A method and system for controlling the operation of a pesticide production line

Abstract

The application relates to the technical field of automatic control of pesticide production, and discloses a pesticide production line operation process control method and system, which comprises the following steps: obtaining operation state parameters of a material conveying system, at least including driving parameters, fluid physical parameters and execution parameters; determining a distribution stage of the material in a conveying pipeline during a material switching process based on the fluid physical parameters; carrying out signal decoupling on the fluid physical parameters to obtain a trend characteristic component reflecting global flow resistance evolution and a disturbance characteristic component reflecting local material characteristic difference; identifying a pressure abnormality source type in combination with the distribution stage and the two types of characteristic components; determining target values of the driving parameters and the execution parameters according to the abnormality type; synchronously issuing a correction instruction according to the target values and limiting a change rate of the correction instruction, so that the filling end inlet pressure is maintained within a safe range. Through signal decoupling and hierarchical control, the application effectively suppresses pressure pulses in the switching process, and guarantees the consistency of filling metering.

Description

Technical Field

[0001] This invention relates to the field of pesticide production automation control technology, and in particular to a method and system for controlling the operation of a pesticide production line. Background Technology

[0002] The co-production of multiple pesticide varieties on the same production line requires frequent material switching. Existing conventional solutions use a fixed feed pump speed-filling speed matching table, simply calling a preset speed setting when switching products to maintain stable feed pressure and flow rate. This method is suitable for continuous production of a single product. However, when switching from low-viscosity to high-viscosity pesticides, the actual material viscosity may be higher than the nominal value. In addition, residual low-viscosity liquid in the pipeline forms a mixing transition zone, gradually increasing pipeline flow resistance and causing a continuous rise in pump outlet pressure. The system continues to use a fixed speed setting without adjusting according to the actual pipeline pressure. The rising pressure causes the actual feed flow rate to exceed expectations, resulting in excessive filling inlet pressure, triggering intermittent pressure relief from the safety valve, and causing periodic pulse fluctuations in the feed pressure: excessively rapid feeding before pressure relief and insufficient feeding during the pressure relief phase. The existing control relies solely on fixed factory parameters, which cannot be dynamically adjusted according to the actual state of the material and the complex flow resistance in the transition section. It cannot identify the hidden pressure accumulation caused by viscosity differences and pipeline residues, which ultimately causes the supply pressure to repeatedly touch the safety valve opening pressure, resulting in pulsed supply, which leads to fluctuations in the filling volume of a single bottle and poor consistency in filling metering. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention discloses a method and system for controlling the operation of a pesticide production line. The aim is to solve the technical problems of inconsistent feeding pressure pulses and filling metering caused by differences in material viscosity and pipeline residues during the switching process of multi-variety pesticide production on the same line.

[0004] The technical solution of the present invention is as follows: In a first aspect, the present invention discloses a method for controlling the operation process of a pesticide production line, used in a material conveying system of a pesticide production line. The material conveying system includes a power end, a conveying pipeline, and a filling end, comprising: Obtain the operating status parameters of the material conveying system. The operating status parameters include at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. Based on fluid physical parameters, the distribution stages in the conveying pipeline during the material switching process are determined. The distribution stages include the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. The fluid physical parameters are decoupled from the signal to obtain the trend characteristic components that reflect the evolution of the global flow resistance of the pipeline, as well as the disturbance characteristic components that reflect the differences in local material properties. By combining the distribution stage, trend characteristic components, and disturbance characteristic components, the current source type of pressure anomaly can be identified; Based on the type of pressure anomaly source, determine the target values ​​of the driving parameters and the target values ​​of the execution parameters; Based on the target values ​​of the driving parameters and the execution parameters, correction commands are issued synchronously to the power end and the filling end, and the rate of change of the correction commands is limited so that the inlet pressure of the filling end is maintained within the preset safe range.

[0005] This technical solution enables precise location of pressure anomalies by dynamically identifying material distribution stages and decoupling fluid signals. It also allows for dynamic adjustment of drive and execution parameters based on the anomaly type, effectively suppressing pressure pulse fluctuations during switching and ensuring consistent filling and metering.

[0006] Furthermore, the steps for determining the distribution stage within the delivery pipeline during material switching based on fluid physical parameters include: Calculate the cumulative conveying volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters; The cumulative transport volume is compared with the preset pipeline volume; When the cumulative conveying volume is greater than the second proportion of the pipeline volume, and the pressure fluctuation amplitude in the fluid physical parameters is less than the stable threshold within a preset time, the distribution stage is determined to be the new material-dominated conveying stage. When the cumulative conveying volume is between the first and second proportions of the pipeline volume, or when the cumulative conveying volume is greater than the second proportion of the pipeline volume but the pressure fluctuation amplitude is not less than the stable threshold, the distribution stage is determined to be the new and old mixed transition material stage; the first proportion is less than the second proportion. Otherwise, the distribution stage is determined to be the residual old material stage.

[0007] Furthermore, the steps for calculating the cumulative conveying volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters include: Acquire the output torque fluctuation signal at the power end and the inlet pressure pulsation signal at the filling end; Calculate the phase hysteresis angle between the output torque fluctuation signal and the inlet pressure pulsation signal; The fluid compressibility compensation coefficient is determined based on the magnitude of the phase lag angle. The instantaneous flow rate is weighted and corrected using the fluid compressibility compensation coefficient to obtain the corrected instantaneous flow rate; Using the start time of material switching as the integration starting point, a trapezoidal integral is performed on the corrected instantaneous flow rate to obtain the cumulative conveying volume.

[0008] Furthermore, the steps for obtaining the preset pipeline volume include: The internal volume of the straight pipe section is calculated based on the physical topology of the pipeline, and the pipeline volume is obtained by combining the fluid dynamic equivalent volume compensation value of the pipeline accessories. The steps for determining the distribution stage within the delivery pipeline during material switching based on fluid physical parameters also include: The variance of pressure values ​​in the fluid physical parameters within the dynamic observation window is calculated, and dynamic cross-validation is performed on the distribution stage by combining the first derivative of the output torque at the power end.

[0009] Furthermore, the steps of decoupling the fluid physical parameters to obtain the trend characteristic components reflecting the evolution of the global flow resistance of the conveying pipeline, and the disturbance characteristic components reflecting the differences in local material properties include: A dual-channel morphological filtering algorithm is used to decouple the pressure signal in the fluid physical parameters into a low-frequency baseline component and a high-frequency pulse component; The low-frequency baseline component was identified as the trend characteristic component, and the high-frequency pulse component was identified as the disturbance characteristic component.

[0010] Furthermore, the types of sources of pressure anomalies include persistent increases in flow resistance, short-term local disturbances, and composite anomalies resulting from the superposition of the two. The steps to identify the current source type of pressure anomaly by combining distribution stage, trend characteristic components, and disturbance characteristic components include: Calculate the first and second derivatives of the pressure difference between the two ends of the conveying pipeline and the flow surplus rate of the material conveying system; By combining distribution stage, trend characteristic components, disturbance characteristic components, first derivative, second derivative, and flow surplus rate, the source type of pressure anomaly can be identified.

