Flow control method

Abstract

The invention belongs to the field of fluid control, and particularly relates to a flow control method. Comprising the following steps that a constant-pressure loop is constructed, and a pressure gauge is installed in the constant-pressure loop and used for detecting the pressure in a pipeline in real time; collecting a pressure signal output by the pressure meter, and collecting an actual output flow value of the diaphragm pump; establishing a pressure-flow compensation model according to the pressure signal and the actual output flow value; inputting a pressure signal acquired in real time into the pressure-flow compensation model to obtain a compensated flow predicted value; and according to the difference between the compensated flow prediction value and the set flow value, the operation parameters of the diaphragm pump are adjusted, and closed-loop control over the output flow is achieved. According to the method, the flow is controlled by detecting the pressure in real time, and the problem that closed-loop control cannot be carried out due to the fact that the flow cannot be accurately detected due to pulsation output of the diaphragm pump is solved.

Description

Technical Field

[0001] This invention belongs to the field of fluid control, specifically a flow control method. Background Technology

[0002] In many stages of modern industrial production, such as chemical, pharmaceutical, and food processing, precise fluid flow control is a key factor in ensuring stable production processes and reliable product quality. Diaphragm pumps, with their excellent sealing performance, strong self-priming capability, and ability to adapt to various fluid transport needs, are widely used in these industries. However, during operation, the unique working principle of diaphragm pumps—relying on the reciprocating motion of a diaphragm to achieve fluid intake and discharge—results in an inherent discontinuity in their output flow, exhibiting a significant pulsating output characteristic. This pulsating output causes drastic pressure fluctuations within the pipeline, posing a significant challenge to accurate real-time flow measurement and stable control.

[0003] Currently, mainstream technologies for flow control of diaphragm pumps typically employ ultrasonic, positive displacement, and impeller flow measurement techniques. Ultrasonic flow meters suffer from significant time difference or frequency shift errors due to the instantaneous changes in fluid velocity and direction caused by pulsed flow in the diaphragm pump, which alter the ultrasonic wave propagation path and characteristics. Positive displacement flow meters exhibit inaccurate measurements because the pulsed output leads to irregular fluid entry into the metering chamber, resulting in incomplete or overfilling. Impeller flow meters suffer from large discrepancies between calculated and actual flow rates due to uneven impact of the pulsed fluid on the impeller, causing abnormal impeller speed. Because precise measurement of the diaphragm pump's output flow rate is impossible, closed-loop control based on flow feedback is difficult to implement effectively. Existing closed-loop flow control systems are not effectively applicable to diaphragm pumps, limiting their advantages in high-precision flow rate production and impacting product quality and production efficiency. Therefore, a flow control method is urgently needed to achieve stable flow control. Summary of the Invention

[0004] The existing diaphragm pumps cause unstable output flow pulsation during operation, which makes it impossible for various flow meters to accurately measure instantaneous flow, thus causing the closed-loop control system based on flow feedback to fail and making it impossible to achieve high-precision flow control.

[0005] This invention provides a flow control method. This method establishes a mathematical model to compensate for pressure and flow rate changes by real-time monitoring of pipeline pressure. Based on this model, a predicted flow rate is calculated from the real-time pressure value and compared with the target flow rate. This allows for dynamic adjustment of pump operating parameters, achieving precise closed-loop control of the flow rate and overcoming the measurement and control challenges caused by pulsation.

[0006] To achieve the above objectives, the present invention specifically employs the following technical means: A flow control method includes the following steps: S1: Construct a constant pressure loop and install a pressure gauge in the constant pressure loop to detect the pressure in the pipeline in real time; S2: Collect the pressure signal output by the pressure gauge and collect the actual output flow rate value of the diaphragm pump; S3: Establish a pressure-flow compensation model based on the pressure signal and the actual output flow rate; The specific steps for establishing the pressure-flow compensation model include: S31: Under different operating conditions of the diaphragm pump, multiple sets of pressure values ​​and corresponding actual flow values ​​are collected; the collected multiple sets of pressure values ​​and flow values ​​are divided into a training subset and a validation subset; the data of the training subset is used to participate in the fitting; the validation subset is used to evaluate the rationality of the pressure-flow relationship model. S32: Outlier identification and processing are performed on the collected data using statistical methods based on the mean and standard deviation; S33: The pressure and flow rates are fitted using a multinomial regression method to obtain a pressure-flow rate relationship model; The specific steps of S33 are as follows: A second-order polynomial is used to fit the data. The form of the second-order polynomial is:

