include the primary concentration, the secondary concentration, and the tertiary concentration inside the sea cucumber. The diffusion coefficients at each level include the first-order diffusion coefficient, the second-order diffusion coefficient, and the third-order diffusion coefficient.

[0017] Compared with existing technologies, the advantages of this invention lie in its ability to achieve closed-loop control of the entire sea cucumber feeding process through a complete system architecture comprising a data acquisition module, a multi-level feeding analysis module, a multi-level feeding control module, a compensation feeding analysis module, and a diffusion coefficient update module. The data acquisition module simultaneously collects three different physical quantities: the sea cucumber's elastic modulus, real-time feed concentration, and surface electrical impedance. These represent the sea cucumber's physical state, process parameters, and individual permeability differences, providing a multi-dimensional data foundation for subsequent decision-making. The multi-level feeding analysis module uses the elastic modulus to calculate the activation rate and determine the start-up timing, ensuring that the feeding program is initiated based on a thorough assessment of the sea cucumber's state, avoiding surface damage caused by blind initiation. The multi-level feeding control module establishes a gentle concentration gradient through real-time difference adjustment, ensuring a smooth concentration change in the initial stage and protecting the sea cucumber's surface structure from damage. The multi-stage feeding analysis module then uses the deviation rate between the measured and estimated concentrations to correct internal concentrations, making the calculation of secondary parameters more reliable. The multi-stage feeding control module introduces a hysteresis coefficient to dynamically adjust the initial gradient of the secondary process, achieving real-time compensation for permeation hysteresis characteristics and making the actual permeation process closer to expectations. The compensation feeding analysis module determines compensation parameters based on the secondary results and corrects the compensation time based on real-time temperature, balancing target achievement rate and protection of heat-sensitive components in the final feeding stage. Through the coordinated work of each module, the control accuracy of the multi-stage feeding process in sea cucumber processing is effectively improved.

[0018] Furthermore, this invention determines whether to initiate the primary feeding procedure based on the percentage of sea cucumbers with a surface porosity greater than a preset threshold, thereby achieving quantitative assessment of material uniformity within a batch and start-up control based on the group state. By introducing start-up judgment based on actual state, the timing of the feeding procedure's initiation matches the actual state of the material.

[0019] Furthermore, this invention determines the target primary concentration change rate based on the average activation rate of qualified sea cucumbers, achieving dynamic matching between the feeding rate and the degree of pore opening in the sea cucumbers. The average activation rate calculation only includes individuals with qualified activation rates, making the average value more accurately reflect the state of the core population. The target primary concentration change rate is set based on the average activation rate, allowing the feeding rate to adaptively adjust according to the actual state of the material. This avoids surface burns caused by feeding batches with low activation rates too quickly at a fixed rate, and also avoids inefficiency waste caused by feeding batches with high activation rates too slowly. A preset primary target concentration sequence serves as a baseline for subsequent real-time tracking. Its overall slope is directly related to the state of the sea cucumbers, providing a quantitative basis and material adaptability for the gentleness of the entire primary feeding process, thus achieving refined control of the feeding process.

[0020] Furthermore, this invention introduces a real-time rate adjustment coefficient and constructs a real-time first-level concentration change rate based on the ratio of the real-time difference degree to a preset threshold, achieving precise tracking of the target concentration curve while maintaining a gentle concentration change. The real-time difference degree reflects the extent and direction of the actual concentration deviating from the target. Dividing it by the preset difference degree threshold yields a dimensionless ratio, which physically represents the multiple of the current deviation degree relative to the allowable deviation range. The real-time first-level concentration change rate is obtained by multiplying the target rate by one and summing it with this ratio, achieving a linear response of the rate to the degree of deviation; the greater the deviation, the greater the rate adjustment. This adjustment mechanism allows the actual concentration to smoothly return to the target curve. Throughout the feeding process, the actual concentration change fluctuates slightly around the target rate, with the fluctuation range limited within the threshold range, thus achieving a gentle control effect—that is, although the concentration change is dynamically adjusted, there are no abrupt changes or sudden jumps.

[0021] Furthermore, this invention corrects the estimated internal concentration by multiplying the deviation rate between the measured and estimated concentrations of the feed solution by the primary internal concentration of the sea cucumber, effectively compensating for prediction bias and the influence of external disturbances. The deviation rate between the measured and estimated concentrations is a dimensionless comprehensive indicator. Applying this error as a multiplicative factor to the predicted internal concentration is equivalent to assuming that the error sources affect both the feed solution concentration and the internal concentration in the same proportion. The corrected internal concentration retains the dynamic information of the prediction while incorporating the verification information from the actual measurement, making it closer to the true value. Using this corrected concentration as the benchmark for calculating the secondary feeding parameters ensures that subsequent concentration and temperature gradients are established on a relatively accurate basis, avoiding the propagation and amplification of primary errors to secondary levels.