[0011] Furthermore, the steps for determining the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source include: When the source of the pressure anomaly is a continuous increase in flow resistance or a complex anomaly, the maximum safe feed flow rate of the conveying pipeline is calculated back based on the trend characteristic components. The theoretical execution parameters are calculated based on the maximum safe feeding flow rate, and then rounded down to the discrete level allowed at the filling end to obtain the target value of the execution parameters. Determine the target values ​​of the drive parameters based on the target values ​​of the execution parameters and the maximum safe feed flow rate; When the source of the pressure anomaly is a composite anomaly, a dynamic bias is also superimposed on the target value of the driving parameter according to the disturbance characteristic component to suppress local disturbances, thereby obtaining the final target value of the driving parameter. When the source of the pressure anomaly is a short-term local disturbance, the target value of the execution parameter remains unchanged, and a limited-amplitude mitigation bias is applied to the driving parameter.

[0012] Furthermore, the steps of synchronously issuing correction commands to the power end and filling end based on the target values ​​of the driving parameters and the execution parameters, and limiting the rate of change of the correction commands to maintain the inlet pressure of the filling end within a preset safety range, include: Correction instructions for the power end and filling end are generated based on the target values ​​of the driving parameters and the execution parameters. Correction commands for the power supply and filling ends are issued synchronously within the same programmable logic controller scan cycle; The correction instruction is subjected to a first-order low-pass filter to limit its rate of change and make the correction instruction transition smoothly. The execution effect is continuously monitored through a closed-loop feedback correction mechanism, and the Smith predictive compensator is used to compensate for pipeline lag, so that the inlet pressure at the filling end is maintained within the preset safe range.

[0013] Furthermore, the method also includes: When the pressure difference between the two ends of the delivery pipeline increases exponentially and the flow rate drops to the preset blockage threshold, the safety fallback control is triggered. Safety fallback controls include: cutting off the power end's operation enable signal, stopping the filling end's operation, closing the filling end's inlet shut-off valve, and opening the pressure relief bypass.

[0014] Secondly, the present invention also discloses a pesticide production line operation process control system for performing the steps in the above method, including: The parameter acquisition module is used to acquire the operating status parameters of the material conveying system. The operating status parameters include at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. The distribution determination module is used to determine the distribution stage in the conveying pipeline during the material switching process based on fluid physical parameters. The distribution stages include the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. The feature processing module is used to decouple the fluid physical parameters from the signal, obtain the trend feature components that reflect the evolution of the global flow resistance of the conveying pipeline, and the disturbance feature components that reflect the differences in local material properties. The anomaly identification module is used to identify the current source type of pressure anomaly by combining the distribution stage, trend feature components, and disturbance feature components. The dynamic adjustment module is used to determine the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source. The instruction issuing module is used to synchronously issue correction instructions to the power end and the filling end based on the target values ​​of the driving parameters and the execution parameters, and to limit the rate of change of the correction instructions so that the inlet pressure of the filling end is maintained within a preset safe range.

[0015] In summary, the present invention provides a method and system for controlling the operation of a pesticide production line. The method acquires the operating status parameters of the material conveying system and dynamically identifies the distribution stages during material switching based on fluid physical parameters. Combined with signal decoupling technology, it obtains global flow resistance trends and local disturbance characteristics, enabling accurate identification of pressure anomaly sources. Based on this, it dynamically adjusts the parameters of the power and filling ends according to the anomaly type and simultaneously issues correction commands. By limiting the rate of change and using closed-loop feedback compensation, it effectively solves the problem of pressure pulse fluctuations caused by viscosity differences and pipeline residues during the switching of multiple pesticide varieties in a co-production line. This significantly improves the stability and consistency of filling and metering, and extends the service life of equipment such as safety valves. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a pesticide production line operation control method provided in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the structure of a pesticide production line operation process control system provided in an embodiment of the present invention.

[0018] Labeling Explanation: 210, Parameter Acquisition Module; 220, Distribution Determination Module; 230, Feature Processing Module; 240, Anomaly Identification Module; 250, Dynamic Adjustment Module; 260, Command Issuance Module. Detailed Implementation

[0019] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some, not all, of the embodiments of this invention. The components of this invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] In the co-production of multiple pesticides, when switching from low-viscosity aqueous solutions to high-viscosity suspensions, the new material has a higher viscosity, while low-viscosity old material remains in the pipeline, creating a mixing transition zone. As the high-viscosity material is advanced, the frictional resistance of the pipeline foundation continuously accumulates, leading to an increase in the outlet pressure of the feed pump. This forces the liquid to enter the filling line at an excessively high flow rate, causing the safety valve to open intermittently to release pressure, resulting in periodic pulse fluctuations in the feed pressure. The filling machine's feed speed fluctuates erratically, severely disrupting the consistency of single-bottle filling and metering. Simultaneously, the incompletely mixed low-viscosity old material clumps in the pipeline form localized blockage flows, causing the pressure difference signal to exhibit a composite pattern of overall monotonic increase superimposed with high-frequency pulses. Conventional fixed-position control logic cannot recognize this implicit pressure accumulation and localized disturbance caused by viscosity differences and pipeline residues, easily inducing high-frequency oscillations in the actuator.

[0022] Firstly, please see Figure 1 This invention discloses a method for controlling the operation of a pesticide production line, used in the material conveying system of a pesticide production line. The material conveying system includes a power end, a conveying pipeline, and a filling end, comprising: 1. Obtain the operating status parameters of the material conveying system. The operating status parameters include at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. 2. Based on fluid physical parameters, determine the distribution stages in the conveying pipeline during the material switching process. The distribution stages include the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. 3. Decouple the fluid physical parameters from the signal to obtain the trend characteristic components that reflect the evolution of the global flow resistance of the conveying pipeline, as well as the disturbance characteristic components that reflect the differences in local material properties. 4. Identify the current source type of pressure anomaly by combining the distribution stage, trend characteristic components, and disturbance characteristic components; 5. Determine the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source; 6. Based on the target values ​​of the driving parameters and the execution parameters, synchronously issue correction commands to the power end and the filling end, and limit the rate of change of the correction commands so that the inlet pressure of the filling end is maintained within the preset safety range.

[0023] Specifically, the power end refers to the material feeding drive equipment, such as variable frequency conveying pumps or screw pumps, which provide power for material flow and determine the initial pressure head and propulsion rhythm of the pipeline. Its driving parameters can reflect the equipment's workload and indirectly indicate changes in the material state within the pipeline. For example, an increase in current or enhanced torque fluctuations usually means increased flow resistance or poorer material uniformity.

[0024] The filling end refers to the final filling equipment, such as an automatic filling machine, which is responsible for quantitatively injecting materials. Its execution rhythm, valve opening, and servo speed changes will inversely affect the instantaneous back pressure of the upstream pipeline. Therefore, the filling end is both a controlled object and a factor involved in pressure stability.

[0025] Signal decoupling is the process of separating a composite pressure signal into slow-changing and fast-changing components. The slow-changing component corresponds to the overall flow resistance evolution caused by the advancement of new material, while the fast-changing component corresponds to short-term disturbances caused by residual old material, slug flow, and transient valve actions. By separating these components, the control logic can separately determine persistent risks and local transient risks, avoiding the confusion between the two types of anomalies.

[0026] In practical implementation, regarding the core technical means of acquiring the operating status parameters of the material conveying system, in one possible example, the output frequency and current can be read by a frequency converter as drive parameters; high-frequency pressure transmitters and mass flow meters can be installed at the pump outlet and filling machine inlet to acquire pressure and instantaneous flow rate as fluid physical parameters; and the valve opening or servo speed can be read by the filling machine PLC as execution parameters. In actual deployment, the drive parameters are preferably read from the feedback values ​​of the communication registers of the frequency converter or servo drive, rather than manually entered set values, to truly reflect load changes and transient deviations. The pump outlet pressure transmitter reflects the upstream pressure establishment, and the filling machine inlet pressure transmitter reflects the end-pressure state; the two work together to form a pressure judgment along the pipeline. The mass flow meter is preferably placed at a stable full-pipe position to reduce interference from bubbles or backflow. The execution parameters at the filling end are read from the corresponding feedback values ​​according to the structural type. To avoid data misalignment, the control system sets a unified refresh cycle to synchronously package the latest feedback values ​​of each device, ensuring that the data used for subsequent stage judgments and anomaly identification belong to the same moment.

