Veterinary drug production intelligent control method and system

Abstract

The present application belongs to the technical field of industrial automation and factory integrated control, and particularly relates to a kind of intelligent control method and system for veterinary drug production, and the method comprises: obtaining main pipeline pressure, drug liquid density and total volume flow of the whole factory;Based on the transient disturbance calculation of the total volume flow change of the whole factory to the main pipeline pressure, the comprehensive loss index is obtained, and the node mass flow is obtained combined with the change rate of the main pipeline pressure;According to the sliding average momentum characteristics of node mass flow, the advance valve closing time is calculated;The excess volume is predicted using the advance valve closing time for truncation control, and the comprehensive loss index is combined for pipe network global oscillation interlock protection control.The present application reduces the periodic error caused by the control command issued to the distributed filling machine deviating from the real motion state of veterinary drugs, and improves the overall filling precision under multi-line parallel high-speed beat.

Description

Technical Field

[0001] This invention relates to the field of industrial automation and integrated factory control technology. More specifically, this invention relates to an intelligent control method and system for veterinary drug production. Background Technology

[0002] In modern veterinary drug production scenarios, the factory control system is responsible for the centralized scheduling of the entire manufacturing system. The main liquid supply network is responsible for distributing the liquid medicine to multiple parallel distributed filling machines. The whole plant control system centrally monitors the flow rate of each distributed filling machine branch pipe and uniformly controls the opening and closing of each distributed filling machine's filling valve, ensuring that the filling volume of each distributed filling machine meets the specified target filling volume. This process is a key link in ensuring the overall plant production efficiency and product compliance.

[0003] In existing veterinary drug production and filling control technologies, there are centralized control schemes for multi-machine parallel filling. For example, Chinese invention patent application with publication number CN117234168A and titled "An Intelligent Management System for a Liquid Veterinary Drug Production Workshop" has a main classification number of G05B. This application discloses a centralized control architecture for liquid veterinary drug production workshops. It monitors multiple distributed filling machines in a unified manner through a factory control system, collects the flow data of each filling machine branch pipe in real time, and controls the opening and closing of filling valves based on flow thresholds. At the same time, to adapt to the mechanical action delay of filling valves, a fixed mechanical delay time is introduced as a static compensation parameter.

[0004] However, under the conditions of large-scale and high-speed production of veterinary drugs, the above-mentioned application documents have certain shortcomings. The high-frequency asynchronous start-stop of multiple filling machines can easily cause pressure oscillations in the main liquid supply network. The existing solution only uses independent flow threshold control and fixed delay compensation for each machine, which does not achieve global coordinated management and control of the entire plant's pipeline network. It cannot adapt to the dynamic changes in the flow state of the liquid medicine, and it does not fully consider the impact of operating condition fluctuations on metering accuracy. It also lacks a global prevention and control mechanism for pipeline operation risks, which can easily lead to consistency deviations and periodic deviations in filling volume, affecting both production efficiency and product compliance. Summary of the Invention

[0005] To address the aforementioned technical problem of errors arising from concurrent multi-line filling, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides an intelligent control method for veterinary drug production, comprising: The system collects the original main pipeline pressure, original drug temperature, and original branch volumetric flow rate of each distributed filling machine from the main liquid supply network. After smoothing, it obtains the main pipeline pressure, drug temperature, and branch volumetric flow rate of each distributed filling machine. The branch volumetric flow rates are then aggregated to obtain the total volumetric flow rate of the entire plant. The drug temperature is converted to drug density using a pre-defined drug property mapping table. Based on the transient disturbance of the main pipeline pressure caused by changes in the total volumetric flow rate of the entire plant, the comprehensive loss index of the veterinary drugs in the main liquid supply network is calculated. Based on the comprehensive loss index and the rate of change of the main pipeline pressure, the branch volumetric flow rate and drug density of each distributed filling machine are compensated to obtain the node mass flow rate of each distributed filling machine. Based on the moving average momentum characteristics of the node mass flow rate, the advance valve closing time of each distributed filling machine is calculated. The surplus volume is predicted based on the advance valve closing time to implement cutoff control for each distributed filling machine, and the system combines the comprehensive loss index with global network oscillation interlock protection control.

[0007] This invention obtains the main pipeline pressure, drug density, and total plant volumetric flow rate, calculates the comprehensive loss index based on transient disturbances, and obtains the node mass flow rate by combining the pressure change rate, thereby obtaining the advance valve closing time. Finally, it combines the predicted surplus volume and comprehensive loss index to execute cutoff control and global oscillation interlock protection control. Furthermore, it integrates the macroscopic energy dissipation assessment of the entire plant's pipeline network with the transient forward surge simulation of local branch veterinary drugs, dynamically compensates for the time difference of veterinary drug inertia and mechanical hysteresis, transforms the closed-loop cutoff process into a forward-looking volumetric simulation, and actively implements interlock interception when the system cross-interference is severe, reducing the deviation of the single-bottle capacity of veterinary drugs under complex veterinary drug operating conditions and reducing the risk of destructive impact triggered by concurrent startup.