[0022] Furthermore, this invention determines the real-time hysteresis influence coefficient based on the ratio of the maximum to minimum surface impedance of each sea cucumber per unit time, achieving online quantification of the hysteresis degree in the permeation process and providing a reliable basis for dynamic compensation. Electrical impedance reflects the ion concentration and moisture state of the sea cucumber surface, and is closely related to the depth of feed solution penetration. Measuring the impedance values ​​of multiple sea cucumbers at the same time, the ratio of the maximum to minimum value directly characterizes the dispersion of the permeation state among individuals. When all sea cucumbers absorb synchronously, the ratio is close to one; when there are large differences between individuals, the ratio is significantly greater than one. This difference is precisely the manifestation of permeation hysteresis at the group level, i.e., some sea cucumbers absorb quickly while others absorb slowly, leading to an inconsistent overall permeation process. The hysteresis influence coefficient is calculated based on this ratio, providing a clear physical quantitative indicator of the hysteresis degree. This coefficient is used for subsequent gradient adjustment, determining the compensation intensity based on the real-time perceived severity of the hysteresis.

[0023] Furthermore, this invention achieves dynamic compensation for the permeation drive by multiplying the real-time lag effect coefficient by the secondary initial feed concentration gradient and the secondary initial temperature gradient, enabling the actual gradient to adaptively increase according to the degree of lag. The secondary initial feed concentration gradient and the secondary initial temperature gradient are the optimal initial values ​​calculated based on the primary correction concentration, representing the driving force that should be used under ideal conditions without lag. The real-time lag effect coefficient quantifies the degree of lag in the actual process. The actual driving force is proportional to the degree of lag; the more severe the lag, the stronger the driving force. This adaptive adjustment mechanism effectively solves the problem of actual permeation lagging behind expectations due to lag, by enhancing the driving force to offset the time delay caused by lag, making the final permeation effect as close as possible to the ideal target.

[0024] Furthermore, this invention optimizes the compensation feed parameters by determining the compensation solution concentration based on the mapping relationship between the current concentration difference and the historical difference distribution, and by determining the compensation time based on the compensation concentration and historical deviation values. The concentration difference reflects the compliance status at the end of the secondary stage. Mapping this difference to the historical difference distribution draws on empirical values ​​from previous batches under similar conditions. The historical deviation value reflects the statistical deviation between the actual effect and the target at a given compensation concentration. Introducing this into the compensation time determination process is equivalent to calibrating the theoretical compensation time using past actual effects.

[0025] Furthermore, this invention achieves dynamic compensation for the volatilization loss of heat-sensitive liquid components by correcting the compensation time based on the difference between real-time temperature and preset temperature, thus improving the effectiveness of compensation feeding. The difference between real-time temperature and preset temperature directly quantifies the potential impact of the current environment on volatilization loss. Multiplying this difference by a temperature correction coefficient yields the adjustment ratio of the compensation time, making the compensation time positively correlated with the risk of volatilization loss. When the temperature is too high, the actual amount entering the sea cucumber at the same liquid concentration will decrease due to volatilization, requiring a longer treatment time to allow the sea cucumber more opportunities to absorb it; when the temperature is too low, the volatilization loss is small, and the time can be appropriately shortened without affecting the total absorption. The corrected compensation time enables the compensation feeding to maintain consistent effectiveness under different temperature conditions, improving adaptability to environmental changes.

[0026] Furthermore, this invention updates and adjusts the diffusion coefficients at each stage based on the residual between the measured and predicted concentrations within the sea cucumber, achieving batch-to-batch self-learning optimization of the control process and continuously improving system accuracy with accumulated production. The residual reflects the systematic deviation between the predicted and actual values. Using this residual to update the diffusion coefficients is equivalent to using the measured results of each production run to correct the parameters, gradually bringing the prediction closer to the actual mass transfer law. As the number of production batches increases, the deviation between the prediction and the measured results gradually decreases. Attached Figure Description

