Intelligent control system and method for filling production line

Abstract

The invention belongs to the technical field of intelligent manufacturing and industrial automation, and provides an intelligent control system and method for a filling production line, and the method comprises the steps: deploying an ultrasonic guided wave sensor array and a power analyzer, and achieving the high-frequency synchronous collection of the flow velocity and density of materials at multiple pipeline positions and the real-time power of a servo motor in a filling period; based on a Pearson's correlation coefficient and trend consistency value double-index system, quantifying synchronous characteristics of material residues and energy consumption; constructing a linear correlation model in combination with a least square method, and verifying a fitting degree through a decision coefficient; turbulence dead angle areas are dynamically divided; and fusion control is carried out through a flow velocity compensation algorithm and a self-adaptive PID. The system comprises a data analysis module, a fitting division module, a dead angle optimization module and an energy consumption regulation and control module. And control over material residues and optimization of energy efficiency of the filling system are effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing and industrial automation technology, specifically an intelligent control system and method for a filling production line. Background Technology

[0002] Automated filling production is widely used in many modern technological fields. High-precision material control and low-energy operation have become key indicators for measuring the core performance of filling systems. As production lines continue to improve their requirements for product qualification rate, raw material utilization rate and green production, the problem of coordinated optimization of the two core modules of filling system, namely material residue in pipelines and servo motor drive, is becoming increasingly prominent. Material residue not only causes raw material waste but also affects the energy consumption waste during the start-up and shutdown of servo motors. In the pipeline transmission process of the filling system, material residue not only causes raw material waste, but also exacerbates the energy consumption burden of the servo motor through a chain reaction of "residue-load-energy consumption". When the amount of material residue in the pipeline increases, the total mass of the actual material in the pipeline increases, which will significantly increase the load resistance of the servo motor during the start-up and stop phases. This will cause the motor to output higher power to complete the start-up and stop actions, directly causing an additional energy consumption increase of more than 30% during the start-up and stop phases. Currently, there are obvious deficiencies in the industry's control over pipeline material residue and the handling of its correlation with energy consumption.

[0003] Therefore, the present invention provides an intelligent control system and method for a filling production line. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0005] The technical solution adopted by this invention to solve its technical problem is: an intelligent control method for a filling production line, comprising: Collect pipeline material-related data and servo motor start-stop values ​​during multiple filling cycles. Perform residual analysis based on pipeline material-related data to obtain material residual values. Determine whether there is a synchronous characteristic between the pipeline material residual values ​​and energy consumption values ​​based on the pipeline material residual values ​​and energy consumption values. If synchronization characteristics exist, a model relating residual amount to motor energy consumption is established and the fit is evaluated. If the fit is good, dead zone analysis is performed based on the model relating residual amount to motor energy consumption to obtain turbulent dead zone values. Turbulent dead zone regions are then divided based on the turbulent dead zone values. Based on the defined turbulent dead zone regions, velocity compensation analysis is performed to obtain compensation coefficients, and material flow velocity is optimized based on these compensation coefficients; the material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. If the expected results are met, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor regulation is calculated. The regulation of the servo motor is then evaluated based on the energy consumption optimization rate.

[0006] Furthermore, the process of performing the residue analysis is as follows: Obtain the total effective length of the pipeline, the pipeline radius, the material density, and the residual coefficient; Input the total effective length and radius of the pipeline into the formula for calculating the total effective volume of the pipeline to obtain the total effective volume of the pipeline; The residual value of the material is obtained by multiplying the total effective volume of the pipeline, the material density, and the residual coefficient.

[0007] Furthermore, the process of performing the aforementioned synchronous characteristic analysis is as follows: Obtain the material residual value sequence and energy consumption value sequence, and perform Pearson correlation analysis to obtain the Pearson correlation coefficient; A consistent trend was obtained by analyzing the co-directional changes of the material residual value sequence and the energy consumption value sequence. If both the Pearson correlation coefficient and the trend consistency value meet the requirements, it indicates that the material residual value sequence and the energy consumption value sequence have synchronous characteristics.

[0008] Furthermore, the process of performing the aforementioned same-direction change analysis is as follows: The residual change in each filling cycle is obtained by performing difference processing on the residual values ​​of adjacent materials in the material residual value sequence. The energy consumption change for each filling cycle is obtained by performing difference processing on adjacent energy consumption values ​​in the energy consumption value sequence. The ratio of the residual change to the energy consumption change in the corresponding filling cycle is used to obtain the same-direction change rate. Based on the same-direction change rate of each filling cycle, if the same-direction change rate is positive, the filling cycle is marked as a same-direction change filling cycle. The trend consistency value is obtained by comparing the total number of filling cycles with the total number of filling cycles that change in the same direction.

