A method and system for dynamically controlling moisture content in a feed pelleting process

Abstract

This invention discloses a method and system for dynamic control of moisture content in feed pelleting, relating to the field of automatic control technology. The method includes: collecting material moisture and equipment operation data through distributed sensing nodes and wirelessly transmitting the data to a remote control platform; establishing a spatial propagation function of material moisture along the pelleting process on the platform to describe the moisture evolution of the material from upstream to downstream; predicting future moisture values ​​at downstream nodes based on this function, generating advance control commands, measuring communication delays, and compensating for command delays by calculating moisture changes within the delay using the spatial propagation function; further calculating the moisture coupling effect after execution at downstream nodes, generating collaborative control commands for adjacent nodes, and achieving precise dynamic control of moisture in the pelleting process.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, specifically to a method and system for dynamic control of moisture content in the feed pelleting process. Background Technology

[0002] Feed pelleting is a crucial production step in forming pellets from powdered raw materials through processes such as conditioning and pressing. The moisture content of the material directly affects the pellet's formability and strength. In existing technologies, most production lines use moisture detection devices installed at key nodes in the conditioner or pellet mill to collect moisture data in real time. Based on the deviation between the detected values ​​and the preset target moisture content, closed-loop control is achieved by adjusting the steam or liquid injection volume. This method is a typical feedback control approach, relying primarily on single-point data and localized adjustments. It fails to effectively reflect the dynamic moisture changes of the material throughout the entire pelleting process and cannot predict the moisture status at downstream nodes.

[0003] As granulation production lines become longer and more automated, significant spatial propagation and time lag effects occur during material transport, while coupling effects exist between operations at different nodes. Traditional control methods cannot adequately account for communication delays and inter-node coupling, easily leading to problems such as adjustment lag, over-adjustment, or increased moisture fluctuations. This makes single-point feedback control insufficient to meet the high-precision, high-stability moisture control requirements of granulation.

[0004] Therefore, the pelleting process urgently needs an automated control method based on distributed moisture data. By predicting future moisture changes at downstream nodes and combining communication delay compensation with multi-node coupled collaborative control, dynamic and precise control of moisture in the pelleting process can be achieved. Optimizing pelleting moisture control through automation, predictive methods, and multi-node collaboration can improve production process stability and pellet feed quality. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for dynamic control of moisture content in the feed pelleting process, aiming to solve at least one of the technical problems existing in the prior art.

[0006] The technical solution of this invention is: a method for dynamically controlling the moisture content in the feed pelleting process, comprising the following steps: Material moisture data and equipment operation data are collected through distributed sensor nodes and transmitted to a remote control platform via a wireless communication network. Based on material moisture data at different spatial locations, a spatial propagation function for material moisture along the granulation process is established on a remote control platform. This spatial propagation function describes the evolution of moisture values ​​as the material flows from upstream to downstream nodes. Based on the spatial propagation function and the material moisture data of the upstream node at the current moment, calculate the predicted moisture value of the material when it arrives at the downstream node at a future moment, and generate an advance control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value. The communication delay time from the generation of the advanced control command to its issuance to the downstream node for execution is measured. The change in material moisture within the communication delay time is calculated based on the spatial propagation function. The control quantity of the advanced control command is compensated and corrected, and a delay compensation control signal is generated and issued to the downstream node for execution. The coupling effect of downstream nodes executing delay compensation control signals on the moisture content of materials in adjacent nodes is calculated based on the spatial propagation function, and coordinated control instructions for adjacent nodes are generated based on the coupling effect.

[0007] Based on material moisture data from different spatial location nodes on the remote control platform, a spatial propagation function for material moisture along the granulation process is established, including: Obtain material moisture data and location coordinates of nodes at different spatial locations in the granulation process, sort the nodes at different spatial locations according to the material flow direction based on the location coordinates, and generate a node sequence; Select two adjacent spatial location nodes from the node sequence to form a node pair, and extract the material moisture data and location coordinates of the two spatial location nodes in the node pair; The material moisture difference is calculated based on the material moisture data of the two spatial nodes in the node pair. The position coordinate difference is calculated based on the position coordinates of the two spatial nodes in the node pair. The unit distance moisture change rate of the node pair is calculated based on the material moisture difference and the position coordinate difference. The unit distance moisture change rate of all node pairs in the node sequence is associated with the corresponding location coordinates to construct a discrete data set in which the moisture change rate varies with the location coordinates. The discrete data set is fitted with a function to generate a continuous function of the rate of change of moisture with respect to the location coordinates. The continuous function is then integrated along the location coordinates to generate a spatial propagation function.

[0008] The discrete data set is fitted with a function to generate a continuous function of the rate of change of moisture with respect to location coordinates. This continuous function is then integrated along the location coordinates to generate a spatial propagation function, including: Identify node pairs in a discrete data set that exhibit abnormal fluctuations in the rate of change of moisture per unit distance, extract the rate of change of moisture per unit distance from adjacent normal node pairs of the abnormal node pairs, smooth them out, replace the rate of change of moisture per unit distance from the abnormal node pairs, and generate a corrected discrete data set. Extract the position coordinates of each node pair in the corrected discrete data set and the corresponding rate of change of moisture per unit distance. Establish a polynomial fitting equation with position coordinates as the independent variable and rate of change of moisture per unit distance as the dependent variable. Generate a continuous function of rate of change of moisture per unit distance with respect to position coordinates by performing function fitting on the corrected discrete data set. Obtain the initial moisture value of the material at the starting point of the granulation process, perform an integral operation on the continuous function along the position coordinates from the starting point, and superimpose the integral result with the initial moisture value of the material to generate a spatial propagation function.

[0009] Based on the spatial propagation function and the material moisture data of the upstream node at the current moment, the predicted moisture value of the material when it arrives at the downstream node at a future moment is calculated. Then, based on the deviation between the predicted moisture value and the preset target moisture value, advanced control instructions for the downstream node are generated, including: Obtain the material moisture data and location coordinates of the upstream node at the current moment, and obtain the location coordinates and target moisture value of the downstream node; The spatial propagation function is input with the coordinates of the upstream node as the starting position, the current moisture content of the upstream node as the starting moisture value, and the coordinates of the downstream node as the ending position. The moisture evolution value of the material flowing from the upstream node to the downstream node is calculated by the spatial propagation function, and is used as the predicted moisture value of the material when it arrives at the downstream node in the future. The difference between the predicted moisture value and the target moisture value is calculated to obtain the deviation. The direction of moisture regulation at downstream nodes is determined based on the positive or negative direction of the deviation, including increasing or decreasing moisture. Based on the direction of moisture regulation, extract the controllable process parameters corresponding to the downstream nodes, and determine the adjustment amount of the process parameters based on the value of the deviation and the degree of influence of the controllable process parameters on the moisture content of the material. The control direction and process parameter adjustment amount are encapsulated to generate advanced control instructions for downstream nodes, which are then sent to downstream nodes for execution before a future time.

[0010] The process involves measuring the communication delay from the generation of the advanced control command to its execution by downstream nodes, calculating the change in material moisture content within this delay period based on the spatial propagation function, compensating and correcting the control quantity of the advanced control command, generating a delay compensation control signal, and sending it to downstream nodes for execution. Record the start time of the advance control command generation and the arrival time of the command sent to the downstream node for execution, and calculate the time difference between the start time and the arrival time as the communication delay duration; Obtain the material location coordinates and material moisture data corresponding to the starting point of the communication delay duration. Use the material location coordinates as the delay start position, the material moisture data as the delay start moisture value, and the material arrival position coordinates corresponding to the communication delay end point determined according to the communication delay duration and material flow direction as the delay termination position input into the spatial propagation function. The moisture evolution difference of the material from the beginning position to the end position of the delay within the communication delay time is calculated by the spatial propagation function, and the moisture evolution difference is used as the change in material moisture within the communication delay time. Extract the control quantity from the advance control command, and superimpose the change in material moisture content within the communication delay time with the control quantity. When the change is positive, perform incremental compensation; when the change is negative, perform decremental compensation to obtain the compensated control quantity. The compensated control quantity is encapsulated to generate a delay-compensated control signal and sent to downstream nodes for execution.