[0027] Figure 1 This is a module connection diagram of an intelligent multi-level feeding control system for sea cucumber processing according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the process of initiating the primary feeding procedure in an embodiment of the present invention. Figure 3 This is a flowchart illustrating the process of determining a preset primary target concentration sequence in an embodiment of the present invention. Figure 4 A flowchart illustrating the workflow for constructing a gentle feed concentration gradient in an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments. It should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0029] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0030] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0031] Please see Figure 1 The diagram shown is a module connection diagram of an intelligent multi-stage feeding control system for sea cucumber processing according to an embodiment of the present invention. The intelligent multi-stage feeding control system for sea cucumber processing according to an embodiment of the present invention includes: The data acquisition module is used to obtain the initial elastic modulus of sea cucumbers, the real-time liquid concentration sequence in the first feeding process, and the surface impedance of each sea cucumber per unit time in the second feeding process. A multi-level feeding analysis module, which is connected to the data acquisition module, is used to determine the surface porosity activation rate of the sea cucumber based on the initial elastic modulus of the sea cucumber, to determine whether to start the first-level feeding program based on the surface porosity activation rate of the sea cucumber, and to start the second-level feeding program based on the internal first-level concentration of the sea cucumber determined by the deviation rate between the measured first-level concentration of the feed liquid and the estimated feed liquid concentration. A multi-stage feeding control module, which is connected to the data acquisition module and the multi-stage feeding analysis module, is used to construct a mild feeding concentration gradient, predict the primary concentration inside the sea cucumber based on the mild feeding concentration gradient and the primary diffusion coefficient, determine the secondary process liquid concentration gradient and the secondary process temperature gradient based on the real-time lag effect coefficient, and predict the secondary concentration inside the sea cucumber based on the secondary process liquid concentration gradient and the secondary diffusion coefficient. The compensation feeding analysis module, which is connected to the data acquisition module, the multi-level feeding analysis module, and the multi-level feeding control module, is used to determine the compensation liquid concentration and compensation time based on the difference between the secondary concentration and the tertiary target concentration inside the sea cucumber, correct the compensation time based on the real-time temperature in the compensation feeding program, and predict the tertiary concentration inside the sea cucumber based on the compensation liquid concentration, the compensation correction time, and the tertiary diffusion coefficient.

[0032] In this embodiment, the data acquisition module is responsible for acquiring key data in the production process in real time. Before the first feeding procedure is started, the initial elastic modulus of the sea cucumber is collected by the texture sensor. This parameter reflects the tissue hardness of the sea cucumber before any treatment. In the first feeding procedure, the concentration value of the liquid in the reactor is continuously collected by the online concentration meter at a frequency of once per second to form a real-time liquid concentration sequence. In the second feeding procedure, the impedance value of the surface of each sea cucumber is collected every minute by the array impedance probe. No less than fifty individual sea cucumbers are tested in each sampling cycle.

[0033] Please see Figure 2 As shown, it is a flowchart of the process of determining the start of the first-level feeding procedure in an embodiment of the present invention.

[0034] Specifically, the multi-level feeding analysis module determines to start the first-level feeding procedure based on the fact that the proportion of sea cucumbers whose surface pore activation rate is greater than the preset sea cucumber surface pore activation rate threshold meets the first preset condition. The first preset condition is that the proportion of the ocean parameter quantity is greater than a preset proportion threshold.

[0035] In this embodiment, the data acquisition module measures the elastic modulus of at least fifty individual sea cucumbers in the reactor. Each sea cucumber is measured three times, and the average value is taken as its current elastic modulus. After receiving this data, the multi-stage feeding analysis module calculates the surface porosity activation rate for each individual sea cucumber using the formula: the difference between the current elastic modulus and the initial elastic modulus divided by the initial elastic modulus. A preset threshold for the sea cucumber surface porosity activation rate is set at 15%. When the activation rate reaches 15%, it indicates that the collagen network on the sea cucumber surface has sufficiently relaxed, and the pores are open enough to support subsequent feed penetration. The module further counts the number of individuals with an activation rate greater than 15% among all tested sea cucumbers and calculates the ratio of this number to the total number of tested individuals to obtain the percentage of sea cucumbers meeting the standards. A preset threshold for this percentage is set at 90%. When the percentage of sea cucumbers meeting the standards exceeds 90%, it indicates that the vast majority of sea cucumbers are in a suitable state for receiving feed. If the percentage of sea cucumbers meeting the standards is less than or equal to 90%, the system will maintain monitoring until the percentage meets the requirements. The multi-level feeding analysis module only issues a trigger signal to initiate the first-level feeding procedure when the proportion of sea cucumbers meeting the standards exceeds 90%. This process avoids surface damage caused by forcibly feeding when most sea cucumbers are not yet ready, ensuring that the feeding procedure is initiated based on sufficient material preparation.

[0036] Please see Figure 3 As shown, it is a flowchart of the process for determining the preset primary target concentration sequence in an embodiment of the present invention.