[0008] Preferably, the step of collecting the original main pipeline pressure, original drug temperature, and original branch pipe volumetric flow rate of each distributed filling machine from the main liquid supply network, and performing smoothing processing to obtain the main pipeline pressure, drug temperature, and branch pipe volumetric flow rate of each distributed filling machine includes: High-frequency micro-pressure sensors and temperature sensors are deployed at the end of the main supply pipeline of veterinary drugs, and volumetric flow meters are deployed at the branch nodes of all parallel distributed filling machines. The factory control system synchronously collects the original main pipeline pressure, the original drug temperature, and the original branch volumetric flow of each distributed filling machine at fixed sampling time intervals. The original main pipeline pressure, original liquid temperature, and original branch pipe volumetric flow rate of each distributed filling machine are smoothed by a moving average filtering algorithm to obtain the main pipeline pressure, liquid temperature, and branch pipe volumetric flow rate of each distributed filling machine.

[0009] Preferably, the step of summing the volumetric flow rates of the branch pipes to obtain the total volumetric flow rate of the entire plant; and converting the liquid temperature into liquid density using a preset liquid property mapping table, includes: The total volumetric flow rate of the entire plant is obtained by summing the branch pipe volumetric flow rates of all parallel distributed filling machines. Using a drug property mapping table, the drug temperature is converted to drug density via cubic spline interpolation. It should be noted that the drug property mapping table is constructed as follows: within the normal production temperature range of the veterinary drug solution, exemplarily 10℃-40℃, temperature test points are set in 5℃ increments. A liquid density meter is used to test the drug density at each temperature point, and the corresponding data for each temperature-density pair are recorded to obtain the drug property mapping table. The cubic spline interpolation algorithm employs natural boundary conditions, meaning the second derivative of the interpolation polynomial is zero at both ends of the temperature range, to avoid severe oscillations in the interpolation curve at the boundaries.

[0010] Preferably, the comprehensive loss index satisfies the following relationship: ; In the formula, This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; Indicates The total number of sampling points contained within the analysis window ending at a given time. Indicates the first in the analysis window The pressure of the main pipeline at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; This indicates the preset rated main pipeline pressure; This indicates the preset total rated volumetric flow rate for the entire plant.

[0011] This invention extracts and analyzes the transient disturbance characteristics of the main pipeline pressure and the total volumetric flow rate gradient within the analysis window, and combines the rated main pipeline pressure and the rated total volumetric flow rate of the entire plant to obtain a comprehensive loss index. This allows for the assessment of the degree of transient energy loss in the conversion of veterinary drugs into internal frictional dissipation at the entire plant's pipeline level. It also restores the power fluctuations caused by the asynchronous operation of multiple distributed filling machines on the main liquid supply network, reduces the blind spots in the assessment of internal energy loss in the system under multi-machine coupling conditions, and provides benchmark physical parameters for the accurate capture of subsequent dynamic forward surge characteristics.

[0012] Preferably, the node quality flow rate satisfies the following relationship: ; In the formula, Indicates the first A distributed filling machine in Node quality flow at any given moment; Indicates in The density of the liquid medicine at any given time; Indicates the first A distributed filling machine in The branch volumetric flow rate at any given time; This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; This represents the preset pipe deformation coefficient; express The pressure in the main pipeline at any given time; express The pressure in the main pipeline at any given time; This indicates the preset rated pressure of the main pipeline.

[0013] This invention integrates the comprehensive loss index and basic veterinary drug parameters, and combines pipeline deformation and pressure fluctuation characteristics to obtain the node mass flow rate of each distributed filling machine. This combines global energy dissipation parameters with local instantaneous pressure waveform attenuation factors to restore the actual physical thrust of veterinary drugs at the junction nodes. It makes up for the deficiency of a single sensor in not being able to detect global resistance and pipe wall conduction attenuation, and truly reflects the ratio of the basic mass of veterinary drugs entering a specific branch after deducting cross oscillation dissipation and additional forward driving force, reducing the sluggishness of branch input momentum assessment under disturbed conditions.

[0014] Preferably, the advance valve closing time satisfies the following relationship: ; In the formula, Indicates the first A distributed filling machine in The valve closing time should be set in advance. Indicates the first The inherent mechanical delay time of a distributed filling machine; Indicates The total number of sampling points included within the evaluation window ending at time 1; Indicates the first The distributed filling machine in the first Node mass flow rate at each sampling point; Indicates the first Rated volumetric flow rate of a distributed filling machine; This indicates the density of the reference drug solution.

