Intelligent digital dosing method and system based on edge computing

Abstract

The application belongs to the technical field of intelligent control, and particularly relates to an intelligent digital dosing method and system based on edge computing, which comprises the following steps: calculating a system dynamic inertia period representing a comprehensive time lag of a dosing process by using pipeline volume, real-time flow rate and reference reaction time, correcting a static deviation of a pH value from a target set value by using the inertia period and a real-time change rate of the pH value to obtain a forward-looking state deviation, calculating a dynamic control gain modulation factor by using a nonlinear mapping function, applying the dynamic control gain modulation factor to a feedback control algorithm to generate and drive an accurate control amount of a digital dosing pump. The application suppresses oscillation of a control process, and significantly improves stability, dosing accuracy and self-adaptive capacity to working condition changes of the system.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology. More specifically, this invention relates to an intelligent digital dosing method and system based on edge computing. Background Technology

[0002] In industrial processes such as chemical production and water treatment, accurately controlling the dosage of chemical agents is a key step in ensuring product quality, treatment effectiveness, and cost control.

[0003] Traditional dosing systems often rely on manual timed and quantitative control or simple open-loop control, which is difficult to cope with real-time changes in operating conditions, leading to waste of chemicals or substandard treatment.

[0004] To address this issue, intelligent digital dosing systems integrating sensors and control algorithms have emerged, particularly edge computing-based dosing systems. These systems achieve low latency and high autonomy in control by performing data processing and decision-making on edge devices located close to the data source. Typically, these systems acquire fluid parameters in real time through sensors, calculate the required dosage using algorithms within the edge controller, and then drive the digital dosing pump to perform the dosing.

[0005] However, when implementing intelligent control, existing technologies primarily rely on snapshot data collected by sensors at the current moment for decision-making. This approach ignores the inherent physical and chemical lags in the dosing process itself. For example, there is a transmission delay from the dosing point to the downstream sensor detection point, as well as the mixing and reaction time required for the chemical to react with the fluid and reach a stable state. If the controller cannot effectively distinguish between the two states of insufficient dosage and ineffective efficacy, and only adjusts based on instantaneous errors, it is prone to frequent adjustments of control commands, causing severe overshoot and continuous oscillations in the system near the target setpoint. This not only results in additional consumption of chemicals but also makes it difficult to maintain stable final water quality. Summary of the Invention

[0006] To address the technical problem of accurately modeling the temporal dynamic characteristics of the drug delivery process at the edge to overcome control instability caused by system inertia, this invention provides solutions in the following aspects.

[0007] In a first aspect, the present invention provides an intelligent digital dosing method based on edge computing, comprising: acquiring the real-time flow rate and pH value of the fluid in the dosing pipeline; calculating the system dynamic inertia period, characterizing the comprehensive time lag of the dosing process, using the pipeline volume, the real-time flow rate, and a preset baseline reaction time; calculating a forward-looking state deviation for predicting the future state of the system based on the static deviation between the pH value and the target setpoint of the pH value, and the product of the rate of change of the pH value and the system dynamic inertia period; calculating a dynamic control gain modulation factor based on the absolute value of the forward-looking state deviation using a nonlinear mapping function, wherein the nonlinear mapping function makes the dynamic control gain modulation factor positively correlated with the absolute value of the forward-looking state deviation; and incorporating the dynamic control gain modulation factor into the system's feedback control algorithm to obtain the control quantity of the digital dosing pump at the next moment, driving the digital dosing pump to perform the dosing operation.

[0008] This invention introduces a system dynamic inertial period to dynamically quantify the combined lag of drug delivery and chemical reaction, and calculates a forward-looking state deviation that can predict the future state of the system. A nonlinear function is used to dynamically adjust the gain of the control algorithm according to the magnitude of the predicted deviation, realizing intelligent regulation that intervenes strongly when the deviation is large, makes fine adjustments when the deviation is small, and slows down in advance when the trend is positive. Combining forward-looking and adaptive gain, this invention fundamentally solves the problem of control instability caused by lag, making the dosing process more stable and precise, and effectively avoiding drug waste and water quality fluctuations.