[0037] Specifically, the multi-stage feeding control module determines a preset first-stage target concentration sequence based on the target first-stage concentration change rate; wherein, the target first-stage concentration change rate is determined based on the average value of the sea cucumber surface pore activation rate that is greater than the preset sea cucumber surface pore activation rate threshold.

[0038] In this embodiment, after the conditions for starting the primary feeding procedure are met, the multi-level feeding control module obtains the specific activation rate values ​​of all sea cucumber individuals with an activation rate greater than 15% from the multi-level feeding analysis module, and calculates the arithmetic mean of these values ​​as the average activation rate. This average activation rate reflects the overall state of the sea cucumbers in the current batch that are ready to receive feeding. A higher activation rate indicates a better degree of pore opening and a stronger tolerance to concentration changes. The target primary concentration change rate is determined by obtaining the success rate used by historical batches that match the current average activation rate within a historical period. Specifically, the ten most recent successful batches whose absolute difference from the current average activation rate does not exceed 2% are retrieved from the historical data, and the target primary concentration change rates used by these batches are extracted. Their arithmetic mean is calculated as the target primary concentration change rate for the current batch, ensuring that the concentration change rate matches the pore state of the sea cucumbers. After determining the target primary concentration change rate, the multi-level feeding control module uses this rate as the slope and starts from an initial feed concentration of zero to generate a target concentration curve that increases linearly with time, i.e., the preset primary target concentration sequence. The concentration value at any time t in this sequence is equal to the rate of change of the target primary concentration multiplied by time t. It can be understood that the preset primary target concentration sequence can adaptively adjust according to the actual physical state of the sea cucumbers. Batches with high activation rates can withstand a slightly faster feeding rate, while batches with low activation rates use a more conservative rate, achieving a precise match between process parameters and material characteristics.

[0039] Please see Figure 4 The diagram shown illustrates the workflow for constructing a mild feed concentration gradient according to an embodiment of the present invention.

[0040] Specifically, the multi-level feeding control module constructs a gentle feeding concentration gradient based on the real-time first-level concentration change rate; The real-time first-level concentration change rate is determined based on the target first-level concentration change rate and the real-time rate adjustment coefficient. The real-time rate adjustment coefficient is determined based on the ratio of the real-time difference between the real-time feed concentration sequence and the preset first-level target concentration sequence in the first-level feeding program to the preset difference threshold.

[0041] In this embodiment, during the primary feeding process, the multi-stage feeding control module acquires the real-time liquid concentration value from the data acquisition module once per second, compares this value with the concentration value at the corresponding moment in the preset primary target concentration sequence, and calculates the difference between the two, i.e., the real-time difference degree. The preset difference degree threshold is set to 1% of the target concentration value. This value represents the allowable concentration deviation range. When the real-time difference degree is within this threshold, the system considers the tracking effect to be good and no significant adjustment is required. The real-time rate adjustment coefficient is calculated by dividing the real-time difference degree by the preset difference degree threshold to obtain the ratio between the two. Since the real-time difference degree may be positive (actual concentration is lower than the target) or negative (actual concentration is higher than the target), this ratio also carries a positive or negative sign, reflecting the direction and degree of deviation. The real-time primary concentration change rate is determined by multiplying the target primary concentration change rate by this ratio and adding one times the target primary concentration change rate. That is, when the actual concentration is low, the ratio increases the rate to catch up with the target; when the actual concentration is high, the ratio decreases the rate to wait for the concentration to fall back. Through this continuous adjustment mechanism, the rate of change of the actual concentration is dynamically fine-tuned around the target rate, ensuring tracking accuracy, avoiding sudden rate changes, and achieving gentle feeding. Throughout the process, the trajectory of the actual concentration change smoothly fluctuates around the target curve, with the fluctuation amplitude controlled within a preset threshold range, thus constructing a gentle feeding concentration gradient.

[0042] In this embodiment, the primary concentration inside the sea cucumber is predicted based on the mild feeding concentration gradient and the primary diffusion coefficient. Specifically, the real-time concentration value of the liquid in the vessel after the completion of the primary feeding procedure is obtained, which is measured by an online concentration meter. The expected concentration in the vessel is calculated based on the currently established mild feeding concentration gradient. The primary diffusion coefficient is obtained from the diffusion coefficient update module after the completion of the previous batch feeding procedure, and it characterizes the rate constant of the diffusion of the liquid from the solution to the surface of the sea cucumber under the primary feeding conditions. The calculation process for predicting the primary concentration inside the sea cucumber is to multiply the primary diffusion coefficient by the integral of the effective driving force over time during the entire primary feeding process, where the effective driving force is the difference between the liquid concentration in the vessel and the estimated internal concentration within one minute. The estimated internal concentration is set to zero at the initial moment, i.e., when the primary feeding procedure starts. This prediction provides a theoretical basis for monitoring the primary feeding process and subsequent secondary correction.

