Method and apparatus for predicting density of limestone slurry, storage medium and electronic device

Abstract

The present application relates to the technical field of wet desulfurization, and relates to a method and apparatus for predicting the density of a limestone slurry, a storage medium, and an electronic device. The method comprises: first acquiring feature data related to a limestone slurry, wherein the feature data comprises a limestone slurry tank level, a limestone slurry pump current value, an instantaneous limestone slurry flow rate, and an instantaneous limestone powder flow rate; and then, on the basis of the feature data, using a particle swarm optimized support vector machine model to predict the density of the limestone slurry. By means of the technical solution of the present application, on the basis of feature data related to a limestone slurry, a particle swarm optimized support vector machine algorithm is used to predict the density of the limestone slurry in real time, thereby improving the accuracy of measuring the density of the limestone slurry, providing a basis for production, and further improving the desulfurization efficiency of a unit and the economy in a desulfurization process by controlling the density of the limestone slurry.

Description

Methods, apparatus, storage media, and electronic equipment for predicting the density of limestone slurry. Technical Field

[0001] This application relates to the field of wet desulfurization technology, specifically to a method, apparatus, storage medium, and electronic equipment for predicting the density of limestone slurry. Background Technology

[0002] In limestone wet desulfurization systems at thermal power plants, limestone slurry acts as an absorbent, reacting with sulfur dioxide in the flue gas to produce calcium sulfite or calcium sulfate, thereby achieving desulfurization. If the density of the limestone slurry is not properly controlled during desulfurization operation, it will severely impact the unit's desulfurization efficiency and the economic viability of the process.

[0003] Currently, the density of limestone slurry can be measured using methods such as nuclear radiation densitometers, differential pressure densitometers, mass densitometers, and tuning fork densitometers. However, these methods are easily affected by corrosion, suspended particles, and scaling in the operating environment, resulting in poor measurement accuracy. Summary of the Invention

[0004] In view of this, this application provides a method, apparatus, storage medium and electronic device for predicting the density of limestone slurry, the main purpose of which is to improve the technical problem that the existing limestone slurry density measurement method is easily affected by the environment, resulting in poor measurement accuracy.

[0005] In a first aspect, this application provides a method for predicting the density of limestone slurry, comprising:

[0006] Acquire characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current value, limestone slurry instantaneous flow rate, and limestone powder instantaneous flow rate;

[0007] Based on the aforementioned feature data, the density of the limestone slurry is predicted using a particle swarm optimization support vector machine (SVM) model.

[0008] Optionally, before acquiring the characteristic data related to the limestone slurry, the method further includes:

[0009] Pierce correlation analysis was used to determine the factors affecting the density of limestone slurry;

[0010] The acquisition of characteristic data related to limestone slurry includes:

[0011] Based on the factors affecting the density of limestone slurry, the characteristic data are retrieved from the Distributed Control System (DCS).

[0012] Optionally, the training process of the support vector machine model includes:

[0013] Multiple sets of characteristic data under different time periods and loads are retrieved from the DCS to construct the target dataset;

[0014] Based on the target dataset, the support vector machine model is obtained by training the model using the particle swarm optimization (PSO) algorithm.

[0015] Optionally, the step of training the support vector machine model using the particle swarm optimization algorithm based on the target dataset to obtain the model includes:

[0016] Based on the regularization parameter and kernel function parameter, the particle swarm optimization algorithm is used for iterative optimization to obtain the optimal regularization parameter and optimal kernel function parameter;

[0017] Based on the target dataset, the model is trained using the optimal regularization parameters and the optimal kernel function parameters to obtain the support vector machine model.

[0018] Optionally, the step of iteratively optimizing the regularization parameters and kernel function parameters using a particle swarm optimization algorithm to obtain the optimal regularization parameters and kernel function parameters includes:

[0019] Based on the regularization parameter and the kernel function parameter, the various parameters in the particle swarm optimization algorithm are initialized.

[0020] Based on the various parameters set in the initialization, the global optimal value of the particles in the particle swarm is obtained by iteratively updating them.

[0021] Based on the optimization parameters corresponding to the global optimum, the optimal regularization parameter and the optimal kernel function parameter are determined.

[0022] Optionally, the iterative update of various parameters based on the initial settings to obtain the global optimal value of the particles in the particle swarm includes:

[0023] In each iteration update, the parameters corresponding to the particle are evaluated by absolute residual, mean square error and / or model determination coefficient to obtain the fitness value of each particle;

[0024] Update the position and velocity of individual particles in the particle swarm, and calculate the updated fitness value;

[0025] The global optimum of the particle is updated based on the fitness value.