[0009] Preferably, the calculation of the system dynamic inertial period, which characterizes the overall time lag of the dosing process, using the pipeline volume, the real-time flow rate, and the preset reference reaction time, includes: In the formula: express The system's dynamic inertial period at any given moment; This indicates the pipe volume from the reagent dosing point to the downstream sensor installation location; express Real-time flow rate at any given moment; , These represent the baseline reaction time and the baseline flow rate, respectively. The coefficient representing the influence of mixing efficiency; This represents the natural exponential function.

[0010] This invention determines the system dynamic inertial period, which characterizes the overall time lag of the dosing process, based on both physical transmission delay and chemical reaction delay. This can more accurately reflect the different effects of flow rate changes on transmission time and mixing reaction efficiency, providing a more reliable basis for subsequent predictive control.

[0011] Preferably, the step of calculating the forward-looking state deviation for predicting the future state of the system based on the static deviation between the pH value and the target set value, and the product of the rate of change of the pH value and the dynamic inertia period of the system, includes: In the formula: express Foresight bias at any given moment; This indicates the target setpoint for pH value; , They represent time, pH value after fluid mixing reaction at a given time; This indicates the static deviation of the pH value from the target pH setting. Indicates the time interval for data collection; express The rate of change of pH value after the fluid mixing reaction at a given time; express The system's dynamic inertial period at any given moment.

[0012] This invention obtains a forward-looking state deviation for predicting the future state of the system by subtracting a predicted value representing the future trend from the current static deviation. When the system pH value is rapidly approaching the target value, the forward-looking deviation is made much smaller than the static deviation, thereby weakening the control effect in advance and effectively suppressing overshoot.

[0013] Preferably, the step of calculating the dynamic control gain modulation factor based on the absolute value of the forward state deviation using a nonlinear mapping function includes: In the formula: express Dynamic control gain modulation factor at any given time; express Foresight bias at any given moment; Indicates taking the absolute value; Represents the natural exponential function; This represents the gain sensitivity coefficient.

[0014] This invention provides a smooth and continuous curve from 0 to 1 through a nonlinear mapping function, mapping the forward state deviation to a dynamic control gain modulation factor. This makes the adjustment process of the control force smooth, enabling rapid response when the deviation is large and fine-tuning when the deviation is small. It effectively avoids system jitter caused by sudden gain changes and ensures the smoothness and stability of the control process.

[0015] Preferably, the step of incorporating the dynamic control gain modulation factor into the system's feedback control algorithm to obtain the control quantity of the digital dosing pump at the next moment includes: In the formula: express The control quantity of the digital dosing pump at any given time; express The basic feedforward dosage at any given time; , These are the proportional gain coefficient and integral gain coefficient in the PID controller, respectively. express Dynamic control gain modulation factor at any given time; express The static deviation of pH value after fluid mixing reaction at time t, and , express pH value of the fluid after the reaction at a given time. This indicates the target set value for pH.

[0016] This invention adds feedforward control based on real-time flow rate to the feedback control, which enables the system to respond more quickly and with smaller deviations when dealing with changes in major operating conditions such as flow rate, thereby enhancing the system's anti-disturbance capability and dynamic performance.

[0017] Preferably, the basic feedforward dosage , express Real-time flow rate at any given moment This indicates the target concentration that the drug needs to achieve in the pipeline. This indicates the concentration of the pharmaceutical stock solution itself.

[0018] Preferably, the driving of the digital dosing pump to perform the dosing operation includes: transmitting the calculated control quantity of the digital dosing pump through an edge computing device. The signal is converted into a hardware control signal, which drives the digital dosing pump to precisely execute the dosing operation, thereby completing a closed-loop control.

[0019] Preferably, the preset baseline reaction time, target pH value, and pipe volume are... The target concentration of the agent to be achieved in the pipeline and reference flow rate These are all system calibration parameters; among them, the target setpoint for pH and the reference flow rate are... The target concentration of the agent to be achieved in the pipeline The system's human-machine interface allows operators to set parameters according to process requirements; and to consider the pipeline distance between the reagent dosing point and the downstream sensor installation location. and pipe diameter Calculate the pipe volume from the reagent dosing point to the downstream sensor installation location. The baseline reaction time was obtained through experimental calibration.

[0020] Preferably, the method for obtaining the reference reaction time is as follows: at a reference flow rate Next, perform a single dosing operation according to process requirements; continuously monitor and record the pH value change curve using a downstream pH sensor; measure the time elapsed from the start of reagent dosing until the downstream pH sensor reading reaches and remains at the target set value, and use this as the reaction time obtained in one experiment; obtain multiple reaction times through multiple trials, and use the average of the multiple reaction times as the baseline reaction time. .

