A Stable Liquid-Adding Integrated System and Method for a Test Solution

Abstract

The present invention relates to the technical field of stable liquid addition, and specifically discloses an integrated system and method for stable liquid addition of test solutions, including: obtaining basic data; real-time identifying the working conditions of sudden changes in solution temperature and triggering an adaptive control process; through four-dimensional state space modeling and optimal estimation, fusing basic data, suppressing multi-physical field coupling errors, and providing high-precision state estimation; calculating the viscosity compensation coefficients of each fluid type, dynamically adjusting the liquid addition process parameters, and compensating for the fluctuations in flow resistance caused by viscosity changes, so that the experimental solution can be stably added; through a four-dimensional dynamic control model, a fluid type differential compensation strategy, and an adaptive PID parameter adjustment technology, combined with a feedforward-feedback collaborative architecture and a high-precision data acquisition system, the present invention realizes the precise and stable control of the liquid addition process of test solutions under multi-physical field coupling conditions, breaking through the adaptability bottleneck of traditional liquid addition systems to sudden changes in solution temperature, fluid property differences, and complex disturbances.

Description

Technical Field

[0001] The present invention relates to the technical field of stable liquid addition, and particularly to a stable liquid addition integration system and method for test solutions. Background Art

[0002] The stable liquid addition of test solutions means that in laboratories or industrial production, through technical means, the solution is ensured to be added to the target container or system with precise and constant flow rate or volume, avoiding liquid addition errors caused by flow rate fluctuations, pressure changes or external disturbances. It is widely used in fields such as chemical analysis, biopharmaceuticals, and material synthesis. Its core goal is to improve the reliability of experimental results and the stability of production processes.

[0003] However, although existing liquid addition equipment uses multiple sensors (such as a combination of flow + pressure), there is a lack of a dynamic fusion model in the existing technology, and it mostly stays at the data acquisition level, resulting in the inability to effectively correlate viscosity compensation when the solution temperature changes; the traditional two-dimensional control model does not construct a four-dimensional state space (lacking the coupling relationship between the flow rate change rate and the solution temperature), and does not incorporate the Bernoulli equation into the state transition matrix, making it difficult to predict the impact of viscosity changes on the dynamic characteristics of the flow rate; the existing technology uses simple mean filtering, which cannot effectively suppress high-frequency noise (such as pump head vibration), resulting in insufficient data credibility; the existing compensation strategies do not distinguish fluid types and force the Newtonian fluid exponential model to be applied to non-Newtonian fluids, resulting in deviations in apparent viscosity calculation (for example, the yield stress of tomato sauce is not included in the model, and the error after compensation is still up to ±2.0%); the above technical defects make it difficult for traditional liquid addition systems to achieve long-term stable high-precision liquid addition control under sudden changes in solution temperature, fluid property differences, and complex interference conditions. Summary of the Invention

[0004] The purpose of the present invention is to provide a stable liquid addition integration system and method for test solutions to solve the above technical problems in the background.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] In the first aspect, the present invention provides a stable liquid addition integration method for test solutions, including:

[0007] Step 1: Multidimensional real-time monitoring to obtain basic data;

[0008] Step 2: Based on the basic data, identify the sudden change condition of the solution temperature in real time and trigger the adaptive control process;

[0009] Step 3: When the adaptive control process is triggered, through four-dimensional state space modeling and optimal estimation, fuse the basic data, suppress the multi-physical field coupling error, and provide high-precision state estimation;

[0010] Step 4: Based on the state estimation results, calculate the viscosity compensation coefficients for each fluid type, so as to dynamically adjust the liquid addition process parameters, compensate for the flow resistance fluctuations caused by viscosity changes, and enable the stable addition of the experimental solution.

[0011] As a further solution of the present invention: The basic data includes: real-time actual flow rate, pipeline back pressure, and solution temperature.

[0012] As a further solution of the present invention: The process of obtaining the real-time actual flow rate is as follows:

[0013] High-frequency collect the original signal through a flow sensor, generate a linear compensation coefficient after zero-offset deduction and three-point calibration, and then perform moving average filtering on the recent multiple sampling points, and apply the linear compensation coefficient to correct the non-linear error to obtain the real-time actual flow rate.