[0015] This invention combines the sliding average momentum characteristics of node mass flow rate with inherent mechanical delay time to obtain the advance valve closing time of each distributed filling machine. It integrates the inherent action hysteresis attribute of electromagnetic filling valve with the local transient veterinary drug forward impulse in the near valve closing stage, and deduces the additional inertial thrust ratio of the current veterinary drug exceeding the standard working condition. It dynamically adjusts the advance amount of the cutoff control, compensates for the timing deviation caused by the hysteresis of the hardware actuator, suppresses the interference of veterinary drug inertial fluctuations on the cutoff control, and enables the timing of the output control command to adaptively match the actual mechanical state of the veterinary drug.

[0016] Preferably, the step of predicting the surplus volume based on the advance valve closing time to perform cutoff control on each distributed filling machine includes: Multiply the advance valve closing time of each distributed filling machine at the current moment with the corresponding branch pipe volume flow rate to predict the surplus volume that each distributed filling machine will continue to flow out during the period when the advance valve closing time expires; when the sum of the cumulative volume of the filled containers of any distributed filling machine and the surplus volume is greater than or equal to the target filling volume of a single bottle, the factory control system issues a power-off control command to the distributed filling machine through the distributed digital control network bus, so that the corresponding filling valve completes physical closure within the advance valve closing time.

[0017] This invention predicts surplus volume based on the filling characteristics of the branch pipe volume flow rate and the valve closing time in advance. When the comprehensive volume meets the target filling volume of a single bottle, a power-off control command is issued through a distributed digital control network bus. This fully utilizes the dynamic lead time and the instantaneous filling rate of the branch pipe to deduce the additional inflow volume caused by the inertia of veterinary drugs during the physical closure of the valve. It transforms the lagging physical execution process into a pre-judgment of volume control, overcomes the defect of passive threshold control strategy that cannot predict the overflow of veterinary drugs during the action, improves the interception accuracy of the single machine end under complex pipeline interference, and reduces the phenomenon of overfilling and out-of-range capacity of veterinary drugs in high-speed assembly line operations.

[0018] Preferably, obtaining the cumulative volume includes: The factory control system combines the sampling time interval to synchronously and discretely accumulate the branch pipe volume flow of each distributed filling machine, thereby obtaining the cumulative volume of each distributed filling machine that has been filled into the container at the current moment.

[0019] Preferably, the method of combining the comprehensive loss index for global pipeline oscillation interlock protection control includes: The factory control system acquires the current comprehensive loss index in real time and compares it with the preset pipeline oscillation safety threshold. When the comprehensive loss index is less than the pipeline oscillation safety threshold, it indicates that the cross-oscillation of veterinary drugs in the main liquid supply pipeline of the entire plant is too violent and the internal friction dissipation is serious. The factory control system temporarily intercepts and suspends the valve opening commands of all distributed filling machines in standby state until the comprehensive loss index rises back to a level greater than or equal to the pipeline oscillation safety threshold. Then, the factory control system releases the interlock and resumes issuing valve opening commands.

[0020] Secondly, the present invention provides an intelligent control system for veterinary drug production, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned intelligent control method for veterinary drug production is implemented.

[0021] By adopting the above technical solution, a computer program is generated from the above-mentioned intelligent control method for veterinary drug production and stored in a memory so that it can be loaded and executed by a processor. In this way, a terminal device can be made based on the memory and the processor for convenient use.

[0022] The beneficial effects of this invention are as follows: This invention addresses the veterinary drug production scenario where multiple distributed filling machines operate collaboratively. By synchronously processing the transient disturbances of the total volumetric flow rate and main pipeline pressure of the entire plant, it obtains a comprehensive loss index, integrates the branch pipe pressure change rate to deduce the node mass flow rate, and then combines the moving average momentum characteristics to extract the dynamic advance valve closing time. Finally, it implements pre-cutoff control based on the predicted surplus volume and implements global oscillation interlock protection control of the pipeline network in conjunction with the comprehensive loss index. This breaks through the limitations of isolated scheduling of single machines in traditional flexible manufacturing systems, dynamically compensates for the macroscopic water hammer cross-interference of the entire plant's pipeline network and the inherent mechanical lag of the terminal filling valves, transforms the lag of passively waiting for the cumulative threshold into a volume prediction that conforms to the real veterinary drug kinetic laws, and actively implements system-level interlock interception under the critical condition of a surge in pipeline network kinetic energy. This reduces the periodic error in machine output caused by high-frequency concurrent flow operations, reduces the physical risk of the entire plant's pipeline network being subjected to destructive veterinary drug impacts, and improves the global allocation accuracy and centralized liquid supply scheduling stability of veterinary drug group control production under complex and disturbed conditions. Attached Figure Description

[0023] Figure 1 This is a flowchart of an intelligent control method for veterinary drug production according to the present invention; Figure 2 This is a diagram illustrating node quality flow. Figure 3 This is a schematic diagram of cumulative volume dynamic tracking and pre-truncation control; Figure 4 This is a schematic diagram of the monitoring status of the global oscillation interlock protection of the pipeline network. Detailed Implementation

[0024] This invention discloses an intelligent control method for veterinary drug production, referring to... Figure 1 This includes steps S100-S500: S100, data acquisition and simulation of total plant volumetric flow rate and liquid density.