[0021] Secondly, the present invention provides an intelligent digital dosing system based on edge computing, 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 digital dosing method based on edge computing is implemented.

[0022] By adopting the above technical solution, a computer program for the intelligent digital dosing method based on edge computing is generated and stored in a memory for loading and execution by a processor. This allows for the creation of a terminal device based on the memory and processor, making it convenient to use.

[0023] The beneficial effects of this invention are as follows:

[0024] This invention introduces a system dynamic inertial period to dynamically quantify the combined lag of drug delivery and chemical reaction, and calculates a forward-looking state deviation that can predict the future state of the system. A nonlinear function is used to dynamically adjust the gain of the control algorithm according to the magnitude of the predicted deviation, realizing intelligent regulation that intervenes strongly when the deviation is large, makes fine adjustments when the deviation is small, and slows down in advance when the trend is positive. Combining forward-looking and adaptive gain, this invention fundamentally solves the problem of control instability caused by lag, making the dosing process more stable and precise, and effectively avoiding drug waste and water quality fluctuations. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating an intelligent digital dosing method based on edge computing according to the present invention;

[0026] Figure 2 It is a pH curve schematically showing the pH values ​​at multiple points in time;

[0027] Figure 3 It is a flow velocity curve schematically showing the real-time flow velocities at multiple moments;

[0028] Figure 4 It is a schematic diagram showing the curves composed of the control quantities of the digital dosing pump at various times. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] This invention discloses an intelligent digital dosing method based on edge computing, referring to... Figure 1 This includes steps S1-S5:

[0032] S1. Real-time flow rate and pH value are collected during the dosing process through multiple sensors connected to edge computing devices.

[0033] It should be noted that in order to achieve precise closed-loop control of the dosing process, it is first necessary to obtain the real-time status information and basic physical parameters of the system.

[0034] Specifically, the real-time flow rate of the fluid in the pipeline at various times is obtained by a flow sensor deployed on the dosing pipeline; and the real-time chemical parameters, i.e. pH value, of the fluid after mixing and reaction are obtained by a pH sensor deployed at a designated location downstream of the dosing point.

[0035] Furthermore, the system calibration parameters are obtained, including the target setpoint for pH value. Pipeline volume from the reagent dosing point to the downstream sensor installation location The target concentration of the agent to be achieved in the pipeline Reference reaction time and reference flow rate .

[0036] Among them, the target set value of pH value Reference flow rate The target concentration of the agent to be achieved in the pipeline Operators can set the parameters according to process requirements through the system's human-machine interface.

[0037] Additionally, based on the pipeline distance between the reagent dosing point and the downstream sensor installation location... and pipe diameter Calculate the pipe volume from the reagent dosing point to the downstream sensor installation location. .

[0038] The reference reaction time These are inherent system parameters obtained through experimental calibration, and the baseline reaction time represents the time required for the chemical reaction to reach stability at a baseline flow rate. The specific method is as follows: at the baseline flow rate... Next, perform a single dosing operation according to process requirements; continuously monitor and record the pH value change curve using a downstream pH sensor; measure the time elapsed from the start of reagent dosing until the downstream pH sensor reading reaches and remains at the target set value, and use this as the reaction time obtained in one experiment; obtain multiple reaction times through multiple trials, and use the average of the multiple reaction times as the baseline reaction time. .

[0039] S2. Calculate the system dynamic inertial period, which characterizes the overall time lag of the dosing process, using the pipeline volume, real-time flow rate, and preset baseline reaction time.

[0040] It should be noted that the total system delay is not a fixed constant; it is closely related to both the physical transport velocity of the fluid and the chemical reaction kinetics. Higher flow rates result in shorter times for the reagent to reach the detection point, but increased fluid turbulence may promote mixing, thus shortening the reaction settling time. Therefore, an index that dynamically reflects the overall effect, namely the system dynamic inertial period, is needed to characterize the overall time lag effect during the dosing process.