[0014] As a further solution of the present invention: The process of obtaining the pipeline back pressure is as follows:

[0015] Sample through a piezoresistive pressure sensor, perform moving average filtering on the recent multiple sampling points after sampling, and calculate the pipeline back pressure through quadratic polynomial solution temperature compensation and zero-full scale calibration.

[0016] As a further solution of the present invention: The process of obtaining the solution temperature is as follows:

[0017] Collect the solution temperature of the flowing solution body in real time through a solution temperature sensor to obtain the solution temperature.

[0018] As a further solution of the present invention: The process of real-time identifying the solution temperature mutation condition based on the basic data and triggering the adaptive control process is as follows:

[0019] Calculate the average solution temperature change rate of consecutive sampling periods through moving average filtering;

[0020] If the average solution temperature change rate is greater than the preset solution temperature change rate, generate a parameter regulation signal.

[0021] As a further solution of the present invention: The process of suppressing the multi-physical field coupling error and providing high-precision state estimation by fusing multi-sensor data through four-dimensional state space modeling and optimal estimation is as follows:

[0022] When receiving the parameter regulation signal, define a four-dimensional state vector, where the four-dimensional state vector includes the real-time actual flow rate, flow rate change rate, pipeline back pressure, and solution temperature;

[0023] Then, based on the state of the previous moment, predict the state of the current moment to obtain the prior state estimation vector, and at the same time calculate the prior covariance matrix;

[0024] Compose the collected basic data into a measurement vector;

[0025] Through Kalman gain calculation, the weights of the measurement vector and the prior state estimation vector are dynamically adjusted, and the confidence is flexibly allocated according to different situations;

[0026] Then, the prior covariance matrix is corrected by the Kalman gain to obtain the posterior state estimation vector, thereby suppressing the multi-physical-field coupling error and providing a high-precision state estimation.

[0027] As a further solution of the present invention: the process of obtaining the viscosity compensation coefficient is as follows:

[0028] Identify the fluid type;

[0029] If the fluid is a Newtonian fluid, based on the exponential effect of the solution temperature on the viscosity, the ratio of the viscosity at the current solution temperature to the viscosity at the reference solution temperature is calculated in real time. When the solution temperature rises, the viscosity decreases exponentially according to the preset solution temperature coefficient, and the viscosity compensation coefficient is generated;

[0030] If the fluid is a non-Newtonian fluid, the real-time shear rate and the initial calibration parameters are extracted, the apparent viscosity is calculated, and the apparent viscosity is compared with the reference viscosity to generate the viscosity compensation coefficient.

[0031] As a further solution of the present invention: the process of dynamically adjusting the liquid addition process parameters is as follows:

[0032] Based on the viscosity compensation coefficient, the PID parameters are adjusted;

[0033] The PID parameters include: proportional coefficient, integral coefficient, and differential coefficient;

[0034] The adjusted proportional coefficient is calculated by multiplying the pre-adjustment proportional coefficient by the viscosity compensation coefficient, and the adjusted integral coefficient is calculated by dividing the pre-adjustment integral coefficient by the viscosity compensation coefficient. The adjusted differential coefficient is calculated by multiplying the pre-adjustment differential coefficient by the square root of the viscosity compensation coefficient.

[0035] In a second aspect, the present invention provides a stable liquid addition integration system for a test solution, and the system includes:

[0036] Real-time monitoring module: multi-dimensional real-time monitoring to obtain basic data;

[0037] Trigger condition detection module: based on the basic data, the sudden change condition of the solution temperature is identified in real time to trigger the adaptive control process;

[0038] Data fusion module: after the adaptive control process is triggered, the basic data is fused through four-dimensional state space modeling and optimal estimation to suppress the multi-physical-field coupling error and provide a high-precision state estimation;

[0039] Parameter regulation module: Based on the state estimation results, calculate the viscosity compensation coefficients for each fluid type, and then dynamically adjust the liquid addition process parameters to compensate for the fluctuations in flow resistance caused by viscosity changes, so that the experimental solution can be added stably.