[0027] Regarding the core technical approach of determining the distribution stages of materials during material switching within the conveying pipeline based on fluid physics parameters, in a possible example, instantaneous flow data from a mass flow meter is integrated over time to estimate the cumulative conveying volume. This volume is then compared with a pre-recorded standard pipeline volume to initially divide the process into three stages: residual old material, a transitional mixture of old and new material, and new material dominating. The standard pipeline volume is established during the commissioning phase, statistically analyzing the effective volume of all pipe sections and accessories from the power outlet to the filling inlet. Complex components can be calibrated through liquid injection. The switching start point is defined as the moment the upstream switching valve completes and is confirmed by the control system. The system first makes a preliminary stage judgment based on volume ratios, then corrects it by incorporating subsequent pressure fluctuation characteristics. This approach retains the intuitiveness of the volumetric method while avoiding misjudgments in high-viscosity residual scenarios.

[0028] The core technical approach of signal decoupling for fluid physics parameters can be illustrated by the following example: After acquiring the continuous pressure signal from a high-frequency pressure transmitter, a standard digital filter can be used for processing: a low-pass filter extracts slowly changing trend components (reflecting global flow resistance evolution), and a high-pass filter extracts rapidly changing disturbance components (reflecting local material characteristic differences). The boundary frequency is tuned according to the actual production line and set as a configurable parameter. Implementation steps include: effectiveness check, smoothing preprocessing, dual-channel parallel filtering, and extraction of characteristic quantities (mean, rate of rise, etc. for trend components; peak and valley amplitude, frequency, etc. for disturbance components). The control system makes judgments based on the refined characteristics to improve recognition consistency. Furthermore, the filling cycle signal can be used to mark regular pulsations, avoiding misjudging normal periodic pressure fluctuations as abnormal disturbances.

[0029] To identify the source type of current pressure anomalies by combining distribution stage, trend characteristic components, and disturbance characteristic components, some implementations can pre-construct a multi-dimensional state mapping table, using the distribution stage, the rate of change of trend characteristic components, and the amplitude of disturbance characteristic components as input conditions. The mapping table is built by collecting actual data from representative switching processes: recording the stage determination, trend and disturbance performance, and on-site confirmed anomalies during each switch, and mapping feature combinations to anomaly types to form structured judgment rules. For example, a continuously rising trend with weak disturbances is marked as a sustained increase in flow resistance; a stable trend but frequent peaks in disturbances are marked as short-term local disturbances; and both are marked as a composite anomaly. During runtime, the system discretizes the real-time features and matches them with the mapping table to output the anomaly source type. This method relies on the joint judgment of stage, trend, and disturbance, which is more accurate than a single threshold and can effectively distinguish between overall flow resistance increases and local disturbances, two anomalies with drastically different control strategies.

[0030] Regarding the core technical approach of determining the target values ​​of driving parameters and execution parameters based on the type of pressure anomaly source, in a possible example, the system selects the corresponding strategy based on the identified pressure anomaly source type. When a persistent increase in flow resistance or a complex anomaly is identified, the maximum safe feed flow rate is calculated based on the trend characteristic components, divided by the single-bottle filling volume to obtain the theoretical cycle time, and then rounded down to the allowable discrete level at the filling end as the target value of the execution parameter, thereby determining the target value of the driving parameter. In the case of complex anomalies, a dynamic bias is also applied to the driving parameters based on the peak-to-peak value of the disturbance characteristic components. When a short-term local disturbance is identified, the execution parameters remain unchanged, and only a limited-amplitude, gently changing mitigation bias is applied to the driving parameters. Through the determination of target values ​​by type and level, a combination of control parameters matching the anomaly source is generated.

[0031] Regarding the core technical approach of synchronously issuing correction commands to both the power and filling ends and limiting the rate of change of these commands, more specifically, after determining the target values ​​for the parameters at both ends, the central PLC sends speed and valve opening correction commands to the frequency converter and servo drive synchronously via industrial Ethernet such as PROFINET in the same communication cycle, ensuring unified control decisions at both ends. To avoid pressure shocks and mechanical disturbances caused by sudden parameter changes, the system uses a PLC ramp function module to limit the rate of change of commands, preventing a one-time jump in the target value. Independent upper limits R1 and R2 for the lifting and lowering rates are configured for the power and filling ends, respectively, and are individually tuned based on the pump's mechanical load-bearing capacity, filling accuracy, and pipeline pressure stability, abandoning the crude control method of a uniform fixed rate of change. The system also adds parameter boundary protection, verifying the target speed and valve opening before issuing commands, cutting off any exceeding values ​​to the safe operating range of the equipment, and then smoothly transitioning in small steps cycle by cycle through a ramp mechanism. This control method, which combines synchronous decision-making, speed limiting for individual equipment, and boundary fallback, can effectively avoid secondary pressure disturbances caused by rapid unilateral adjustments, ensuring a stable and controllable material conveying and filling process.

[0032] In actual pesticide pipeline transportation, relying solely on simple volume estimation to delineate distribution stages is often insufficiently accurate. Because high-viscosity pesticides adhere strongly to the pipe walls, the actual length of the mixing transition section dynamically varies depending on the material's properties. Sometimes, even when the transported volume exceeds the pipeline's physical capacity, incompletely replaced clumps of old material remain inside the pipeline, causing continuous pressure fluctuations. To more accurately pinpoint the true distribution of the material, this specific implementation plan introduces a dual verification mechanism based on both volume and pressure fluctuation amplitude.

[0033] Optionally, the step of determining the distribution stage within the delivery pipeline during material switching based on fluid physical parameters includes: Calculate the cumulative conveying volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters; The cumulative transport volume is compared with the preset pipeline volume; When the cumulative conveying volume is greater than the second proportion of the pipeline volume, and the pressure fluctuation amplitude in the fluid physical parameters is less than the stable threshold within a preset time, the distribution stage is determined to be the new material-dominated conveying stage. When the cumulative conveying volume is between the first and second proportions of the pipeline volume, or when the cumulative conveying volume is greater than the second proportion of the pipeline volume but the pressure fluctuation amplitude is not less than the stable threshold, the distribution stage is determined to be the new and old mixed transition material stage; the first proportion is less than the second proportion. Otherwise, the distribution stage is determined to be the residual old material stage.

[0034] In this specific implementation scheme, the system calculates the cumulative transport volume since the start of the switchover by integrating the instantaneous flow rate. The preset pipeline volume is calculated based on the pipeline's inner diameter and length. A first proportional value, such as 80%, represents the arrival of new material in the middle to later section of the pipeline; a second proportional value, such as 120%, represents the complete replacement of old material by new material; the stability threshold is the allowable fluctuation range of pressure background noise during normal single-material transport. The above parameters are obtained through production line pre-calibration: representative products are selected for switchover tests, the cumulative transport volume range and the boundary of pressure recovery to stability are recorded, and the first and second proportional values ​​are determined; the stability threshold is determined by the pressure fluctuation range during stable transport plus a safety margin.