[0041] Specifically, by using the pipeline volume, real-time flow rate at each moment, and baseline reaction time, the physical transport delay caused by changes in flow rate and the chemical reaction delay caused by changes in mixing reaction efficiency are obtained. Then, the system dynamic inertial period at each moment is calculated, characterizing the comprehensive time lag effect during the dosing process. The formula for calculating the system dynamic inertial period at each moment is as follows:

[0042] ;

[0043] In the formula: express The system dynamic inertia period at any given moment, in seconds, represents the estimated total time from the application of control to the generation of a stable response under the current operating conditions; This indicates the pipe volume from the reagent dosing point to the downstream sensor installation location; express Real-time flow rate at any given moment; , These represent the baseline reaction time and the baseline flow rate, respectively. The mixing efficiency influence coefficient is used to characterize the degree of influence of flow rate change on reaction time, and the value range of the mixing efficiency influence coefficient is [0.1, 0.5]. When the value is small, the effect of flow rate changes on the correction of reaction delay is weak. When the value is large, the correction effect is too sensitive, therefore, Setting it to 0.2 can achieve good dynamic response under most operating conditions; This represents the natural exponential function.

[0044] It should be noted that when real-time flow rate is detected... When the value is zero, the system suspends the output update of the control quantity of the digital dosing pump until the flow rate returns to normal.

[0045] Among them, based on real-time flow rate The inertial period, which characterizes the speed of the entire system's response, is calculated. System dynamic inertia period It consists of two parts: Part 1 This represents the physical transmission delay, when the flow rate... As the flow rate increases, this term decreases; the second part represents the chemical reaction delay, which occurs when the real-time flow rate increases. greater than the reference flow rate At that time, the exponent term Reducing the flow rate decreases the reaction delay, simulating a scenario where mixing efficiency is improved and the reaction is accelerated at high flow rates; conversely, when the flow rate decreases, the reaction delay will increase accordingly.

[0046] S3. Calculate the forward-looking state deviation used to predict the future state of the system based on the static deviation between the pH value and the target pH value, and the product of the rate of change of pH value and the dynamic inertial period of the system.

[0047] It should be noted that the pH value collected by the sensor after the fluid mixing reaction is actually the result of control action at a certain point in the past. To make accurate current control decisions, the controller needs to anticipate the future state of the system and consider the system state in inertial time. By analyzing the internal trends and combining them with the system's dynamic inertial period, we can calculate the system's forward-looking state deviation and avoid overreacting based on lagging information.

[0048] Specifically, based on the difference between the pH value after the fluid mixing reaction at each time point and the target pH value, and the product of the rate of change of the pH value after the fluid mixing reaction and the system's dynamic inertial period, the forward-looking state deviation at each time point is calculated to predict the future state of the system. The specific calculation formula is as follows:

[0049] ;

[0050] In the formula: express Foresight bias at any given moment; This indicates the target setpoint for pH value; , They represent time, pH value after fluid mixing reaction at a given time; Indicates the time interval for data collection; express The rate of change of pH value after the fluid mixing reaction at a given time; express The system's dynamic inertial period at any given moment.

[0051] in, express The static deviation of pH value after fluid mixing reaction at a given time; express A trend prediction of pH value after fluid mixing reaction at a given time, which estimates the future pH value. The change in pH value over a given time period, solely due to the current trend; this invention addresses static deviations. Based on this, the trend prediction item was subtracted. ;when Approaching rapidly hour, For a large positive value, such that Forward-looking state bias at any moment The deviation is much smaller than the current static deviation, thus sending a signal to the controller to slow down in advance, thereby effectively suppressing overshoot.

[0052] S4. Based on the forward-looking state deviation, the dynamic control gain modulation factor is calculated through a nonlinear mapping function.

[0053] It should be noted that the adjustment force of the controller should match the estimated future deviation: when the forward state deviation is large, strong control should be used to respond quickly to the deviation; when the forward state deviation is small, or the system is rapidly approaching the target, gentle control should be used to ensure stability.

[0054] Specifically, based on the forward-looking state deviation at each time point, the dynamic control gain modulation factor at each time point is calculated using a nonlinear mapping function, wherein the nonlinear mapping function makes the dynamic control gain modulation factor positively correlated with the absolute value of the forward-looking state deviation.

[0055] The specific calculation formula is as follows:

[0056] ;

[0057] In the formula: express The dynamic control gain modulation factor at time t, whose value ranges between [0,1); express Foresight bias at any given moment; Indicates taking the absolute value; Represents the natural exponential function; This represents the gain sensitivity coefficient, used to adjust the dynamic control gain modulation factor. With forward-looking state bias The degree of change The value range is [1, 5], and the smaller the value... The value makes the gain change gradual, the system response is soft but slow, and a larger value... The value makes the gain response more sensitive, therefore, Setting it to 2.5 allows for a faster response time while maintaining stability.