[0009] Furthermore, the process of performing the goodness-of-fit verification is as follows: Obtain the predicted energy consumption value for each filling cycle; The residual value is obtained by calculating the difference between the energy consumption value of each filling cycle and the predicted energy consumption value, and the residual sum of squares is obtained by calculating the sum of squares of the residual values ​​of each filling cycle. The energy consumption deviation value is calculated by comparing the energy consumption value of each filling cycle with the average energy consumption value. The total sum of squares is calculated by summing the energy consumption deviation values ​​of each filling cycle. The coefficient of determination is obtained by comparing the sum of squared residuals with the sum of squared totals. The goodness of fit is determined based on the coefficient of determination.

[0010] Furthermore, the process of performing the aforementioned blind spot region analysis is as follows: The absolute residual value is obtained by performing absolute value processing on the residual value, and the residual rate is obtained by processing the ratio of the absolute residual value to the predicted energy consumption value. A clustering model is constructed based on historical filling data within the historical filling cycle; The residual rate of each filling cycle is input into the clustering model to identify different filling cycles as abnormal energy consumption cycles and normal energy consumption cycles. The average material flow rate is obtained by summing and averaging the material flow velocities at different pipeline locations within a normal energy consumption cycle. The material flow rate deviation rate is obtained by deviation analysis based on the average material flow rate, and the density deviation value is obtained by density deviation analysis of the material density at different pipe outlets. The material flow velocity deviation rate and density deviation value at different pipeline locations are dimensionless and then multiplied to obtain the turbulence dead angle value.

[0011] Furthermore, the process of performing the aforementioned deviation analysis is as follows: The material flow rate deviation value is obtained by processing the difference between the material flow rate at different pipe locations and the average material flow rate during the abnormal energy consumption cycle. The material flow velocity deviation rate at different pipe locations is calculated by comparing the material flow velocity deviation value with the average material flow velocity.

[0012] Furthermore, the process of performing the density deviation analysis is as follows: The density deviation value is obtained by calculating the difference between the material density at different pipeline locations and the normal material density.

[0013] Furthermore, the process of performing the aforementioned flow velocity compensation analysis is as follows: Obtain the material flow velocity in the turbulent dead zone region of all abnormal energy consumption cycles; The material flow velocity in the pipeline in the turbulent dead zone is obtained during all normal energy consumption cycles. The average value of the material flow velocity in the pipeline during all normal energy consumption cycles is summed to obtain the normal flow velocity benchmark in the turbulent dead zone. Obtain the material velocity deviation rate of all turbulent dead zones within all abnormal energy consumption cycles, and take the maximum value as the maximum material velocity deviation rate; The maximum deviation threshold is obtained by calculating the difference between the maximum material flow rate deviation rate and the normal flow rate benchmark. The real-time material flow rate in the turbulent dead zone during the filling cycle is obtained, and the real-time flow rate deviation value is calculated by the difference between the real-time material flow rate and the normal flow rate benchmark. The compensation coefficient is calculated by multiplying the ratio of the real-time flow velocity deviation value to the maximum deviation threshold with the maximum compensation ratio.

[0014] An intelligent control system for a filling production line includes the following modules: Data analysis module: Collects relevant data on pipeline materials and energy consumption values ​​during the start-stop phase of the servo motor in multiple filling cycles; performs residual analysis based on the relevant data to obtain the residual value of the material; and determines whether there is a synchronous characteristic between the residual value of the material in the pipeline and the energy consumption value based on the residual value of the material in the pipeline and the energy consumption value. Fitting and partitioning module: If synchronization characteristics exist, a model related to residual amount and motor energy consumption is established and the fit is evaluated; if the fit is good, dead zone analysis is performed based on the model related to residual amount and motor energy consumption to obtain turbulent dead zone values; turbulent dead zone regions are partitioned based on turbulent dead zone values. Dead Zone Optimization Module: Based on the defined turbulent dead zone regions, flow velocity compensation analysis is performed to obtain compensation coefficients, and material flow velocity is optimized based on the compensation coefficients; the material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. Energy consumption control module: If the expected results are met, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor control is calculated. The control of the servo motor is evaluated based on the energy consumption optimization rate.