[0043] Specifically, the multi-stage feeding analysis module determines the first-stage correction concentration inside the sea cucumber based on the product of the deviation rate between the measured first-stage concentration and the estimated first-stage concentration and the first-stage concentration inside the sea cucumber. It then determines the second-stage initial concentration gradient and the second-stage initial temperature gradient based on the difference between the first-stage correction concentration inside the sea cucumber and the second-stage target concentration inside the sea cucumber, thereby initiating the second-stage feeding procedure.

[0044] In this embodiment, at the end of the first-stage feeding procedure, the multi-stage feeding analysis module first obtains the measured first-stage concentration value of the feed liquid after stabilization from the online concentration meter, which reflects the final actual concentration in the reactor. The multi-stage feeding analysis module calculates the theoretically estimated concentration based on the cumulative feeding amount and initial water volume during the first-stage feeding process. The cumulative feeding amount is obtained by integrating the flow meter of the feeding pump, and the initial water volume is the water volume retained in the reactor before the first-stage feeding procedure begins. The sum of these two is the total liquid volume. The theoretically estimated concentration is equal to the product of the cumulative feeding amount and the concentration in the storage tank, divided by the total liquid volume. The difference between the measured concentration and the estimated concentration is calculated and then divided by the estimated concentration to obtain the dimensionless deviation rate. This error reflects the deviation between the actual process and the ideal process. The multi-stage feeding analysis module obtains the predicted final value of the first-stage concentration inside the sea cucumber during the first-stage feeding process from the multi-stage feeding control module. The calculation method for the corrected first-stage concentration inside the sea cucumber is: multiply the first-stage concentration inside the sea cucumber by the deviation rate and add one times the first-stage concentration inside the sea cucumber. Understandably, the deviation between the measured feed concentration and the theoretical value reflects the overall error of the entire primary feeding system. Proportioning this error to the internal concentration prediction value can compensate for the impact of external disturbances, making the internal concentration basis for the secondary feeding decision-making stage closer to the actual situation. The corrected concentration serves as the benchmark for calculating the secondary feeding parameters, ensuring that the determination of the subsequent secondary initial gradient is based on an accurate evaluation of the primary results.

[0045] In this embodiment, the initial concentration gradient and initial temperature gradient of the secondary feed solution are determined based on the difference between the primary correction concentration and the secondary target concentration inside the sea cucumber. Specifically, the multi-stage feeding analysis module calculates the difference between the secondary target concentration and the primary correction concentration inside the sea cucumber, denoted as the concentration difference. This difference reflects the concentration gap that needs to be filled between the end of the primary stage and the secondary target. The specific calculation method for the initial concentration gradient of the secondary feed solution is: the primary correction concentration inside the sea cucumber plus the driving force coefficient multiplied by the concentration difference. The driving force coefficient typically ranges from 1.5 to 2.5, and in this embodiment, it is set to 2.0. The specific calculation method for the initial temperature gradient of the secondary feed solution is: the temperature inside the vessel at the end of the primary feeding stage plus the temperature compensation coefficient multiplied by the concentration difference. The temperature compensation coefficient typically ranges from 3.0 to 5.0, and in this embodiment, it is set to 4.0 to ensure that the heating process matches the concentration gap. The initial concentration gradient and initial temperature gradient of the secondary feed solution determined in this way provide an initial driving force for the secondary feeding program that matches the primary results.

[0046] Specifically, the multi-level feeding control module determines the real-time lag influence coefficient based on the ratio of the maximum to the minimum surface resistivity of each sea cucumber per unit time in the two-level feeding program.

[0047] In this embodiment, after the secondary feeding procedure is initiated, the data acquisition module measures the surface impedance of at least fifty individual sea cucumbers in the reactor using an array of impedance probes once per minute, recording the impedance values ​​of all detection points within each measurement cycle. These impedance values ​​reflect the ion concentration and moisture state of the sea cucumber surface and are highly correlated with the degree of material penetration. After each unit time, i.e., one-minute measurement cycle, the multi-stage feeding control module extracts the maximum and minimum values ​​of all impedance values ​​within that cycle and calculates their ratio. The ratio increases accordingly as the difference between individuals increases. The real-time hysteresis effect coefficient is calculated based on this ratio, by multiplying the sum of 1 and the sensitivity coefficient by the difference between the ratio and 1. The sensitivity coefficient is preset to 0.5 to adjust the intensity of hysteresis compensation. When there are significant differences in osmotic states among individual sea cucumbers, it indicates that the mass transfer process is affected by material inhomogeneity, with some sea cucumbers absorbing nutrients quickly while others absorb them slowly. This difference is an external manifestation of osmotic lag, requiring a stronger driving force to propel the overall osmotic process. Conversely, when the impedance values ​​of all sea cucumbers are similar, it indicates good synchronicity in the osmotic process and a smaller lag effect. Therefore, the lag effect coefficient dynamically quantifies the degree of lag through the ratio of impedance extreme values; the larger the ratio, the larger the lag coefficient, and the stronger the subsequent compensation force.

