Uniform flow pump flow control method and system and storage medium

Abstract

The invention discloses a uniform flow pump flow control method and system and a storage medium, and relates to the technical field of uniform flow pumps. The method comprises the steps that a curve mathematical model between the flow of a pump and the rotating speed and the pressure is established, a target flow value is set, an actual pressure value is collected, and the flow of the pump is calculated based on the target flow value, the actual pressure value and model parameters; the rotating speed of the pump is obtained according to the determined curve mathematical model, open-loop control is conducted on the pump according to the rotating speed, open-loop accurate control of the uniform flow pump under the condition that an external flow meter is not needed is achieved by establishing a high-precision cubic curve mathematical model of flow, rotating speed and pressure, and the flow control precision can reach + / -2%; the model dynamically compensates the influence of the rotating speed and the pressure, the flow can be rapidly stabilized under different working conditions, the stabilization time is shortened to be within 5 seconds, and the method is suitable for precise liquid conveying scenes with high requirements for the flow precision and the response speed such as biopharmacy.

Description

Technical Field

[0001] This invention relates to the field of uniform flow pump technology, and in particular to a method, system and storage medium for controlling the flow rate of a uniform flow pump. Background Technology

[0002] Flow-regulating pumps, such as peristaltic pumps and diaphragm pumps, are widely used in liquid transfer and process control in industries such as biopharmaceuticals, fine chemicals, and food due to their ability to provide relatively stable flow rates. However, in practical applications, existing technologies generally suffer from the following shortcomings:

[0003] First, the flow output of a uniform flow pump exhibits significant nonlinear characteristics. Ideally, the flow rate per revolution (single-revolution flow rate) should remain constant. However, actual tests show that as the rotational speed increases, the single-revolution flow rate gradually decreases due to factors such as pipeline rebound hysteresis. Simultaneously, the pressure in the process piping also significantly affects the output flow rate, and the rate of decrease in flow rate exhibits a complex coupling relationship with both rotational speed and pressure. Traditional methods of compensation through table lookup or linear approximation are insufficient to accurately reflect this nonlinear relationship, resulting in flow control accuracy often exceeding the ±5% process requirement in high-precision applications such as biopharmaceuticals.

[0004] Secondly, to ensure flow accuracy, existing technologies typically rely on online flow meters for real-time detection and feedback adjustment. While this approach improves accuracy to some extent, it also increases system hardware and maintenance costs. Furthermore, the flow meter's response delay, measurement errors, and long-term drift directly affect the dynamic performance and final accuracy of closed-loop control, and may even become a point of failure for the system.

[0005] Several attempts have emerged that do not directly rely on flow meters. For example, patent publication number CN101033748B discloses a method for determining pump flow rate without using traditional sensors. This method estimates the flow rate at the current operating point by creating a calibrated power curve under valve-closed conditions, calculating the pump's power ratio, and finally solving a polynomial power equation. Although this patent provides a solution that eliminates the need for flow meters, it primarily targets centrifugal pumps and centrifugal blowers. Its technical solution struggles to address the rebound hysteresis effect and pressure sensitivity issues of uniform flow pumps at high speeds, making it unsuitable for or unable to guarantee the accuracy of uniform flow pumps. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a high-precision uniform flow pump control method, system and storage medium that does not require a flow sensor.

[0007] The technical solution adopted by this invention to solve its technical problem is: a flow control method for a uniform flow pump, which establishes a mathematical model of the relationship between the pump's flow rate, rotational speed, and pressure, wherein the mathematical model is Q = cx 3 + x 2 (Fp + Gp) 2 +b)+ x(a + fp +gp 2 Where Q is the flow rate, p is the pressure, x is the rotational speed, and a, b, c, G, F, g, and f are model parameters.

[0008] During operation, at least seven different sets of speed, pressure, and flow rate values ​​are collected to determine the model parameters a, b, c, G, F, g, and f in the mathematical model of the curve.

[0009] Set a target flow rate value, collect the actual pressure value, and based on the target flow rate value, the actual pressure value, and model parameters, obtain the pump speed according to the determined curve mathematical model, and perform open-loop control of the pump according to the speed.