[0007] in To predict the flow rate, P is the pressure, a0, a1, a2 are coefficients to be determined, and ε is the error term; The least squares method is used to determine the polynomial fitting coefficients, and the formula for calculating the sum of squared errors S is as follows:

[0008] Q i For the first i Group measurement flow rate values; For the first i Group predicted flow values; P i For the first i Group measurement pressure values; Taking the partial derivatives of S with respect to a0, a1, and a2, and setting the partial derivatives to zero, we obtain the following system of equations:

[0009] The values ​​of the coefficients a0, a1, and a2 are obtained and substituted into the second-order polynomial equation to obtain the pressure-flow compensation model: =25.143P 2 - 229.12P + 466.81; S34: Use the validation set to validate and optimize the model until the model error is within a preset range; S4: Input the real-time collected pressure signal into the pressure-flow compensation model to obtain the compensated flow prediction value; adjust the operating parameters of the diaphragm pump according to the difference between the compensated flow prediction value and the set flow value to achieve closed-loop control of the output flow.

[0010] Furthermore, the constant pressure circuit is constructed as follows: downstream of the outlet of the diaphragm pump, a check valve, a pressure stabilizing tank, a pressure gauge, and a back pressure valve are connected in series to form the main pressure regulating path; at the same time, a pressure relief valve is connected in parallel to the pipeline between the outlet of the check valve and the inlet of the back pressure valve to form a safety pressure relief path; the main pressure regulating path and the safety pressure relief path together form the constant pressure circuit.

[0011] Furthermore, in step S2, the analog current signal output by the pressure gauge is acquired in real time by the PLC module and converted into an actual pressure value; after the diaphragm pump is running stably, the volume of the outflowing fluid is measured within a preset time period using a graduated cylinder and a stopwatch, and the average instantaneous flow rate is calculated.

[0012] Furthermore, the specific steps of S34 are as follows: S341: Input the pressure values ​​in the validation subset into the fitted pressure-flow compensation model, and calculate the corresponding predicted flow values; S342: Calculate the error index between the predicted flow value and the actual flow value corresponding to the verification subset; S343: Determine whether the error index is less than or equal to a preset error threshold; If yes, the model validation is successful; if not, the model is refitted and validated by adjusting the order of the multinomial regression model until the error index meets the requirements.

[0013] Furthermore, the specific steps of S4 are as follows: using the PLC module to cyclically read the pressure gauge signal, inputting the pressure signal into the pressure-flow compensation model, outputting the control parameters of the diaphragm pump according to the flow set value and the pressure signal, and outputting a current signal to the driver of the diaphragm pump through the PLC module to realize closed-loop control of the output flow of the diaphragm pump.

[0014] Furthermore, the final output current of the PLC = M × I; M = Q1 / Q2; Where Q1 is the set flow rate, Q2 is the predicted flow rate, and M is the correction coefficient; by inputting the measured pressure P into the model, the predicted flow rate Q2 can be obtained from the output.

[0015] Correspondingly, the present invention also provides a dosing device and a flow control method for the device, including a frame, a control box disposed on the top of the frame, and a storage tank located at the bottom of the frame, wherein a plurality of dosing pipes are integrated within the frame; Each dosing pipeline includes a diaphragm pump, a check valve, a pressure stabilizing tank, a pressure relief valve, and a back pressure valve connected by a pipeline; after the diaphragm pump outlet, a check valve, a pressure relief valve, and a manual valve are connected in series in sequence; at the same time, the pressure relief valve is connected in parallel to the pressure stabilizing tank, and a pressure gauge and a back pressure valve are installed at the bottom. The control box is equipped with a PLC module, which is electrically connected to the pressure gauge and the diaphragm pump via a cable. The PLC module is used to adjust the operating frequency of the diaphragm pump based on the pressure feedback data, thereby controlling the dosing accuracy at the liquid outlet.