[0011] The coupling effect of downstream nodes executing delayed compensation control signals on the material moisture content of adjacent nodes is calculated based on the spatial propagation function. Based on this coupling effect, coordinated control instructions for adjacent nodes are generated, including: Obtain the position coordinates of the downstream node and the position coordinates of the adjacent node located after the downstream node; The measured material moisture data after the delay compensation control signal is sent to the downstream node for execution is obtained, along with the material moisture data before execution. The difference between the measured material moisture data and the material moisture data is calculated as the material moisture adjustment amount. The spatial propagation function is input with the coordinates of the downstream node as the coupling start point, the material moisture adjustment amount as the coupling start moisture change value, and the coordinates of the adjacent node as the coupling end point. The moisture coupling transfer value of the material moisture adjustment amount is calculated from the coupling start position to the coupling end position by the spatial propagation function. The moisture coupling transfer value is used as the coupling influence of the downstream node on the material moisture of the adjacent node after the downstream node executes the delay compensation control signal. Obtain the target moisture value of adjacent nodes, calculate the moisture deviation between the coupling influence amount and the target moisture value, and determine the direction and amount of coordinated regulation of adjacent nodes based on the value and positive / negative direction of the moisture deviation. The direction and amount of coordinated control are encapsulated to generate coordinated control instructions for adjacent nodes and then sent to the adjacent nodes for execution.

[0012] The moisture coupling transfer value, calculated using the spatial propagation function, from the initial coupling position to the final coupling position, includes: Calculate the material flow distance between the coupling start position and the coupling end position, obtain the material flow velocity, and calculate the ratio of the material flow distance to the material flow velocity to obtain the material propagation time; Based on the material propagation time, extract the integral value of the rate of change of moisture from the coupling start position to the coupling end position from the spatial propagation function, and calculate the preliminary moisture coupling transfer value based on the integral value of the rate of change of moisture and the material moisture adjustment amount. Extract the measured material moisture data of adjacent nodes at multiple past times, calculate the fluctuation difference between the measured material moisture data at adjacent times, and statistically analyze the fluctuation difference to obtain the fluctuation amplitude; The ratio of the fluctuation amplitude to the initial moisture coupling transfer value is used as a correction coefficient. The initial moisture coupling transfer value is then corrected based on the correction coefficient to obtain the moisture coupling transfer value.

[0013] This invention provides a dynamic moisture content control system for feed pelleting process, the system comprising: The data acquisition module is used to collect material moisture data and equipment operation data through distributed sensor nodes and transmit them to the remote control platform through a wireless communication network. The function establishment module is used to establish a spatial propagation function of material moisture along the granulation process based on material moisture data at different spatial location nodes on the remote control platform. The spatial propagation function describes the evolution relationship of moisture value during the process of material flowing from upstream node to downstream node. The advanced control module is used to calculate the predicted moisture value of the material when it arrives at the downstream node in the future based on the spatial propagation function and the material moisture data of the upstream node at the current time, and to generate an advanced control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value. The delay compensation module is used to measure the communication delay time from the generation of the advanced control command to its execution by the downstream node, calculate the change in material moisture within the communication delay time according to the spatial propagation function, compensate and correct the control quantity of the advanced control command, generate a delay compensation control signal and send it to the downstream node for execution. The collaborative control module is used to calculate the coupling effect of the downstream node's execution of the delay compensation control signal on the material moisture of the adjacent node based on the spatial propagation function, and to generate collaborative control instructions for the adjacent nodes based on the coupling effect.

[0014] One technical solution provided in this embodiment of the invention is an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in any of the aforementioned methods.

[0015] One technical solution provided in this embodiment of the invention is a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the steps in any of the aforementioned methods.

[0016] This invention establishes a spatial propagation function for material moisture along the granulation process, enabling the prediction of the material's moisture state upon arrival at the downstream node based on current moisture data from upstream nodes. This proactive control fundamentally overcomes the lag problem of traditional feedback control, reducing the generation of defective products. By measuring communication delay and calculating the change in material moisture during the delay period, control commands are compensated and corrected, ensuring the timeliness and accuracy of control actions and improving moisture control precision. Furthermore, by quantifying the coupling effect of downstream node control actions on the material moisture of adjacent nodes and generating coordinated control commands accordingly, collaborative control among multiple nodes is achieved. This avoids negative impacts of single-point control on other nodes, ensuring the stability of the moisture content throughout the entire granulation process. Attached Figure Description

[0017] Figure 1 A flowchart of a method for dynamically controlling the moisture content in a feed pelleting process provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a dynamic moisture content control system for feed pelleting process according to an embodiment of the present invention. Detailed Implementation

[0018] like Figure 1 As shown, Figure 1 A flowchart of a method for dynamically controlling the moisture content in a feed pelleting process, provided by an embodiment of the present invention, is included in the following steps: Step 101: Collect material moisture data and equipment operation data through distributed sensor nodes, and transmit them to the remote control platform through a wireless communication network.

[0019] Step 102: Based on the material moisture data of different spatial location nodes, establish a spatial propagation function of material moisture along the granulation process on the remote control platform. The spatial propagation function describes the evolution relationship of moisture value during the process of material flowing from upstream node to downstream node.

[0020] In some embodiments of the present invention, step 102 may specifically include the following sub-steps: Sub-step 1021: Obtain material moisture data and location coordinates of nodes at different spatial locations in the granulation process; sort the nodes at different spatial locations according to the material flow direction based on the location coordinates to generate a node sequence. Sub-step 1022: Select two adjacent spatial location nodes from the node sequence to form a node pair, and extract the material moisture data and location coordinates of the two spatial location nodes in the node pair. Sub-step 1023: Calculate the material moisture difference based on the material moisture data of the two spatial location nodes in the node pair; calculate the position coordinate difference based on the position coordinates of the two spatial location nodes in the node pair; and calculate the unit distance moisture change rate of the node pair based on the material moisture difference and the position coordinate difference. Sub-step 1024: Associate the unit distance moisture change rate of all node pairs in the node sequence with the corresponding location coordinates to construct a discrete data set in which the moisture change rate varies with the location coordinates. Sub-step 1025: Perform function fitting on the discrete data set to generate a continuous function of the rate of change of moisture with respect to the location coordinates. Integrate the continuous function along the location coordinates to generate a spatial propagation function.

[0021] First, moisture sensors are installed at key locations in the granulation process to collect material moisture data in real time, such as the raw material inlet, conditioner outlet, before the pressure rollers, and the granulator outlet. The position coordinates of each sensor are also recorded. These coordinates can be represented by a three-dimensional Cartesian coordinate system, where the x-axis represents the horizontal direction, the y-axis represents the vertical direction, and the z-axis represents the depth direction. Based on the material flow direction in the granulation process, all collection points are sorted to generate a node sequence. The sorting principle is to arrange them sequentially from upstream to downstream according to the order of material flow, ensuring that the node sequence accurately reflects the material's flow path in space.

[0022] From the sorted node sequence, two adjacent spatially located nodes are selected sequentially to form node pairs. For each node pair, the material moisture data and location coordinates of the two nodes are extracted. The material moisture data is usually expressed as a percentage, representing the proportion of water mass to the total mass of the material. The location coordinates are expressed in distance units, recording the specific spatial location of the node.

[0023] For each node pair, calculate the change in moisture content of the material as it flows from the upstream node to the downstream node. First, calculate the moisture difference, which is the moisture content of the material at the downstream node minus the moisture content of the material at the upstream node. A positive difference indicates an increase in moisture content during the flow; a negative difference indicates a decrease in moisture content. Next, calculate the position coordinate difference, i.e., the spatial distance between the downstream and upstream nodes. This spatial distance can be calculated using the Euclidean distance between the two points. Where x1, y1, z1 are the coordinates of the upstream node, x2, y2, z2 are the coordinates of the downstream node, and d is the spatial distance between the two nodes.

