A method for evaluating resin bed life based on multiple regression model

Abstract

The present invention provides a resin bed life assessment method based on a multiple regression model, comprising the following steps: 1) resin bed operation monitoring test: test equipment, test parameters and test results; 2) prediction model: mathematical modeling, numerical analysis, least squares method, polynomial regression and multiple regression analysis; 3) prediction model construction and verification: model establishment, prediction model verification and prediction model application; mathematical modeling plays a good role in solving practical problems and involves a wide range of application fields; it requires the integration and application of various knowledge; it requires the cooperation of various technical means, etc., and a method of using multiple nonlinear regression to solve the model generally uses variable interchange to convert the nonlinear model into a linear model. This method improves the efficiency of calculation, increases the accuracy of calculation results, and realizes the life construction model processing and evaluation of the resin bed during its operation.

Description

Technical Field

[0001] The present invention relates to the field of resin bed life prediction, in particular to a resin bed life evaluation method based on a multiple regression model. Background Art

[0002] Ion exchange resins have exchange capacity and a series of other excellent properties. To put them into practical use, it is advisable to adopt a dynamic continuous exchange method and corresponding equipment to exchange, separate, concentrate and adsorb the electrolyte solution to be treated. Research, design and practical application have shown that columns or towers filled with ion exchange resins are the most effective. Such columns or towers are called resin beds, also often called exchange columns. The basic operation process of the resin bed is divided into two steps, upward and downward. During the upward flow, the filtered waste acid flows from the bottom to the top of the bed, the acid is adsorbed by the resin particles, and the metal salt solution is discharged from the top. During the downward flow, water flows from the top to the bottom through the bed to desorb the acid. The iron ion concentration of the purified acid liquid is very low and is returned to the pickling tank for reuse.

[0003] During use, the absorption and filtration performance of the resin bed (ion exchange resin) will be affected by its own usage time. Therefore, to ensure the high efficiency of the resin bed during use, it is necessary to constantly evaluate and test the life of the resin bed. However, the existing technology does not have a good technical method to evaluate and test the life of the resin bed. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a resin bed life assessment method based on a multiple regression model to solve the problems raised in the above background technology. The present invention has a novel structure. The running resin bed is converted into model data by a computer, and the model data is plotted and the error is reduced to obtain the life assessment of the resin bed.

[0005] In order to achieve the above object, the present invention is implemented by the following technical solution: a resin bed life evaluation method based on a multiple regression model, the evaluation method comprising the following steps:

[0006] 1) Resin bed operation monitoring test;

[0007] Test device: The water from the sampling port flows into the cooler for cooling, then enters the conductivity meter circulation pool, and finally flows back to the water tank. The water enters the water inlet at the top of the resin column through the circulation pump from the water tank, flows out from the water outlet at the bottom, and flows back to the water tank.

[0008] Experimental parameters: Artificial seawater was added to the raw water tank using a peristaltic pump. The dosing flow rate was controlled by controlling the peristaltic pump speed based on the resin bed outlet water conductivity and outlet flow rate, so that the inlet water conductivity was stabilized between 1 and 2 μS / cm. The inlet water flow rate was switched between 20 / 40 / 80 / 120 / 160 m / h, and the inlet water temperature was maintained at 70°C. During the experiment, it was necessary to ensure that the inlet water conditions remained unchanged between the failure of two sampling ports. When two consecutive sampling ports failed, the inlet water conditions were immediately changed. In addition to the inlet water conductivity monitoring meter, a total of three conductivity meters were used, so the outlet water of the three sampling ports could be monitored in real time at the same time. At the beginning of the experiment, the inlet water, port 1, port 2, and port 3 were monitored. When port 1 failed, ports 2, 3, and 4 were monitored instead, and so on. The inlet water conductivity and the outlet water conductivity of the three sampling ports were recorded in real time using a data acquisition device.

[0009] Test results: With the failure criterion set at a sampling port outlet conductivity of 0.6 μS / cm, the system operated for a total of 522 hours until sampling port No. 14 failed. During this operation, the inlet water conductivity was controlled between 1 and 2 μS / cm, and the inlet flow rate was switched between 20 and 160 m / h. This round of experiments was repeated several times to obtain more raw data sets. The real-time monitoring data allowed the failure time of each sampling port to be determined, and the failure point movement speed under these operating conditions to be calculated accordingly.

[0010] 2) Prediction model;

[0011] Mathematical modeling: model preparation, model assumptions, model building, model solving, model analysis and model testing;

[0012] Numerical analysis: Run calculations on a computer to obtain results, and then verify the feasibility of the mathematical model based on the calculation results;

[0013] Least squares method: draw a scatter plot in a two-dimensional coordinate system;

[0014] Multiple regression analysis: Obtain the various parameters of the mathematical model based on a series of sample data, and then establish a mathematical model based on these parameters.