[0040] Advantages of the present invention:

[0041] Through a four-dimensional dynamic modeling system, the present invention breaks through the limitations of traditional two-dimensional control models, constructs a four-dimensional dynamic model including flow rate, flow rate change rate, pressure, and solution temperature, realizes the accurate capture of the coupling relationship of multiple physical fields, integrates the Bernoulli equation and the viscosity-solution temperature exponential model into the state transition matrix, and can predict the dynamic impact of solution temperature changes on flow characteristics in real time;

[0042] Through an intelligent compensation mechanism for fluid constitutive characteristics, establish a solution temperature-viscosity exponential model for Newtonian fluids, and develop a shear rate-yield stress calculation model for non-Newtonian fluids, breaking through the limitations of traditional single compensation algorithms;

[0043] Through the PID parameter dynamic adjustment method of viscosity compensation coefficients, and through the non-linear adjustment strategy of the proportional term being positively correlated with the viscosity coefficient, the integral term being negatively correlated, and the differential term being square root correlated, realize the real-time matching of control parameters and fluid characteristics;

[0044] Through the multi-sensor joint calibration technology, through the three-point calibration of the flow sensor (non-linear error correction), the quadratic polynomial solution temperature compensation of the pressure sensor (eliminating the influence of the ambient solution temperature), and the dynamic response optimization of the solution temperature sensor, construct a basic data acquisition system with an error magnitude lower than 50% of the industry standard, providing reliable input for intelligent control algorithms. Brief description of the drawings

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 is the flowchart of a method for stable liquid addition integration of a test solution in Embodiment 1 of the present invention;

[0047] Figure 2 is the system block diagram of a system for stable liquid addition integration of a test solution in Embodiment 2 of the present invention. Detailed implementation manners

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention. Embodiment

[0049] Please refer to Figure 1 As shown, a stable liquid addition integration method for a test solution according to an embodiment of the present invention includes the following steps:

[0050] Step 1: Multi-dimensional real-time monitoring to obtain basic data;

[0051] In some embodiments, after the power-on of a stable liquid addition integration system for a test solution, it continuously runs, presets a sampling period, and obtains basic data for each sampling period;

[0052] Among them, the sampling period can be: 10ms, 20ms, 50ms;

[0053] The basic data includes: real-time actual flow rate, pipeline back pressure, solution temperature;

[0054] Exemplarily, the process of obtaining the real-time actual flow rate is as follows:

[0055] One flow sensor is connected in series to each liquid addition channel (such as the main pump channel 1, the slave pump channels 2-4), and is installed between the outlet of the peristaltic pump and the inlet of the microfluidic channel to ensure the measurement of the real-time flow rate flowing through this channel;

[0056] Among them, the type of the flow sensor is selected according to the measurement range: for small flow rates (0.1-10 mL / min), electromagnetic type (accuracy ±0.1% FS, such as E+H Promag L) is used, and for large flow rates (10-1000 mL / min), turbine type (accuracy ±0.5% FS, such as Omega FTB-100) is used;

[0057] When the flow sensor is first used or regularly maintained, the flow sensor is initialized and calibrated. Specifically:

[0058] The pump group is turned off, the flow sensor runs idly (no liquid flow), and 10 cycle data are collected and averaged as the zero offset (such as the zero noise of the electromagnetic sensor ≤±0.05 mL / min);

[0059] The pump group runs at 50% of the rated flow rate (such as the calibration flow rate of the 100 mL / min channel is 50 mL / min), and the deviation between the output value of the flow sensor and the measured value of the standard measuring cylinder is recorded (calibration error ≤±0.3% FS);

[0060] The linear compensation coefficient of the flow sensor is generated through three-point calibration (25%, 50%, 75% of the rated flow rate) ;

[0061] Among them, is the calibrated flow rate value, that is, the true flow rate measured by the standard measuring cylinder; is the original flow rate, that is, the uncalibrated flow rate directly output by the sensor;