[0024] Based on the difference in material moisture content and the difference in location coordinates, calculate the rate of change in moisture content per unit distance for each node pair. This rate of change in moisture content per unit distance represents the magnitude of change in moisture content within a unit distance. The calculation formula is as follows: Where r is the rate of change of moisture per unit distance, m1 is the material moisture data of the upstream node, m2 is the material moisture data of the downstream node, and d is the spatial distance between the two nodes.

[0025] By associating the rate of change in moisture per unit distance for all node pairs in the node sequence with their corresponding location coordinates, a discrete data set is constructed showing the change in moisture rate as a function of location coordinates. This data set contains multiple data points, each consisting of location coordinates and the corresponding rate of change in moisture per unit distance. To simplify calculations, the main direction of the material flow path can be selected as a reference axis, and the three-dimensional spatial coordinates can be projected onto this axis to obtain a one-dimensional coordinate representation.

[0026] The discrete data set is subjected to function fitting to generate a continuous function of the moisture change rate with respect to the location coordinates. Function fitting methods can include polynomial fitting, spline interpolation, or least squares. Considering the complexity of the granulation process, a piecewise function fitting method can be chosen, using different function forms for different location intervals to improve fitting accuracy.

[0027] The continuous function obtained by fitting is integrated along the position coordinates to generate a spatial propagation function. The physical meaning of the integral is to calculate the cumulative change in moisture content of the material from the initial position to any position.

[0028] In practical applications, integral calculations can be performed using numerical integration methods, such as the rectangular method, trapezoidal method, or Simpson's method. For piecewise functions, integration is performed within each segment interval, and the results are summed. The calculated spatial propagation function can describe the spatial distribution of material moisture along the granulation process, providing a theoretical basis for subsequent dynamic moisture control.

[0029] This invention establishes a spatial propagation function of material moisture along the granulation process, enabling a precise description of the moisture variation law of materials during granulation. This effectively solves the problem of traditional methods' difficulty in accurately predicting the spatial distribution of material moisture. This method considers the dynamic changes in moisture during material flow, accurately predicting the moisture content at any location, providing a scientific basis for precise moisture control during granulation, and improving the quality stability and production efficiency of granulated products.

[0030] In sub-step 1025, the discrete data set is subjected to function fitting to generate a continuous function of the water change rate with respect to the location coordinates. Integrating this continuous function along the location coordinates to generate the spatial propagation function further includes: Identify node pairs in a discrete data set that exhibit abnormal fluctuations in the rate of change of moisture per unit distance, extract the rate of change of moisture per unit distance from adjacent normal node pairs of the abnormal node pairs, smooth them out, replace the rate of change of moisture per unit distance from the abnormal node pairs, and generate a corrected discrete data set. Extract the position coordinates of each node pair in the corrected discrete data set and the corresponding rate of change of moisture per unit distance. Establish a polynomial fitting equation with position coordinates as the independent variable and rate of change of moisture per unit distance as the dependent variable. Generate a continuous function of rate of change of moisture per unit distance with respect to position coordinates by performing function fitting on the corrected discrete data set. Obtain the initial moisture value of the material at the starting point of the granulation process, perform an integral operation on the continuous function along the position coordinates from the starting point, and superimpose the integral result with the initial moisture value of the material to generate a spatial propagation function.

[0031] Before performing function fitting on the discrete data set, it is necessary to identify node pairs with abnormal fluctuations in the rate of change of moisture per unit distance. During the granulation process, due to factors such as equipment vibration, sensor errors, or instantaneous material fluctuations, the rate of change of moisture in some node pairs may exhibit abnormal fluctuations, which do not conform to the basic laws of material moisture change. The method for identifying abnormal node pairs is to calculate the absolute value of the difference in the rate of change of moisture per unit distance between adjacent node pairs in the discrete data set, and compare this difference with a preset threshold. The preset threshold can be determined based on historical data statistics, and is usually taken as 1.5 times the fluctuation range of historical data. If the absolute value of the difference exceeds the preset threshold, it is determined to be an abnormal node pair.

[0032] Once an anomalous node pair is identified, the unit distance moisture change rate of the adjacent normal node pairs is extracted and smoothed. The smoothing process uses a weighted average method: the unit distance moisture change rate of the two normal node pairs before and after the anomalous node pair is taken, and different weights are assigned based on distance for each pair. Closer distances result in higher weights, and farther distances result in lower weights. The weight allocation uses the reciprocal of the distance as the basis, and after normalization, the final weight value is obtained. The result of the weighted average calculation is used as the new unit distance moisture change rate for the anomalous node pair, replacing the original anomalous value.

[0033] By smoothing all abnormal node pairs, a corrected discrete data set is generated. This corrected discrete data set exhibits better continuity and smoothness, more accurately reflecting the spatial variation of material moisture content. If areas of significant fluctuation still exist in the corrected discrete data set, a moving average method can be used for secondary smoothing to further improve data continuity. The moving average window size can be set to 3 to 5 node pairs, with the specific value determined based on the data size and degree of fluctuation.

[0034] The position coordinates and corresponding rate of change of moisture per unit distance for each node pair are extracted from the corrected discrete data set. To simplify the fitting calculation, the three-dimensional spatial coordinates can be projected onto the principal axis of the material flow to obtain a one-dimensional coordinate representation. A polynomial fitting equation is established with the position coordinates as the independent variable and the rate of change of moisture per unit distance as the dependent variable.

[0035] The general form of the polynomial fitting equation is: Where r(s) is the rate of change of moisture per unit distance, s is the location coordinate, a0, a1, a2, ..., a n Here, represents the polynomial coefficients, and n is the polynomial order. The choice of polynomial order needs to balance data complexity and fitting accuracy; typically, order 3 to 5 is chosen. Too low an order may lead to insufficient fitting accuracy, while too high an order may result in overfitting.

[0036] The fitting coefficients are calculated using the least squares method. The objective function is the sum of squares of the differences between the actual moisture change rate and the fitted moisture change rate. The optimal fitting result is obtained by finding the coefficients that minimize the objective function. The calculation process can be implemented using Gaussian elimination or gradient descent. In practical applications, if the data distribution exhibits obvious piecewise characteristics, piecewise polynomial fitting can be used, that is, different polynomial forms are used in different intervals to improve the overall fitting accuracy.

[0037] By performing function fitting on the corrected discrete data set, a continuous function r(s) representing the rate of change of moisture per unit distance with respect to location coordinates is generated. This function can describe the rate of change of material moisture at any location in the granulation process, laying the foundation for the next step of integral calculation. After fitting, the fitting results are verified by calculating the root mean square error and coefficient of determination between the fitted values ​​and the actual values ​​to evaluate the fitting accuracy. If the fitting accuracy does not meet the requirements, the polynomial order can be adjusted or other fitting methods, such as exponential functions or power functions, can be tried.

[0038] The initial moisture content of the material at the starting point of the granulation process is obtained as the reference point for integration calculation. This initial moisture content is acquired in real-time by a moisture sensor at the granulation process inlet, or determined according to the formula settings. The continuous function r(s) is integrated along the position coordinates from the starting point to calculate the cumulative moisture change at any position. The integration formula is: Where I(s) is the cumulative change in moisture from the starting point s0 to the position s, r(t) is the rate of change of moisture per unit distance, and t is the integral variable.

[0039] The integral result I(s) is superimposed with the initial moisture content M0 of the material to generate the spatial propagation function M(s). The expression of the spatial propagation function is: M(s) = M0 + I(s), where M(s) is the moisture content of the material at position s, M0 is the initial moisture content of the material, and I(s) is the cumulative moisture change. The spatial propagation function describes the evolution of the moisture content of the material as it flows from the starting point of the granulation process to any position, and can accurately predict the moisture content of the material at any position in the granulation process.