[0010] Furthermore, the model parameters a, b, c, G, F, g, and f are calculated using the formula: w = (X T *X) -1 *X T *y is determined, where w is the parameter set [a,b,c,G,F,g,f], X is a matrix containing speed and pressure data, and y is a vector of measured flow values.

[0011] Furthermore, the pump speed is obtained through the cubic equation solving function in the PLC, or through a multi-step approximation method, or through Newton's iteration method.

[0012] Furthermore, the establishment of the mathematical model for the curve includes the following steps:

[0013] The first step is to determine the relationship between the flow rate q per unit revolution of the pump and the rotational speed x. This relationship conforms to the quadratic equation: q = a + bx + cx², where a, b, and c are model parameters.

[0014] The second step is to derive the relationship between the pump's total flow rate Q, the flow rate q per revolution, and the rotational speed x, based on the relationship Q=q*x, and the model of the relationship between the pump's total flow rate Q and the rotational speed x under no-pressure conditions: Q=ax+bx²+cx³.

[0015] The third step involves introducing a relationship model between pressure p and rotational speed x, and then optimizing it to obtain the mathematical model of the curve.

[0016] Furthermore, in the third step, after introducing pressure p, the total flow rate Q of the pump and the rotational speed x satisfy the relationship model described in the second step, and are affected by the square of pressure p.

[0017] Furthermore, in the third step, a relationship model between pressure p and rotational speed x is introduced, and the optimized mathematical model of the curve is obtained through mathematical modeling.

[0018] Furthermore, in the first step, the relationship between the flow rate q and the rotational speed x of one pump revolution is determined, and this relationship conforms to the quadratic equation: q = a + bx + cx².

[0019] A flow control system for a uniform flow pump, used to implement the aforementioned flow control method for a uniform flow pump, comprising:

[0020] The control unit is configured to store and run the mathematical model of the curve.

[0021] The data acquisition unit, connected to the control unit, is configured to acquire the actual pressure and rotational speed values ​​of the pump.

[0022] The parameter determination unit, connected to the control unit and the data acquisition unit, is configured to determine the model parameters a, b, c, G, F, g, and f in the curve mathematical model based on at least seven different sets of acquired rotational speed, pressure, and flow rate values.

[0023] The control calculation unit, integrated in the control unit, is configured to receive the target flow rate value in the operating mode, and calculate and output the corresponding target speed control signal in real time according to the curve mathematical model based on the actual pressure value fed back by the data acquisition unit and the determined model parameters.

[0024] The drive unit, connected to the control unit, is configured to drive the pump to operate according to the target speed control signal.

[0025] A readable storage medium having a program stored thereon for causing a computer to execute the described flow rate regulation method for a uniform flow pump.

[0026] The beneficial effects of this invention are as follows: By establishing a high-precision cubic curve mathematical model of "flow rate-rotation speed-pressure", this invention achieves open-loop precise control of the uniform flow pump without the need for an external flow meter, with a flow control accuracy of ±2%. This method can complete the model parameter identification with a small number of calibration points, has strong versatility, and significantly reduces the system cost, maintenance cost, and response delay caused by reliance on flow meters. At the same time, the model dynamically compensates for the influence of rotation speed and pressure, and can quickly stabilize the flow rate under different operating conditions, with the stabilization time shortened to within 5 seconds. It is suitable for precision liquid transportation scenarios such as biopharmaceuticals where high flow accuracy and response speed are required. Attached Figure Description

[0027] Figure 1 —Local trend diagrams of flow rate and rotational speed under different pressures according to this invention;

[0028] Figure 2—Complete trend diagram of flow rate and rotational speed under different pressures according to this invention;

[0029] Figure 3 —A trend graph of flow rate versus rotational speed for a single revolution of the pump of this invention;

[0030] Figure 4 —Trend graph of pressure and flow rate under fixed rotation speed according to the present invention;

[0031] Figure 5 —Flow accuracy deviation curve of the present invention; Detailed Implementation

[0032] Reference Figures 1-5 This embodiment provides a flow control method for a peristaltic pump, using the peristaltic pump as the controlled object. In other embodiments, this method can be applied to diaphragm pumps, providing a high-precision flow control method without a flow meter. This invention achieves a flow accuracy of ±2% under open-loop control by establishing a nonlinear mathematical model between flow rate, rotational speed, and pressure. Figure 5 As shown in the figure, it significantly reduced system costs and improved response speed.