[0015] 3) Construction and validation of prediction models;

[0016] Model establishment: The conductivity of the resin bed inlet water is set to x 1i , the water flow rate into the resin bed is set to x 2i , the moving speed of the resin bed failure point is set to y i , three data are constructed into a data point (x 1i ,x 2i ,y i ), and then the least squares method is used to construct the fitting function.

[0017] Prediction model validation: Failure point movement speed and working layer obtained using binary linear regression model.

[0018] Application of the prediction model: The calculated multiple linear regression model is linked to the monitoring of the inlet water conductivity and the inlet water flow rate. Whenever either of the two changes beyond the set limit, a calculation is performed.

[0019] In the above technical solution, in step 1), the experimental device is the resin bed portion of a small mobile multifunctional deaerator bed-resin bed sampling and monitoring device. The ion exchange column body is made of stainless steel, and the top end cover and water distribution device are made of 304 stainless steel. A sampling port is opened every 8 mm on the side of the resin column, for a total of 14 sampling ports, numbered 1 to 14 from top to bottom.

[0020] In the above technical solution, in step 1), the lipid bed inlet water is condensate prepared by mixing high-purity water and artificial seawater in a certain proportion. According to the corresponding standards of artificial seawater, 24.53 g NaCl, 5.20 g MgCl2, 4.09 g Na2SO4, and 1.16 g CaCl2 are dissolved in 1 L of desalted water. The conductivity of artificial seawater is about 400 mS / cm.

[0021] In the above technical solution, in step 3), the steps of constructing the fitting function are as follows:

[0022] 1) Determine the type of fitting curve: Draw a scatter plot based on the data points, observe the position of the data in the scatter plot, use curves containing different basis functions to fit, select the most suitable curve as the fitting type, and construct an approximate fitting equation based on the type of basis function selected for the fitting curve;

[0023] 2) Solve the fitting curve equation: Import the measured data into MATLAB, draw the corresponding fitting curve, and solve the fitting curve equation. Then convert the obtained fitting curve equation into a function expression of the actual parameter variables in the experiment;

[0024] 3) Use fitting curves to solve practical problems: Use fitting equations in the experiment to compensate the final experimental results and observe whether the compensated experimental results meet the error requirements. If not, it is necessary to change the fitting curve type, that is, increase the degree of the polynomial equation until the compensation results meet the error requirements.

[0025] Beneficial effects of the present invention: The resin bed life evaluation method based on the multivariate regression model of the present invention is

[0026] 1. This resin bed life assessment method based on the multiple regression model plays a good role in solving practical problems through mathematical modeling, involving a wide range of application fields; it requires the integration of various knowledge and the coordination of various technical means, etc.

[0027] 2. This resin bed life assessment method based on the multivariate regression model uses the basic principle of least squares method to solve the fitting curve, which is an important basis for establishing a prediction model.

[0028] 3. This resin bed life assessment method based on a multiple regression model uses multiple nonlinear regression to solve the model. Generally, the nonlinear model is converted into a linear model by using variable interchange. This method improves the efficiency of the calculation, increases the accuracy of the calculation results, and realizes the life modeling and evaluation of the resin bed during its operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the structure of a test device for a resin bed life assessment method based on a multiple regression model according to the present invention;

[0030] Figure 2 This is a schematic diagram of calculation results of the failure point moving speed of a resin bed life assessment method based on a multiple regression model of the present invention;

[0031] Figure 3 The present invention is a mathematical modeling process principle diagram of a resin bed life assessment method based on a multiple regression model. DETAILED DESCRIPTION

[0032] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0033] See also Figures 1 to 3 The present invention provides a technical solution: a resin bed life evaluation method based on a multiple regression model, the evaluation method comprising the following steps:

[0034] 1) Resin bed operation monitoring test;

[0035] Purpose of the test: Utilize multiple sampling ports on a small device and use an online conductivity meter to monitor the water outlet from the sampling ports in real time to obtain the specific time of failure of each sampling port, thereby obtaining the failure point movement speed under different inlet water conductivities and inlet water flow rates. Based on the historical and real-time failure point movement speeds, a mathematical model is established to predict the remaining life of the resin.

[0036] Test device: The water outlet from the sampling port flows into the cooler for cooling and then enters the conductivity meter circulation pool, and finally flows back to the water tank. The water inlet enters the water inlet on the top of the resin column through the circulation pump from the water tank, flows out from the water outlet at the bottom, and flows back to the water tank. The test device is the resin bed part on the small mobile multifunctional deaerator bed-resin bed sampling and monitoring device. The ion exchange column body is made of stainless steel, and the top end cover and water distribution device are made of 304 stainless steel. A sampling port is opened every 8 mm on the side of the resin column, with a total of 14 sampling ports, numbered 1 to 14 from top to bottom; a needle valve is connected to the sampling port to control the opening, and a ball valve is connected to control the valve opening and closing.