[0040] This invention utilizes data smoothing and function fitting integration techniques to accurately describe and predict the spatial propagation of moisture in feed pelleting, solving the technical problem of traditional methods' inability to accurately capture dynamic moisture changes. Compared to traditional methods, this approach effectively identifies and corrects abnormal data fluctuations, improves fitting accuracy, and accurately reflects the continuous spatial changes in material moisture. It provides a theoretical basis for the dynamic and precise control of moisture content in feed pelleting, significantly improving the stability of pelleted product quality, reducing energy consumption, and increasing production efficiency.

[0041] Step 103: Based on the spatial propagation function and the material moisture data of the upstream node at the current time, calculate the predicted moisture value of the material when it arrives at the downstream node at a future time, and generate an advance control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value.

[0042] In some embodiments of the present invention, step 103 may specifically include the following sub-steps: Sub-step 1031: Obtain the material moisture data and location coordinates of the upstream node at the current moment, and obtain the location coordinates and target moisture value of the downstream node; Sub-step 1032: Input the spatial propagation function with the position coordinates of the upstream node as the starting position, the material moisture data of the upstream node at the current moment as the starting moisture value, and the position coordinates of the downstream node as the ending position. Sub-step 1033: The moisture evolution value of the material flowing from the upstream node to the downstream node is calculated by the spatial propagation function, and used as the moisture prediction value of the material when it arrives at the downstream node in the future. Sub-step 1034: Calculate the difference between the predicted moisture value and the target moisture value to obtain the deviation amount. Determine the direction of moisture regulation at downstream nodes based on the positive or negative direction of the deviation amount, including increasing or decreasing moisture. Sub-step 1035: Extract the adjustable process parameters corresponding to the downstream nodes according to the direction of moisture control, and determine the adjustment amount of the process parameters based on the value of the deviation and the degree of influence of the adjustable process parameters on the moisture content of the material. Sub-step 1036: Package the regulation direction and the adjustment amount of process parameters to generate an advanced regulation instruction for the downstream node, and issue the advanced regulation instruction for the downstream node to the downstream node for execution before the future moment.

[0043] First, obtain the material moisture data and position coordinates of the upstream node at the current moment, as well as the position coordinates and target moisture value of the downstream node, for moisture prediction and regulation parameter calculation. The upstream node usually refers to the material input position in the pelleting process, such as the inlet of the conditioner or the outlet of the mixer; the downstream node is the position where moisture regulation is required, such as the outlet of the conditioner, the pressing chamber, or the inlet of the cooler. The material moisture data is collected in real time by moisture detection devices installed at the corresponding positions. The target moisture value is set according to the technological requirements of the feed product, and the target moisture values of different feed varieties are different, generally set within the optimal moisture range suitable for pellet forming.

[0044] Take the position coordinates of the upstream node as the starting position, the material moisture data of the upstream node at the current moment as the starting moisture value, and the position coordinates of the downstream node as the ending position, and input them into the established space propagation function. Calculate the moisture evolution value of the material flowing from the upstream node to the downstream node through the space propagation function, which is used as the moisture prediction value when the material reaches the downstream node at the future moment. When calculating, substitute the above parameters into the space propagation function, solve the definite integral by the numerical integration method, obtain the cumulative value of moisture change, and then add it to the starting moisture value to get the moisture prediction value. The numerical integration uses the Simpson integration method, divides the integration interval into an even number of small intervals, and approximately replaces the original function curve with a quadratic curve to improve the integration accuracy. The number of divisions of the integration interval is determined according to the distance, generally taking 10 to 20 intervals.

[0045] Calculate the difference between the moisture prediction value and the target moisture value to obtain the deviation amount. The positive or negative of the deviation amount indicates the moisture regulation direction: if the deviation amount is positive, it means the predicted moisture is higher than the target value and moisture needs to be reduced; if the deviation amount is negative, it means the predicted moisture is lower than the target value and moisture needs to be increased. The absolute value of the deviation amount represents the required regulation intensity: the larger the absolute value of the deviation amount, the greater the required regulation intensity.

[0046] Extract the adjustable process parameters corresponding to the downstream node according to the moisture regulation direction. Different downstream nodes correspond to different adjustable process parameters: for the conditioner position, the adjustable parameters include steam injection amount, water injection amount, and stirring speed; for the pressing chamber position, the adjustable parameters include die hole diameter, pressing speed, and roll gap; for the cooler position, the adjustable parameters include cooling air volume and cooling time. Select appropriate process parameters according to the moisture regulation direction: when moisture needs to be increased, the steam injection amount, water injection amount can be increased, the die hole diameter can be reduced, and the cooling air volume can be slowed down; when moisture needs to be reduced, the steam injection amount, water injection amount can be reduced, the die hole diameter can be increased, and the cooling air volume can be increased.

[0047] The adjustment amount of the process parameters is determined based on the numerical value of the deviation and the degree of influence of the adjustable process parameters on the moisture content of the material. The calculation of the process parameter adjustment amount is based on the moisture control sensitivity coefficient, which represents the change in moisture caused by a one-unit change in the process parameter. The moisture control sensitivity coefficient of each process parameter is determined through historical data analysis or professional experience. The formula for calculating the process parameter adjustment amount is: Where ΔP is the adjustment amount of the process parameter, ΔM is the moisture deviation, and K is the moisture control sensitivity coefficient. When multiple process parameters are adjusted simultaneously, the interaction between parameters needs to be considered, and a weighted allocation method is used to determine the adjustment ratio of each parameter.

[0048] The control direction and process parameter adjustment amount are encapsulated to generate advanced control instructions for downstream nodes. These instructions include node identifier, control parameters, adjustment direction, adjustment amount, and execution time. The instruction encapsulation uses a standardized format to ensure communication consistency between the remote control platform and the execution equipment. The execution time is determined based on the transmission time required for material to reach the downstream node from the upstream node. This transmission time is calculated using the material flow rate and the distance between nodes. Control instructions are typically sent 5 to 10 seconds in advance to ensure that the control parameters are adjusted before the material arrives.

[0049] Ahead of time, advance control instructions for downstream nodes are issued to them for execution via real-time communication through an industrial communication network. Upon receiving the instructions, the execution equipment automatically adjusts the corresponding process parameters to achieve precise control of material moisture content. If the execution equipment reports adjustment failure, the system automatically triggers a backup control scheme to ensure the moisture control target is achieved.

[0050] This invention achieves accurate prediction of material moisture content through a spatial propagation function, and generates advanced control commands based on the prediction results, effectively solving the lag problem of traditional passive control methods. Employing an integrated prediction-control architecture, it realizes intelligent management and control of moisture throughout the entire pelleting process, significantly improving the moisture uniformity and stability of feed products, reducing energy consumption, improving product quality, extending mold life, and reducing maintenance costs.

[0051] Step 104: Measure the communication delay time from the generation of the advanced control command to its execution by the downstream node, calculate the change in material moisture within the communication delay time based on the spatial propagation function, compensate and correct the control quantity of the advanced control command, generate a delay compensation control signal and send it to the downstream node for execution.

[0052] In some embodiments of the present invention, step 104 may specifically include the following sub-steps: Sub-step 1041: Record the start time of the advanced control instruction generation and the arrival time of the instruction sent to the downstream node for execution, and calculate the time difference between the start time and the arrival time as the communication delay duration. Sub-step 1042: Obtain the material position coordinates and material moisture data corresponding to the starting point of the communication delay duration; use the material position coordinates as the delay start position, the material moisture data as the delay start moisture value, and use the material arrival position coordinates corresponding to the communication delay end point determined according to the communication delay duration and the material flow direction as the delay termination position input into the spatial propagation function. Sub-step 1043: The moisture evolution difference of the material from the start position to the end position of the delay within the communication delay time is calculated by the spatial propagation function, and the moisture evolution difference is used as the change in material moisture within the communication delay time. Sub-step 1044: Extract the control quantity from the advance control instruction, and superimpose the change in material moisture content within the communication delay time with the control quantity. When the change is positive, perform incremental compensation; when the change is negative, perform decremental compensation to obtain the compensated control quantity. Sub-step 1045 encapsulates the compensated control quantity to generate a delay compensation control signal and sends it to the downstream node for execution.

