Pulping degree soft measurement and control method based on internet of things edge computing

Abstract

The application discloses a beating degree soft measurement and control method based on Internet of Things edge computing, and the method realizes online beating degree soft measurement and accurate control by combining cloud server calculation, edge box calculation and PLC. The cloud server stores a large amount of data related to the beating degree, has strong functions for processing complex calculations, and thus calculates important intermediate coefficients. Edge computing has strong computing power, high real-time data exchange with PLC, and provides real-time data support for PLC control. The method has achieved good results in practical application, not only reduces the workload of workers and saves costs, but also improves the grinding efficiency and greatly improves the grinding quality, thereby laying a solid foundation for producing high-quality paper.

Description

Technical Field

[0001] This invention relates to the field of papermaking technology, specifically to a soft measurement and control method for beating degree based on Internet of Things edge computing. Background Technology

[0002] The pulper is a key piece of equipment in the papermaking pulping process. Its function is to induce physical changes in the pulp fibers through mechanical action, thereby acquiring specific properties. The main indicator is "freezing degree," which comprehensively reflects the degree to which fibers are cut, swollen, fibrillated, and separated. Whether the freezing degree meets the standard and its stability plays a decisive role in the final paper quality. However, in my country, the method for measuring "freezing degree" still relies on manual methods: manual sampling, manual testing in the laboratory, and manual adjustments if deviations occur. This is inefficient and the control and adjustment time is too long, resulting in unstable freezing degree.

[0003] There are more than ten factors affecting freeness, and the forms and data of their influence vary. Previously, the automatic control methods of Industry 3.0 could not overcome this challenge. A small number of "online freeness measuring instruments" that were supposedly capable of online automatic measurement of freeness were imported from abroad. However, in actual trials, it was found that the real-time measurement and control of freeness had significant errors, and the equipment was expensive. Summary of the Invention

[0004] The purpose of this invention is to provide a soft measurement and control method for beating degree based on Internet of Things edge computing, in order to solve the problem that the "online beating degree measuring instrument" in the prior art mentioned in the background has large errors in real-time measurement and control of beating degree and is expensive in actual use.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A soft measurement and control method for pulping degree based on IoT edge computing includes the following steps:

[0007] Acquire parameters related to the degree of beating during the pulping process and transmit the acquired parameters to the edge box;

[0008] The relevant parameters obtained by the edge box are transmitted to the cloud server. The cloud server calculates the intermediate coefficients required for edge computing and transmits the intermediate coefficients to the edge box.

[0009] The edge box calculates the soft measurement value SRr of the beating degree using the acquired data and intermediate coefficients;

[0010] The grinding disc advance and retraction motors are controlled based on the calculated beating degree to achieve constant beating degree grinding.

[0011] More preferably, the parameters related to the degree of beating during the refining process include the pulp concentration entering the mill, the pulp flow rate entering the mill, the mill disc operating power, the beating specific pressure, the beating time, the beating temperature, the distance between the mill discs, the performance parameters of the mill discs, the lifespan of the mill discs, the properties of the wood chips, and the pH value of the pulp.

[0012] More preferably, the pulp concentration, pulp flow rate, grinding disc operating power, beating specific pressure, beating time, beating temperature, distance between grinding discs, and pulp pH value are obtained through calculation and instrument measurement; the characteristics of the grinding discs, the lifespan of the grinding discs, and the origin of the wood chips are fixed values, obtained through manual input.

[0013] More preferably, the cloud server stores the performance parameters of grinding discs from different manufacturers, the lifespan of grinding discs, and the properties of wood chips from different regions. Based on the performance parameters of the grinding discs, the lifespan of the grinding discs, and the properties of the wood chips, an intermediate coefficient affecting the degree of beating can be calculated. Then, the calculated intermediate coefficient is sent back to the edge box to participate in the edge box's calculation of the soft measurement value SRr for the degree of beating.

[0014] More preferably, when controlling the grinding disc advance and retraction motor according to the calculated beating degree to achieve constant beating degree of grinding disc, the soft measurement value SRr of the beating degree obtained by the edge box is compared with the set value, and the deviation value Δ is calculated by the control unit to obtain the adjustment amount, thereby adjusting the advance amount of the grinding disc advance and retraction motor, thereby adjusting the gap of the grinding disc blades and changing the beating degree of the slurry until Δ→0.