[0026] Optionally, predicting the density of the limestone slurry using a particle swarm optimization support vector machine model based on the feature data includes:

[0027] The feature data and the range corresponding to the feature data are input into the support vector machine model to obtain the density of the limestone slurry.

[0028] Secondly, this application provides a device for predicting the density of limestone slurry, comprising:

[0029] The acquisition module is configured to acquire characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current value, limestone slurry instantaneous flow rate, and limestone powder instantaneous flow rate.

[0030] The prediction module is configured to predict the density of the limestone slurry based on the feature data using a particle swarm optimization support vector machine model.

[0031] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0032] Fourthly, this application provides an electronic device including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect.

[0033] By employing the above technical solution, this application provides a method, apparatus, storage medium, and electronic device for predicting the density of limestone slurry. Specifically, it first acquires characteristic data related to the limestone slurry, including the limestone slurry tank level, limestone slurry pump current, instantaneous flow rate of limestone slurry, and instantaneous flow rate of limestone powder. Then, based on the characteristic data, it uses a particle swarm optimization support vector machine model to predict the density of the limestone slurry. Compared with existing technologies, by applying the technical solution of this application, based on the characteristic data related to limestone slurry, and using a particle swarm optimization support vector machine algorithm to predict the density of limestone slurry in real time, the accuracy of limestone slurry density measurement is improved, providing a basis for production. Furthermore, by controlling the density of limestone slurry, the desulfurization efficiency of the unit and the economy of the desulfurization process can be improved.

[0034] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0035] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 shows a flowchart illustrating a method for predicting the density of limestone slurry according to an embodiment of this application;

[0038] Figure 2 shows a flowchart illustrating another method for predicting the density of limestone slurry provided in an embodiment of this application;

[0039] Figure 3 shows a schematic diagram of an application example provided in an embodiment of this application;

[0040] Figure 4 shows a schematic diagram of a limestone slurry density prediction device provided in an embodiment of this application. Detailed Implementation

[0041] To better understand the above-mentioned objectives, features, and advantages of this application, the solution of this application will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0042] To address the problem that current limestone slurry density measurement methods are easily affected by environmental factors, leading to poor measurement accuracy, this embodiment provides a method for predicting limestone slurry density, as shown in Figure 1. The method includes:

[0043] Step 101: Obtain characteristic data related to limestone slurry.

[0044] The characteristic data include the limestone slurry tank level, the limestone slurry pump current, the instantaneous flow rate of the limestone slurry, and the instantaneous flow rate of the limestone powder.

[0045] In some cases, the level in a limestone slurry tank can affect the density of the limestone slurry. The level in a limestone slurry tank can be measured using methods such as ultrasonic level gauges, radar level gauges, and float level gauges. Ultrasonic level gauges measure the level by emitting and receiving ultrasonic signals and are suitable for corrosive and conductive liquids; radar level gauges measure the level by emitting and receiving radar waves and are suitable for high-temperature and high-pressure environments; float level gauges measure the level by the position of a float as the liquid level changes and are suitable for smaller ranges of level variation. If the slurry density increases, the weight of the same volume of slurry increases, and the level may decrease; if the slurry density decreases, the weight of the same volume of slurry decreases, and the level may rise.

[0046] In some examples, the current value of the limestone slurry pump is related to the density of the slurry being pumped; the higher the density, the greater the pump load and the higher the current value. The limestone slurry pump current passing through the motor can be measured using a current transformer installed on the motor cable, or the motor current value can be read directly using an ammeter. Abnormal changes in the current value can provide early warning of changes in slurry density, allowing for timely adjustment of the slurry density and preventing overload or damage to the limestone slurry pump.

[0047] In some examples, the instantaneous flow rate of limestone slurry affects its density; a higher flow rate generally results in a higher slurry density. The instantaneous flow rate can be calculated using an electromagnetic flowmeter by measuring the electromotive force generated when the fluid passes through a magnetic field, an ultrasonic flowmeter by measuring the propagation time of ultrasound waves in the fluid, or a vortex flowmeter by measuring the frequency generated by the fluid passing through a vortex. Abnormal changes in the instantaneous flow rate of limestone slurry may indicate unstable slurry density, requiring adjustments to the slurry preparation system. Too low a flow rate may lead to decreased desulfurization efficiency, while too high a flow rate may increase energy consumption.