[0016] Furthermore, the inlet of the diaphragm pump is connected to the medicine storage tank via a pre-pump hose. The outlet of the back pressure valve is connected to an overpressure return hose, the end of which is connected to the medicine storage tank; when the pipeline pressure is too high, the medicine flows back to the medicine storage tank through the back pressure valve.

[0017] Furthermore, the control box is also equipped with an industrial Ethernet physical interface for connecting to the host AI system.

[0018] Furthermore, the one-way valve, pressure stabilizing tank, pressure relief valve, and back pressure valve are connected by modular movable joints and transparent PVC pipes.

[0019] Furthermore, a touch screen is provided on the front of the control box. The touch screen is communicatively connected to the PLC module and is used for parameter setting and status display.

[0020] Furthermore, the four dosing pipelines are distributed at a 90° angle with respect to the control box.

[0021] Furthermore, a counterweight and a filter are provided at the end of the pre-pump hose.

[0022] Compared with the prior art, the present invention has the following beneficial technical effects: Traditional methods suffer from large flow measurement errors and low control accuracy due to the discontinuous flow and unstable pressure of diaphragm pumps. This invention utilizes a pressure-flow compensation model to compensate for flow meter measurements in real time using pressure signals, while simultaneously eliminating pressure fluctuation interference through a pressure stabilization system, thus improving the accuracy of flow measurement. The closed-loop control implemented on this basis improves flow control accuracy to within ±2%, a significant improvement over traditional methods. In chemical production, this allows for precise control of the flow ratios of various raw materials, effectively improving product quality stability and consistency, and reducing product quality problems caused by flow errors.

[0023] The pressure stabilization system effectively solves the problem of unstable pipeline pressure caused by the pulse output of the diaphragm pump, providing a stable pressure environment for flow measurement and control. Based on a pressure closed-loop combined with PLC control strategy, the system can sense pressure changes in real time and quickly adjust the operating parameters of the diaphragm pump through a compensation model. When the output pressure of the diaphragm pump fluctuates, the flow rate is compensated and adjusted in a timely manner, effectively suppressing flow fluctuations and making the flow more stable.

[0024] After successfully solving the flow control problem of diaphragm pumps, diaphragm pumps can be widely used in fields with extremely high flow accuracy requirements. The method of this invention is also applicable to special working conditions such as high viscosity fluid transportation and high-precision control of small flow rates, further expanding the application scenarios of diaphragm pumps and providing more reliable technical support for the application of diaphragm pumps in different industrial fields.

[0025] Compared to traditional high-precision flow measurement equipment and complex flow control schemes, this invention only requires installing a pressure gauge at the diaphragm pump outlet, adding a pressure stabilization system, and using a PLC for data processing and control. It eliminates the need to purchase expensive dedicated flow measurement instruments, thus reducing equipment procurement costs. Simultaneously, the high-precision flow control and improved system stability help reduce production accidents and product defect rates caused by improper flow control, thereby lowering production costs and improving production efficiency and economic benefits. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of an apparatus using the control method of the present invention; Figure 2 This is a schematic diagram of the internal structure of a device using the control method of the present invention; Figure 3 This is a schematic diagram of a dosing pipeline using the control method of the present invention; Figure 4 This is a schematic diagram of the internal integration of a device using the control method of the present invention.

[0027] Figure 5 This invention presents a pressure-flow rate curve plotted based on test data.

[0028] The numbers in the diagram are: 1. Device door panel; 2. Control box; 3. Dosing pipeline; 4. Touch screen; 5. Frame; 6. Liquid outlet; 7. Manual valve; 8. Pressure stabilizing tank; 9. Pressure relief valve; 10. Flexible joint; 11. Transparent PVC pipe; 12. Pressure gauge; 13. Back pressure valve; 14. Connecting hose; 15. Check valve; 16. Diaphragm pump; 17. Overpressure return hose; 18. Pump inlet hose; 19. Drug storage tank. Detailed Implementation

[0029] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0030] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures, and should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions are omitted in the drawings.