[0015] The beneficial effects of this invention are as follows: 1. Collect pipeline material-related data and servo motor start-up and shutdown phase energy consumption values ​​during multiple filling cycles. Perform residue analysis based on pipeline material-related data to obtain material residue values. Determine whether there is a synchronous characteristic between the pipeline material residue values ​​and energy consumption values. By correlating the synchronous characteristics of material residue and servo motor energy consumption, identify directions to reduce material waste and optimize motor energy consumption, thereby improving production efficiency and cost control. Utilize data models to locate and delineate pipeline turbulence dead zones, providing a scientific basis for subsequent targeted solutions to material residue problems and production process optimization. 2. Based on the defined turbulent dead zone regions, velocity compensation analysis is performed to obtain compensation coefficients. Material flow velocity is then optimized based on these compensation coefficients. The material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. Velocity compensation optimization effectively reduces the material residue in the turbulent dead zone regions and verifies whether it meets the standards, improving production quality and process stability. If it meets expectations, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters and calculate the energy consumption optimization rate after servo motor adjustment. The servo motor adjustment is evaluated based on the energy consumption optimization rate. Multi-objective optimization can be achieved by dynamically adjusting servo motor parameters, and the control effect can be clearly evaluated through the energy consumption optimization rate, ultimately achieving a dual guarantee of energy consumption reduction and control effectiveness. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart illustrating the steps of an intelligent control method for a filling production line according to an embodiment of the present invention. Figure 2 This is a logic diagram of an intelligent control system and method for a filling production line according to an embodiment of the present invention; Figure 3 This is a module diagram of an intelligent control system for a filling production line according to an embodiment of the present invention. Detailed Implementation