[0025] It should be noted that, considering that temperature changes can cause nonlinear drift in veterinary drug density, and that high-frequency sampling combined with filtering can separate the baseline physical trend of the signal, this invention simultaneously collects the original main pipeline pressure, the original drug solution temperature, and the original branch pipe volumetric flow rate, and uses moving average filtering and temperature-density interpolation conversion to reduce the deviation caused by temperature drift and signal noise in subsequent kinetic calculations.

[0026] Specifically, high-frequency micro-pressure sensors and temperature sensors are deployed at the end of the main veterinary drug supply pipeline, and volumetric flow meters are deployed at the branch pipe nodes of all parallel distributed filling machines. The factory control system synchronously collects the original main pipeline pressure, the original drug solution temperature, and the original branch pipe volumetric flow rate of each distributed filling machine at fixed sampling time intervals. For example, the sampling time interval is 10ms to capture high-frequency transient oscillations of the veterinary drug within the pipeline network.

[0027] The original main pipeline pressure, original liquid temperature, and original branch pipe volumetric flow rates of each distributed filling machine are smoothed to obtain the main pipeline pressure, liquid temperature, and branch pipe volumetric flow rates of each distributed filling machine. For example, the smoothing process uses a moving average filtering algorithm. The core parameters of the moving average filtering algorithm include a sliding window length of 3-5 sampling points and a sliding step size of 1 sampling point. The moving average filtering algorithm is existing technology and will not be elaborated upon here.

[0028] The total volumetric flow rate of the entire plant is obtained by summing the branch pipe volumetric flow rates of all parallel distributed filling machines. Using a drug property mapping table, the drug temperature is converted into drug density using a cubic spline interpolation algorithm. It should be noted that the cubic spline interpolation algorithm is existing technology and will not be elaborated upon here.

[0029] Thus, the main pipeline pressure, liquid density, and total flow rate of the entire plant were obtained.

[0030] S200, Calculate the comprehensive loss index of veterinary drugs in the main liquid supply network.

[0031] It should be noted that since the flow of each branch in the centralized liquid supply system is supplied by the main liquid supply network, the change in the total volumetric flow of the entire plant will inevitably be reflected through the pressure response of the main pipeline. Therefore, the pressure of the main pipeline can be equivalently characterized as the unified driving force of the entire plant's veterinary drug system. This invention utilizes the product and accumulation of the gradient of the main pipeline pressure and the total volumetric flow of the entire plant within the analysis window to construct a comprehensive loss index, thereby characterizing the equivalent work done by veterinary drugs on flow disturbances at the pipeline network scale, reducing the blind spot of unknown internal losses in the system.

[0032] Specifically, based on the transient disturbance of the main pipeline pressure caused by the change in the total volumetric flow rate of the entire plant, the comprehensive loss index of veterinary drugs in the main liquid supply network is calculated, including: An analysis window is set with the current time as the endpoint. For example, the analysis window contains 50 sampling points to cover a complete typical veterinary drug oscillation cycle. If the analysis window is too large, for example, reaching two hundred sampling points, it will introduce stationary data from historical periods or decaying irrelevant oscillations, causing a delay in the calculation of the current transient frictional dissipation. If the analysis window is too small, for example, only five sampling points, it cannot cover the complete physical oscillation waveform and is easily affected by local random noise, leading to deviations in loss assessment.

[0033] It should be further explained that in fluid engineering, the product of pressure and volumetric flow rate represents the hydraulic power. This invention multiplies the main pipeline pressure by the transient gradient of the total plant volumetric flow rate, which characterizes the transient power fluctuations caused by asynchronous valve start-stop. Simultaneously, dividing the transient power fluctuations by the product of the rated reference parameters completes the dimensionless processing, allowing the calculated comprehensive loss index to escape the limitations of specific dimensions and objectively reflect the degree of internal friction loss when multiple machines operate concurrently. (The main fluid supply network for veterinary drugs...) The overall loss index at time t satisfies the following relationship: ; In the formula, This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; Indicates The total number of sampling points contained within the analysis window ending at a given time. Indicates the first in the analysis window The pressure of the main pipeline at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; The preset rated main pipeline pressure is set to 0.3 MPa in this embodiment. The rated main pipeline pressure is determined through preliminary rheological testing to avoid high-speed frictional foaming of the liquid and maintain stable laminar flow. The preset rated total volumetric flow rate of the entire plant is set to 5L / s in this embodiment. It is calculated by multiplying the total number of parallel distributed filling machines in the plant by the peak flow rate of a single machine, and then multiplying by a safety factor. For example, the safety factor is set to 1.2.