[0062] During the continuous operation after the power-on of a stable liquid addition integration system for a test solution, each channel flow sensor collects flow signals at a frequency of 200 Hz (such as the turbine sensor outputs a pulse signal, and the frequency is proportional to the flow rate);

[0063] Take the average of the last 10 sampling points (data within 50 ms) to suppress high-frequency noise (such as pulse jitter caused by pump head vibration), and output the smoothed flow rate value ;

[0064] Apply the linear compensation coefficient of the flow sensor to correct the non-linear error of the sensor (such as the non-linear deviation of the turbine sensor at low flow rates ≤ ±0.5%), and obtain the real-time actual flow rate , where is the zero offset;

[0065] Exemplarily, the process of obtaining the pipeline back pressure is as follows:

[0066] Vertically install the piezoresistive pressure sensor 20 cm away from the outlet end of the peristaltic pump (to avoid interference from pump head vibration), and connect it in series with the microfluidic channel through a high-pressure resistant hose (inner diameter 3 mm);

[0067] Preset the acquisition rate (to meet the requirements for detecting pressure mutations, such as a change rate of 0.3 MPa / s requires at least 500 Hz sampling), and take the average of the last 10 sampling points (10 ms) as the smoothed pressure value ;

[0068] Through the built-in NTC thermistor, the temperature of the chip solution is monitored in real time, and quadratic polynomial compensation is performed: , and obtain the pressure after compensating the solution temperature , where the coefficient , , both a and b are obtained through calibration in a thermostat;

[0069] Then through the formula: , obtain the pipeline back pressure , where is the zero pressure (ambient pressure when there is no liquid, unit kPa), is the full-scale calibration coefficient (such as for a piezoresistive pressure sensor with a range of 1 MPa, after calibration );

[0070] Exemplarily, the process of obtaining the solution temperature is as follows:

[0071] Integrate the PT100 solution temperature sensor on the outer wall of the pipeline at the outlet end of the peristaltic pump (20 - 50 cm away from the pump head), and closely adhere it to the pipeline through heat-conducting silica gel to ensure that the sensor synchronizes with the solution temperature quickly, so as to obtain the solution temperature of the test solution body flowing through the liquid addition channel;

[0072] It should be noted that the solution temperature directly reflects the solution temperature of the liquid in the flowing state, rather than the ambient solution temperature or the solution temperature of the pump body, because the change in the solution temperature will directly affect its viscosity, and further affect the flow control accuracy;

[0073] Step 2: Real-time identify the sudden change working condition of the solution temperature, trigger the adaptive control process, and avoid the control lag caused by passive adjustment;

[0074] In some embodiments, through moving average filtering, calculate the average solution temperature change rate of consecutive sampling periods. The specific process is as follows:

[0075] Through the formula: , calculate the solution temperature change rate , where is the solution temperature at the i-th moment, is the solution temperature at the (i - 1)-th moment, is the sampling period;

[0076] Then calculate the average value of all solution temperature change rates within consecutive sampling periods and output the average solution temperature change rate;

[0077] If the average solution temperature change rate is greater than the preset solution temperature change rate, generate a parameter regulation signal;

[0078] Step 3: Based on the parameter regulation signal, through four-dimensional state space modeling and optimal estimation, fuse multi-sensor data, suppress the multi-physical field coupling error, and provide high-precision state estimation;

[0079] In some embodiments, when receiving the parameter regulation signal, define the state vector: , where is the real-time actual flow rate (mL / min) at the i-th moment, is the flow rate change rate (mL / min²) at the i-th moment, is the pipeline back pressure (MPa) at the i-th moment, is the solution temperature (°C) at the i-th moment;

[0080] It should be noted that the defined state vector is used to comprehensively describe the system state and provide a basis for state estimation and control;

[0081] It should be noted that the real-time actual flow rate is derived through the Bernoulli equation: , where is the pipeline damping coefficient (Pa·s / m), m is the liquid mass (kg), r is the pipeline radius (m), L is the pipeline length (m), is the liquid viscosity (Pa·s), is the pressure difference between the two ends of the pipeline at the i-th moment (MPa), is the pi;