[0018] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0019] Example 1: Please refer to Figures 1-2 As shown in the embodiment of the present invention, an intelligent control method for a filling production line includes: Step 1: Collect relevant data on pipeline materials and energy consumption values ​​during the start-stop phase of the servo motor in multiple filling cycles. Perform residual analysis based on the relevant data on pipeline materials to obtain the residual material value. Determine whether there is a synchronous characteristic between the residual material value and the energy consumption value in the pipeline. In step one, the process of collecting pipeline material-related data and energy consumption values ​​during the start-up and shutdown phases of the servo motor across multiple filling cycles, and then performing residual analysis based on the pipeline material-related data to obtain the material residual value, is as follows: An array of ultrasonic guided wave sensors is deployed inside the pipeline to collect relevant data on pipeline materials during the filling cycle; Among them, pipeline material-related data includes material flow rate and material density at different points in the pipeline during the filling cycle; The sampling frequency of the ultrasonic guided wave sensor array is set to 1Hz (one set of data per second) based on the material flow characteristics of the pipeline. A power analyzer is connected to the servo motor to collect real-time power data during the start-stop phase of the servo motor in multiple filling cycles; The sampling frequency of the power analyzer is set to 1Hz. Preprocessing is performed on the collected pipeline material-related data and real-time power data; Understandably, the preprocessing process involves using the 3σ criterion to identify outliers in the collected pipeline material-related data and real-time power data, and then removing these outliers. Obtaining pipe structural parameters includes the pipe's inner diameter and total effective pipe length L. 总 The pipe radius r is calculated based on the pipe's inner diameter. It should be noted that the total effective length L of the pipeline 总This refers to the effective pipe length used to calculate the total effective volume of the pipe and the actual material being transported. Input the pipe radius and total effective pipe length into the formula for calculating the total effective pipe volume to obtain the total effective pipe volume V. 管 ; The formula for calculating the total effective volume of the pipeline is as follows: ; Where r is the pipe radius, L 总 This represents the total effective length of the pipeline; The material residual value is obtained by multiplying the total effective volume of the pipeline, the material density, and the residual coefficient. It should be noted that the residual coefficient was obtained by those skilled in the art through analysis of historical filling data for different materials; Obtain the material residue value for each filling cycle and construct a material residue value sequence; The real-time power data of each sampling point during the start-stop phase of each filling cycle after preprocessing are accumulated to obtain the energy consumption value of the start-stop phase in each filling cycle. An energy consumption value sequence is constructed based on the energy consumption values ​​during the start-up and shutdown phases of each filling cycle; In step one, the process of determining whether there is a synchronous characteristic between the residual material value and the energy consumption value in the pipeline is as follows: Based on the material residue value and energy consumption value corresponding to each filling cycle, a one-to-one correspondence is made between the material residue value sequence and the energy consumption value sequence; Based on the material residual value sequence and energy consumption value sequence, the Pearson correlation coefficient is calculated using the Pearson correlation coefficient calculation formula to quantify the linear correlation strength between the material residual value sequence and the energy consumption value sequence. If the Pearson correlation coefficient is positive, it indicates that there is a positive correlation between the material residual value series and the energy consumption value series; The residual value change for each filling cycle is obtained by performing difference processing on the residual values ​​of adjacent materials in the residual value sequence. The energy consumption change for each filling cycle is obtained by subtracting adjacent energy consumption values ​​in the energy consumption value sequence. The ratio of the residual change to the energy consumption change in the corresponding filling cycle is used to obtain the same-direction change rate. Based on the same-direction change rate of each filling cycle, if the same-direction change rate is positive, the filling cycle is marked as a same-direction change filling cycle; if the same-direction change rate is negative, the filling cycle is marked as a non-same-direction change filling cycle. The trend consistency value is obtained by comparing the total number of filling cycles with the total number of filling cycles that change in the same direction. It should be noted that the physical meaning of the trend consistency value is as follows: by statistically analyzing the proportion of filling cycles with the same direction of change, the trend consistency value quantifies the directional consistency between changes in material residue in the pipeline and changes in servo motor energy consumption during dynamic adjustment; it reflects the coordinated fluctuation characteristics of the two in most cycles of "increasing and decreasing together", and also forms a complementary verification with the Pearson correlation coefficient - when the correlation coefficient is positive and the trend consistency value exceeds the threshold, it indicates that residue and energy consumption not only have a linear correlation, but also show a stable same-direction change pattern in periodic changes. This dual characteristic provides data support for optimizing filling strategies, which helps to improve system energy efficiency and control material residue; If the Pearson correlation coefficient is positive and the trend consistency value is greater than the preset trend consistency threshold, it indicates that the residual value of materials in the pipeline and the energy consumption value have synchronous characteristics. Conversely, it does not possess synchronous characteristics; It should be noted that the preset trend consistency threshold was obtained by those skilled in the art through experiments based on historical data; Step 2: If synchronization characteristics exist, establish a model relating residual amount to motor energy consumption and evaluate the fit; if the fit is good, perform dead zone analysis based on the model relating residual amount to motor energy consumption to obtain turbulent dead zone values; divide turbulent dead zone regions based on turbulent dead zone values. In step two, if synchronization characteristics exist, the process of establishing a model relating residual amount to motor energy consumption and evaluating the goodness of fit is as follows: Based on the obtained material residual value sequence and energy consumption value sequence, and their synchronous characteristics, a linear correlation model is constructed using the least squares method. The linear correlation model constructed using the least squares method is as follows: The material residue value sequence and the energy consumption value sequence are averaged to obtain the average material residue value and the average energy consumption value, respectively. Input the material residue value and energy consumption value of different filling cycles, as well as the average material residue value and average energy consumption value, into the regression coefficient calculation formula to obtain the regression coefficient. The influence value is obtained by multiplying the residual value of the material with the regression coefficient, and the intercept is obtained by subtracting the average energy consumption value from the influence value. The core expression of the model relating material residue and motor energy consumption is obtained based on the regression coefficients and intercepts. The goodness of fit of the obtained model related to material residue and motor energy consumption was verified. The goodness-of-fit verification process is as follows: Input the material residue value of each filling cycle into the material residue amount and motor energy consumption correlation model to obtain the predicted energy consumption value of each filling cycle. The residual value is obtained by calculating the difference between the energy consumption value of each filling cycle and the predicted energy consumption value, and the residual sum of squares is obtained by calculating the sum of squares of the residual values ​​of each filling cycle. The energy consumption deviation value is calculated by comparing the energy consumption value of each filling cycle with the average energy consumption value. The total sum of squares is calculated by summing the energy consumption deviation values ​​of each filling cycle. The coefficient of determination is obtained by comparing the sum of squared residuals with the total sum of squares. Compare the coefficient of determination with a preset threshold for the coefficient of determination; If the coefficient of determination is greater than or equal to the preset threshold, it indicates that the model relating material residue and motor energy consumption fits well. If the coefficient of determination is less than the preset threshold, it indicates that the model relating material residue and motor energy consumption is poorly fitted. It should be noted that the preset decision coefficient threshold was set by those skilled in the art based on the technical characteristics of the field; Understandably, the benefits of obtaining the coefficient of determination are: the coefficient of determination can quantitatively evaluate the degree of fit of the model related to material residue and motor energy consumption to the actual data, and intuitively reflect the proportion of variation explained by the model through the ratio of the residual sum of squares to the total sum of squares; when the coefficient of determination reaches a preset threshold, it can be confirmed that the model has good predictive ability; at the same time, it supports the comparative verification of different models or parameters, avoiding misjudgments caused by insufficient fitting. In step two, if the fit is good, the dead zone region analysis is performed based on the residual amount and motor energy consumption correlation model to obtain the turbulent dead zone value; the process of dividing the turbulent dead zone region based on the turbulent dead zone value is as follows: The absolute residual value is obtained by performing absolute value processing on the obtained residual value, and the residual rate is obtained by performing ratio processing on the absolute residual value and the predicted energy consumption value. A clustering model is constructed based on historical filling data within the historical filling cycle; The residual rate of each filling cycle is input into the clustering model to identify different filling cycles as abnormal energy consumption cycles and normal energy consumption cycles. The average material flow rate is obtained by summing and averaging the material flow rates at different pipeline locations within a normal energy consumption cycle. The material flow rate deviation value is obtained by processing the difference between the material flow rate at different pipelines and the average material flow rate during the abnormal energy consumption cycle, and the material flow rate deviation rate at different pipelines is calculated by the ratio of the material flow rate deviation value to the average material flow rate. The density