[0033] First, establish a mathematical model for the flow rate-speed-pressure curve of the uniform flow pump. The mathematical model for the curve is Q = cx. 3 + x 2 (Fp + Gp) 2 +b)+ x(a + fp +gp 2 ); where Q is the flow rate, p is the pressure, x is the rotational speed, and a, b, c, G, F, g, f are model parameters.

[0034] During operation, at least seven different sets of speed, pressure, and flow rate values ​​are collected to determine the model parameters a, b, c, G, F, g, and f in the mathematical model of the curve.

[0035] Set a target flow rate value, collect the actual pressure value, and based on the target flow rate value, the actual pressure value, and model parameters, obtain the pump speed according to the determined curve mathematical model, and perform open-loop control of the pump according to the speed.

[0036] The establishment of the mathematical model for the curve includes the following steps:

[0037] The first step is to determine the relationship between the flow rate q per unit revolution of the pump and the rotational speed x. This relationship conforms to the quadratic equation: q = a + bx + cx², where a, b, and c are model parameters. Figure 3 Statistical analysis of multiple experimental test data, with the horizontal axis representing rotational speed and the vertical axis representing the flow rate per pump revolution, revealed that the flow rate per revolution decreases as the speed increases. This is primarily because the pipe's rebound speed cannot keep up with the extrusion speed at higher speeds. Furthermore, the relationship between the flow rate per pump revolution and rotational speed almost perfectly fits a quadratic curve. Figure 3 It forms the basis of the final curve mathematical model;

[0038] The second step is to derive the relationship between the pump's total flow rate Q, the flow rate per unit revolution q, and the rotational speed x, based on the relationship Q=q*x, and the model of the relationship between the pump's total flow rate Q and the rotational speed x under no-pressure conditions: Q=ax+bx²+cx³.

[0039] The third step involves introducing a model relating pressure p to rotational speed x. Tests revealed that as pressure increases, the set flow rate decreases. The table below shows the experimental data for pressure and flow rate of pump 5 running at 500 r / min:

[0040]

[0041] Draw the above table as follows Figure 4 As shown in the figure, under a pipeline pressure of 0-3 bar, the pump operates at 500 r / min, and the flow rate decreases linearly. This illustrates that the relationship between the pump's flow rate and rotational speed still satisfies: Q = ax + bx² + cx³, but it is affected by the square of the pressure. Finally, the optimized mathematical model of the curve is obtained through mathematical modeling.

[0042] The mathematical model of the curve contains seven unknown model parameters: a, b, c, G, F, g, and f. At least seven sets of sampling data are required, including pressure, rotation speed, and flow rate. The rotation speed and pressure are preferably within the range of daily use, such as pressure between 0-3 bar and rotation speed between 0-1000 r / min.

[0043] In this embodiment, the sampling scheme is a statistical analysis of the flow rate over 2-3 minutes. The sampling points are shown in the table below, with pressure P and rotational speed X as control variables and flow rate Q as the measured variable. Variables with the same name are identical. If either pressure P or rotational speed X is different, it is considered a new set of data.

[0044]

[0045] The constraints are as follows:

[0046] 1) p0_ series, which is the system's built-in pressure at the set speed x.

[0047] 2) p2 is the maximum range

[0048] 3) p1_series = (p2 + p0_series) / 2

[0049] 4) x1 = x2-k, x3 = x2+k. [x1,x2,x3] is an arithmetic sequence, and k is the difference.

[0050] Note the following when measuring Q:

[0051] 1) A mass flow meter or weighing device should be used.

[0052] 2) After the flow rate stabilizes, collect the liquid mass N for at least 2-3 minutes (T).

[0053] 3) Q = N / T.

[0054] 4) T is related to the rotational speed, and the time must satisfy the condition that the stopping time is an integer number of revolutions.