[0031] Please see Figure 1 , Figure 2 and Figure 4 The four-way dosing pipeline 3 is arranged at a 90° angle with the control box 2 as the center, and is fixed on the square steel frame around the dosing device frame 5, with a maintenance passage of not less than 200mm reserved around the perimeter.

[0032] Please see Figure 2 Electrical components are integrated. The touchscreen 4 of the dosing device is connected to the PLC module (Siemens S7-1200 model) in control box 2 via an Ethernet cable for human-machine interaction, status display, and parameter setting. The pressure gauge 12 is connected to the PLC's analog input module via a 4-20mA analog signal line to receive real-time pressure data. The diaphragm pump 16 is connected to the PLC's pulse output module via a pulse signal line to receive frequency control signals. Control box 2 also integrates an industrial Ethernet module (supporting Profinet / Modbus TCP protocol), which connects to the workshop network via a physical network port to establish communication with the upper-level AI system, configuring IP addresses and protocol parameters to ensure stable data upload and command reception.

[0033] Please see Figures 1-3 First, assemble the hardware. Take the dosing device frame 5 and adjust the frame level using the bottom adjusting feet, requiring a horizontal deviation of ≤2mm. Install the dosing device control box 2 on the top of the frame and place the dosing device storage tank 19 in the bottom area.

[0034] Taking dosing pipeline 3 as an example, the pipeline system assembly is performed. Connect the dosing pipeline 3 to the pump inlet hose 18, which is equipped with a counterweight and filter at one end. Place one end of the hose into the storage tank 19, and connect the other end to the inlet of the diaphragm pump 16 via a flexible connector 10. After the outlet of the diaphragm pump 16, connect the check valve 15, pressure relief valve 9, and hand valve 7 in series. Simultaneously, connect the pressure stabilizing tank 8 in parallel, and install the pressure gauge 12 and back pressure valve 13. Connect and lock all components together via modular flexible connectors 10 and transparent PVC pipes 11 to ensure a leak-free seal. After completion, a pressure test is required: maintain pressure at 0.6 MPa for 30 minutes, with a pressure drop ≤ 0.02 MPa.

[0035] Please see Figure 3 The flow of the chemical solution in the device is as follows: the solution starts from the storage tank 19, is drawn into the diaphragm pump 16 via the pre-pump hose 18 and pressurized for output; then it flows through the check valve 15 to prevent backflow; subsequently, it enters the transparent PVC dosing pipeline 11 through the connecting hose 14; it flows through the manual valve 7 for manual flow adjustment or pipeline isolation; and finally, it is output from the outlet 6. The back pressure valve 13 is a mechanical valve used to maintain a constant system outlet pressure. When its outlet pressure is too high, excess solution can return to the storage tank 19 through the overpressure return hose 17 connected to it. The pressure relief valve 9 serves as a safety redundancy, opening to release pressure when the pressure is abnormally high; the pressure stabilizing tank 8 is used to buffer the pressure pulsations generated by the operation of the diaphragm pump 16, creating stable pressure conditions for the pipeline.

[0036] The flow control method includes the following steps: First, a constant pressure loop is constructed downstream of the outlet of the diaphragm pump 16. This loop is connected in series with the check valve 15, the pressure stabilizing tank 8, the pressure gauge 12 and the back pressure valve 13. At the same time, the pressure relief valve 9 is connected in parallel to the pipeline between the outlet of the check valve 15 and the inlet of the back pressure valve 13 to form a closed-loop environment with stable pressure.

[0037] The pressure signal in the pipeline is collected in real time by pressure gauge 12, and its output 4-20mA analog signal is connected to the analog input module of PLC module. The current signal is converted into the actual pressure value according to the preset range and conversion formula.