of material at different pipe locations is collected using an ultrasonic guided wave sensor array, and the density deviation value is calculated by comparing the material density at different pipe locations with the normal material density. Dimensionless processing was performed based on the material flow velocity deviation rate and density deviation value at different pipeline locations; The turbulence dead angle value is obtained by multiplying the material flow velocity deviation rate and density deviation value at different pipe locations. Compare the turbulence dead angle value with the preset turbulence dead angle threshold; If the turbulence dead angle value is greater than or equal to the preset turbulence dead angle threshold, then the pipe is marked as a pre-turbulence dead angle region; If the turbulence dead angle value is less than the preset turbulence dead angle threshold, then the pipe is marked as a normal turbulence area; Obtain the pre-turbulent dead zone regions marked by all abnormal energy consumption cycles; The number of times different pipe locations were marked as pre-turbulent dead zones was counted, and the results were arranged in descending order of marking frequency to obtain a frequency sequence; The top 20% of pipes that were marked as pre-turbulent dead zones were selected and officially marked as turbulent dead zones. The technical solution of this embodiment is as follows: Collect relevant data on pipeline materials and energy consumption values ​​during the start-stop phase of the servo motor within multiple filling cycles; perform residue analysis based on the relevant data to obtain the material residue value; determine whether there is a synchronous characteristic between the material residue value and energy consumption value in the pipeline; identify directions to reduce material waste and optimize motor energy consumption by associating the synchronous characteristics of material residue and servo motor energy consumption, thereby improving production efficiency and cost control; and locate and delineate pipeline turbulence dead zones based on the data model, providing a scientific basis for subsequent targeted solutions to material residue problems and optimization of the production process. Example 2: Please refer to Figure 1 As shown in the embodiment of the present invention, an intelligent control method for a filling production line includes: Step 3: Based on the defined turbulent dead zone regions, perform velocity compensation analysis to obtain compensation coefficients, and optimize material flow velocity based on the compensation coefficients; evaluate whether the material residue in the turbulent dead zone regions after optimization meets expectations. In step three, based on the defined turbulent dead zone regions, velocity compensation analysis is performed to obtain compensation coefficients. The process of optimizing material flow velocity based on these compensation coefficients is as follows: A velocity compensation benchmark database is established based on the velocity characteristics of all turbulent dead zones. The process of establishing the velocity compensation benchmark database is as follows: Extract the material flow velocity in all abnormal energy consumption cycle turbulent dead zone regions; The material flow velocity in the pipeline in the turbulent dead zone is obtained during all normal energy consumption cycles. The average value of the material flow velocity in the pipeline during all normal energy consumption cycles is summed to obtain the normal flow velocity benchmark in the turbulent dead zone. Obtain the material velocity deviation rate of all turbulent dead zones within all abnormal energy consumption cycles, and take the maximum value as the maximum material velocity deviation rate; The maximum deviation threshold V is obtained by calculating the difference between the maximum material flow rate deviation rate and the normal flow rate benchmark. max ; An ultrasonic guided wave sensor array is used to collect the real-time material flow rate in the turbulent dead zone during the filling cycle. The real-time material flow rate is calculated by the difference between the real-time material flow rate and the normal flow rate benchmark to obtain the real-time flow rate deviation value ∆V. If the real-time flow rate deviation is greater than or equal to 80% of the maximum deviation threshold, the flow rate compensation algorithm is triggered. The compensation coefficient is calculated by multiplying the ratio of the real-time flow velocity deviation value to the maximum deviation threshold with the maximum compensation ratio. It should be noted that the maximum compensation ratio is to avoid new turbulence problems caused by a sudden increase in material flow velocity, and to ensure that the compensation range is controlled within a safe range; The obtained real-time flow velocity deviation value and the maximum deviation threshold are input into the compensation coefficient calculation formula to obtain the compensation coefficient; The obtained compensation coefficients are input into the PLC controller, and the PLC controller generates a flow rate adjustment command. The real-time material flow rate in the turbulent dead zone is controlled by controlling the frequency conversion module of the material conveying pump. After control and adjustment, the real-time material flow rate in the turbulent dead zone area is continuously collected; If the flow rate stabilizes within ±5% of the normal reference value, compensation is stopped; if the target is still not met, the compensation coefficient is recalculated and iteratively adjusted until the target is met. In step three, the process of evaluating whether the material residue in the optimized turbulent dead zone region meets expectations is as follows: Obtain real-time material flow velocity data and real-time material density data for the turbulent dead zone region in multiple filling cycles after optimization; The optimized average material flow rate and optimized average material density are obtained by averaging the real-time material flow rate data and real-time material density data of the turbulent dead zone regions in multiple filling cycles. The difference between the optimized average material flow rate and the unoptimized average material flow rate is processed, and then the ratio of the optimized average material flow rate to the unoptimized average material flow rate is processed to obtain the material flow rate improvement rate in the turbulent dead zone region. Obtain the material residue value of the turbulent dead zone region before optimization and the material residue value of the turbulent dead zone region after optimization; The material residue value of the turbulent dead zone before optimization and the material residue value of the turbulent dead zone after optimization are processed by difference, and then the ratio is processed with the material residue value of the turbulent dead zone before optimization to obtain the residue reduction rate of the turbulent dead zone. The average residual reduction rate is obtained by summing the residual reduction rates of all turbulent dead zones and taking the mean. The material flow rate increase rate is compared with the preset increase rate threshold, and the average residue reduction rate is compared with the preset reduction rate threshold. If the material flow rate increase rate is greater than or equal to the preset increase rate threshold, and the average residue reduction rate is greater than or equal to the preset reduction rate threshold, then the optimized turbulent dead zone region meets expectations. Conversely, it indicates that the optimized turbulent dead zone region does not meet expectations; Step 4: If the expected results are met, construct a multi-objective optimization model to dynamically adjust the servo motor control parameters and calculate the energy consumption optimization rate after servo motor adjustment; evaluate the servo motor adjustment based on the energy consumption optimization rate. In step four, if the expected results are met, the process of constructing a multi-objective optimization model to dynamically adjust the servo motor control parameters and calculating the energy consumption optimization rate after servo motor adjustment is as follows: Based on the optimized material flow rate improvement rate and average residue reduction rate, the core basis for servo motor parameter adjustment is used, and a multi-objective optimization model is constructed using an adaptive PID control algorithm based on neural networks. The architecture of the multi-objective optimization model is designed as follows: A three-layer feedforward neural network (input layer - hidden layer - output layer) is adopted. The input layer receives key parameters such as the material flow rate increase rate and the average residue reduction rate. The hidden layer enhances the nonlinear fitting capability through the ReLU activation function, and the output layer generates dynamic compensation coefficient S and servo motor control parameter adjustment. An integrated PID control module is used to fuse the compensation coefficients output by the neural network with the PID parameters in real time to form an adaptive control strategy. Obtain the servo motor control parameter adjustment amount output by the multi-objective optimization model; Input the servo motor control parameter adjustment amount into the PLC controller, so that the PLC controller controls the servo driver to modify the servo motor motion state in real time. For example, the target power during the start-stop phase of the servo motor can be adjusted to avoid redundant energy consumption caused by excessive power, while ensuring reliable start-stop action; and the acceleration time of the servo motor can be shortened to reduce excessive power output loss. The real-time power data of the servo motor during the start-stop phase after regulation is collected by a power analyzer. The real-time power data of each sampling point is accumulated and calculated to obtain the energy consumption value of the servo motor during the start-stop phase after regulation. The energy consumption optimization rate of the servo motor after regulation is obtained by calculating the difference between the energy consumption value before and after regulation and then proportionally calculating the energy consumption value before regulation. In step four, the process of evaluating the control of the servo motor based on the energy consumption optimization rate is as follows: The energy consumption optimization rate is compared with the preset optimization benchmark value; If the energy consumption optimization rate is greater than or equal to the preset optimization benchmark value, it means that the energy consumption of the servo motor is qualified. If the energy consumption optimization rate is less than the preset optimization benchmark value, it means that the energy consumption of the servo motor is not up to standard. The technical solution of this embodiment is as follows: Based on the divided turbulent dead zone region, flow velocity compensation analysis is performed to obtain the compensation coefficient, and the material flow velocity is optimized based on the compensation coefficient; the material residue in the turbulent dead zone region after optimization is evaluated to see if it meets expectations; the flow velocity compensation optimization can effectively reduce the material residue in the turbulent dead zone region and verify whether it meets the standard, thereby improving production quality and process stability; if it meets expectations, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor regulation is calculated. The regulation of the servo motor is evaluated based on the energy consumption optimization rate; multi-objective optimization can be achieved by dynamically adjusting the servo motor parameters, and the regulation effect can be clearly evaluated through the energy consumption optimization rate, ultimately achieving a dual guarantee of energy consumption reduction and regulation effectiveness.