[0035] When the cumulative conveying volume has not yet reached the first proportion value, it indicates that the main body of the pipeline is still occupied by old material, and it is judged as the residual old material stage. When the cumulative conveying volume is between the first and second proportion values, it is physically in the core area of ​​the transition between new and old materials, and it must be judged as the new and old mixed transition material stage. The key is that when the cumulative conveying volume has exceeded the second proportion value, that is, theoretically the replacement has been completed, the system does not blindly switch states, but checks the preset time, such as the pressure fluctuation amplitude in the past 10 seconds. If the pressure fluctuation amplitude is still greater than or equal to the stability threshold, it indicates that there is still stubborn local blockage flow in the pipeline causing disturbance. At this time, the system will forcibly maintain the judgment of the new and old mixed transition material stage; only when the volume reaches the standard and the pressure fluctuation truly subsides (below the stability threshold) is it confirmed as the new material-dominated conveying stage. Considering that using the past 10 seconds may not always be optimal under different filling cycles and different pipeline lengths, the preset time can be set as a time window T, and adjusted through trial operation to cover the duration of at least several complete filling cycles. If the time window is too short, only occasional stable segments may be observed, leading to misjudgment as the new material-dominated conveying stage; if the time window is too long, it will delay stage switching and reduce control response speed. By setting the time window to an observation interval that matches the actual cycle time, both judgment stability and timely response can be balanced.

[0036] To address the fluid compression error under high-pressure pulses when calculating cumulative volume using instantaneous flow rate integration, this solution proposes a dynamic compensation mechanism based on phase lag analysis. By analyzing the phase lag angle between torque fluctuations at the power end and pressure pulsations at the filling end, the degree of fluid compression is quantified, and a compensation coefficient is generated to correct the instantaneous flow rate, thus eliminating the distortion in cumulative volume calculation caused by the compression effect.

[0037] Furthermore, the steps for calculating the cumulative conveying volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters include: Acquire the output torque fluctuation signal at the power end and the inlet pressure pulsation signal at the filling end; Calculate the phase hysteresis angle between the output torque fluctuation signal and the inlet pressure pulsation signal; The fluid compressibility compensation coefficient is determined based on the magnitude of the phase lag angle. The instantaneous flow rate is weighted and corrected using the fluid compressibility compensation coefficient to obtain the corrected instantaneous flow rate; Using the start time of material switching as the integration starting point, a trapezoidal integral is performed on the corrected instantaneous flow rate to obtain the cumulative conveying volume.

[0038] In this specific implementation scheme, the system acquires the motor torque fluctuation signal at the power end through the high-speed interface of the frequency converter, and obtains the inlet pressure pulsation signal from the high-frequency pressure transmitter at the end of the pipeline. Due to the transmission delay of fluid pressure waves and the compressibility of the fluid, the pressure pulsation at the end lags behind the torque fluctuation. The system uses a cross-correlation algorithm to solve for the phase lag angle of the two periodic signals. The larger the phase lag angle, the more significant the fluid compression deformation and transmission delay. Before phase analysis, the two signals are preprocessed in the same frequency band to retain common characteristic frequency bands and suppress low-frequency drift and high-frequency electrical noise. The peaks and troughs of the two signals are compared within a sliding time window to obtain the time lag and convert it into a phase lag angle, thereby quantifying the degree of delay in the transmission of power disturbances to the filling end. A small lag indicates good fluid transmission rigidity; a large lag indicates a prominent fluid elastic energy storage effect, and direct flow integral will overestimate the actual material propulsion volume.

[0039] The system has a preset curve corresponding to the phase lag angle and the fluid compressibility compensation coefficient. Based on the real-time phase lag angle, a compensation coefficient less than 1 is obtained by looking up a table or fitting. The instantaneous flow rate of the flow meter is multiplied by this coefficient to eliminate the false flow rate caused by fluid compression and obtain the corrected instantaneous flow rate. Using the material switching command as the time zero point, trapezoidal integral is used to continuously integrate the corrected flow rate to obtain a high-precision cumulative conveying volume.

[0040] The corresponding curve is obtained through offline calibration: During the commissioning phase, torque, pressure, and flow data are simultaneously collected for different materials, viscosities, and pressure conditions. The actual displacement volume is obtained through actual discharge measurements, and the compensation coefficients corresponding to each phase lag interval are deduced by comparing the uncompensated integral flow rate, forming a lookup mapping table. When new materials are added, calibration samples can be added to update this mapping relationship.

[0041] To address the issues of inaccurate pipeline volume calculation and susceptibility to interference from single pressure fluctuations, a liquid injection calibration method is used to accurately obtain the pipeline volume. This is combined with cross-validation using multi-dimensional parameters such as torque derivative and pressure variance to improve the accuracy and robustness of distribution stage determination.

[0042] Optionally, the steps for obtaining the preset pipeline volume include: The internal volume of the straight pipe section is calculated based on the physical topology of the pipeline, and the pipeline volume is obtained by combining the fluid dynamic equivalent volume compensation value of the pipeline accessories. The steps for determining the distribution stage within the delivery pipeline during material switching based on fluid physical parameters also include: The variance of pressure values ​​in the fluid physical parameters within the dynamic observation window is calculated, and dynamic cross-validation is performed on the distribution stage by combining the first derivative of the output torque at the power end.

[0043] Specifically, the preset pipeline volume refers to the effective fluid capacity established and stored before system commissioning based on the actual pipeline structure, accessory types, and installation status. The internal volume of straight pipe sections is calculated from the inner diameter and length; the internal flow field of pipeline accessories is complex, requiring the introduction of fluid dynamic equivalent volume compensation values ​​for correction. These compensation values ​​can be calibrated through historical databases or actual liquid displacement tests. The physical topology is derived from 3D modeling, as-built drawings, and on-site verification data, and structural verification is required after importation.

[0044] The dynamic cross-validation mechanism distinguishes between actual material distribution changes and external disturbances by simultaneously observing the pressure signal at the filling end and the load changes at the power end. Pressure variance reflects the degree of pressure dispersion within a time window, providing a more consistent representation than a single fluctuation amplitude; the dynamic observation window continuously captures the most recent sampling duration, avoiding misjudgments due to instantaneous spikes. The first derivative of the output torque characterizes the rate of change of the power end load, and its combined analysis with pressure variance demonstrates that slug flow simultaneously increases pressure dispersion and causes abrupt changes in the torque derivative, while external mechanical vibration only affects the pressure side, thus improving the robustness of distribution stage determination.

[0045] In a specific embodiment, during system initialization, the system reads the three-dimensional topology model of the pipeline, automatically extracts the length and inner diameter of each straight pipe segment, and accumulates them using the cylindrical volume formula to obtain the straight pipe volume. For pipeline accessories such as Y-type filters and pneumatic diaphragm valves, the system retrieves the equivalent volume compensation value from a pre-set accessory equivalent volume database. The straight pipe volume is then superimposed with the accessory compensation value to obtain the high-precision total pipeline volume. To more clearly and completely describe this acquisition process, in some preferred embodiments, the control system parses the three-dimensional model, extracts pipe segment numbers, nodes, inner diameters, pipe lengths, and accessory information, filters the main delivery channel pipe segments, and excludes non-fluid-bearing structures such as insulation layers and supports. The system matches database entries based on accessory codes; if a match is successful, the compensation value is directly read; if a match fails, manual confirmation is initiated, with the operator selecting a similar model or entering the on-site calibration volume. The database distinguishes between horizontal and vertical installation postures of accessories and provides correction coefficients for flow-sensitive accessories suitable for high-viscosity suspending agents and low-viscosity aqueous agents.