[0038] Select a graduated cylinder with an appropriate range. The graduated cylinder range should be selected based on the expected output flow range of the diaphragm pump 16. After the diaphragm pump 16 is running stably, open the valve at the measuring point to allow fluid to flow into the graduated cylinder. At the same time, start the stopwatch. When the timer reaches its limit, quickly close the valve and record the volume V of the fluid in the graduated cylinder at this time.

[0039] During the initial stage of system operation, data collection is conducted for a period of time, acquiring both pressure values ​​measured by the PLC and flow values ​​measured manually, ensuring that the collected data covers different operating conditions of the diaphragm pump 16. The collected pressure and flow data are analyzed to check for outliers. If outliers are found, they are identified and processed using standard deviation and mean-based methods, with options for either removing outliers or making corrections.

[0040] The pump was calibrated at a pressure of 1.7 bar, resulting in a flow rate of 150 ml / s. The pipeline pressure was then adjusted by 0.01 bar at a time, and all pressure values ​​from 1.7 bar to 1.6 bar and from 1.7 bar to 1.8 bar were recorded. The corresponding pressure value P and flow rate value Q were also recorded. The test data are as follows: Table 1 - Measurement values ​​of pressure P and flow rate Q

[0041] See Figure 5 Pressure-flow curves are plotted based on the measurement data.

[0042] Perform polynomial fitting on the curve data: A second-order polynomial is used to fit the curve data. The form of the second-order polynomial is:

[0043] in To predict the flow rate, P is the pressure, a0, a1, a2 are coefficients to be determined, and ε is the error term; The least squares method is used to determine the polynomial fitting coefficients, and the formula for calculating the sum of squared errors S is as follows:

[0044] Q i For the first i Group measurement flow rate values; For the first i Group predicted flow values; P i For the first i Group measurement pressure values; To find the a0, a1, a2 values ​​that minimize S, we take the partial derivatives of S with respect to a0, a1, a2 respectively, and set the partial derivatives to zero, resulting in the following system of equations:

[0045] Solving this system of equations yields the values ​​of coefficients a0, a1, and a2: a0 = 466.81, a1 = -229.12, and a2 = 25.143. Substituting these coefficients into the second-order polynomial equation, we obtain the pressure-flow compensation model: =25.143P2 -229.12P+466.81.

[0046] Test using unfitted stress, calculate the error between the model's predicted value and the actual value. If the error is large, it indicates that the model may not be accurate enough. It is necessary to consider increasing the order of the polynomial, refitting and validating until the model's error is within an acceptable range.

[0047] The pressure was adjusted to 1.55 bar and 1.85 bar, and the error between the calculated predicted flow rate and the measured actual flow rate was 1 ml, which is within the acceptable range.

[0048] Table 2 - Comparison of Predicted and Measured Flow Rates

[0049] During the control phase, the PLC module reads the real-time pressure value cyclically and inputs it into the established pressure-flow compensation model to calculate the predicted flow value under the current pressure.

[0050] P is the measured pressure, Q1 is the set flow rate, Q2 is the predicted flow rate, and M is the correction coefficient. By inputting the measured pressure P into the model, the model outputs the predicted flow rate Q2.

[0051] If M = Q1 / Q2, then the final output current of the PLC is M × the current value when the model is not applied. Through the above steps, based on the test data, a pressure-flow compensation model was established using polynomial fitting. This model can compensate for the output flow of the diaphragm pump 16 according to the real-time detected changes in pipeline pressure, thereby achieving precise flow control.

[0052] The pressure gauge 12 signal is read cyclically, the pressure signal is processed and input into the model, and the model outputs the pump control parameters according to the flow setpoint and pressure value.

[0053] The analog output of the PLC is connected to the driver of the diaphragm pump 16 using a shielded cable. A 4-20mA current signal is output to the driver of the diaphragm pump 16 through the analog output module, thereby realizing the control of the pump output flow.

[0054] Through the above steps, precise flow control of the diaphragm pump 16 based on pressure closure is achieved, ensuring that the diaphragm pump 16 can stably and accurately output the target flow rate even under complex operating conditions.