[0015] More preferably, the edge box integrates data fitting and neural network self-learning algorithms.

[0016] More preferably, the calculation of the edge box includes the following steps:

[0017] Enter the beating degree measured in the laboratory;

[0018] The edge box derives an initial beating degree algorithm model based on laboratory measurements of beating degree using a data fitting algorithm. The formula for beating degree is:

[0019]

[0020] SR represents the degree of beating, K is the coefficient, W is the active power of the disc mill, C is the feed concentration to the disc mill, F is the discharge flow rate of the disc mill, and T1 is the percentage of the disc mill's operating time to its total lifespan.

[0021] Further analysis reveals a non-linear relationship between power and freeness, exhibiting a curvilinear relationship. Therefore, the freeness formula is:

[0022]

[0023] B is a constant;

[0024] By fitting the data using this formula model, relevant parameters can be obtained, and thus an accurate model can be derived.

[0025] Train the initial pulping degree algorithm model;

[0026] The trained beating degree algorithm is then further optimized through the Back Propagation neural network self-learning algorithm to obtain an accurate soft measurement value SRr for beating degree.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] This method utilizes a combination of cloud server computing, edge computing, and PLC to achieve online soft measurement and precise control of freeness. The cloud server stores a large amount of data related to freeness and possesses powerful capabilities for handling complex calculations, enabling the calculation of crucial intermediate coefficients. Edge computing offers powerful computing capabilities and high real-time data exchange with the PLC, providing real-time data support for PLC control. This method has achieved excellent results in practical applications, not only reducing worker workload and saving costs in factories, but also significantly improving pulping efficiency and quality, laying a solid foundation for producing high-quality paper. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the soft measurement and control method for pulping degree based on Internet of Things edge computing of the present invention;

[0030] Figure 2 This is a flowchart of the edge box calculation process of the present invention;

[0031] Figure 3 This is a schematic diagram of the control system structure of the present invention;

[0032] Figure 4 This is a flowchart of the neural network algorithm of the present invention; Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Please see Figure 1-4 The present invention provides a technical solution:

[0035] A soft measurement and control method for pulping degree based on IoT edge computing includes the following steps:

[0036] Acquire parameters related to the degree of beating during the pulping process and transmit the acquired parameters to the edge box;

[0037] The relevant parameters obtained by the edge box are transmitted to the cloud server. The cloud server calculates the intermediate coefficients required for edge computing and transmits the intermediate coefficients to the edge box.

[0038] The edge box calculates the soft measurement value SRr of the beating degree using the acquired data and intermediate coefficients;

[0039] The grinding disc advance and retraction motors are controlled based on the calculated beating degree to achieve constant beating degree grinding.

[0040] In this invention, the parameters related to the degree of beating during the pulping process include the pulp concentration entering the mill, the pulp flow rate entering the mill, the mill disc operating power, the beating specific pressure, the beating time, the beating temperature, the distance between mill discs, the performance parameters of the mill discs, the lifespan of the mill discs, the properties of the wood chips, and the pH value of the pulp. The pulp concentration entering the mill, the pulp flow rate entering the mill, the mill disc operating power, the beating specific pressure, the beating time, the beating temperature, the distance between mill discs, and the pH value of the pulp are obtained through calculation and instrument measurement; the characteristics of the mill discs, the lifespan of the mill discs, and the origin of the wood chips are fixed values, obtained through manual input.

[0041] In this invention, the cloud server stores the performance parameters of grinding discs from different manufacturers, the life cycle of grinding discs, and the properties of wood chips from different regions. Based on the performance parameters of the grinding discs, the life cycle of the grinding discs, and the properties of the wood chips, an intermediate coefficient affecting the degree of beating can be calculated. Then, the calculated intermediate coefficient is sent back to the edge box to participate in the edge box's calculation of the soft measurement value SR of the degree of beating.