[0020] Example 3: Please refer to Figure 3 As shown in the embodiment of the present invention, an intelligent control system for a filling production line includes the following modules: Data analysis module: Collects relevant data on pipeline materials and energy consumption values ​​during the start-stop phase of the servo motor in multiple filling cycles; performs residual analysis based on the relevant data to obtain the residual value of the material; and determines whether there is a synchronous characteristic between the residual value of the material in the pipeline and the energy consumption value based on the residual value of the material in the pipeline and the energy consumption value. Fitting and partitioning module: If synchronization characteristics exist, a model related to residual amount and motor energy consumption is established and the fit is evaluated; if the fit is good, dead zone analysis is performed based on the model related to residual amount and motor energy consumption to obtain turbulent dead zone values; turbulent dead zone regions are partitioned based on turbulent dead zone values. Dead Zone Optimization Module: Based on the defined turbulent dead zone regions, flow velocity compensation analysis is performed to obtain compensation coefficients, and material flow velocity is optimized based on the compensation coefficients; the material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. Energy consumption control module: If the expected results are met, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor control is calculated. The control of the servo motor is evaluated based on the energy consumption optimization rate.

[0021] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent control of a filling production line, characterized in that: include: Collect pipeline material-related data and servo motor start-stop values ​​during multiple filling cycles. Perform residual analysis based on pipeline material-related data to obtain material residual values. Determine whether there is a synchronous characteristic between the pipeline material residual values ​​and energy consumption values ​​based on the pipeline material residual values ​​and energy consumption values. If synchronization characteristics exist, a model relating residual amount to motor energy consumption is established and the fit is evaluated. If the fit is good, dead zone analysis is performed based on the model relating residual amount to motor energy consumption to obtain turbulent dead zone values. Turbulent dead zone regions are then divided based on the turbulent dead zone values. Based on the defined turbulent dead zone regions, velocity compensation analysis is performed to obtain compensation coefficients, and material flow velocity is optimized based on these compensation coefficients; the material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. If the expected results are met, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor regulation is calculated. The regulation of the servo motor is then evaluated based on the energy consumption optimization rate.