[0046] During system operation, a dynamic observation window is set (5 seconds in the example, configurable to parameter T). The variance of the filling inlet pressure signal within the window is calculated, and the first derivative of the torque is obtained by differentiating the output torque of the frequency converter. This window is adapted to pesticide filling conditions, taking into account both pump pulsation cycle and short-term change capture. When initially determined to be in the transition stage of mixed old and new materials, the pressure and torque signals are first time-aligned to eliminate sampling misalignment. If the pressure variance suddenly increases or the first derivative of the torque shows continuous alternating positive and negative peaks, it is determined that there is a serious local slugging flow. The transition stage determination is maintained, and the transition stage control strategy is not switched to the new material-dominated stage. The determination threshold is obtained by collecting historical data of normal and abnormal switching during the debugging phase and stored in the parameter table for later use.

[0047] During the switching between multiple pesticides, the pressure signal generated by the mixing of high-viscosity and low-viscosity materials is extremely complex. It includes both low-frequency baseline drift caused by the overall increase in flow resistance and high-frequency irregular pulses caused by localized material clumps. Traditional linear filtering methods, such as moving averages or Butterworth filters, are prone to phase delays or blurring of the true inflection points of the low-frequency baseline when processing such non-stationary and nonlinear signals, leading to distortion of characteristic components. Therefore, it is necessary to introduce nonlinear morphological signal processing techniques to accurately extract trend and disturbance characteristic components.

[0048] In some preferred embodiments, the steps of decoupling the fluid physical parameters to obtain trend characteristic components reflecting the evolution of the global flow resistance of the conveying pipeline, and disturbance characteristic components reflecting local material property differences, include: A dual-channel morphological filtering algorithm is used to decouple the pressure signal in the fluid physical parameters into a low-frequency baseline component and a high-frequency pulse component; The low-frequency baseline component was identified as the trend characteristic component, and the high-frequency pulse component was identified as the disturbance characteristic component.

[0049] Specifically, the trend component refers to the signal component that characterizes the direction and degree of slow change in the overall flow resistance of the entire pipeline over a period of time; the disturbance component refers to the short-term, sudden, and sharply shaped signal component caused by events such as local material clumps, local blockages, and local detachment impacts. The dual-channel morphological filtering algorithm processes the pressure signal in parallel using structural elements of different scales: large-scale structural elements extract the low-frequency baseline component (trend), and small-scale structural elements extract the high-frequency pulse component (disturbance). The structural element is a sliding comparison template of fixed time length, scanned along the time axis, retaining slowly changing parts and removing peaks narrower than the template width. The dual-channel parallel processing ensures strict time correspondence between the two components, facilitating joint identification. Preprocessing includes timestamp alignment, distortion point removal, broken line value filtering, and mild amplitude limiting to remove invalid data without altering the true trend and impact.

[0050] In a specific embodiment, the control system signal processing module sets up two parallel morphological processing channels to decouple the pressure signal from the high-frequency pressure transmitter. Channel 1 uses a 2-second wide linear structure element to perform a morphological combination operation of opening and closing operations, filtering out pressure fluctuations lasting less than 2 seconds and outputting a smooth low-frequency baseline as a trend feature component characterizing the evolution of global flow resistance. This operation first removes short-term upward spikes and then fills in local signal valleys, ensuring the baseline does not follow short-term pulsation distortion and adapts to slow changes in flow resistance. Channel 2 uses a 0.1-second narrow pulse structure element. After performing a morphological opening operation on the original signal, it performs a top-hat transformation on the difference between the original signal and the operation result to extract transient pressure spikes, obtaining a high-frequency pulse component as a disturbance feature component characterizing local material differences, highlighting short-term pressure pulses caused by material clumps and instantaneous blockages. The pressure signal is converted from analog to digital and stored in a buffer. The control system reads the signal at fixed step sizes, and each processing cycle extracts the most recent original pressure sequence for dual-channel operation. 2s and 0.1s are example parameters; under actual operating conditions, the duration of the two structural elements is configurable, requiring that the narrow scale be much smaller than the wide scale, and the wide scale be significantly smaller than the overall trend change time. During the commissioning phase, multiple sets of material switching pressure signals are collected. Through testing multiple length combinations, the baseline smoothness, peak retention integrity, and phase hysteresis are used as evaluation criteria to associate the optimal parameters with material type, pump type, and pipe diameter, forming a preset parameter table. During production operation, the control system automatically retrieves the corresponding parameters according to the current formula and material category, ensuring stable signal decoupling effect.

[0051] After extracting trend and disturbance components, the system needs to identify the types of pressure anomalies, including persistent increases in flow resistance, short-term local disturbances, and combined anomalies. Relying solely on distribution stages and static characteristic components is insufficient to capture dynamic trends and is prone to misjudgment. Therefore, higher-order derivative features and flow balance indices are introduced to construct a multi-dimensional comprehensive identification model.

[0052] In practical applications, the types of sources of pressure anomalies include persistent increases in flow resistance, short-term local disturbances, and composite anomalies resulting from the superposition of the two. The steps to identify the current source type of pressure anomaly by combining distribution stage, trend characteristic components, and disturbance characteristic components include: Calculate the first and second derivatives of the pressure difference between the two ends of the conveying pipeline and the flow surplus rate of the material conveying system; By combining distribution stage, trend characteristic components, disturbance characteristic components, first derivative, second derivative, and flow surplus rate, the source type of pressure anomaly can be identified.

[0053] Specifically, a sustained increase in flow resistance refers to a continuous rise in the total pressure differential of the pipeline over a period of time, and this rise is not caused by short-term spikes, but reflects an increase in the overall base value of flow resistance; short-term local disturbances refer to short-term pressure anomalies caused by events such as local agglomerates, instantaneous blockages, or valve port deposits falling off; and compound anomalies refer to the simultaneous existence of overall flow resistance deterioration and local disturbances, with both a rise in base pressure differential and frequent spikes or pulses.

[0054] The pressure difference is the difference between the outlet pressure at the power end and the inlet pressure at the filling end, reflecting the friction loss along the pipeline. The first derivative characterizes the rate of change of the pressure difference, while the second derivative characterizes the convergence or divergence of the resistance deterioration trend. The flow surplus rate is the percentage of the difference between the pump inflow and the filling consumption flow to the pump inflow, reflecting the degree of material accumulation in the pipeline. The outlet pressure at the power end is collected by the pump outlet pressure sensor, and the inlet pressure at the filling end is collected by the pressure transmitter before the filling machine; the pump inflow can be obtained through a mass flow meter or electromagnetic flow meter, and the filling consumption flow is calculated by converting the filling cycle time and the single filling volume. The first and second derivatives are obtained by differentiating and lightly smoothing the pressure difference sequence within a continuous control cycle. A persistently high flow surplus rate indicates that material is prone to stagnation and accumulation, while a flow surplus rate close to zero and with isolated pressure peaks indicates that the disturbance is mostly local. The above multi-dimensional features are input into a pre-built decision tree model, which is trained by labeling historical operating data to generate tree-like decision rules and is then embedded in the control system.