[0082] Based on the state vector, obtain the state transition matrix , where is the acquisition period (s), is the solution temperature of the liquid viscosity (Pa·s), is the viscosity (Pa·s) at the reference solution temperature (such as 25°C), is the viscosity - solution temperature exponential coefficient ( , typical value of Newtonian fluid -0.03), is the reference solution temperature (25°C), e is the natural constant, with a value of 2.71828;

[0083] It should be noted that the state transition matrix needs to consider the exponential effect of the solution temperature on the viscosity and update the matrix elements in real time;

[0084] It needs to be further explained that the state transition matrix describes the transfer relationship of the state vector from the (i - 1)-th moment to the i-th moment, incorporates the effect of the solution temperature on the viscosity in real time, and is used to predict the change of the system state. Among them, the first row is the transfer of the real-time actual flow rate, the second row is the transfer of the flow rate change rate, and the third and fourth rows regard the pressure and the solution temperature as slow-varying parameters and directly inherit the values of the previous moment (assuming that the pressure and the solution temperature remain unchanged without mutation);

[0085] Predict the state at the current moment through the state and control input at the previous moment. The specific process is as follows:

[0086] A priori state estimation vector: , where is the pump drive pulse increment, is the state transition matrix at the previous moment, is the posteriori state estimation vector at the previous moment, is the input matrix, and ;

[0087] It should be noted that the a priori state estimation vector , uses the state transition matrix Predict the natural evolution of the previous state (e.g., the flow rate changes due to viscosity changes), and superimpose the control input pump drive pulse increment , reflecting the impact of manual adjustment on the flow rate;

[0088] It should be noted that the prior state estimation vector represents the predicted state at the current time i;

[0089] It should be noted that the input matrix , the first element represents the conversion of the number of pulses to the flow rate (e.g., 1 pulse corresponds to , 60 is the conversion factor from minutes to seconds); the second element represents the impact of the pulse increment on the flow rate change rate (considering the acquisition period ), determined based on the pulse control principle of the pump, and used to predict the current state by combining the natural evolution and manual control of the system;

[0090] Prior covariance matrix: , where is the noise matrix;

[0091] It should be noted that the noise matrix includes: flow vibration noise, solution temperature drift noise;

[0092] It should be further noted that is to map the covariance matrix of the previous moment of the state transition matrix to the current moment, reflecting the natural propagation of the uncertainty of state estimation. Among them, the left multiplication is to propagate the mean uncertainty according to the physical model (such as the flow rate change rate formula), and the right multiplication is to ensure the symmetry and dimensional matching of the covariance matrix;

[0093] Fuse the basic data measured by the sensor with the predicted state at the current moment to correct the prediction error. The specific process is as follows:

[0094] Measurement vector: , where is the observation matrix, and , is the measurement noise vector;

[0095] It should be noted that the measurement vector represents the basic data collected by the sensor at the i-th moment, that is ; the observation matrix represents the mapping state vector The linear transformation matrix to the measurement space, which describes the measurement range and accuracy of the sensor; the measurement noise vector represents the random noise in the sensor measurement process, which follows a zero-mean Gaussian distribution, and the covariance matrix is , , where , , are the measurement noise standard deviations of flow rate, pressure, and solution temperature respectively;

[0096] It is calculated through the Kalman gain: , dynamically adjusting the weights of the sensor data and the predicted value. That is, when the solution temperature suddenly changes (the credibility of the sensor data is high), increase the weight and give priority to believing the measurement value of the solution temperature; when the pressure is abnormal (possibly a pipeline blockage), suppress the flow gain to avoid incorrect adjustment;