[0037] Test parameters: The resin bed inlet water is condensate prepared by mixing high-purity water and artificial seawater in a certain proportion. According to the corresponding standards of artificial seawater, 24.53 g NaCl, 5.20 g MgCl2, 4.09 g Na2SO4, and 1.16 gCaCl2 are dissolved in 1L of demineralized water. The conductivity of artificial seawater is about 400 mS / cm. A peristaltic pump is used to add artificial seawater to the raw water tank. According to the conductivity and flow rate of the resin bed outlet water, the dosing flow rate is controlled by controlling the peristaltic pump speed to make the inlet water conductivity stable between 1~2 μS / cm and the inlet flow rate is 20 / 40 / 80 / 120 / 160 m / h, and the inlet water temperature is kept at 70℃. During the test, it is necessary to ensure that the inlet water condition remains unchanged between the failures of two sampling ports. When two consecutive sampling ports fail, the inlet water condition is changed immediately. In addition to the inlet water conductivity monitoring meter, there are a total of three conductivity meters. Therefore, the outlet water of the three sampling ports can be monitored in real time at the same time. At the beginning of the test, the inlet water, port 1, port 2, and port 3 are monitored. When port 1 fails, port 2, port 3, and port 4 are monitored instead, and so on. The inlet water conductivity and the outlet water conductivity of the three sampling ports are recorded in real time using a data acquisition device.

[0038] Test results: With the failure criterion set at a sampling port outlet conductivity of 0.6 μS / cm, the system operated for a total of 522 hours until sampling port No. 14 failed. During this operation, the inlet water conductivity was controlled between 1 and 2 μS / cm, and the inlet flow rate was switched between 20 and 160 m / h. This round of experiments was repeated several times to obtain more raw data sets. The real-time monitoring data allowed the failure time of each sampling port to be determined, and the failure point movement speed under these operating conditions to be calculated accordingly.

[0039] 2) Prediction model;

[0040] Mathematical modeling: model preparation, model assumptions, model building, model solving, model analysis and model testing (e.g. Figure 3 shown).

[0041] Numerical analysis: Find a calculation method for the mathematical model based on the knowledge learned, then design a program, run the program on the computer to calculate the results, and then verify the feasibility of the mathematical model based on the calculation results;

[0042] Least squares method: Observed discrete data is plotted as a scatter plot in a two-dimensional coordinate system. When all observation points are connected in sequence, it is found that the connected curve is an irregular curve. In this case, it is necessary to construct an approximate curve φ(x) to approximate all observation points (xi, yi). It is not required to pass through all observation points, but to require the constructed curve φ(x) to approximate the objective function so that the error or distance between φ(x) and y is minimized. At this time, φ(x) can become a fitting curve.

[0043] Multiple regression analysis: Obtain the various parameters of the mathematical model based on a series of sample data, and then establish a mathematical model based on these parameters.

[0044] 3) Construction and validation of prediction models;

[0045] Model establishment: The conductivity of the resin bed inlet water is set to x 1i , the water flow rate into the resin bed is set to x 2i , the moving speed of the resin bed failure point is set to y i , three data are constructed into a data point (x 1i ,x 2i ,y i ), and then use the least squares method to construct the fitting function;

[0046] Verification of the prediction model: The failure point movement speed and working layer obtained by the binary linear regression model can be concluded that the multivariate linear regression model can well predict the remaining operating time of the resin bed under the operating conditions of inlet water conductivity of 1~2 μS / cm and inlet water flow rate of 20~160m / h.

[0047] Prediction model application: The calculated multivariate linear regression model is linked to the monitoring of inlet water conductivity and inlet water flow rate. Whenever either of the two changes exceeds the set limit (such as conductivity change exceeding 0.10 μS / cm, inlet water flow rate change exceeding 10 m / h), a calculation is performed.

[0048] In this embodiment, in step 3, the steps of constructing the fitting function are as follows:

[0049] 1) Determine the type of fitting curve: Draw a scatter plot based on the data points, observe the position of the data in the scatter plot, use curves containing different basis functions to fit, select the most suitable curve as the fitting type, and construct an approximate fitting equation based on the type of basis function selected for the fitting curve;

[0050] 2) Solve the fitting curve equation: Import the measured data into MATLAB, draw the corresponding fitting curve, and solve the fitting curve equation. Then convert the obtained fitting curve equation into a function expression of the actual parameter variables in the experiment;

[0051] 3) Use fitting curves to solve practical problems: Use fitting equations in the experiment to compensate the final experimental results and observe whether the compensated experimental results meet the error requirements. If not, it is necessary to change the fitting curve type, that is, increase the degree of the polynomial equation until the compensation results meet the error requirements.