[0034] In this relation, This indicates the transient power fluctuation caused by the sudden changes in flow rate of the liquid medicine in the main pipeline due to the start and stop of valves of multiple distributed filling machines. The larger the absolute value of the transient power fluctuation, the stronger the impact of the asynchronous operation of multiple filling machines on the main liquid supply network, and the more violent the transient kinetic energy fluctuation inside the main liquid supply network. Conversely, it indicates that the veterinary drug delivery state in the main liquid supply network is more stable and the mechanical work is more stable. This indicates the percentage of cumulative energy loss in the main liquid supply network of the entire plant, caused by disordered cross-oscillation of veterinary drugs, which is converted into internal frictional dissipation. The larger the percentage of cumulative energy loss, the more severe the water hammer cross-interference caused by the coordinated operation of multiple distributed filling machines, and the higher the proportion of kinetic energy originally used for transportation by the veterinary drugs is lost due to residual oscillation. Conversely, the lower the percentage of overall veterinary drug energy transmission loss in the main liquid supply network of the entire plant, the less severe the degree of multi-machine coupling interference.

[0035] Thus, the comprehensive loss index of veterinary drugs in the main supply pipeline network was obtained.

[0036] S300, calculate the node mass flow rate of each distributed filling machine.

[0037] It should be noted that, considering the global energy loss and the pressure change rate of the main pipeline after material attenuation, the actual physical thrust of the veterinary drug at the junction node can be restored. Therefore, this invention introduces the pipeline deformation coefficient and drug density to correct the branch pipe volume flow rate to the node mass flow rate, thereby compensating for the defect that a single sensor cannot detect the global resistance of the pipeline network and the local pipe wall conduction attenuation, and reducing the error in calculating the input momentum of a single branch.

[0038] It should be further explained that multiplying the drug density by the branch pipe volumetric flow rate only represents the local flow rate at the pipe opening under ideal conditions. However, in the main liquid supply network with multiple distributed filling machines operating concurrently, the disorderly collision of veterinary drugs will generate internal friction, and the original transport kinetic energy will be dissipated, resulting in an inertially high flow velocity. Based on this, the present invention multiplies by a comprehensive loss index to obtain the basic mass flow rate of veterinary drugs, which is equivalent to introducing a dynamic conversion factor for energy attenuation, thereby removing the false flow rate that lacks real kinetic thrust. In addition, the instantaneous water hammer impact is also accompanied by the elastic expansion and absorption of the pipe wall. By using the normalized pressure change rate combined with the pipe deformation coefficient, a local transient forward driving force ratio is constructed. Finally, the node mass flow rate is calculated by combining the basic mass flow rate of veterinary drugs and the transient forward driving force ratio, thus restoring the true node mass flow rate under obstructed conditions.

[0039] Specifically, based on the dynamic compensation logic of pipeline network comprehensive loss and pressure wave propagation attenuation, the node mass flow rate of each distributed filling machine is calculated. For any distributed filling machine... The node quality flow at any given time satisfies the following relationship: ; In the formula, Indicates the first A distributed filling machine in Node quality flow at any given moment; Indicates in The density of the liquid medicine at any given time; Indicates the first A distributed filling machine in The branch volumetric flow rate at any given time; This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; The preset pipe deformation coefficient is set to 0.15 in this embodiment. It is obtained from the material elastic modulus test of the main liquid supply network of the whole plant and is used to characterize the degree of attenuation of the pipe wall for the transmission of instantaneous pressure waves of veterinary drugs. express The pressure in the main pipeline at any given time; express The pressure in the main pipeline at any given time; This indicates the preset rated pressure of the main pipeline.