Claims

1. A flow control method, characterized in that, Includes the following steps: S1: Construct a constant pressure loop and install a pressure gauge (12) in the constant pressure loop to detect the pressure in the pipeline in real time; S2: Collect the pressure signal output by the pressure gauge and collect the actual output flow rate of the diaphragm pump (16); S3: Establish a pressure-flow compensation model based on the pressure signal and the actual output flow rate; The specific steps for establishing the pressure-flow compensation model include: S31: Under different operating conditions of the diaphragm pump, multiple sets of pressure values ​​and corresponding actual flow values ​​are collected; the collected multiple sets of pressure values ​​and flow values ​​are divided into a training subset and a validation subset; the data of the training subset is used to participate in the fitting; the validation subset is used to evaluate the rationality of the pressure-flow relationship model. S32: Outlier identification and processing are performed on the collected data using statistical methods based on the mean and standard deviation; S33: The pressure and flow rates are fitted using a multinomial regression method to obtain a pressure-flow relationship model; the specific steps are as follows: A second-order polynomial is used to fit the data. The form of the second-order polynomial is: ,in To predict the flow rate, P is the pressure, a0, a1, a2 are coefficients to be determined, and ε is the error term; The least squares method is used to determine the polynomial fitting coefficients, and the formula for calculating the sum of squared errors S is as follows: Q i For the first i Group measurement flow rate values; For the first i Group predicted flow values; P i For the first i Group measurement pressure values; Taking the partial derivatives of S with respect to a0, a1, and a2, and setting the partial derivatives to zero, we obtain the following system of equations: The values ​​of the coefficients a0, a1, and a2 are obtained and substituted into the second-order polynomial equation to obtain the pressure-flow compensation model: =25.143P 2 - 229.12P + 466.81; S34: Use the validation set to validate and optimize the model until the model error is within a preset range; S4: Input the pressure signal collected in real time into the pressure-flow compensation model to obtain the compensated flow prediction value; adjust the operating parameters of the diaphragm pump (16) according to the difference between the compensated flow prediction value and the set flow value to realize closed-loop control of the output flow.

2. The flow control method according to claim 1, characterized in that, The constant pressure circuit is constructed as follows: downstream of the outlet of the diaphragm pump, a check valve, a pressure stabilizing tank, a pressure gauge, and a back pressure valve are connected in series to form the main pressure regulation path; at the same time, a pressure relief valve is connected in parallel to the pipeline between the outlet of the check valve and the inlet of the back pressure valve to form a safety pressure relief path; the main pressure regulation path and the safety pressure relief path together form the constant pressure circuit.

3. The flow control method according to claim 1, characterized in that, In step S2, the analog current signal output by the pressure gauge (12) is collected in real time by the PLC module and converted into the actual pressure value. After the diaphragm pump is running stably, the volume of the outflowing fluid is measured within a preset time period using a measuring cylinder and a stopwatch, and the average instantaneous flow rate is calculated.

4. The flow control method according to claim 1, characterized in that, The specific steps of S34 are as follows: S341: Input the pressure values ​​in the validation subset into the fitted pressure-flow compensation model, and calculate the corresponding predicted flow values; S342: Calculate the error index between the predicted flow value and the actual flow value corresponding to the verification subset; S343: Determine whether the error index is less than or equal to a preset error threshold; If yes, the model validation is successful; if not, the model is refitted and validated by adjusting the order of the multinomial regression model until the error index meets the requirements.

5. The flow control method according to claim 1, characterized in that, The specific steps of S4 are as follows: use the PLC module to read the pressure gauge signal in a loop, input the pressure signal into the pressure-flow compensation model, output the control parameters of the diaphragm pump according to the flow set value and the pressure signal, and output the current signal to the driver of the diaphragm pump through the PLC module to realize closed-loop control of the output flow of the diaphragm pump.

6. The flow control method according to claim 5, characterized in that, The final output current of the PLC = M × I; M = Q1 / Q2; Where Q1 is the set flow rate, Q2 is the predicted flow rate, and M is the correction coefficient; by inputting the measured pressure P into the model, the predicted flow rate Q2 can be obtained from the output.

Images