[0052] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.

[0053] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A resin bed life assessment method based on a multiple regression model, characterized by: The evaluation method comprises the following steps: 1) Resin bed operation monitoring test; Test device: The water from the sampling port flows into the cooler for cooling, then enters the conductivity meter circulation pool, and finally flows back to the water tank. The water enters the water inlet at the top of the resin column through the circulation pump from the water tank, flows out from the water outlet at the bottom, and flows back to the water tank; Test parameters: Artificial seawater was added to the raw water tank using a peristaltic pump. The dosing flow rate was controlled by controlling the peristaltic pump speed according to the conductivity and flow rate of the resin bed outlet water, so that the inlet water conductivity was stabilized between 1 and 2 μS / cm. The inlet water flow rate was switched between 20 / 40 / 80 / 120 / 160 m / h, and the inlet water temperature was maintained at 70°C. During the test, it was necessary to ensure that the inlet water conditions remained unchanged between the failures of two sampling ports. When two consecutive sampling ports failed, the inlet water conditions were immediately changed. In addition to the inlet water conductivity monitoring meter, a total of three conductivity meters were used, so the outlet water of the three sampling ports could be monitored in real time at the same time. At the beginning of the test, the inlet water, port 1, port 2, and port 3 were monitored. When port 1 failed, port 2, port 3, and port 4 were monitored instead, and so on. The inlet water conductivity and the outlet water conductivity of the three sampling ports were recorded in real time using a data acquisition device. Test results: The failure criterion was a sample outlet water conductivity of 0.6 μS / cm. The system operated for a total of 522 hours until the failure of sample outlet No.

14. During this operation, the inlet water conductivity was controlled between 1 and 2 μS / cm, and the inlet flow rate varied between 20 and 160 m / h. This round of experiments was repeated several times to obtain more raw data sets. Based on real-time monitoring data, the failure time of each sample outlet was determined, and the failure point movement speed under these operating conditions was calculated. 2) Prediction model; Mathematical modeling: model preparation, model assumptions, model building, model solving, model analysis and model testing; Numerical analysis: Run calculations on a computer to obtain results, and then verify the feasibility of the mathematical model based on the calculation results; Least squares method: draw a scatter plot in a two-dimensional coordinate system; Multiple regression analysis: Obtain the parameters of the mathematical model based on a series of sample data, and then establish the mathematical model based on these parameters; 3) Construction and validation of prediction models; Model establishment: The conductivity of the resin bed inlet water is set to x 1i , the water flow rate into the resin bed is set to x 2i , the moving speed of the resin bed failure point is set to y i , three data are constructed into a data point (x 1i ,x 2i ,y i ), and then use the least squares method to construct the fitting function; Prediction model verification: Failure point movement speed and working layer obtained using binary linear regression model; Application of the prediction model: The calculated multiple linear regression model is linked to the monitoring of the inlet water conductivity and the inlet water flow rate. Whenever either of the two changes beyond the set limit, a calculation is performed.

2. The resin bed life assessment method based on a multiple regression model according to claim 1, characterized in that: The test device is the resin bed part of a small mobile multifunctional deaeration bed-resin bed sampling and monitoring device. The ion exchange column body is made of stainless steel, and the top end cover and water distribution device are made of 304 stainless steel. A sampling port is opened every 8 mm on the side of the resin column, with a total of 14 sampling ports, numbered 1 to 14 from top to bottom.

3. The resin bed life assessment method based on a multiple regression model according to claim 2, characterized in that: In step 1), the resin bed inlet water is condensate prepared by mixing high-purity water and artificial seawater in a certain proportion. According to the corresponding standard of artificial seawater, 24.53 g NaCl, 5.20 g MgCl2, 4.09 g Na2SO4, and 1.16 g CaCl2 are dissolved in 1 L of demineralized water. The conductivity of artificial seawater is about 400 mS / cm.

4. The resin bed life assessment method based on a multiple regression model according to claim 1, characterized in that: In step 3), the steps of constructing the fitting function are as follows: 1) Determine the type of fitting curve: Draw a scatter plot based on the data points, observe the position of the data in the scatter plot, use curves containing different basis functions to fit, select the most suitable curve as the fitting type, and construct an approximate fitting equation based on the type of basis function selected for the fitting curve; 2) Solve the fitting curve equation: Import the measured data into MATLAB, make the corresponding fitting curve, solve the fitting curve equation, and convert the obtained fitting curve equation into a function expression of the actual parameter variables in the experiment; 3) Use fitting curves to solve practical problems: Use fitting equations in the experiment to compensate the final experimental results and observe whether the compensated experimental results meet the error requirements. If not, it is necessary to change the fitting curve type, that is, increase the degree of the polynomial equation until the compensation results meet the error requirements.

Images