[0055] In a specific embodiment, within the control cycle, the system collects the pump outlet pressure and the filling machine inlet pressure in real time and calculates the difference between them. Then, it performs two consecutive differential operations on the pressure difference signal to obtain the first and second derivatives of the pressure difference. Simultaneously, the system reads the mass flow meter data at the pump end and the instantaneous filling speed of the filling machine to calculate the current flow surplus rate. When an anomaly is identified, the system inputs these dynamic parameters into a pre-constructed multi-dimensional decision tree model. For example, when the system is in the transition stage between new and old materials, the trend characteristic component rises slowly, but the first derivative of the pressure difference is large and the second derivative is positive (indicating that the resistance is rapidly deteriorating), while the flow surplus rate exceeds 10% (indicating severe material accumulation in the pipeline), and the disturbance characteristic component frequently exhibits high-amplitude pulses, the decision tree model integrates these characteristics and accurately outputs that the current pressure anomaly source type is a composite anomaly.

[0056] To more clearly and completely describe the identification process, in a specific example, the control system first completes signal acquisition according to a fixed control cycle, and then performs time alignment on the trend feature components, disturbance feature components, and differential pressure derivative features within the same cycle to ensure that each feature input to the decision tree model corresponds to the same actual flow state. If individual features are missing in the current cycle due to different sensor refresh rates, the control system preferably uses the most recent valid value for a short period of time and triggers a data quality alarm when data is missing for multiple consecutive cycles, avoiding the model from outputting unreliable results under conditions of significant data loss. The example threshold of a flow surplus rate exceeding 10% is applicable to typical pesticide filling conditions where the pump supply is continuous, the filling cycle is stable, and the pipeline volume is large, and is used to characterize the significant risk of material accumulation in the pipeline. If there are significant differences in the pump type, pipe diameter, buffer volume, or filling cycle of the actual production line, this threshold can also be reset to parameter R based on historical data during the commissioning phase. When performing judgments, the multidimensional decision tree model can filter layer by layer according to the following logic: First, it excludes ordinary fluctuations that have not yet entered the critical switching stage based on the distribution stage; then, it judges whether there is a tendency for overall flow resistance to increase based on the trend characteristic components and the first derivative of the pressure difference; subsequently, it judges whether the increase tends to stabilize or continues to worsen based on the second derivative; finally, it confirms whether there are local shocks and systemic backlogs based on the disturbance characteristic components and the flow surplus rate. If only the trend-related characteristics are significant and the disturbance pulses are not obvious, it is judged as a continuous increase in flow resistance; if the trend change is not significant but the disturbance characteristic components frequently burst, it is judged as a short-term local disturbance; if both types of characteristics are met simultaneously, it is judged as a composite anomaly. Through the above processing, the system can not only give the anomaly type, but also give the key characteristic sources that form the anomaly type, which is convenient for subsequent control strategy selection and operation and maintenance traceability.

[0057] After identifying the type of pressure anomaly, the system sets target values ​​according to the anomaly type: for continuous increases in flow resistance, the pump speed and filling cycle are adjusted synchronously; for short-term local disturbances, the execution parameters remain unchanged, with only a small mitigation applied to the pump end; for complex anomalies, a dynamic bias is superimposed on the synchronous adjustment. Theoretical calculation parameters need to be rounded down to the discrete levels allowed at the filling end to ensure that the target values ​​are actually executable, achieving a precise match between the control strategy and the anomaly type.

[0058] In a preferred embodiment of the present invention, the step of determining the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source includes: When the source of the pressure anomaly is a continuous increase in flow resistance or a complex anomaly, the maximum safe feed flow rate of the conveying pipeline is calculated back based on the trend characteristic components. The theoretical execution parameters are calculated based on the maximum safe feeding flow rate, and then rounded down to the discrete level allowed at the filling end to obtain the target value of the execution parameters. Determine the target values ​​of the drive parameters based on the target values ​​of the execution parameters and the maximum safe feed flow rate; When the source of the pressure anomaly is a composite anomaly, a dynamic bias is also superimposed on the target value of the driving parameter according to the disturbance characteristic component to suppress local disturbances, thereby obtaining the final target value of the driving parameter. When the source of the pressure anomaly is a short-term local disturbance, the target value of the execution parameter remains unchanged, and a limited-amplitude mitigation bias is applied to the driving parameter.

[0059] Specifically, the driving parameters preferably refer to the adjustable controllable quantities of the power-end equipment, such as the set frequency of the variable frequency pump and the speed of the servo pump; the execution parameters preferably refer to the adjustable controllable quantities of the filling-end equipment, such as the filling cycle time, valve opening and closing rhythm, or discrete levels. The maximum safe feed flow rate is the upper limit of the feed that the pipeline can continuously withstand without triggering the safe pressure under the basic flow resistance reflected by the current trend characteristic component. It is obtained by back-calculation through a pre-established trend characteristic component-acceptable feed flow rate mapping table. Rounding down to discrete levels means mapping the theoretical execution parameters to the nearest level that the equipment can actually execute without exceeding the theoretical value, ensuring that the end consumption does not exceed the safe feed capacity. Dynamic bias is used to address complex anomalies by superimposing a small adjustment amount opposite to the disturbance on the basic driving parameters to offset local impacts; mild suppression bias is used to address short-term local disturbances by maintaining the filling rhythm and applying only a very small and very mild suppression signal to the pump end. Both biases have limits on the rate of change, duration, and maximum amplitude to ensure smooth adjustment and reversibility.

[0060] In a specific embodiment, when the system identifies a continuous increase in flow resistance, the control system calculates the maximum safe feed flow rate of the current pipeline as 80 liters / minute based on the trend characteristic component. The theoretical filling speed is calculated to be 85 bottles / minute based on the single bottle capacity. The system queries the available speed table of the filling machine, rounds down the theoretical speed to a discrete speed of 80 bottles / minute, and uses this as the target value for the execution parameters. Subsequently, based on the actual consumption of 80 bottles / minute, the target value for the drive parameters of the variable frequency screw pump is recalculated and determined, such as setting the frequency to 35Hz to ensure strict matching of supply and demand. When a composite anomaly is identified, the system extracts the instantaneous amplitude of the disturbance characteristic component from the aforementioned 35Hz base target value and calculates a dynamic bias of ±1.5Hz. When a local high-pressure material mass passes through, the pump frequency momentarily drops slightly to 33.5Hz to absorb the impact, and then quickly recovers, thus obtaining the final target value for the drive parameters. When a short-term local disturbance is detected, the system determines that the overall flow resistance has not worsened, and therefore maintains the current target values ​​of the filling machine's execution parameters, such as 90 bottles / minute, without changing them. At the same time, a mild suppression bias with an amplitude limited to -2Hz is applied to the variable frequency screw pump, and the pump speed is only slightly reduced at the moment the disturbance peak occurs to buffer it and avoid triggering system-wide oscillations.

[0061] To more clearly and completely describe the implementation process, in a specific example, the control system pre-stores three types of basic data: a trend characteristic component-maximum safe feed flow rate mapping table, a filling machine available speed range table, and a calibration relationship between pump frequency and stable output flow rate. During operation, the maximum safe feed flow rate is first obtained based on the trend characteristic component, converted into theoretical execution parameters, and then rounded down to the available speed range of the filling machine to obtain the target value of the execution parameters; then, the target value of the drive parameters is matched through the pump frequency-flow rate calibration relationship. All three types of data are established through on-site commissioning before production. The amplitudes of dynamic bias and mitigation bias can be preset as bias parameters P1 and P2, and adjusted during trial operation based on the suppression effect and stability. The final target value, constrained by equipment speed range, supply and demand matching, and on-site calibration, can be directly issued for execution.

[0062] After determining the target values ​​for the driving and execution parameters, the final step in ensuring control effectiveness is to smoothly and accurately translate these target values ​​into the actual actions of the underlying equipment while overcoming the inherent delay characteristics of pipeline fluid transmission. In actual pesticide production lines, if there are timing misalignments in the issuance of instructions, or if the control loop cannot adapt to the pure lag effect caused by long pipelines, even if the target values ​​are calculated accurately, new pressure oscillations will occur during the execution phase.