[0048] Specifically, the multi-stage feeding control module determines the secondary process liquid concentration gradient and the secondary process temperature gradient based on the product of the real-time lag effect coefficient and the secondary initial liquid concentration gradient and the secondary initial temperature gradient.

[0049] In this embodiment, during the secondary feeding process, the multi-level feeding control module obtains the initial secondary feed liquid concentration gradient and initial secondary temperature gradient from the multi-level feeding analysis module, as well as the current real-time lag influence coefficient. The secondary process feed liquid concentration gradient is determined by multiplying the initial secondary feed liquid concentration gradient by the real-time lag influence coefficient to obtain the actual concentration gradient at the current moment. It can be understood that the actual concentration gradient is proportionally amplified from the initial value based on the lag influence coefficient, driving penetration with a stronger concentration difference. The secondary process temperature gradient is determined in the same way: the initial secondary temperature gradient is multiplied by the real-time lag influence coefficient to obtain the actual heating rate at the current moment. The more severe the lag, the faster the heating rate, enhancing penetration through temperature coordination. Through the lag adjustment mechanism, the driving force of the secondary feeding process can dynamically change according to the real-time perceived lag level. When slow penetration is detected, the driving force is automatically strengthened, achieving adaptive compensation for fluctuations in unknown materials.

[0050] In this embodiment, the secondary concentration inside the sea cucumber is predicted based on the secondary process feed concentration gradient and the secondary diffusion coefficient. Specifically, after the secondary feeding procedure is completed, the real-time feed concentration sequence in the vessel, measured by a concentration meter, is obtained throughout the secondary feeding process, along with the secondary process feed concentration gradient adjusted according to the real-time lag effect coefficient. The secondary diffusion coefficient is obtained from the diffusion coefficient update module after the previous batch feeding procedure. The calculation of the predicted secondary concentration inside the sea cucumber is as follows: the secondary diffusion coefficient is multiplied by the integral of the effective driving force over time throughout the secondary feeding process. The driving force is taken as the difference between the concentration in the vessel and the internal concentration at the start and end of the time period. The internal concentration at the start of each time period is passed from the calculation result of the previous time period. The internal concentration at the initial moment, i.e., when the secondary feeding procedure starts, is the corrected primary concentration inside the sea cucumber after the primary feeding procedure is completed. The system obtains the predicted value of the secondary concentration inside the sea cucumber at the end of the secondary feeding procedure. This predicted value is used by the compensation feeding analysis module to determine the compensation feeding parameters.

[0051] Specifically, the compensation feeding analysis module determines the compensation liquid concentration based on the difference between the secondary concentration inside the sea cucumber and the target tertiary concentration inside the sea cucumber, combined with the historical difference distribution. The compensation time is determined based on the compensation liquid concentration and the deviation between the historical target tertiary concentration inside the sea cucumber and the historical tertiary concentration inside the sea cucumber.

[0052] In this embodiment, after the secondary feeding procedure is completed, the compensation feeding analysis module obtains the predicted secondary concentration inside the sea cucumber from the multi-level feeding control module, and obtains the target tertiary concentration inside the sea cucumber, calculating the difference between the two, which is recorded as the concentration difference. The module retrieves the feeding records of the most recent fifty batches from the historical database, analyzes the correspondence between the concentration difference at the end of the secondary feeding in these batches and the final compensation liquid concentration used, forming a difference distribution curve. The compensation liquid concentration of the current batch is determined by mapping the current concentration difference to the mean of the historical compensation concentrations corresponding to the same difference in the historical difference distribution curve. After determining the compensation liquid concentration, the module further analyzes the distribution of deviation values ​​between the target tertiary concentration inside the sea cucumber and the final measured tertiary internal concentration under the same compensation liquid concentration conditions in historical data. This deviation value reflects the gap between the actual effect achieved and the target under a given compensation concentration. The compensation time is determined based on the compensation time corresponding to the minimum historical deviation value corresponding to the currently determined compensation liquid concentration. Through this nonlinear mapping method based on historical data distribution, the determination of compensation feeding fully utilizes the accumulated experience of past production.