[0097] It should be explained that is the Kalman gain matrix, which is used to balance the weights of the previous state estimate value and the current measurement value to minimize the estimation error, is the prior estimation error covariance matrix at the i-th moment, is the observation noise covariance matrix, which describes the statistical characteristics of the sensor measurement noise, is the observation matrix, is the transpose matrix of the observation matrix . The observation matrix is used to map the state vector to the measurement space to show the measurement range and accuracy of the sensor. The transpose matrix is obtained by interchanging the rows and columns of the observation matrix . The transpose matrix can adjust the weights of the measurement vector and the prior state estimate vector during data fusion. In the case of a sudden change in the solution temperature, etc., the measurement data is more reliable, and the transpose matrix will increase the influence of the measurement vector and make the estimation result closer to the measurement value; if the pressure is abnormal, it will reduce the weight of the corresponding measurement value to prevent incorrect adjustment. For example, in an experiment, when the solution temperature suddenly changes, the transpose matrix can make the system respond quickly and correct the estimated state according to the measurement value. The elements and arrangement of the transpose matrix reflect the contribution degree of each component of the measurement vector to the state estimation; if the element at the corresponding position of a certain component of the measurement vector in the transpose matrix is large, it means that this component has a great influence on the state estimation; otherwise, the influence is small, is the observation prediction error covariance matrix, which synthesizes the uncertainty of the prediction error and the measurement noise;

[0098] Then through the formula: , perform posterior state correction to obtain the posterior state estimation vector ;

[0099] Step 4: The change in the solution temperature will significantly affect the liquid viscosity, which in turn leads to flow fluctuations. By dynamically adjusting the PID parameters (the PID parameters refer to the three control parameters of the proportional coefficient (Proportional), integral coefficient (Integral), and derivative coefficient (Derivative) used for flow control), the influence of viscosity changes is offset, so that the experimental solution can be added stably. The specific process is as follows:

[0100] In some embodiments, through the formula: , calculate the viscosity compensation coefficient ; where specifically refers to the viscosity of the liquid at 25°C, which is a fixed reference value used to normalize the viscosity at different solution temperatures, is the yield stress of the non-Newtonian fluid, is the shear rate representing the relative movement speed between adjacent two layers inside the fluid, is the consistency coefficient of the non-Newtonian fluid, is the flow index of the non-Newtonian fluid;

[0101] Based on the viscosity compensation coefficient , adjust the PID parameters, so that the experimental solution can be added stably;

[0102] It should be explained that during the process of adding the test solution, the solution viscosity will change due to factors such as temperature, affecting the flow stability. By adjusting the PID parameters based on the viscosity compensation coefficient , the influence brought by viscosity changes can be effectively compensated, and stable liquid addition can be achieved. Specifically, it is manifested as the adjustment of the proportional coefficient, integral coefficient, and derivative coefficient, and the roles of these adjustments in compensating for the fluctuation of flow resistance, eliminating the steady-state error, and suppressing the flow change;

[0103] Exemplarily, the adjustment of the proportional coefficient: Multiply the proportional coefficient before adjustment by the viscosity compensation coefficient, and output the proportional coefficient after adjustment; when the solution viscosity changes due to factors such as temperature, the viscosity compensation coefficient will change accordingly. If the viscosity increases, the flow resistance increases, the viscosity compensation coefficient becomes larger, and the proportional coefficient after adjustment increases. The control signal output by the controller is enhanced, so that the driving device (such as a peristaltic pump) of the liquid addition device increases the driving force and improves the liquid addition flow rate to overcome the increased flow resistance and ensure the stability of the liquid addition flow rate; conversely, if the viscosity decreases, the viscosity compensation coefficient becomes smaller, and the proportional coefficient decreases, avoiding excessive liquid addition flow rate and maintaining stable liquid addition;

[0104] Integral coefficient adjustment: Calculate the ratio of the integral coefficient before adjustment to the viscosity compensation coefficient, and output the integral coefficient after adjustment; the integral action is mainly used to eliminate the steady-state error, so that the liquid addition flow rate can finally stabilize at the set value. When the viscosity change causes a deviation in the flow rate, if the viscosity increases, the viscosity compensation coefficient increases, the integral coefficient decreases, and the accumulation speed of the integral term slows down, avoiding system overshoot caused by too strong integral action; if the viscosity decreases, the viscosity compensation coefficient decreases, the integral coefficient increases, accelerating the accumulation of the integral term, and eliminating the steady-state error caused by viscosity change more quickly, ensuring that the liquid addition flow rate is stable at the set value;