[0063] In a preferred embodiment of the present invention, the step of synchronously issuing correction commands to the power end and the filling end based on the target values ​​of the driving parameters and the execution parameters, and limiting the rate of change of the correction commands to maintain the inlet pressure of the filling end within a preset safety range includes: Correction instructions for the power end and filling end are generated based on the target values ​​of the driving parameters and the execution parameters. Correction commands for the power supply and filling ends are issued synchronously within the same programmable logic controller scan cycle; The correction instruction is subjected to a first-order low-pass filter to limit its rate of change and make the correction instruction transition smoothly. The execution effect is continuously monitored through a closed-loop feedback correction mechanism, and the Smith predictive compensator is used to compensate for pipeline lag, so that the inlet pressure at the filling end is maintained within the preset safe range.

[0064] To ensure absolute temporal coordination between the spatially separated power and filling ends, the above steps involve synchronously issuing commands within the same programmable logic controller (PLC) scan cycle. Industrial PLC scan cycles typically range from a few milliseconds to tens of milliseconds. After calculating the target value, the control system packages the speed correction command for the power end and the cycle time correction command for the filling end, sending them to the corresponding lower-level driver via an industrial Ethernet network, such as PROFINET or EtherCAT, in the same communication message. This is because adjusting the pump speed before adjusting the filling valve, even with a difference of only a few tenths of a second, can create a transient supply-demand mismatch at the end of the pipeline, potentially triggering unnecessary pressure hammer. First-order low-pass filtering transforms the step-like target value into a smooth asymptotic curve. In practice, the control system configures a first-order inertial element in the command output channel, setting an appropriate time constant so that the correction command does not jump instantaneously but transitions smoothly according to an exponential law. This effectively avoids the impact of sudden changes in motor torque on mechanical transmission components, and also prevents cavitation or local high pressure from the fluid due to excessive acceleration.

[0065] Furthermore, the combination of closed-loop feedback correction mechanism and Smith Predictor is the core means to solve the problem of long-distance pesticide pipeline transportation control. In pesticide production lines, the physical transmission of fluid in the pipeline from the pumping end to the pressure change felt at the filling end requires a certain amount of time, which is called dead time in control theory. If only conventional PID closed-loop feedback is used, when the pressure at the filling end deviates, the controller adjusts the pump speed, but due to the dead time, the end pressure does not change immediately. At this time, the conventional controller will mistakenly think that the adjustment is insufficient and continue to increase the output. When the fluid actually reaches the end, the pressure will inevitably be severely overshooted, thus causing high-frequency oscillations in the system. To overcome this physical limitation, the control system introduces Smith Predictor in the closed-loop loop. This predictor pre-embeds a pipeline dynamic model without dead time and an equivalent dead time model. In actual operation, the predictor can calculate the theoretical pressure response of the current correction command in advance under the condition of no dead time, and feed back this predicted real-time response to the controller in advance. The controller adjusts based on this hysteresis-free feedback signal, just as if directly controlling an ideal system without delay. Simultaneously, the predictor compares the hysteresis-laden model output with the actual measured pressure, using the error between the two to correct for inaccuracies in the model itself.

[0066] While sophisticated control strategies can handle most switching anomalies, in extreme cases, such as pipeline blockage, conventional adjustments are ineffective and may even damage the equipment. Therefore, this invention further proposes corresponding safety protection mechanisms.

[0067] Specifically, the method also includes: When the pressure difference between the two ends of the delivery pipeline increases exponentially and the flow rate drops to the preset blockage threshold, the safety fallback control is triggered. Safety fallback controls include: cutting off the power end's operation enable signal, stopping the filling end's operation, closing the filling end's inlet shut-off valve, and opening the pressure relief bypass.

[0068] To address the above steps, the system continuously monitors the first and second derivatives of the pressure differential to determine if it is increasing exponentially. When the pressure differential is not only increasing but also increasing at an increasingly rapid rate (i.e., the second derivative is significantly positive), and simultaneously the instantaneous flow rate reported by the mass flow meter drops sharply below the preset blockage threshold for maintaining normal flow, the system determines that a severe physical blockage has occurred inside the pipeline. At this point, the conventional closed-loop control logic is immediately suspended, and the system forcibly triggers the highest priority safety fallback control. In practical execution, the control system first cuts off the power-end inverter's operation enable signal directly through hardware interlocking or safety communication messages, forcing the material pump to instantly lose power output and stop energy injection at the source. Simultaneously, it issues an emergency stop command to the filling machine to stop all filling operations. Next, it quickly closes the inlet shut-off valve located in front of the filling machine to block any potential high-pressure shock waves from reaching the precision filling valve, protecting the expensive end actuator. Finally, it automatically opens the pressure relief bypass solenoid valve installed on the pipeline to safely divert the high-pressure material accumulated in the pipeline to a dedicated recovery tank or buffer tank, quickly releasing the pressure inside the pipeline.

[0069] Secondly, please see Figure 2 The present invention also proposes a pesticide production line operation process control system for executing the steps in any of the above methods, including: The parameter acquisition module 210 is used to acquire the operating status parameters of the material conveying system. The operating status parameters include at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. The distribution determination module 220 is used to determine the distribution stage in the conveying pipeline during the material switching process based on fluid physical parameters. The distribution stages include the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. The feature processing module 230 is used to decouple the fluid physical parameters from the signal, obtain the trend feature components that reflect the evolution of the global flow resistance of the conveying pipeline, and the disturbance feature components that reflect the differences in local material properties. Anomaly identification module 240 is used to identify the current source type of pressure anomaly by combining distribution stage, trend feature components and disturbance feature components. The dynamic adjustment module 250 is used to determine the target values ​​of the driving parameters and the execution parameters according to the type of pressure anomaly source. The instruction issuing module 260 is used to synchronously issue correction instructions to the power end and the filling end according to the target values ​​of the driving parameters and the execution parameters, and to limit the rate of change of the correction instructions so as to keep the inlet pressure of the filling end within a preset safe range.

[0070] Through the above technical solution, the pesticide production line operation process control system highly integrates complex data acquisition, stage judgment, signal decoupling, anomaly identification, and collaborative execution. Data flow between functional modules is clear, and logical connections are tight, forming a complete closed-loop control architecture. The parameter acquisition module 210 provides a solid data foundation, while the distribution determination module 220 and feature processing module 230 act as the system's eyes and brain, penetrating complex surface signals to perceive the true flow field state inside the pipeline. Based on these insights, the anomaly identification module 240 and dynamic adjustment module 250 formulate refined adjustment strategies best suited to the current operating conditions. Finally, the command issuance module smoothly implements the strategies. This system completely breaks through the limitations of traditional pesticide production lines that rely on manual experience or fixed-gear control during multi-variety switching. It achieves adaptive and feedforward suppression of feed pressure pulse fluctuations, significantly improving the consistency and production efficiency of single-bottle pesticide filling and metering, and significantly extending the service life of pipeline equipment. It provides a reliable system-level solution for the intelligent upgrading of pesticide co-production lines.