[0055] The parameters are solved using the least squares method: w = (X T *X) -1 *X T *y, where T is the transpose, -1 is the inverse, * is the dot product, y is the flow set [Q1 Q2 Q3 Q4 Q5 Q6 Q7], and w is the parameter set [abc FG fg], to determine the model parameters a, b, c, G, F, g, f in the mathematical model of the curve.

[0056] Control scenario: a, b, c, G, F, g, f, which are already determined, can be substituted into the open-loop control equation:

[0057] Q 理论 =cx 3 + x 2 (Fp + Gp) 2 +b)+ x(a + fp +gp 2 ).

[0058] Set a target flow rate value, collect actual pressure values ​​(with the pressure gauge as close to the pump as possible), obtain the pump speed based on the determined curve mathematical model, and perform open-loop control of the pump based on the speed. The table below shows the experimental data regarding the flow characteristics:

[0059] Draw the table above as follows Figure 2 , Figure 1 yes Figure 2 Based on the partial diagram, the conclusion is that under the same pressure but different speeds, the pump flow rate is very linear, and under the same speed but different pressures, the pump's attenuation is very small. Figure 1 and Figure 2 This demonstrates that a deterministic, modelable nonlinear relationship exists between flow rate (Q), rotational speed (x), and pressure (p).

[0060] Different pump models require parameter determination during initial installation to determine the model parameters a, b, c, G, F, g, and f in the curve mathematical model. Each time a pump part is replaced, it may affect the model parameters, requiring parameter determination again. For example, replacing the hose of a peristaltic pump or the diaphragm of a diaphragm pump.

[0061] In other embodiments, the parameters can be determined by the following method:

[0062] In the first step, we determine the relationship between the flow rate q per unit revolution of the pump and the rotational speed x. This relationship conforms to the quadratic equation: q = a + bx + cx², where a, b, and c are all unknowns. Therefore, we... Figure 3 By intercepting three points (x, y) on the curve, the values ​​of constants a, b, and c can be calculated. Multiple experiments have shown that pump speeds of 200 r / min, 500 r / min, and 800 r / min, with 100 revolutions per pump, are optimal.

[0063] Substitute the corresponding three points: (S1,V1), (S2,V2), (S3,V3), where S is the rotational speed, V is the volume, and n is the number of revolutions.

[0064] q1= =a+S1*b+S1²*C

[0065] q2= =a+S2*b+S2²*C

[0066] q3= =a+S3*b+S3²*C

[0067] Substituting the three points, we get:

[0068] c=

[0069] b=

[0070] a= -(S1*b+S1²*C)

[0071] In the third step, extensive testing revealed that increased pressure leads to decreased traffic, as shown in Figures 1 and 2. Figure 2 As shown. Introduce a function K that varies with pressure and rotational speed, where K is the percentage decrease in flow rate, K = (K≤1), through calculation, the general formula for K is:

[0072] K = (GP² + FP + 1) + (gp² + fp) * x (K ≤ 1, and p = 0, then k = 1)

[0073] G, F, g, and f are all unknowns, p is the pressure, and x is the rotational speed.

[0074] To find the four unknowns mentioned above, we can take six points each and solve for the unknowns.

[0075] The six points are:

[0076] V1: The common point for solving the unknowns of the q equation using the original three-point calibration. The volume corresponding to a pressure of 0 bar, a rotational speed of S1, and a revolution count of n1.

[0077] V2: The volume corresponding to pressure P1, rotation speed S1, and number of revolutions n1.

[0078] V3: The volume corresponding to pressure P2, rotation speed S1, and number of revolutions n1.

[0079] V4: The common point for the unknowns in the q equation, calibrated at the original three points. The volume corresponding to a pressure of 0 bar, a rotational speed of S3, and a revolution count of n3.

[0080] V5: The volume corresponding to pressure P1, rotation speed S3, and number of revolutions n3.

[0081] V6: The volume corresponding to pressure P2, rotation speed S3, and number of revolutions n3.

[0082] The pressure values ​​are P1=0.5 and P2=1.

[0083] G

[0084] f=

[0085] F= -1

[0086] G=

[0087] Q compensation = Q / K = ax + bx² + Cx³

[0088] Input Q to solve for X.