[0058] Among them, by forward state deviation The input is fed into a nonlinear mapping function to obtain the dynamic control gain modulation factor. :when When the absolute value of the denominator approaches 0, it means that the system state, after prediction, is very close to the target setpoint. Approaching 2, making Approaching 0, the controller's output will be significantly reduced, thus playing a role in fine-tuning and stabilizing the system; when When the absolute value of is large, it indicates that the system state, after prediction, will still be far from the target setpoint. In this case, the denominator approaches 1, making As the value approaches 1, the controller will adjust more aggressively to respond quickly to deviations.

[0059] S5. Integrate the dynamic control gain modulation factor into the system's feedback control algorithm to obtain the control quantity of the digital dosing pump at the next moment, and drive the digital dosing pump to perform the dosing operation.

[0060] Specifically, the dynamic control gain modulation factor is incorporated into the system's feedback control algorithm to obtain the control quantity of the digital dosing pump at the next moment; then, the calculated control quantity of the digital dosing pump is transmitted through an edge computing device. The signal is converted into a hardware control signal, which drives the digital dosing pump to precisely execute the dosing operation, thereby completing a closed-loop control.

[0061] The formula for calculating the control quantity of the digital dosing pump at the next moment is as follows:

[0062] ;

[0063] In the formula: express The control quantity of the digital dosing pump at any given time; express The base feedforward dosage at each time step, and the base feedforward dosage , express Real-time flow rate at any given moment This indicates the target concentration that the drug needs to achieve in the pipeline. Indicates the concentration of the pharmaceutical stock solution itself; , These are the proportional gain coefficient and integral gain coefficient in the PID controller, respectively. express Dynamic control gain modulation factor at any given time; express The static deviation of pH value after fluid mixing reaction at time t, and , express pH value of the fluid after the reaction at a given time. This indicates the target set value for pH.

[0064] The final control quantity of the digital dosing pump consists of a basic feedforward component and a dynamically modulated feedback component. The feedback component includes the proportional term of the PID controller. and integral terms Furthermore, the overall effect strength of the feedback component was... Real-time adjustment: When the system tends to stabilize, When the setpoint is close to zero, the feedback regulation is suppressed, preventing unnecessary control actions and oscillations caused by minor disturbances near the setpoint. When the system deviation is large, When the value approaches 1, the feedback regulation effect is fully released, ensuring the system's rapid response capability.

[0065] In a PID controller, the proportional gain coefficient is... and integral gain coefficient The settings directly affect the system's response speed, stability, and steady-state accuracy. Among them, the proportional gain coefficient... The control quantity is output directly based on the current error magnitude, resulting in a fast response and... The larger the integral gain coefficient, the faster the response; however, an excessively large coefficient can easily cause overshoot, oscillation, or even instability. By accumulating historical errors, steady-state errors can be eliminated, and A larger proportional gain coefficient eliminates steady-state errors faster, but excessively large coefficients can easily cause overshoot, integral saturation, or system oscillations. Therefore, the proportional gain coefficient is determined using the critical proportional method (Ziegler-Nichols method). and integral gain coefficient The specific steps are as follows: Let the integral gain coefficient... =0, only the proportional gain coefficient is used. To control, gradually increase Continue until the system output exhibits constant-amplitude oscillations, then record the critical gain. and oscillation period Set according to the ZN formula , .

[0066] For example, when the system calibration parameters include: the target setpoint for pH value =8.5, Pipeline volume from the reagent dosing point to the downstream sensor installation location =9.8m 3 The target concentration of the agent to be achieved in the pipeline =0.05mol / m 3 Reference reaction time =2min and baseline flow rate =10m 3 / min, the intrinsic concentration of the reagent stock solution =1000 mol / m 3 And the proportional gain coefficient of the PID controller =0.08 and integral gain coefficient When the pH value is 0.02, the pH curve composed of pH values ​​at multiple moments throughout the entire regulation process is as follows: Figure 2 As shown, the obtained pH curve smoothly approaches the target pH value, and the oscillations near the target value are greatly reduced; the flow velocity curve composed of real-time flow velocities at multiple times is shown in the figure. Figure 3 As shown, the curves composed of the control quantities of the digital dosing pump at each moment are as follows: Figure 4 As shown, the dosage can be adjusted quickly and smoothly during flow rate changes, rather than reacting drastically after deviations occur.