[0053] In actual implementation, the start time of the advanced control command generation and the arrival time of its execution at the downstream node are recorded. The time difference between the two is calculated as the communication delay. The start time refers to the system time when the control command is generated, which is recorded by the clock module of the central control unit; the arrival time refers to the time when the execution device of the downstream node receives the command and sends back confirmation of receipt, which is recorded by the clock module of the execution device. The two clock modules need to be synchronized in advance to ensure that the time base is consistent.

[0054] After calculating the communication delay duration, the material location coordinates and moisture data corresponding to the starting point of the communication delay duration are obtained. The starting point of the communication delay duration is usually the moment the advance control command is generated, at which time the material is located at a certain position between the upstream and downstream nodes. This location coordinate can be calculated by multiplying the material flow rate by the time elapsed between the command generation time and the material at the upstream node. The material flow rate is measured in real time by a flow meter or estimated based on equipment operating parameters. The material moisture data is calculated using an interpolation method: if there is a moisture detection device at the starting point of the communication delay duration, the device's measurement value is directly used; if there is no moisture detection device, linear interpolation is performed based on the data from the two most recent moisture measurement points.

[0055] The material location coordinates are used as the starting position of the delay, the material moisture data is used as the starting moisture value of the delay, and the material arrival position coordinates corresponding to the end point of the communication delay, determined by the communication delay duration and the material flow direction, are used as the ending position of the delay. These are input into the spatial propagation function. The method for calculating the material arrival position coordinates corresponding to the end point of the communication delay is: the starting position coordinates of the delay plus the distance the material flows within the communication delay duration. The material flow distance is equal to the product of the material flow velocity and the communication delay duration. If the material flow velocity changes during the communication delay, a segmented calculation method is required. The communication delay duration is divided into multiple small time intervals, with the flow velocity assumed to be constant within each interval. The total flow distance is obtained by summing the material flow distances of each interval.

[0056] The moisture evolution difference of the material as it flows from the starting position to the ending position of the delay within the communication delay time is calculated using a spatial propagation function. This moisture evolution difference is then used as the change in material moisture within the communication delay time. The calculation process utilizes the spatial propagation function, namely: ΔM 延迟 =M(s 终 )−M(s 起 Wherein, ΔM 延迟 M(s) represents the change in material moisture content during the communication delay period. 终 M(s) represents the moisture content of the material at the delayed termination position. 起 The values ​​represent the material moisture content at the initial delay position, all calculated using a spatial propagation function. The moisture evolution difference can be positive or negative; a positive value indicates an increase in moisture content during the delay, while a negative value indicates a decrease. The magnitude of the change is related to factors such as the communication delay duration, material characteristics, and granulation process parameters.

[0057] The control quantity is extracted from the advance control command. The change in material moisture content during the communication delay period is then superimposed with the control quantity to obtain the compensated control quantity. The control quantity in the advance control command refers to the parameter values ​​that need to be adjusted to achieve the target moisture content, such as the steam injection rate adjustment value and the pressure roller gap adjustment value. The basic principle of the superposition operation is: if the material moisture content changes during the communication delay period, the original control quantity needs to be adjusted accordingly to compensate for this change. When the change is positive (moisture content increases), incremental compensation is performed; when the change is negative (moisture content decreases), decremental compensation is performed. The formula for calculating the compensated control quantity is: P 补 =P 原 +α⋅ΔM 延迟 Among them, P 补 To compensate for the control quantity, P 原 The original control variable is α, the compensation coefficient is ΔM. 延迟This represents the change in material moisture content within the communication delay period. The compensation coefficient α is related to specific process parameters and reflects the sensitivity relationship between changes in process parameters and changes in moisture content. It is determined through historical data analysis or expert experience. The compensation coefficient value varies for different process parameters. For example, the compensation coefficient for steam injection rate might be 2.5, meaning that 2.5 units of steam injection rate need to be adjusted for every 1% change in moisture content.

[0058] The compensated control quantity is encapsulated to generate a delay-compensated control signal and sent to downstream nodes for execution. The encapsulation format of the delay-compensated control signal is the same as that of the lead control instruction, including node identifier, control parameters, adjustment direction, adjustment amount (after compensation), and execution priority. To distinguish the delay-compensated control signal from ordinary control instructions, a compensation identifier field is added to the signal. Upon receiving a signal with the compensation identifier, the execution device treats it as the highest priority to ensure timely execution. The delay-compensated control signal is transmitted through a high-priority communication channel to reduce the risk of secondary delays. Upon receiving the delay-compensated control signal, the execution device at the downstream node immediately interrupts the currently executing ordinary control instruction (if any) and prioritizes the parameter adjustment in the compensation control signal.

[0059] After the executing equipment completes parameter adjustments, it sends execution completion information to the central control unit, including the actual parameter values ​​and execution time. The central control unit receives this feedback and updates the material moisture control status, providing a reference for the next round of proactive control. If the executing equipment reports execution failure or inadequate parameter adjustment, the central control unit immediately initiates an emergency response procedure, using other adjustable parameters to compensate and ensure that the material moisture content is controlled within the target range.

[0060] This invention dynamically adjusts control parameters by measuring the delay time in real time and calculating the change in material moisture during the delay period, ensuring the accuracy of the control commands at the moment of execution. Compared with traditional static control methods, this method considers the dynamic change characteristics of material moisture during pelleting, improves the timeliness and accuracy of control, reduces moisture fluctuations caused by control lag, significantly improves the moisture uniformity and product quality stability of feed pellet products, and reduces energy consumption and scrap rate.

[0061] Step 105: Calculate the coupling effect of the downstream node's execution of the delay compensation control signal on the material moisture of the adjacent node based on the spatial propagation function, and generate a coordinated control command for the adjacent node based on the coupling effect.

[0062] In some embodiments of the present invention, step 105 may specifically include the following sub-steps: Sub-step 1051: Obtain the position coordinates of the downstream node and the position coordinates of the adjacent node located after the downstream node; Sub-step 1052: Obtain the material moisture measured data after the delay compensation control signal is sent to the downstream node and the material moisture data before execution, and calculate the difference between the material moisture measured data and the material moisture data as the material moisture adjustment amount. Sub-step 1053: Input the spatial propagation function with the position coordinates of the downstream node as the coupling start position, the material moisture adjustment amount as the coupling start moisture change value, and the position coordinates of the adjacent node as the coupling end position. Sub-step 1054: The moisture coupling transfer value of the material moisture adjustment amount is calculated from the coupling start position to the coupling end position through the spatial propagation function. The moisture coupling transfer value is used as the coupling influence of the downstream node on the material moisture of the adjacent node after the downstream node executes the delay compensation control signal. Sub-step 1055: Obtain the target moisture value of adjacent nodes, calculate the moisture deviation between the coupling influence amount and the target moisture value, and determine the direction and amount of coordinated regulation of adjacent nodes based on the value and positive / negative direction of the moisture deviation. Sub-step 1056: Encapsulate the coordinated control direction and coordinated control amount to generate coordinated control instructions for adjacent nodes and send them to adjacent nodes for execution.

[0063] The location coordinates of the downstream node and its adjacent node are obtained. The downstream node refers to the node that receives and executes delay compensation control signals, such as the conditioner outlet. The adjacent node refers to the next control node after the downstream node, such as the pressing chamber inlet. The location coordinates are determined through pre-measurement and are represented using a rectangular coordinate system or a one-dimensional coordinate system along the material flow direction, with the unit being meters. In a feed pelleting production line, the distance between adjacent nodes is typically between 0.5 meters and 3 meters, depending on the equipment layout and process flow design.