[0053] Specifically, the compensation feeding analysis module corrects the compensation time based on the difference between the real-time temperature and the preset temperature to determine the compensation correction time.

[0054] In this embodiment, at the start of the compensation feeding procedure, the compensation feeding analysis module reads the real-time temperature value inside the reactor from the data acquisition module. This value is measured by the temperature sensor inside the reactor. The preset temperature is set to 25°C, which is determined based on the critical evaporation temperature of the heat-sensitive liquid. When the temperature is below this value, the evaporation loss is negligible; when the temperature is above this value, the evaporation rate increases significantly. The module calculates the difference between the real-time temperature and the preset temperature. The compensation correction time is determined by multiplying the compensation time by the temperature correction coefficient and the difference, and then adding one time to the product to determine the compensation correction time. Temperature directly affects the retention rate of volatile components. When the temperature is too high, the action time needs to be extended at the same liquid concentration to compensate for evaporation loss and ensure that the actual amount of sea cucumber entering the reactor reaches the expected level. When the temperature is too low, the evaporation loss is small, and the time can be appropriately shortened to improve efficiency.

[0055] In this embodiment, the tertiary concentration inside the sea cucumber is predicted based on the compensation solution concentration, the compensation correction time, and the tertiary diffusion coefficient. Specifically, after the compensation feeding procedure is completed, the actual compensation solution concentration and compensation correction time executed during the compensation feeding process are obtained. The tertiary diffusion coefficient is obtained from the diffusion coefficient update module after the completion of the previous batch feeding procedure. The calculation process for predicting the tertiary concentration inside the sea cucumber is as follows: the secondary concentration inside the sea cucumber plus the increase in the internal concentration of the sea cucumber. The increase in the internal concentration of the sea cucumber is equal to the tertiary diffusion coefficient multiplied by the driving force and then multiplied by the compensation correction time. The driving force, i.e., the difference between the compensation solution concentration and the internal concentration at the end of the secondary stage, remains constant throughout the compensation correction time. This predicted value serves as the estimated internal concentration of the finished product for this batch, used for quality prediction and batch recording, and together with the measured concentration obtained from subsequent sampling and testing, serves as the input to the diffusion coefficient update module.

[0056] Specifically, it also includes a diffusion coefficient update module, which is connected to the data acquisition module, the multi-level feeding analysis module, the multi-level feeding control module and the compensation feeding analysis module, and is used to adjust the diffusion coefficients at each level based on the residual between the measured concentration inside the sea cucumber and the predicted concentration at each level inside the sea cucumber; The measured concentration inside the sea cucumber is obtained by sampling and testing the sea cucumbers after each feeding procedure is completed using the data acquisition module. The concentration levels inside the sea cucumber include the primary concentration, the secondary concentration, and the tertiary concentration inside the sea cucumber. The diffusion coefficients at each level include the first-order diffusion coefficient, the second-order diffusion coefficient, and the third-order diffusion coefficient.

[0057] In this embodiment, after each batch of production, at least five sea cucumbers are randomly selected from the finished product. Samples are taken at the time points corresponding to the end of primary feeding, secondary feeding, and compensation feeding, respectively, to detect the actual concentration of the feed solution inside the sea cucumbers, obtaining the measured concentrations inside the sea cucumbers at each level. The diffusion coefficient update module obtains these measured values ​​from the data acquisition module, and simultaneously obtains the predicted primary, secondary, and tertiary concentrations inside the sea cucumbers for the corresponding batch. For each level, the difference between the measured concentration and the predicted concentration, i.e., the residual, is calculated, reflecting the systematic bias of the prediction. For updating the primary diffusion coefficient, the module analyzes the concentration gradient and time during the primary feeding stage, and combines the residual to back-calculate a diffusion coefficient value that better matches the actual observation through an optimization algorithm. Specifically, by minimizing the sum of squared prediction errors for all sample points, the gradient descent method is used to iteratively correct the primary diffusion coefficient. The updates for the secondary and tertiary diffusion coefficients are performed using the same method. The updated diffusion coefficients at each level are stored in the system as the initial values ​​for the next batch of production. For example, if multiple consecutive batches show that the predicted values ​​are generally low, the system will automatically increase the diffusion coefficient; otherwise, it will decrease it. Through this batch-to-batch closed-loop learning mechanism, the diffusion coefficient continuously approaches the true value under the characteristics of the production line and the raw material, so that the prediction accuracy continues to improve with the increase of production batches, and ultimately achieves self-optimization of control parameters.