[0071] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for controlling the operation of a pesticide production line, used in the material conveying system of the pesticide production line, the material conveying system comprising a power end, a conveying pipeline, and a filling end, characterized in that, include: The operating status parameters of the material conveying system are obtained, including at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. Based on the fluid physical parameters, the distribution stages in the conveying pipeline during the material switching process are determined. The distribution stages include the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. The fluid physical parameters are decoupled from the signal to obtain the trend characteristic component reflecting the global flow resistance evolution of the delivery pipeline, and the disturbance characteristic component reflecting the local material characteristic differences. By combining the distribution stage, the trend feature components, and the disturbance feature components, the current source type of pressure anomaly is identified; Based on the type of pressure anomaly source, determine the target values ​​of the driving parameters and the execution parameters; Based on the target values ​​of the driving parameters and the execution parameters, correction commands are synchronously issued to the power end and the filling end, and the rate of change of the correction commands is limited so that the inlet pressure of the filling end is maintained within a preset safe range.

2. The method for controlling the operation of a pesticide production line according to claim 1, characterized in that, The step of determining the distribution stage within the delivery pipeline during the material switching process based on the fluid physical parameters includes: Calculate the cumulative transport volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters; The cumulative transport volume is compared with the preset pipeline volume; When the cumulative conveying volume is greater than the second proportional value of the pipeline volume, and the pressure fluctuation amplitude in the fluid physical parameters is less than the stable threshold within a preset time, the distribution stage is determined to be the new material-dominated conveying stage. When the cumulative conveying volume is between a first proportional value and a second proportional value of the pipeline volume, or when the cumulative conveying volume is greater than the second proportional value of the pipeline volume but the pressure fluctuation amplitude is not less than the stability threshold, the distribution stage is determined to be a new and old mixed transition material stage; the first proportion is less than the second proportion. Otherwise, the distribution stage is determined to be the residual old material stage.

3. The method for controlling the operation of a pesticide production line according to claim 2, characterized in that, The step of calculating the cumulative transport volume since the start of material switching based on the instantaneous flow rate in the fluid physical parameters includes: The output torque fluctuation signal of the power end and the inlet pressure pulsation signal of the filling end are obtained; Calculate the phase hysteresis angle between the output torque fluctuation signal and the inlet pressure pulsation signal; The fluid compressibility compensation coefficient is determined based on the magnitude of the phase hysteresis angle. The instantaneous flow rate is weighted and corrected using the fluid compressibility compensation coefficient to obtain the corrected instantaneous flow rate; Using the material switching start time as the integration starting point, a trapezoidal integral is performed on the corrected instantaneous flow rate to obtain the cumulative conveying volume.

4. The method for controlling the operation of a pesticide production line according to claim 2, characterized in that, The steps for obtaining the preset pipeline volume include: The internal volume of the straight pipe section is calculated based on the physical topology of the pipeline, and the pipeline volume is obtained by combining the fluid dynamic equivalent volume compensation value of the pipeline accessories. The step of determining the distribution stage within the delivery pipeline during the material switching process based on the fluid physical parameters further includes: Calculate the variance of the pressure values ​​in the fluid physical parameters within the dynamic observation window, and perform dynamic cross-validation on the distribution stage by combining the first derivative of the output torque of the power end.

5. The method for controlling the operation of a pesticide production line according to claim 1, characterized in that, The steps of decoupling the fluid physical parameters to obtain trend characteristic components reflecting the global flow resistance evolution of the delivery pipeline and disturbance characteristic components reflecting local material property differences include: A dual-channel morphological filtering algorithm is used to decouple the pressure signal in the fluid physical parameters into a low-frequency baseline component and a high-frequency pulse component. The low-frequency baseline component is determined as the trend characteristic component, and the high-frequency pulse component is determined as the disturbance characteristic component.

6. The method for controlling the operation of a pesticide production line according to claim 1, characterized in that, The types of pressure anomalies include persistent increases in flow resistance, short-term local disturbances, and composite anomalies resulting from the superposition of the two. The step of identifying the current source type of pressure anomaly by combining the distribution stage, the trend feature component, and the disturbance feature component includes: Calculate the first and second derivatives of the pressure difference between the two ends of the conveying pipeline and the flow surplus rate of the material conveying system; By combining the distribution stage, the trend characteristic component, the disturbance characteristic component, the first derivative, the second derivative, and the flow surplus rate, the source type of the pressure anomaly is identified.

7. The method for controlling the operation of a pesticide production line according to claim 6, characterized in that, The step of determining the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source includes: When the source of the pressure anomaly is a continuous increase in flow resistance or a combination of anomalies, the maximum safe feed flow rate of the conveying pipeline is calculated back based on the trend characteristic components. The theoretical execution parameters are calculated based on the maximum safe feeding flow rate, and the theoretical execution parameters are rounded down to the discrete level allowed by the filling end to obtain the target value of the execution parameters. The target value of the drive parameter is determined based on the target value of the execution parameter and the maximum safe feed flow rate; When the source of the pressure anomaly is a composite anomaly, a dynamic bias is also superimposed on the target value of the driving parameter according to the disturbance characteristic component to suppress local disturbances, thereby obtaining the final target value of the driving parameter. When the source of the pressure anomaly is a short-term local disturbance, the target value of the execution parameter is kept unchanged, and a limited-amplitude mitigation bias is applied to the driving parameter.

8. The method for controlling the operation of a pesticide production line according to claim 1, characterized in that, The step of synchronously issuing correction commands to the power end and the filling end based on the target values ​​of the driving parameters and the execution parameters, and limiting the rate of change of the correction commands to maintain the inlet pressure of the filling end within a preset safety range includes: The correction instructions for the power end and the filling end are generated based on the target values ​​of the driving parameters and the execution parameters. The correction commands for the power end and the filling end are issued synchronously within the same scan cycle of the programmable logic controller; The correction instruction is subjected to a first-order low-pass filter to limit its rate of change and make the correction instruction transition smoothly. The execution effect is continuously monitored through a closed-loop feedback correction mechanism, and the pure lag in the pipeline is compensated by the Smith predictor compensator, so that the inlet pressure at the filling end is maintained within a preset safe range.

9. The method for controlling the operation of a pesticide production line according to claim 1, characterized in that, The method further includes: When the pressure difference between the two ends of the delivery pipeline increases exponentially and the flow rate drops to a preset blockage threshold, the safety fallback control is triggered. The safety fallback control includes: cutting off the operation enable signal of the power end, stopping the operation of the filling end, closing the inlet shut-off valve of the filling end, and opening the pressure relief bypass.

10. A pesticide production line operation process control system, used to execute the steps of the method according to any one of claims 1 to 9, characterized in that, include: The parameter acquisition module is used to acquire the operating status parameters of the material conveying system. The operating status parameters include at least the driving parameters of the power end, the fluid physical parameters in the conveying pipeline, and the execution parameters of the filling end. The distribution determination module is used to determine the distribution stage in the conveying pipeline during the material switching process based on the fluid physical parameters. The distribution stage includes the residual old material stage, the new and old mixed transition material stage, and the new material-dominated conveying stage. The feature processing module is used to decouple the fluid physical parameters from the signal, obtain the trend feature components that reflect the global flow resistance evolution of the conveying pipeline, and the disturbance feature components that reflect the differences in local material properties. Anomaly identification module is used to identify the current source type of pressure anomaly by combining the distribution stage, the trend feature component, and the disturbance feature component; A dynamic adjustment module is used to determine the target values ​​of the driving parameters and the execution parameters based on the type of pressure anomaly source. The instruction issuing module is used to synchronously issue correction instructions to the power end and the filling end according to the target values ​​of the driving parameters and the execution parameters, and to limit the rate of change of the correction instructions so as to keep the inlet pressure of the filling end within a preset safety range.

Images