[0089] If the PLC has a cubic equation solving function, x can be solved directly. If it does not have a cubic equation solving function, a multi-step approximation method is used.

[0090] Let b=0, c=0.

[0091] Then Q compensation = ax

[0092] untie: =x1

[0093] Substitute the solution of x1 into the equation. =x2,

[0094] Solving for x, we get x2.

[0095] Substitute the solution of x2 into the equation. =x3,

[0096] Solving for x, we get x3.

[0097] Following the above method, the x solution obtained from x6 and subsequent steps is very close to the true solution of x, and can be used as the x solution in actual engineering.

[0098] In other embodiments, x can also be calculated online using Newton's iteration method, which has been verified to require only two iterations:

[0099] 1) The governing formula can be viewed as a Taylor expansion, so we take its first-order term and calculate x0 as the starting point for approximation.

[0100] x0=

[0101] 2) Newton's iteration term:

[0102] f(x) / f(x)'=

[0103] 3) x1 = x0 - f(x0) / f(x0)'

[0104] 4) x2 = x1 - f(x1) / f(x1)'

[0105] x2 is the final output control value.

[0106] This invention establishes a mathematical model of "rotation speed - flow rate - pressure," which can maintain a flow accuracy of ±2% without flow meter feedback adjustment. The table below shows the flow accuracy test results:

[0107]

[0108] Plot the above table as a line graph to generate... Figure 5 As shown in the figure, quantitative delivery of 50ml-3000ml and 0-3bar can be achieved with an accuracy within ±2% through one parameter calibration.

[0109] This invention achieves precise control with only a small amount of calibration data and calculations, completely eliminating the need for a flow meter, saving on procurement and maintenance costs. After eliminating the flow meter, the system hardware structure is simplified, and the overall cost can be reduced by 15-30% (depending on the pump specifications and flow meter configuration); the system reliability is improved, avoiding downtime and maintenance caused by flow meter failure.

[0110] This invention introduces pressure compensation, enabling automatic compensation under different pressure and speed conditions. After the user sets the target flow rate, the motor speed can be calculated in real time, and the target flow rate can be stably maintained even if the pressure fluctuates.

[0111] Compared to closed-loop control relying on flow meters (which suffers from measurement delay and signal noise), this invention employs direct open-loop control using a mathematical model, resulting in a faster response time. Compared to adjusting pump speed through flow meter feedback, the time to achieve flow stabilization is reduced from 30-40 seconds to within 5 seconds.

[0112] It is suitable for scenarios with high response time requirements, such as precision infusion, reagent filling and online solution preparation systems.

[0113] A flow control system for a uniform flow pump is provided to implement the aforementioned flow control method for a uniform flow pump. The system specifically includes the following units:

[0114] The control unit is configured to store and run the curve mathematical model, which is Q = cx 3 + x 2 (Fp + Gp) 2 +b)+ x(a + fp +gp 2 Where Q is the flow rate, p is the pressure, x is the rotational speed, and a, b, c, G, F, g, and f are model parameters.

[0115] The data acquisition unit, connected to the control unit, is configured to acquire the actual pressure and speed values ​​of the pump. It includes a pressure acquisition module and a speed feedback module. The pressure acquisition module is a pressure sensor installed on the pump outlet pipeline. The speed feedback module is not integrated into the pump's drive motor, such as an optical encoder or Hall sensor, and is used to monitor the actual speed x of the motor in real time and feed it back to the control unit for system status monitoring and model verification.

[0116] The parameter determination unit, connected to the control unit and the data acquisition unit, is configured to determine the model parameters a, b, c, G, F, g, and f in the curve mathematical model based on at least seven different sets of acquired rotational speed, pressure, and flow rate values.

[0117] The control calculation unit, integrated in the control unit, is configured to receive the target flow rate value in the operating mode, and calculate and output the corresponding target speed control signal in real time according to the curve mathematical model based on the actual pressure value fed back by the data acquisition unit and the determined model parameters.

[0118] The drive unit, connected to the control unit, is configured to drive the pump to operate according to the target speed control signal.

[0119] The above five units work together to form a complete closed-loop system (information flow closed loop, control open loop) that can achieve high-precision flow control without a flow meter.