[0067] This invention also discloses an intelligent digital dosing system based on edge computing, including a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement an intelligent digital dosing method based on edge computing according to the present invention.

[0068] 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.

Claims

1. An intelligent digital dosing method based on edge computing, characterized in that, The method comprises the following steps: acquiring real-time flow rate and pH value of fluid in a dosing pipeline; The system dynamic inertia period representing the comprehensive time lag of the dosing process is calculated by the pipeline volume, the real-time flow rate and the preset reference reaction time, including: ; wherein: represents the system dynamic inertia period at the time t; represents the pipeline volume between the dosing point and the installation position of the downstream sensor; represents the real-time flow rate at the time t; , respectively represent the reference reaction time and the reference flow rate; is the mixing efficiency influence coefficient; represents a natural exponential function; the method for obtaining the reference reaction time is as follows: under the reference flow rate, a dosing operation is performed according to the process requirements; the pH value change curve is continuously monitored and recorded by the downstream pH sensor; the time elapsed from the start of the dosing to the time when the downstream pH sensor reading reaches and maintains the target set value is measured as the reaction time obtained in one experiment; a plurality of reaction times are obtained by multiple implementations, and the mean value of the plurality of reaction times is taken as the reference reaction time;​​​ According to the static deviation of the pH value from a target set value of the pH value and a product of a variation rate of the pH value and a system dynamic inertia period, a prospective state deviation for predicting a future state of the system is calculated, including: ; wherein represents a prospective state deviation at a time point; represents a target set value of the pH value; , respectively represent a pH value at a time point, a pH value after a fluid mixing reaction at a time point, a pH value after a fluid mixing reaction at a time point; represents a static deviation of the pH value from a target set value of the pH value; represents a time interval of data collection; represents a variation rate of the pH value after the fluid mixing reaction at the time point; represents a variation rate of the pH value after the fluid mixing reaction at the time point.​ Based on the absolute value of the prospective state deviation, a dynamic control gain modulation factor is calculated by a nonlinear mapping function, the nonlinear mapping function makes the dynamic control gain modulation factor positively correlated with the absolute value of the prospective state deviation, comprising: ; wherein: represents the dynamic control gain modulation factor at the moment, the value range is between [0, 1); represents taking absolute value; represents a gain sensitivity coefficient, the value range is [1, 5]; The dynamic control gain modulation factor is combined into the feedback control algorithm of the system to obtain the control amount of the digital dosing pump at the next moment, including: ; wherein: represents the control amount of the digital dosing pump at the moment t; represents the basic feedforward dosing amount at the moment t, and , represents the target concentration of the medicament required to be reached in the pipeline, represents the self-concentration of the medicament reserve solution; , are respectively the proportional gain coefficient and the integral gain coefficient in the PID controller; represents the static deviation of the pH value after the fluid mixing reaction at the moment t, and ; and driving the digital dosing pump to perform the dosing operation.

2. The intelligent digital dosing method based on edge computing according to claim 1, characterized in that, the driving digital dosing pump to perform a dosing operation, comprising: The control quantity of the digital dosing pump calculated by the edge computing device is The signal is converted into a hardware control signal, which drives the digital dosing pump to precisely execute the dosing operation, thereby completing a closed-loop control. 3.The intelligent digital dosing method based on edge computing of claim 1, wherein, The preset reference reaction time, target set value of pH value, pipe volume , target concentration of the medicament required to be reached in the pipe , and reference flow rate , are all calibration parameters of the system; The target set value of pH value, the reference flow rate And the target concentration of the medicament required in the pipeline Through the man-machine interface of the system, set by the operator according to the process requirements; According to the pipe distance between the medicament dosing point and the downstream sensor installation location and the pipe diameter the pipe volume between the medicament dosing point and the downstream sensor installation location is calculated ; The reference reaction time is obtained by experimental calibration.

4. An edge computing based intelligent digital dosing system, characterized in that, The method comprises the following steps: a processor and a memory, wherein the memory stores computer program instructions which, when executed by the processor, implement the intelligent digital dosing method based on edge computing according to any one of claims 1-3.

Images