[0042] In this invention, the grinding disc advance and retraction motor is controlled according to the calculated beating degree to achieve constant beating degree of grinding disc. When grinding disc is constantly beating, the soft measurement value SRr of beating degree obtained by the edge box is compared with the set value. The deviation value Δ is calculated by the control unit to obtain the adjustment amount, and the advance amount of the grinding disc advance and retraction motor is adjusted, thereby adjusting the gap of the grinding disc and changing the beating degree of the slurry until Δ→0.

[0043] In this invention, the edge box integrates data fitting and neural network self-learning algorithms. The calculation of the edge box includes the following steps:

[0044] Enter the beating degree measured in the laboratory;

[0045] The edge box derived an initial freeness algorithm model based on laboratory-measured freeness using a data fitting algorithm. Freeness is positively correlated with grinding power, inversely correlated with concentration flow rate, and inversely correlated with the percentage of disc mill running time (T1) of the total lifespan. The formula for freeness is:

[0046]

[0047] SR represents the degree of beating, K is the coefficient, W is the active power of the disc mill, C is the feed concentration to the disc mill, F is the discharge flow rate of the disc mill, and T1 is the percentage of the disc mill's operating time to its total lifespan.

[0048] Further analysis reveals a non-linear relationship between power and freeness, exhibiting a curvilinear relationship. Therefore, the freeness formula is:

[0049]

[0050] B is a constant;

[0051] By fitting the data using this formula model, relevant parameters can be obtained, and thus an accurate model can be derived.

[0052] Train the initial pulping degree algorithm model;

[0053] The trained beating degree algorithm is then further optimized through a BP neural network self-learning algorithm to obtain an accurate soft measurement value SRr for beating degree.

[0054] Example 1

[0055] The PLC acquires parameters related to freeness during the refining process. These parameters include pulp concentration, pulp flow rate, refining disc power, refining pressure, refining time, refining temperature, disc spacing, disc performance parameters, refining disc lifespan, wood chip properties, and pulp pH. Pulp concentration, pulp flow rate, refining disc power, refining pressure, refining time, refining temperature, disc spacing, and pulp pH are obtained through calculation and instrument measurement. Disc characteristics, refining disc lifespan, and wood chip origin are fixed values ​​obtained through manual input. The acquired parameters are then transmitted to the edge box.

[0056] The relevant parameters acquired through the edge box are transmitted to the cloud server. The cloud server calculates the intermediate coefficients required for edge computing. The cloud server stores the performance parameters of grinding discs from different manufacturers, the lifespan of the grinding discs, and the properties of wood chips from different regions. Based on the performance parameters of the grinding discs, the lifespan of the grinding discs, and the properties of the wood chips, the intermediate coefficients affecting the freeness can be calculated. These calculated intermediate coefficients are then transmitted back to the edge box to participate in the edge box's calculation of the soft measurement value SRR for the freeness. The intermediate coefficients are then transmitted back to the edge box.

[0057] The edge box calculates the soft measurement value SRr of the pulping degree using the acquired data and intermediate coefficients; the edge box integrates data fitting and neural network self-learning algorithms. The edge box calculation includes the following steps:

[0058] Enter the beating degree measured in the laboratory;

[0059] The edge box derived an initial freeness algorithm model based on laboratory-measured freeness using a data fitting algorithm. Freeness is positively correlated with grinding power, inversely correlated with concentration flow rate, and inversely correlated with the percentage of disc mill running time (T1) of the total lifespan. The formula for freeness is:

[0060]

[0061] SR represents the degree of beating, K is the coefficient, W is the active power of the disc mill, C is the feed concentration to the disc mill, F is the discharge flow rate of the disc mill, and T1 is the percentage of the disc mill's operating time to its total lifespan.

[0062] Further analysis reveals a non-linear relationship between power and freeness, exhibiting a curvilinear relationship. Therefore, the freeness formula is:

[0063]

[0064] B is a constant;

[0065] By fitting the data using this formula model, relevant parameters can be obtained, and thus an accurate model can be derived.

[0066]

[0067] The data recorded in the table are parameters related to the degree of beating during the pulping process. The degree of beating in the table is the same as the degree of beating mentioned in the text.

[0068] Train the initial pulping degree algorithm model;

[0069] The trained beating degree algorithm is then further optimized through a BP neural network self-learning algorithm to obtain an accurate soft measurement value SRr for beating degree.