[0064] The material moisture content is adjusted by acquiring the measured material moisture data after the delay compensation control signal is sent to the downstream node and comparing it with the material moisture data before execution. The measured material moisture data is collected in real time by a moisture sensor installed at the downstream node, with a collection frequency of 2 to 5 times per second to ensure that moisture change trends can be captured. The material moisture data before execution is obtained by backtracking historical data, using the moisture value from the last collection before the delay compensation control signal is executed. The formula for calculating the moisture adjustment is as follows: ΔM 调整 =M 执行后 -M 执行前 Where, ΔM 调整 M is the material moisture adjustment amount. 执行后 M is the measured material moisture data after the delay compensation control signal is executed. 执行前This is the material moisture data before execution. The material moisture adjustment can be positive or negative; a positive value indicates an increase in moisture, and a negative value indicates a decrease in moisture. The magnitude of the adjustment reflects the actual execution effect of the delay compensation control signal.

[0065] The spatial propagation function is input with the coordinates of the downstream node as the coupling start point, the material moisture adjustment amount as the coupling start moisture change value, and the coordinates of the adjacent node as the coupling end point. The spatial propagation function is used here to calculate the spatial transmission characteristics of moisture changes, reflecting the evolution of moisture as the material flows from one location to another. The parameters input to the spatial propagation function include: the coupling start point (coordinates of the downstream node), the coupling start moisture change value (material moisture adjustment amount), the coupling end point (coordinates of the adjacent node), and process parameters along the propagation path, such as temperature distribution, humidity distribution, and material flow rate.

[0066] The moisture coupling transfer value, calculated from the coupling initiation position to the coupling termination position by the spatial propagation function, is used as the coupling influence on the material moisture of adjacent nodes after the downstream node executes the delay compensation control signal. The calculation uses a modified spatial propagation function: ΔM 耦合 =ΔM 调整 ×C(s 起 s 终 Wherein, ΔM 耦合 ΔM represents the coupling effect quantity. 调整 C(s) is the material moisture adjustment amount. 起 s 终 ) is the coupling transfer coefficient, representing the change in moisture from the initial position s. 起 Pass to the termination position s 终 The coupling transfer coefficient is calculated using the spatial propagation function, taking into account the influence of factors such as path length, material properties, and environmental conditions on moisture transfer. In the feed pelleting process, the coupling transfer coefficient is usually less than 1, indicating that moisture changes are attenuated during spatial transfer. However, under certain special conditions (such as high temperature and high humidity environments), it may be greater than 1, indicating that moisture changes are amplified during transfer.

[0067] Obtain the target moisture values ​​of adjacent nodes, calculate the moisture deviation between the coupling influence and the target moisture value, and determine the direction and amount of coordinated regulation of adjacent nodes based on the value and direction of the moisture deviation. The target moisture value is set according to the feed product process requirements, and different target moisture values ​​correspond to different production stages. Moisture deviation ΔM 偏差 =M 目标 −(M 当前 +ΔM 耦合 Wherein, ΔM 偏差 For moisture deviation, M 目标M represents the target moisture value of adjacent nodes. 当前 Let ΔM be the current moisture value of the adjacent nodes. 耦合 This represents the coupling effect. The sign of the moisture deviation determines the direction of coordinated regulation: a positive value indicates that moisture needs to be increased, and a negative value indicates that moisture needs to be decreased. The absolute value of the moisture deviation determines the magnitude of the coordinated regulation: the larger the deviation, the larger the regulation.

[0068] The coordinated regulation amount between adjacent nodes is determined based on the moisture deviation, and the calculation formula is as follows: P 协同 =β×ΔM 偏差 Among them, P 协同 ΔM is the amount of coordinated regulation, β is the coefficient of coordinated regulation, and ΔM is the variable for coordinated regulation. 偏差 This represents moisture deviation. The synergistic control coefficient reflects the sensitivity of adjustable process parameters at adjacent nodes to the influence of moisture, and is obtained through historical data analysis or experimental determination. Different process parameters correspond to different synergistic control coefficients; for example, the synergistic control coefficient for steam flow rate may be 2.0, while the synergistic control coefficient for die diameter may be 0.5.

[0069] The coordinated control direction and amount are encapsulated to generate coordinated control instructions for adjacent nodes, which are then issued to the adjacent nodes for execution. These instructions include node identifier, coordination type identifier, control parameters, adjustment direction, adjustment amount, priority, and execution time. The coordination type identifier distinguishes between regular control instructions and coordinated control instructions. After receiving the coordinated control instructions, adjacent nodes determine the execution order based on priority. The priority of coordinated control instructions is typically higher than that of regular control instructions but lower than that of emergency control instructions. The execution time is set to the estimated time required for material to flow from the downstream node to the adjacent node, ensuring that coordinated control takes effect before the coupling effects arrive.

[0070] After receiving coordinated control commands, adjacent nodes adjust corresponding process parameters according to the commands, such as conditioner steam flow, cooler wind speed, and pressure roller gap. After adjustment, the adjacent nodes report the execution results to the control center, including the actual adjusted parameter values, adjustment time, and execution status. Upon receiving the feedback, the control center updates the control status of the adjacent nodes, providing a basis for subsequent control decisions. If the coordinated control command fails to execute, the control center immediately initiates an emergency response procedure, using other adjustable parameters for alternative control to ensure the achievement of moisture control targets.

[0071] This invention achieves coordinated operation among multiple nodes in the pelleting process by calculating the spatial coupling effect of moisture regulation. Compared with single-point control, coordinated regulation fully considers the evolution of material moisture during spatial transmission, predicts and compensates for the mutual influence between nodes, and realizes global optimization of the entire pelleting process. This dynamic coordinated regulation method based on spatial propagation functions significantly improves the accuracy and stability of moisture control in the feed pelleting process, reduces moisture fluctuations, enhances the consistency of finished feed quality, and reduces energy consumption and scrap rate.

[0072] In sub-step 1054, the calculation of the moisture coupling transfer value of the material moisture adjustment from the coupling start position to the coupling termination position through the spatial propagation function also includes: Calculate the material flow distance between the coupling start position and the coupling end position, obtain the material flow velocity, and calculate the ratio of the material flow distance to the material flow velocity to obtain the material propagation time; Based on the material propagation time, extract the integral value of the rate of change of moisture from the coupling start position to the coupling end position from the spatial propagation function, and calculate the preliminary moisture coupling transfer value based on the integral value of the rate of change of moisture and the material moisture adjustment amount. Extract the measured material moisture data of adjacent nodes at multiple past times, calculate the fluctuation difference between the measured material moisture data at adjacent times, and statistically analyze the fluctuation difference to obtain the fluctuation amplitude; The ratio of the fluctuation amplitude to the initial moisture coupling transfer value is used as a correction coefficient. The initial moisture coupling transfer value is then corrected based on the correction coefficient to obtain the moisture coupling transfer value.

[0073] The process of calculating the moisture coupling transfer value using the spatial propagation function requires comprehensive analysis considering material flow time, moisture change rate, and actual fluctuations. When calculating the material flow distance between the coupling start and end positions, the coordinate difference method is used, i.e., the absolute value of the coupling end position coordinates minus the absolute value of the coupling start position coordinates. In feed pelleting, the position coordinates can be established along the main direction of material flow in a one-dimensional coordinate system to simplify the calculation process. The unit of material flow distance is meters, and the value range is typically between 0.5 and 5 meters, depending on the equipment layout and structural design. Material flow velocity can be obtained by direct measurement with a flow meter or indirectly calculated based on parameters such as screw conveyor speed and pressure roller speed. The unit of material flow velocity is meters per second, typically between 0.05 and 0.5 meters per second in pelleting. The material propagation time is calculated by dividing the distance by the velocity, i.e., T = distance / velocity. 传播 =D 流动 / V 流动 Among them, T 传播 D represents the duration of material propagation, measured in seconds. 流动 V represents the material flow distance, in meters. 流动The velocity of the material is expressed in meters per second. When the material velocity is not uniform along the propagation path, the path can be divided into segments. Each segment is assumed to have a local constant velocity, and the propagation time of each segment is calculated separately. The total propagation time is then summed.