[0040] In this relation, This indicates that after deducting the cross-oscillation dissipation caused by the asynchronous operation of multiple distributed filling machines in the main liquid supply network of the entire plant, the actual amount that can enter the [stage / process / etc.] is [amount / value]. Veterinary drug basic quality flow rate of a distributed filling machine branch. The larger the value, the more likely it is to enter the [number]th [stage]. The larger the basic flow rate of veterinary drugs for each distributed filling machine, the smaller the veterinary drug kinetic energy dissipation within the main liquid supply network of the entire plant; conversely, it indicates that the flow rate of veterinary drugs entering the first distributed filling machine is smaller. The basic flow rate of veterinary drugs in a distributed filling machine is obstructed, or the kinetic energy of veterinary drugs is dissipated due to friction within the main liquid supply network of the entire plant caused by the coordinated operation of multiple distributed filling machines. This indicates that the instantaneous water hammer pressure surge in the main liquid supply network of the entire plant, after being attenuated by the elasticity of the pipe wall material, is transmitted to the first... The proportion of transient forward driving force of a branch of a distributed filling machine for veterinary drugs. The larger the pressure, the more severe the pressure surge in the plant's main liquid supply network, giving rise to the pressure required to enter the next stage. The distributed filling machine has a strong additional forward momentum for veterinary drugs; conversely, the smaller this factor is, the more it indicates that the main liquid supply network of the entire plant is in a state of sudden pressure drop, weakening the entry into the first... The inherent forward momentum of a distributed filling machine for veterinary drugs.

[0041] For example, Figure 2 This is a schematic diagram of node quality flow. The diagram shows the actual flow fluctuations of each distributed filling machine after dynamic parameter compensation, objectively reflecting the physical thrust that actually enters the local branch after deducting the dissipation of cross-oscillations throughout the plant.

[0042] At this point, the node quality flow of each distributed filling machine was obtained.

[0043] S400, Calculate the advance valve closing time for each distributed filling machine.

[0044] It should be noted that, due to the inherent mechanical time lag between the electromagnetic filling valve receiving the control command and the physical cut-off of the liquid flow, and given the residual oscillations in the veterinary drug pipeline, the forward flow velocity is not constant each time it approaches valve closure. This invention uses the inherent mechanical delay time as a basis, superimposed with a compensation time based on the real-time node mass flow rate ratio, to dynamically output the valve closure time in advance. This combines the hardware response action with the transient veterinary drug dynamics state, suppressing the interference caused by the inertial fluctuations of the veterinary drug on the timing of the cut-off action.

[0045] Specifically, based on the moving average momentum characteristics of the node mass flow rate, the advance valve closing time of each distributed filling machine is calculated, including: An evaluation window is set with the current time as the endpoint. For example, the evaluation window contains 20 sampling points to capture the transient flow fluctuation characteristics of the branch pipe near the valve closing stage. If the evaluation window is too large, for example, reaching fifty sampling points equal to the analysis window, the actual instantaneous forward inertia of the branch pipe opening of the distributed filling machine will be smoothed out due to averaging, resulting in a sluggish calculation of the early valve closing time. If the evaluation window is too small, for example, only three sampling points, it is easily affected by the transient electrical noise of a single sensor, causing logical jitter in the issued valve closing command.

[0046] It should be further explained that this invention divides the average node mass flow rate of veterinary drugs by the product of the reference drug density and the rated volumetric flow rate to construct a dimensionless transient forward inertia ratio, which reflects the relative degree of deviation of the current branch veterinary drug's true momentum from the ideal reference. Multiplying the transient forward inertia ratio by the inherent mechanical delay time represents a dynamic scaling of the time scale based on the static response time of the actuator and the impact inertia of the real-time fluid, so that the output early valve closing time can dynamically adapt to the inertia differences under different flow rates.

[0047] The valve closing time of any distributed filling machine satisfies the following relationship: ; In the formula, Indicates the first A distributed filling machine in The valve closing time should be set in advance. Indicates the first The inherent mechanical delay time of the distributed filling machine is set to 0.05s in this embodiment. This inherent delay benchmark is obtained by testing the physical response time of the electromagnetic filling valve of the distributed filling machine. Indicates The total number of sampling points included within the evaluation window ending at time 1; Indicates the first The distributed filling machine in the first Node mass flow rate at each sampling point; Indicates the first The rated volumetric flow rate of each distributed filling machine is set to 0.1 L / s in this embodiment, which is based on the standard output capacity boundary of the branch pipe design of a single distributed filling machine. The reference drug solution density is set to 1000 kg / m³ in this embodiment, which is obtained by using the physical density of pure water under standard environmental conditions.

[0048] In this relation, This indicates that within the evaluation window, you are entering the [stage / process]. The average node mass flow rate of veterinary drugs in the branch pipelines of a distributed filling machine reflects the proximity of the nodes. The value represents the actual average forward momentum of the veterinary drug at any given moment. A larger value indicates a stronger forward inertia of the veterinary drug in the distributed filling machine branch just before valve closure, resulting in a greater actual inflow momentum of the veterinary drug. Conversely, a smaller value indicates a weaker forward potential energy of the veterinary drug entering the distributed filling machine branch, which tends towards a stable state. The greater the inherent mechanical delay time of a distributed filling machine, the slower the mechanical transmission of the valve, and the more the control system needs to advance the timing of issuing the shut-off command to compensate for the lag in mechanical action; conversely, it indicates that the valve's physical response is rapid and the mechanical execution process takes less time.