2. The intelligent control method for a filling production line according to claim 1, characterized in that: The process of performing the residual analysis is as follows: Obtain the total effective length of the pipeline, the pipeline radius, the material density, and the residual coefficient; Input the total effective length and radius of the pipeline into the formula for calculating the total effective volume of the pipeline to obtain the total effective volume of the pipeline; The residual value of the material is obtained by multiplying the total effective volume of the pipeline, the material density, and the residual coefficient.

3. The intelligent control method for a filling production line according to claim 1, characterized in that: The process of performing the aforementioned synchronization characteristic analysis is as follows: Obtain the material residual value sequence and energy consumption value sequence, and perform Pearson correlation analysis to obtain the Pearson correlation coefficient; A consistent trend was obtained by analyzing the co-directional changes of the material residual value sequence and the energy consumption value sequence. If both the Pearson correlation coefficient and the trend consistency value meet the requirements, it indicates that the material residual value sequence and the energy consumption value sequence have synchronous characteristics.

4. The intelligent control method for a filling production line according to claim 3, characterized in that: The process of performing the aforementioned same-direction change analysis is as follows: The residual change in each filling cycle is obtained by performing difference processing on the residual values ​​of adjacent materials in the material residual value sequence. The energy consumption change for each filling cycle is obtained by performing difference processing on adjacent energy consumption values ​​in the energy consumption value sequence. The ratio of the residual change to the energy consumption change in the corresponding filling cycle is used to obtain the same-direction change rate. Based on the same-direction change rate of each filling cycle, if the same-direction change rate is positive, the filling cycle is marked as a same-direction change filling cycle. The trend consistency value is obtained by comparing the total number of filling cycles with the total number of filling cycles that change in the same direction.