[0120] A readable storage medium, such as a read-only memory (ROM), random access memory (RAM), flash memory, hard disk (HDD), solid-state drive (SSD), or any other form of non-transitory computer-readable storage medium including cloud storage, on which a computer program (i.e. a series of instructions) is stored.

[0121] When the program is loaded and executed by one or more processors (such as a computer, PLC, or CPU of an embedded system), the processor performs the flow control method for the uniform flow pump.

[0122] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for controlling the flow rate of a uniform flow pump, characterized in that: A mathematical model is established to represent the relationship between the pump's flow rate, rotational speed, and pressure. This mathematical model is Q = cx. 3 + x 2 (Fp + Gp) 2 +b)+ x(a + fp +gp 2 Where Q is the flow rate, p is the pressure, x is the rotational speed, and a, b, c, G, F, g, and f are model parameters. During operation, at least seven different sets of speed, pressure, and flow rate values ​​are collected to determine the model parameters a, b, c, G, F, g, and f in the mathematical model of the curve. Set a target flow rate value, collect the actual pressure value, and based on the target flow rate value, the actual pressure value, and model parameters, obtain the pump speed according to the determined curve mathematical model, and perform open-loop control of the pump according to the speed.

2. The flow control method for a uniform flow pump according to claim 1, characterized in that: The model parameters a, b, c, G, F, g, and f are calculated using the formula: w = (X T *X) -1 *X T *y is determined, where w is the parameter set [a,b,c,G,F,g,f], X is a matrix containing speed and pressure data, and y is a vector of measured flow values.

3. The flow control method for a uniform flow pump according to claim 1, characterized in that: The pump speed is obtained through the cubic equation solving function in the PLC, or through a multi-step approximation method, or through Newton's iteration method.

4. The flow control method for a uniform flow pump according to claim 1, characterized in that: The establishment of the mathematical model for the curve includes the following steps: The first step is to determine the relationship between the flow rate q per unit revolution of the pump and the rotational speed x. This relationship conforms to the quadratic equation: q = a + bx + cx², where a, b, and c are model parameters. The second step is to derive the relationship between the pump's total flow rate Q, the flow rate per unit revolution q, and the rotational speed x, based on the relationship Q=q*x, and the model of the relationship between the pump's total flow rate Q and the rotational speed x under no-pressure conditions: Q=ax+bx²+cx³. The third step involves introducing a relationship model between pressure p and rotational speed x, and then optimizing it to obtain the mathematical model of the curve.

5. The flow control method for a uniform flow pump according to claim 4, characterized in that: In the third step, after introducing pressure p, the total flow rate Q of the pump and the rotational speed x satisfy the relationship model described in the second step, and are affected by the square of pressure p.

6. The flow control method for a uniform flow pump according to claim 4, characterized in that: In the third step, a relationship model between pressure p and rotational speed x is introduced, and the optimized mathematical model of the curve is obtained through mathematical modeling.

7. The flow control method for a uniform flow pump according to claim 4, characterized in that: In the first step, the relationship between the flow rate q and the rotational speed x of one pump revolution is determined. This relationship conforms to the quadratic equation: q = a + bx + cx².

8. A flow control system for a uniform flow pump, used to implement the flow control method for a uniform flow pump according to any one of claims 1-6, characterized in that, include: The control unit is configured to store and run the mathematical model of the curve. The data acquisition unit, connected to the control unit, is configured to acquire the actual pressure and rotational speed values ​​of the pump. The parameter determination unit, connected to the control unit and the data acquisition unit, is configured to determine the model parameters a, b, c, G, F, g, and f in the curve mathematical model based on at least seven different sets of acquired rotational speed, pressure, and flow rate values. The control calculation unit, integrated in the control unit, is configured to receive the target flow rate value in the operating mode, and calculate and output the corresponding target speed control signal in real time according to the curve mathematical model based on the actual pressure value fed back by the data acquisition unit and the determined model parameters. The drive unit, connected to the control unit, is configured to drive the pump to operate according to the target speed control signal.

9. A readable storage medium having a program stored thereon, characterized in that: The program is used to enable a computer to execute the flow rate regulation method of the uniform flow pump as described in any one of claims 1-6.

Images