[0049] Thus, the advance valve closing time for each distributed filling machine was obtained.

[0050] S500 implements pre-cutoff control and global interlock protection.

[0051] It should be noted that, due to the physical action time required for the solenoid valve to fully shut off the flow from receiving the signal, and considering the high transport kinetic energy of veterinary drugs within the main supply pipeline, and taking into account that the branch pipe will continue to pour liquid into the container due to inertia during the specific time period of slow valve closure, relying solely on the existing accumulated volume for shut-off would inevitably lead to overflow of the actual filling volume. Therefore, this invention logically simulates the surplus volume during valve action by multiplying the dynamically calculated advance valve closing time by the instantaneous flow rate, thereby transforming the delayed physical shut-off process into a forward-looking volume simulation. At the same time, considering that the instantaneous concurrent startup of multiple machines will cause violent pressure fluctuations within the pipeline, the critical risk of pipeline energy dissipation is determined by the real-time drop of the comprehensive loss index. Therefore, a global interlock protection strategy based on loss characteristics is introduced to suppress the risk of veterinary drug impact under extreme operating conditions.

[0052] Specifically, the factory control system combines the sampling time interval to synchronously and discretely accumulate the branch pipe volume flow of each distributed filling machine, thereby obtaining the cumulative volume of each distributed filling machine that has been filled into the container at the current moment.

[0053] Multiply the advance valve closing time of each distributed filling machine at the current moment with the corresponding branch pipe volume flow rate to predict the surplus volume that each distributed filling machine will continue to flow out during the period when the advance valve closing time expires; when the sum of the cumulative volume of the filled containers of any distributed filling machine and the surplus volume is greater than or equal to the target filling volume of a single bottle, the factory control system issues a power-off control command to the distributed filling machine through the distributed digital control network bus, so that the corresponding filling valve completes physical closure within the advance valve closing time.

[0054] For example, Figure 3 This is a schematic diagram of cumulative volume dynamic tracking and pre-cutoff control. The diagram shows the forward-looking volume prediction process when each distributed filling machine approaches the target filling volume of a single bottle, based on the pre-closed valve time to extrapolate the surplus volume and issue the valve closing command in advance.

[0055] The factory control system acquires the current comprehensive loss index in real time and compares it with a preset pipeline oscillation safety threshold. When the comprehensive loss index is less than the pipeline oscillation safety threshold, it indicates that the cross-oscillation of veterinary drugs within the main liquid supply pipeline of the entire plant is too severe and internal friction dissipation is serious. The factory control system temporarily intercepts and suspends the valve opening commands of all distributed filling machines in standby mode until the comprehensive loss index rises back to a level greater than or equal to the pipeline oscillation safety threshold. Then, the factory control system releases the interlock and resumes issuing valve opening commands. For example, the process for determining the pipeline oscillation safety threshold is as follows: keep all filling machine valves closed, maintain the rated main pipeline pressure, and record the comprehensive loss index at this time; turn on each filling machine to the rated flow rate until the entire plant reaches the rated total volume flow rate, and record the lowest comprehensive loss index under multi-machine concurrent operation; multiply the lowest comprehensive loss index by 0.8 to obtain the final pipeline oscillation safety threshold, with an example value of 0.3.

[0056] For example, Figure 4 This is a schematic diagram of the monitoring status of the pipeline network global oscillation interlock protection. The diagram reflects the trend of the comprehensive loss index dynamically falling with the concurrent interference of the system, and triggering interlock interception to suppress the impact of veterinary drugs when it falls below the pipeline network oscillation safety threshold.

[0057] Thus, the adaptive filling control of veterinary drugs and the global oscillation interlock protection of the pipeline network were completed.

[0058] This invention also discloses an intelligent control system for veterinary drug production, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for intelligent control of veterinary drug production according to the present invention is implemented.

[0059] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0060] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A method for intelligent control of veterinary drug production, characterized in that, include: The original main pipeline pressure, original liquid temperature, and original branch volume flow rate of each distributed filling machine are collected from the main liquid supply network and smoothed to obtain the main pipeline pressure, liquid temperature, and branch volume flow rate of each distributed filling machine. The volumetric flow rates of the branch pipes are summarized to obtain the total volumetric flow rate of the entire plant; the liquid temperature is converted into liquid density using a preset liquid property mapping table. Based on the transient disturbance of the main pipeline pressure caused by the change in the total volumetric flow rate of the entire plant, the comprehensive loss index of veterinary drugs in the main liquid supply network is calculated; based on the comprehensive loss index and the rate of change of the main pipeline pressure, the branch volumetric flow rate and drug density of each distributed filling machine are compensated to obtain the node mass flow rate of each distributed filling machine. Based on the moving average momentum characteristics of the node mass flow rate, calculate the advance valve closing time for each distributed filling machine; The surplus volume is predicted based on the advance valve closing time to cut off control of each distributed filling machine, and the overall network oscillation interlock protection control is combined with the comprehensive loss index.

2. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The process of collecting the original main pipeline pressure, original drug temperature, and original branch volumetric flow rate of each distributed filling machine from the main liquid supply network, and smoothing them to obtain the main pipeline pressure, drug temperature, and branch volumetric flow rate of each distributed filling machine includes: High-frequency micro-pressure sensors and temperature sensors are deployed at the end of the main supply pipeline of veterinary drugs, and volumetric flow meters are deployed at the branch nodes of all parallel distributed filling machines. The factory control system synchronously collects the original main pipeline pressure, the original drug temperature, and the original branch volumetric flow of each distributed filling machine at fixed sampling time intervals. The original main pipeline pressure, original liquid temperature, and original branch pipe volumetric flow rate of each distributed filling machine are smoothed by a moving average filtering algorithm to obtain the main pipeline pressure, liquid temperature, and branch pipe volumetric flow rate of each distributed filling machine.

3. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The process involves summing the volumetric flow rates of the branch pipes to obtain the total volumetric flow rate of the entire plant; and converting the liquid temperature into liquid density using a pre-set liquid property mapping table, including: The total volumetric flow rate of the entire plant is obtained by summing the branch pipe volumetric flow rates of all parallel distributed filling machines; the liquid temperature is converted into liquid density by using a cubic spline interpolation algorithm in conjunction with the liquid property mapping table.

4. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The comprehensive loss index satisfies the following relationship: ； In the formula, This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; Indicates The total number of sampling points contained within the analysis window ending at a given time. Indicates the first in the analysis window The pressure of the main pipeline at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; Indicates the first in the analysis window Total volumetric flow rate of the entire plant at each sampling point; This indicates the preset rated main pipeline pressure; This indicates the preset total rated volumetric flow rate for the entire plant.

5. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The node quality flow rate satisfies the following relationship: ； In the formula, Indicates the first A distributed filling machine in Node quality flow at any given moment; Indicates in The density of the liquid medicine at any given time; Indicates the first A distributed filling machine in The branch volumetric flow rate at any given time; This indicates that the main supply pipeline for veterinary drugs is in The overall loss index at any given moment; This represents the preset pipe deformation coefficient; express The pressure in the main pipeline at any given time; express The pressure in the main pipeline at any given time; This indicates the preset rated pressure of the main pipeline.

6. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The valve closing time in advance satisfies the following relationship: ； In the formula, Indicates the first A distributed filling machine in The valve closing time should be set in advance. Indicates the first The inherent mechanical delay time of a distributed filling machine; Indicates The total number of sampling points included within the evaluation window ending at time 1; Indicates the first The distributed filling machine in the first Node mass flow rate at each sampling point; Indicates the first Rated volumetric flow rate of a distributed filling machine; This indicates the density of the reference drug solution.

7. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The step of predicting surplus volume based on the advance valve closing time to perform cutoff control on each distributed filling machine includes: Multiply the advance valve closing time of each distributed filling machine at the current moment with the corresponding branch pipe volume flow rate to predict the surplus volume that each distributed filling machine will continue to flow out during the period when the advance valve closing time expires; when the sum of the cumulative volume of the filled containers of any distributed filling machine and the surplus volume is greater than or equal to the target filling volume of a single bottle, the factory control system issues a power-off control command to the distributed filling machine through the distributed digital control network bus, so that the corresponding filling valve completes physical closure within the advance valve closing time.

8. The intelligent control method for veterinary drug production according to claim 7, characterized in that, The acquisition of the cumulative volume includes: The factory control system combines the sampling time interval to synchronously and discretely accumulate the branch pipe volume flow of each distributed filling machine, thereby obtaining the cumulative volume of each distributed filling machine that has been filled into the container at the current moment.

9. The intelligent control method for veterinary drug production according to claim 1, characterized in that, The method of combining the comprehensive loss index for global pipeline oscillation interlock protection control includes: The factory control system acquires the current comprehensive loss index in real time and compares it with the preset pipeline oscillation safety threshold. When the comprehensive loss index is less than the pipeline oscillation safety threshold, it indicates that the cross-oscillation of veterinary drugs in the main liquid supply pipeline of the entire plant is too violent and the internal friction dissipation is serious. The factory control system temporarily intercepts and suspends the valve opening commands of all distributed filling machines in standby state until the comprehensive loss index rises back to a level greater than or equal to the pipeline oscillation safety threshold. Then, the factory control system releases the interlock and resumes issuing valve opening commands.

10. An intelligent control system for veterinary drug production, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a smart control method for veterinary drug production according to any one of claims 1-9.

Images