[0105] Differential coefficient adjustment: Multiply the differential coefficient before adjustment by the square root of the viscosity compensation coefficient, and output the differential coefficient after adjustment; the differential action is used to suppress the change trend of the flow rate and enhance the system stability. When the viscosity change causes flow rate fluctuations, if the viscosity increases, the viscosity compensation coefficient increases, the differential coefficient increases, the sensitivity to flow rate changes increases, and the flow rate fluctuations can be suppressed more quickly; if the viscosity decreases, the viscosity compensation coefficient decreases, the differential coefficient decreases, avoiding overreaction to small flow rate changes and maintaining the stability of the liquid addition process;

[0106] The technical solution of the embodiment of the present invention is mainly as follows: Through a four-dimensional dynamic modeling system, the present invention breaks through the limitations of traditional two-dimensional control models, constructs a four-dimensional dynamic model including flow rate, flow rate change rate, pressure, and solution temperature, realizes the accurate capture of the coupling relationship of multiple physical fields, integrates the Bernoulli equation and the viscosity-solution temperature exponential model into the state transition matrix, and can dynamically predict the dynamic impact of solution temperature changes on flow rate characteristics; through the intelligent compensation mechanism of fluid constitutive characteristics, the present invention establishes a solution temperature-viscosity exponential model for Newtonian fluids and develops a shear rate-yield stress calculation model for non-Newtonian fluids, breaking through the limitations of traditional single compensation algorithms; through the dynamic adjustment method of PID parameters of the viscosity compensation coefficient, through the non-linear adjustment strategy that the proportional term is positively correlated with the viscosity coefficient, the integral term is negatively correlated, and the differential term is correlated with the square root, the present invention realizes the real-time matching of control parameters and fluid characteristics; through the multi-sensor joint calibration technology, through the three-point calibration of the flow sensor (non-linear error correction), the quadratic polynomial solution temperature compensation of the pressure sensor (eliminating the influence of the ambient solution temperature), and the dynamic response optimization of the solution temperature sensor, the present invention constructs a basic data acquisition system with an error magnitude lower than 50% of the industry standard, providing reliable input for intelligent control algorithms. Embodiment

[0107] On the basis of Embodiment 1, please refer to Figure 2 As shown, a stable liquid addition integrated system for a test solution described in the embodiment of the present invention includes:

[0108] Real-time monitoring module: multi-dimensional real-time monitoring to obtain basic data;

[0109] Trigger condition detection module: based on the basic data, real-time identify the sudden change condition of the solution temperature, and trigger the adaptive control process;

[0110] Data fusion module: after triggering the adaptive control process, through four-dimensional state space modeling and optimal estimation, fuse the basic data, suppress the multi-physical field coupling error, and provide high-precision state estimation;

[0111] Parameter regulation module: based on the state estimation result, calculate the viscosity compensation coefficient of each fluid type, so as to dynamically adjust the liquid addition process parameters, compensate the flow resistance fluctuation caused by the viscosity change, and thus enable the experimental solution to be added stably.