[0058] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An intelligent multi-stage feeding control system for sea cucumber processing, characterized in that, include: The data acquisition module is used to obtain the initial elastic modulus of sea cucumbers, the real-time liquid concentration sequence in the first feeding process, and the surface impedance of each sea cucumber per unit time in the second feeding process. The multi-level feeding analysis module is used to determine the surface porosity activation rate of sea cucumbers based on the initial elastic modulus of sea cucumbers, to determine whether to start the first-level feeding program based on the surface porosity activation rate of sea cucumbers, and to start the second-level feeding program based on the internal first-level concentration of sea cucumbers determined by the deviation rate between the measured first-level concentration of the feed liquid and the estimated feed liquid concentration. A multi-stage feeding control module is used to construct a mild feeding concentration gradient, predict the primary concentration inside the sea cucumber based on the mild feeding concentration gradient and the primary diffusion coefficient, determine the secondary process liquid concentration gradient and the secondary process temperature gradient based on the real-time lag effect coefficient, and predict the secondary concentration inside the sea cucumber based on the secondary process liquid concentration gradient and the secondary diffusion coefficient. The compensation feeding analysis module is used to determine the compensation liquid concentration and compensation time based on the difference between the secondary concentration and the target tertiary concentration inside the sea cucumber, to correct the compensation time based on the real-time temperature in the compensation feeding program, and to predict the tertiary concentration inside the sea cucumber based on the compensation liquid concentration, the compensation correction time and the tertiary diffusion coefficient.

2. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 1, characterized in that, The multi-level feeding analysis module determines to start the first-level feeding procedure based on the fact that the proportion of sea cucumbers whose surface pore activation rate is greater than the preset sea cucumber surface pore activation rate threshold meets the first preset condition. The first preset condition is that the proportion of the ocean parameter quantity is greater than a preset proportion threshold.

3. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 2, characterized in that, The multi-stage feeding control module determines a preset primary target concentration sequence based on the target primary concentration change rate; The target primary concentration change rate is determined based on the average value of the sea cucumber surface pore activation rate that is greater than the preset sea cucumber surface pore activation rate threshold.

4. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 3, characterized in that, The multi-stage feeding control module constructs a gentle feeding concentration gradient based on the real-time first-stage concentration change rate. The real-time first-level concentration change rate is determined based on the target first-level concentration change rate and the real-time rate adjustment coefficient. The real-time rate adjustment coefficient is determined based on the ratio of the real-time difference between the real-time feed concentration sequence and the preset first-level target concentration sequence in the first-level feeding program to the preset difference threshold.

5. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 4, characterized in that, The multi-stage feeding analysis module determines the first-stage correction concentration inside the sea cucumber based on the product of the deviation rate between the measured first-stage concentration and the estimated first-stage concentration and the first-stage concentration inside the sea cucumber. It then determines the second-stage initial concentration gradient and the second-stage initial temperature gradient based on the difference between the first-stage correction concentration inside the sea cucumber and the second-stage target concentration inside the sea cucumber, thereby initiating the second-stage feeding procedure.

6. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 5, characterized in that, The multi-level feeding control module determines the real-time lag effect coefficient based on the ratio of the maximum to the minimum surface resistivity of each sea cucumber per unit time in the two-level feeding program.

7. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 6, characterized in that, The multi-stage feeding control module determines the secondary process liquid concentration gradient and the secondary process temperature gradient based on the product of the real-time lag effect coefficient and the secondary initial liquid concentration gradient and the secondary initial temperature gradient.

8. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 7, characterized in that, The compensation feeding analysis module determines the compensation solution concentration based on the difference between the secondary concentration inside the sea cucumber and the target tertiary concentration inside the sea cucumber, combined with the historical difference distribution. It then determines the compensation time based on the compensation solution concentration and the deviation between the historical target tertiary concentration inside the sea cucumber and the historical tertiary concentration inside the sea cucumber.

9. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 8, characterized in that, The compensation feeding analysis module corrects the compensation time based on the difference between the real-time temperature and the preset temperature to determine the compensation correction time.

10. The intelligent multi-stage feeding control system for sea cucumber processing according to claim 9, characterized in that, It also includes, The diffusion coefficient update module is used to adjust the diffusion coefficients at each level based on the residuals between the measured concentration and the predicted concentration at each level inside the sea cucumber. The measured concentration inside the sea cucumber is obtained by sampling and testing the sea cucumbers after each feeding procedure is completed using the data acquisition module. The concentration levels inside the sea cucumber include the primary concentration, the secondary concentration, and the tertiary concentration inside the sea cucumber. The diffusion coefficients at each level include the first-order diffusion coefficient, the second-order diffusion coefficient, and the third-order diffusion coefficient.

Images