[0070] The formula for a neural network is:

[0071]

[0072] y: Outputted; x: Outputted i : Input variables; n: Number of variables; w i θ: weights; θ: activation function

[0073] The activation function used is the ReLU function.

[0074] f(x) = max(0,x).

[0075] The PLC controls the grinding disc advance and retraction motors based on the calculated beating degree. When grinding with a constant beating degree, the soft measurement value SRr of the beating degree obtained by the edge box is compared with the set value. The deviation value Δ is calculated by the control unit through PID operation to obtain the adjustment amount, which adjusts the advance amount of the grinding disc advance and retraction motors. This adjusts the gap of the grinding disc blades and changes the beating degree of the pulp until Δ→0.

[0076] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0077] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A soft measurement and control method for pulping degree based on IoT edge computing, characterized in that, Includes the following steps: Acquire parameters related to the degree of beating during the pulping process and transmit the acquired parameters to the edge box; The relevant parameters obtained by the edge box are transmitted to the cloud server. The cloud server calculates the intermediate coefficients required for edge computing and transmits the intermediate coefficients to the edge box. The edge box calculates the soft measurement value SRr of the beating degree using the relevant parameters and intermediate coefficients obtained; The grinding disc advance and retraction motors are controlled based on the calculated beating degree to achieve constant beating degree grinding. The edge box integrates data fitting and neural network self-learning algorithms. The neural network formula is as follows: For output; For input variables; The number of variables; As weight; For activation functions; The activation function uses the ReLU function: The grinding disc advance and retraction motors are controlled according to the calculated beating degree to achieve constant beating degree grinding. When this is done, the soft measurement value SRr of the beating degree obtained from the edge box is compared with the set value, and the deviation value is... After calculation by the control unit, the adjustment amount is obtained, which adjusts the feed rate of the grinding disc advance and retraction motor, thereby adjusting the gap between the grinding discs and changing the beating degree of the slurry until... → up to 0.

2. The method for soft measurement and control of pulping degree based on IoT edge computing according to claim 1, characterized in that: The parameters related to beating degree during the refining process include pulp concentration, pulp flow rate, refining disc operating power, beating specific pressure, beating time, beating temperature, distance between refining discs, refining disc performance parameters, refining disc lifespan, wood chip properties, and pulp pH value.

3. The method for soft measurement and control of pulping degree based on IoT edge computing according to claim 2, characterized in that: The pulp concentration, pulp flow rate, grinding disc operating power, beating specific pressure, beating time, beating temperature, distance between grinding discs, and pulp pH value are obtained through calculation and instrument measurement; the characteristics of the grinding discs, the lifespan of the grinding discs, and the origin of the wood chips are fixed values, obtained through manual input.

4. The method for soft measurement and control of pulping degree based on IoT edge computing according to claim 1, characterized in that: The cloud server stores the performance parameters of grinding discs from different manufacturers, the lifespan of grinding discs, and the properties of wood chips from different regions. Based on the performance parameters of the grinding discs, the lifespan of the grinding discs, and the properties of the wood chips, an intermediate coefficient affecting the degree of beating can be calculated. Then, the calculated intermediate coefficient is sent back to the edge box to participate in the edge box's calculation of the soft measurement value SR of the degree of beating.

5. The method for soft measurement and control of pulping degree based on IoT edge computing according to claim 1, characterized in that: The calculation of the edge box includes the following steps: Enter the beating degree measured in the laboratory; The edge box derives an initial beating degree algorithm model based on laboratory measurements of beating degree using a data fitting algorithm. The formula for beating degree is: SR is the degree of beating, K is the coefficient, W is the active power of the disc mill, C is the concentration at the feed to the disc mill, F is the flow rate at the outlet of the disc mill, and T1 is the percentage of the disc mill's operating time to its total lifespan. Further analysis reveals a non-linear relationship between power and freeness, exhibiting a curvilinear relationship. Therefore, the freeness formula is: B is a constant; By fitting the data using this formula model, relevant parameters can be obtained, and thus an accurate model can be derived. Train the initial pulping degree algorithm model; The trained beating degree algorithm is then further optimized through a neural network self-learning algorithm to obtain an accurate soft measurement value SRr for beating degree.

Images