[0112] The above has described an embodiment of the present invention in detail, but the described content is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for the stable addition of a test solution, characterized in that: include: Step 1: Multi-dimensional real-time monitoring to obtain basic data; Step 2: Based on basic data, identify the sudden change of solution temperature in real time and trigger the adaptive control process; Step 3: Once the adaptive control process is triggered, four-dimensional state space modeling and optimal estimation are used to fuse basic data, suppress multi-physics field coupling errors, and provide high-precision state estimation. Step 4: Based on the state estimation results, the viscosity compensation coefficient of each fluid type is calculated to dynamically adjust the liquid addition process parameters to compensate for the flow resistance fluctuation caused by viscosity changes, so that the experimental solution can be added stably; Basic data include: real-time actual flow rate, pipeline back pressure, solution temperature; The process of obtaining real-time actual traffic is as follows: The original signal is collected by the flow sensor at high frequency, the zero point offset is calculated, and a linear compensation coefficient is generated through three-point calibration. The smoothed flow value is then averaged over the most recent sampling points. The nonlinear error of the smoothed flow value is corrected using the zero point offset and the linear compensation coefficient to obtain the real-time actual flow rate. The process of obtaining pipeline back pressure is as follows: Through the piezoresistive pressure sensor, after sampling, the average of the most recent multiple sampling points is taken to obtain a smoothed pressure value, and the smoothed pressure value is compensated by a quadratic polynomial based on the solution temperature to obtain a compensated pressure value. The compensated pressure value is further corrected based on the zero-point pressure value and the full-scale correction coefficient to obtain the pipeline back pressure; The process of obtaining the solution temperature is: The solution temperature is obtained by collecting the temperature of the flowing solution in real time through the solution temperature sensor; Based on basic data, the process of identifying sudden temperature changes in the solution in real time and triggering the adaptive control process is as follows: First, calculate the solution temperature change rate of each sampling period, and then calculate the average solution temperature change rate by averaging all solution temperature change rates in consecutive sampling periods; If the average solution temperature change rate is greater than the preset solution temperature change rate, a parameter control signal is generated; The process of providing high-precision state estimation by four-dimensional state space modeling and optimal estimation, fusing multi-sensor data, suppressing multi-physics field coupling errors, and providing high-precision state estimation is as follows: When the parameter control signal is received, a four-dimensional state vector is defined, wherein the four-dimensional state vector includes the real-time actual flow rate, flow rate change rate, pipeline back pressure and solution temperature; Then, based on the posterior state estimation vector and the input matrix at the previous moment, the state at the current moment is predicted to obtain the prior state estimation vector and the prior covariance matrix. Combining the collected basic data into a measurement vector; Through Kalman gain calculation, the weights of the measurement vector and the prior state estimation vector are dynamically adjusted to flexibly allocate trust according to different situations; The prior state estimation vector is then corrected by the Kalman gain to obtain the posterior state estimation vector, thereby suppressing the multi-physics field coupling error and providing high-precision state estimation. The process of obtaining the viscosity compensation coefficient is: Identify the fluid type; If the fluid is a Newtonian fluid, based on the exponential effect of solution temperature on viscosity, the ratio of the viscosity at the current solution temperature to the viscosity at the reference solution temperature is calculated in real time. When the solution temperature rises, the viscosity decreases exponentially according to the preset solution temperature coefficient, generating a viscosity compensation coefficient. If the fluid is non-Newtonian, calculate the apparent viscosity, compare the apparent viscosity with the reference viscosity, and generate a viscosity compensation factor; The process of dynamically adjusting the dosing process parameters is as follows: Adjust the PID parameters based on the viscosity compensation coefficient; PID parameters include: proportional coefficient, integral coefficient, differential coefficient; The proportional coefficient before adjustment is multiplied by the viscosity compensation coefficient, and the adjusted proportional coefficient is output; the ratio of the integral coefficient before adjustment and the viscosity compensation coefficient is calculated, and the adjusted integral coefficient is output; the differential coefficient before adjustment is multiplied by the square root of the viscosity compensation coefficient, and the adjusted differential coefficient is output.

2. An integrated system for stable liquid addition of a test solution, applied to the integrated method for stable liquid addition of a test solution according to claim 1, characterized in that: The system includes: Real-time monitoring module: multi-dimensional real-time monitoring to obtain basic data; Trigger condition detection module: Based on basic data, it identifies the sudden change of solution temperature in real time and triggers the adaptive control process; Data fusion module: When the adaptive control process is triggered, it fuses basic data through four-dimensional state space modeling and optimal estimation, suppresses multi-physics field coupling errors, and provides high-precision state estimation; Parameter control module: Based on the state estimation results, the viscosity compensation coefficient of each fluid type is calculated, so as to dynamically adjust the parameters of the liquid addition process to compensate for the flow resistance fluctuation caused by the viscosity change, so that the experimental solution can be added stably.

Images