[0074] The integral value of the moisture change rate between the coupling start position and the coupling termination position is extracted from the spatial propagation function based on the material propagation time. The spatial propagation function describes the change law of material moisture with spatial position and time, and can be expressed as the moisture change rate function f(s,t), where s is the spatial position and t is the time. The integral value of the moisture change rate is calculated using the spatiotemporal integration method. in, I变化率 T is the integral value of the rate of change in moisture content; 传播 Duration of material propagation; s 起 The coordinates of the initial coupling position; s 终 The coordinates represent the coupling termination position; f(s, t) is the moisture change rate function. In practical applications, a numerical integration method can be used, discretizing the time and spatial dimensions, taking 5 to 10 sampling points respectively, and approximating the integral value by summation. The moisture change rate function is obtained by fitting historical data, reflecting the evolution of material moisture under different process conditions.

[0075] The preliminary moisture coupling transfer value is calculated based on the integral value of the moisture change rate and the material moisture adjustment amount. The preliminary moisture coupling transfer value represents the theoretical coupling effect without considering actual fluctuations. The calculation formula is: ΔM 初步 =ΔM 调整 ×(1+I 变化率 Wherein, ΔM 初步 This is the initial moisture coupling and transfer value; ΔM 调整 For material moisture adjustment; I 变化率 This is the integral value of the rate of change of moisture. In this formula, the integral value of the rate of change of moisture represents the natural change ratio of moisture in the material during the propagation process. "1" represents the baseline state with no change. The formula as a whole reflects the theoretical change result of the initial moisture adjustment after spatial propagation.

[0076] Extract measured material moisture data from adjacent nodes over multiple past time points, and calculate the fluctuation difference between these adjacent data points. The measured moisture data for adjacent nodes comes from moisture sensors installed at those locations, with a data acquisition frequency of 6 to 10 times per minute. Extract measured data within a specific time window, typically 3 to 5 times the material propagation time, to ensure a sufficient number of data points are included. Calculate the difference for each pair of adjacent measured data points, and statistically analyze the fluctuation difference to obtain the fluctuation amplitude. The fluctuation amplitude reflects the natural fluctuation degree of material moisture content and can be calculated using the standard deviation method. Where, σ波动 ΔM represents the fluctuation amplitude; n represents the number of fluctuation difference samples; ΔM i This represents the i-th fluctuation difference. A larger fluctuation amplitude indicates more drastic fluctuations in material moisture content, which is usually related to factors such as process stability and raw material uniformity. In a stable granulation process, the fluctuation amplitude is typically within the range of 0.2% to 0.5%.

[0077] The ratio of the fluctuation amplitude to the initial moisture coupling transfer value is used as a correction factor. The initial moisture coupling transfer value is then corrected based on this correction factor to obtain the final moisture coupling transfer value. The formula for calculating the correction factor is: Among them, K 修正 σ is the correction factor; 波动 The fluctuation range; |ΔM 初步 | represents the absolute value of the initial moisture coupling transmission value; λ is a scaling factor used to adjust the correction strength, typically ranging from 0.5 to 1.5. The correction coefficient reflects the degree of influence of actual fluctuations relative to the theoretical coupling transmission; the larger the correction coefficient, the stronger the interference of actual fluctuations on coupling transmission.

[0078] Final water coupling transfer value ΔM' 耦合 =ΔM 初步 ×(1−K 修正 Wherein, ΔM' 耦合 The final moisture coupling transfer value; ΔM 初步 This is the initial value for water coupling and transfer; K 修正 For the correction factor, "1-K 修正 "This represents the coupling attenuation factor after taking into account actual fluctuations, reflecting the effective transmission ratio of the moisture adjustment signal after being disturbed by actual fluctuations during propagation. When the correction coefficient is large, the coupling attenuation factor decreases, indicating that the actual coupling effect is weakened by fluctuations; when the correction coefficient is small, the coupling attenuation factor is close to 1, indicating that the coupling effect is basically unaffected by fluctuations."

[0079] Using the above calculation method, the final moisture coupling transfer value comprehensively considers both the theoretical propagation model and actual fluctuations, accurately reflecting the true coupling impact of downstream nodes executing delayed compensation control signals on the moisture content of materials at adjacent nodes. This coupling impact will be used to subsequently generate coordinated control commands for adjacent nodes, ensuring consistent moisture control throughout the entire granulation process.

[0080] This invention achieves precise quantification of the moisture coupling relationship between nodes in the pelleting process by combining material flow characteristics, moisture variation patterns, and actual fluctuations. This data-driven dynamic correction mechanism effectively solves the problem of mutual interference in moisture regulation at multiple nodes during pelleting, providing key technical support for achieving coordinated regulation throughout the entire process. This ensures the uniform and stable moisture content of feed products, improves finished product quality, and reduces energy consumption and production costs.

[0081] like Figure 2 As shown, Figure 2 This is a schematic diagram of a dynamic moisture content control system for feed pelleting provided in an embodiment of the present invention. The system includes: Data acquisition module 201 is used to collect material moisture data and equipment operation data through distributed sensing nodes and transmit them to the remote control platform through a wireless communication network; The function establishment module 202 is used to establish a spatial propagation function of material moisture along the granulation process based on material moisture data at different spatial location nodes on the remote control platform. The spatial propagation function describes the evolution relationship of moisture value during the process of material flowing from upstream node to downstream node. The advanced control module 203 is used to calculate the predicted moisture value of the material when it reaches the downstream node at a future time based on the spatial propagation function and the material moisture data of the upstream node at the current time, and to generate an advanced control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value. The delay compensation module 204 is used to measure the communication delay time from the generation of the advanced control command to its execution by the downstream node, calculate the change in material moisture within the communication delay time according to the spatial propagation function, compensate and correct the control quantity of the advanced control command, generate a delay compensation control signal and send it to the downstream node for execution. The collaborative control module 205 is used to calculate the coupling effect of the downstream node's execution of the delay compensation control signal on the material moisture of the adjacent node based on the spatial propagation function, and to generate collaborative control instructions for the adjacent node based on the coupling effect.

[0082] One technical solution provided in this embodiment of the invention is an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in any of the aforementioned methods.

[0083] One technical solution provided in this embodiment of the invention is a computer-readable storage medium storing a computer program, wherein the processor executes the computer program to implement the steps in any of the aforementioned methods.

[0084] The specific embodiments described above are preferred embodiments of the present invention and are not intended to limit the specific scope of the present invention. The scope of the present invention includes, but is not limited to, these specific embodiments. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A method for dynamically controlling the moisture content in a feed pelleting process, characterized in that, Includes the following steps: Material moisture data and equipment operation data are collected through distributed sensor nodes and transmitted to a remote control platform via a wireless communication network. Based on material moisture data at different spatial locations, a spatial propagation function for material moisture along the granulation process is established on a remote control platform. This spatial propagation function describes the evolution of moisture values ​​as the material flows from upstream to downstream nodes. Based on the spatial propagation function and the material moisture data of the upstream node at the current moment, calculate the predicted moisture value of the material when it arrives at the downstream node at a future moment, and generate an advance control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value. The communication delay time from the generation of the advanced control command to its issuance to the downstream node for execution is measured. The change in material moisture within the communication delay time is calculated based on the spatial propagation function. The control quantity of the advanced control command is compensated and corrected, and a delay compensation control signal is generated and issued to the downstream node for execution. The coupling effect of downstream nodes executing delay compensation control signals on the moisture content of materials in adjacent nodes is calculated based on the spatial propagation function, and coordinated control instructions for adjacent nodes are generated based on the coupling effect.

2. The method according to claim 1, characterized in that, Based on material moisture data from different spatial location nodes on the remote control platform, a spatial propagation function for material moisture along the granulation process is established, including: Obtain material moisture data and location coordinates of nodes at different spatial locations in the granulation process, sort the nodes at different spatial locations according to the material flow direction based on the location coordinates, and generate a node sequence; Select two adjacent spatial location nodes from the node sequence to form a node pair, and extract the material moisture data and location coordinates of the two spatial location nodes in the node pair; The material moisture difference is calculated based on the material moisture data of the two spatial nodes in the node pair. The position coordinate difference is calculated based on the position coordinates of the two spatial nodes in the node pair. The unit distance moisture change rate of the node pair is calculated based on the material moisture difference and the position coordinate difference. The unit distance moisture change rate of all node pairs in the node sequence is associated with the corresponding location coordinates to construct a discrete data set in which the moisture change rate varies with the location coordinates. The discrete data set is fitted with a function to generate a continuous function of the rate of change of moisture with respect to the location coordinates. The continuous function is then integrated along the location coordinates to generate a spatial propagation function.