5. The intelligent control method for a filling production line according to claim 1, characterized in that: The process of performing the goodness-of-fit verification is as follows: Obtain the predicted energy consumption value for each filling cycle; The residual value is obtained by calculating the difference between the energy consumption value of each filling cycle and the predicted energy consumption value, and the residual sum of squares is obtained by calculating the sum of squares of the residual values ​​of each filling cycle. The energy consumption deviation value is calculated by comparing the energy consumption value of each filling cycle with the average energy consumption value. The total sum of squares is calculated by summing the energy consumption deviation values ​​of each filling cycle. The coefficient of determination is obtained by comparing the sum of squared residuals with the sum of squared totals. The goodness of fit is determined based on the coefficient of determination.

6. The intelligent control method for a filling production line according to claim 1, characterized in that: The process of performing the aforementioned blind spot region analysis is as follows: The absolute residual value is obtained by performing absolute value processing on the residual value, and the residual rate is obtained by processing the ratio of the absolute residual value to the predicted energy consumption value. A clustering model is constructed based on historical filling data within the historical filling cycle; The residual rate of each filling cycle is input into the clustering model to identify different filling cycles as abnormal energy consumption cycles and normal energy consumption cycles. The average material flow rate is obtained by summing and averaging the material flow velocities at different pipeline locations within a normal energy consumption cycle. The material flow rate deviation rate is obtained by deviation analysis based on the average material flow rate, and the density deviation value is obtained by density deviation analysis of the material density at different pipe outlets. The material flow velocity deviation rate and density deviation value at different pipeline locations are dimensionless and then multiplied to obtain the turbulence dead angle value.

7. The intelligent control method for a filling production line according to claim 6, characterized in that: The process of performing the aforementioned deviation analysis is as follows: The material flow rate deviation value is obtained by processing the difference between the material flow rate at different pipe locations and the average material flow rate during the abnormal energy consumption cycle. The material flow velocity deviation rate at different pipe locations is calculated by comparing the material flow velocity deviation value with the average material flow velocity.

8. The intelligent control method for a filling production line according to claim 6, characterized in that: The process of performing the density deviation analysis is as follows: The density deviation value is obtained by calculating the difference between the material density at different pipeline locations and the normal material density.

9. The intelligent control method for a filling production line according to claim 1, characterized in that: The process of performing the aforementioned flow velocity compensation analysis is as follows: Obtain the material flow velocity in the turbulent dead zone region of all abnormal energy consumption cycles; The material flow velocity in the pipeline in the turbulent dead zone is obtained during all normal energy consumption cycles. The average value of the material flow velocity in the pipeline during all normal energy consumption cycles is summed to obtain the normal flow velocity benchmark in the turbulent dead zone. Obtain the material velocity deviation rate of all turbulent dead zones within all abnormal energy consumption cycles, and take the maximum value as the maximum material velocity deviation rate; The maximum deviation threshold is obtained by calculating the difference between the maximum material flow rate deviation rate and the normal flow rate benchmark. The real-time material flow rate in the turbulent dead zone during the filling cycle is obtained, and the real-time flow rate deviation value is calculated by the difference between the real-time material flow rate and the normal flow rate benchmark. The compensation coefficient is calculated by multiplying the ratio of the real-time flow velocity deviation value to the maximum deviation threshold with the maximum compensation ratio.

10. An intelligent control system for a filling production line, used to implement the intelligent control method for a filling production line as described in any one of claims 1-9, characterized in that, Includes the following modules: Data analysis module: Collects relevant data on pipeline materials and energy consumption values ​​during the start-stop phase of the servo motor in multiple filling cycles; performs residual analysis based on the relevant data to obtain the residual value of the material; and determines whether there is a synchronous characteristic between the residual value of the material in the pipeline and the energy consumption value based on the residual value of the material in the pipeline and the energy consumption value. Fitting and partitioning module: If synchronization characteristics exist, a model related to residual amount and motor energy consumption is established and the fit is evaluated; if the fit is good, dead zone analysis is performed based on the model related to residual amount and motor energy consumption to obtain turbulent dead zone values; turbulent dead zone regions are partitioned based on turbulent dead zone values. Dead Zone Optimization Module: Based on the defined turbulent dead zone regions, flow velocity compensation analysis is performed to obtain compensation coefficients, and material flow velocity is optimized based on the compensation coefficients; the material residue in the turbulent dead zone regions after optimization is evaluated to see if it meets expectations. Energy consumption control module: If the expected results are met, a multi-objective optimization model is constructed to dynamically adjust the servo motor control parameters, and the energy consumption optimization rate after servo motor control is calculated. The control of the servo motor is evaluated based on the energy consumption optimization rate.

Images