3. The method according to claim 2, characterized in that, The discrete data set is fitted with a function to generate a continuous function of the rate of change of moisture with respect to location coordinates. This continuous function is then integrated along the location coordinates to generate a spatial propagation function, including: Identify node pairs in a discrete data set that exhibit abnormal fluctuations in the rate of change of moisture per unit distance, extract the rate of change of moisture per unit distance from adjacent normal node pairs of the abnormal node pairs, smooth them out, replace the rate of change of moisture per unit distance from the abnormal node pairs, and generate a corrected discrete data set. Extract the position coordinates of each node pair in the corrected discrete data set and the corresponding rate of change of moisture per unit distance. Establish a polynomial fitting equation with position coordinates as the independent variable and rate of change of moisture per unit distance as the dependent variable. Generate a continuous function of rate of change of moisture per unit distance with respect to position coordinates by performing function fitting on the corrected discrete data set. Obtain the initial moisture value of the material at the starting point of the granulation process, perform an integral operation on the continuous function along the position coordinates from the starting point, and superimpose the integral result with the initial moisture value of the material to generate a spatial propagation function.

4. The method according to claim 1, characterized in that, Based on the spatial propagation function and the material moisture data of the upstream node at the current moment, the predicted moisture value of the material when it arrives at the downstream node at a future moment is calculated. Then, based on the deviation between the predicted moisture value and the preset target moisture value, advanced control instructions for the downstream node are generated, including: Obtain the material moisture data and location coordinates of the upstream node at the current moment, and obtain the location coordinates and target moisture value of the downstream node; The spatial propagation function is input with the coordinates of the upstream node as the starting position, the current moisture content of the upstream node as the starting moisture value, and the coordinates of the downstream node as the ending position. The moisture evolution value of the material flowing from the upstream node to the downstream node is calculated by the spatial propagation function, and is used as the predicted moisture value of the material when it arrives at the downstream node in the future. The difference between the predicted moisture value and the target moisture value is calculated to obtain the deviation. The direction of moisture regulation at downstream nodes is determined based on the positive or negative direction of the deviation, including increasing or decreasing moisture. Based on the direction of moisture regulation, extract the controllable process parameters corresponding to the downstream nodes, and determine the adjustment amount of the process parameters based on the value of the deviation and the degree of influence of the controllable process parameters on the moisture content of the material. The control direction and process parameter adjustment amount are encapsulated to generate advanced control instructions for downstream nodes, which are then sent to downstream nodes for execution before a future time.

5. The method according to claim 1, characterized in that, The process involves measuring the communication delay from the generation of the advanced control command to its execution by downstream nodes, calculating the change in material moisture content within this delay period based on the spatial propagation function, compensating and correcting the control quantity of the advanced control command, generating a delay compensation control signal, and sending it to downstream nodes for execution. Record the start time of the advance control command generation and the arrival time of the command sent to the downstream node for execution, and calculate the time difference between the start time and the arrival time as the communication delay duration; Obtain the material location coordinates and material moisture data corresponding to the starting point of the communication delay duration. Use the material location coordinates as the delay start position, the material moisture data as the delay start moisture value, and the material arrival position coordinates corresponding to the communication delay end point determined according to the communication delay duration and material flow direction as the delay termination position input into the spatial propagation function. The moisture evolution difference of the material from the beginning position to the end position of the delay within the communication delay time is calculated by the spatial propagation function, and the moisture evolution difference is used as the change in material moisture within the communication delay time. Extract the control quantity from the advance control command, and superimpose the change in material moisture content within the communication delay time with the control quantity. When the change is positive, perform incremental compensation; when the change is negative, perform decremental compensation to obtain the compensated control quantity. The compensated control quantity is encapsulated to generate a delay-compensated control signal and sent to downstream nodes for execution.

6. The method according to claim 1, characterized in that, The coupling effect of downstream nodes executing delayed compensation control signals on the material moisture content of adjacent nodes is calculated based on the spatial propagation function. Based on this coupling effect, coordinated control instructions for adjacent nodes are generated, including: Obtain the position coordinates of the downstream node and the position coordinates of the adjacent node located after the downstream node; The measured material moisture data after the delay compensation control signal is sent to the downstream node for execution is obtained, along with the material moisture data before execution. The difference between the measured material moisture data and the material moisture data is calculated as the material moisture adjustment amount. The spatial propagation function is input with the coordinates of the downstream node as the coupling start point, the material moisture adjustment amount as the coupling start moisture change value, and the coordinates of the adjacent node as the coupling end point. The moisture coupling transfer value of the material moisture adjustment amount is calculated from the coupling start position to the coupling end position by the spatial propagation function. The moisture coupling transfer value is used as the coupling influence of the downstream node on the material moisture of the adjacent node after the downstream node executes the delay compensation control signal. Obtain the target moisture value of adjacent nodes, calculate the moisture deviation between the coupling influence amount and the target moisture value, and determine the direction and amount of coordinated regulation of adjacent nodes based on the value and positive / negative direction of the moisture deviation. The direction and amount of coordinated control are encapsulated to generate coordinated control instructions for adjacent nodes and then sent to the adjacent nodes for execution.

7. The method according to claim 6, characterized in that, The moisture coupling transfer value, calculated using the spatial propagation function, from the initial coupling position to the final coupling position, includes: Calculate the material flow distance between the coupling start position and the coupling end position, obtain the material flow velocity, and calculate the ratio of the material flow distance to the material flow velocity to obtain the material propagation time; Based on the material propagation time, extract the integral value of the rate of change of moisture from the coupling start position to the coupling end position from the spatial propagation function, and calculate the preliminary moisture coupling transfer value based on the integral value of the rate of change of moisture and the material moisture adjustment amount. Extract the measured material moisture data of adjacent nodes at multiple past times, calculate the fluctuation difference between the measured material moisture data at adjacent times, and statistically analyze the fluctuation difference to obtain the fluctuation amplitude; The ratio of the fluctuation amplitude to the initial moisture coupling transfer value is used as a correction coefficient. The initial moisture coupling transfer value is then corrected based on the correction coefficient to obtain the moisture coupling transfer value.

8. A dynamic moisture content control system for feed pelleting process, used to implement the method described in any one of claims 1-7, characterized in that, The system includes: The data acquisition module is used to collect material moisture data and equipment operation data through distributed sensor nodes and transmit them to the remote control platform through a wireless communication network. The function establishment module is used to establish a spatial propagation function of material moisture along the granulation process based on material moisture data at different spatial location nodes on the remote control platform. The spatial propagation function describes the evolution relationship of moisture value during the process of material flowing from upstream node to downstream node. The advanced control module is used to calculate the predicted moisture value of the material when it arrives at the downstream node in the future based on the spatial propagation function and the material moisture data of the upstream node at the current time, and to generate an advanced control instruction for the downstream node based on the deviation between the predicted moisture value and the preset target moisture value. The delay compensation module is used to measure the communication delay time from the generation of the advanced control command to its execution by the downstream node, calculate the change in material moisture within the communication delay time according to the spatial propagation function, compensate and correct the control quantity of the advanced control command, generate a delay compensation control signal and send it to the downstream node for execution. The collaborative control module is used to calculate the coupling effect of the downstream node's execution of the delay compensation control signal on the material moisture of the adjacent node based on the spatial propagation function, and to generate collaborative control instructions for the adjacent nodes based on the coupling effect.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions that, when executed by a processor, implement the steps of the method as described in any one of claims 1 to 7.

Images