[0048] In some examples, the instantaneous flow rate of limestone powder affects the density of the limestone slurry; a higher flow rate results in a higher slurry density. The instantaneous flow rate of limestone powder can be calculated using a mass flow meter by measuring the mass flow rate of the fluid; it can also be calculated using a rotor flow meter by measuring the rotational speed of the rotor; or it can be calculated using a screw feeder by measuring the rotational speed of the screw feeder. Abnormal changes in the limestone powder flow rate can affect the control of the slurry density, thus affecting desulfurization efficiency; excessively low flow rates may lead to poor desulfurization results.

[0049] Step 102: Based on the feature data, predict the density of limestone slurry using a particle swarm optimization support vector machine model.

[0050] In some examples, the feature data can first be preprocessed, including normalizing all feature data to eliminate the influence of dimensions, filling or deleting missing values, and identifying and handling outliers to ensure data accuracy and thus effectively predict the density of limestone slurry. This embodiment combines the global search capability of the particle swarm optimization algorithm with the powerful learning capability of the support vector machine model, which can improve the model's prediction accuracy and generalization ability. Deploying the support vector machine model trained using the particle swarm optimization algorithm into the actual system allows for real-time prediction of limestone slurry density. This facilitates timely adjustment of the slurry preparation system parameters based on the prediction results, maintaining the limestone slurry density within the optimal range.

[0051] This embodiment first acquires characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current, instantaneous flow rate of limestone slurry, and instantaneous flow rate of limestone powder. Then, based on this characteristic data, a particle swarm optimization (PSO) support vector machine model is used to predict the density of the limestone slurry. Due to the influence of the working environment, existing limestone slurry density measurement methods have limited lifespans: mass density meters have a service life of less than one year; differential pressure density meters have a long service life but cannot perform continuous measurements and have many equipment defects; and nuclear density meters have many limitations. Compared with current technologies, this embodiment uses a particle swarm optimization (PSO) support vector machine algorithm based on limestone slurry-related characteristic data to predict limestone slurry density in real time, improving the accuracy of limestone slurry density measurement, providing a basis for production, and further enhancing the unit's desulfurization efficiency and the economy of the desulfurization process by controlling the limestone slurry density.

[0052] Furthermore, as a refinement and extension of the above embodiments, and to fully illustrate the specific implementation process of the method in this embodiment, this embodiment provides a specific method as shown in Figure 2, which includes:

[0053] Step 201: Obtain characteristic data related to limestone slurry.

[0054] The characteristic data include the limestone slurry tank level, the limestone slurry pump current, the instantaneous flow rate of the limestone slurry, and the instantaneous flow rate of the limestone powder.

[0055] For example, a DCS (Distributed Control System) is a control system widely used in industrial automation. It manages complex production processes through distributed controllers and centralized monitoring systems. Various process parameters, such as temperature, pressure, flow rate, and liquid level, can be collected in real time using sensors and measuring devices. This data is transmitted to a central control room for operator monitoring and analysis, ensuring real-time data accuracy and enabling timely detection and handling of anomalies. This embodiment can retrieve characteristic data related to the current limestone slurry from the DCS. Specifically, instruments such as level gauges, ammeters, and flow meters used for collecting characteristic data can be connected to a data acquisition system (such as a DCS), and a sampling frequency can be set, for example, collecting data once per second or per minute. The collected characteristic data is stored in a database or file system for subsequent analysis. This process can also remove outliers and noise to ensure data accuracy and reliability. Furthermore, statistical analysis and machine learning methods can be used to uncover patterns and trends in the data and optimize the operating system.

[0056] Optionally, the method in this embodiment may further include: determining the factors affecting the density of limestone slurry through Pierce correlation analysis; correspondingly, step 201 may specifically include: retrieving feature data from the distributed control system (DCS) based on the factors affecting the density of limestone slurry.

[0057] In some examples, Pearson Correlation Analysis can be used in MATLAB to calculate the factors affecting the density of limestone slurry, i.e., to find the set of independent variables that have the strongest explanatory power for the dependent variable. The factors affecting the density of limestone slurry and their ranges, calculated through correlation analysis, are shown in Table 1.

[0058] Table 1

[0059] Step 202: Input the feature data and the corresponding range of the feature data into the support vector machine model to obtain the density of the limestone slurry.

[0060] In some examples, the support vector machine model is a classification and regression tool.

[0061] Optionally, the training process of the support vector machine model may include: retrieving multiple sets of feature data from the DCS under different time periods and different loads to construct a target dataset; and training the model using the particle swarm optimization algorithm based on the target dataset to obtain the support vector machine model.

[0062] For example, 60-100 sets of data can be retrieved from the desulfurization control DCS system under different time periods and loads, including the values ​​of four variables (limestone slurry tank level, limestone slurry pump current, limestone slurry instantaneous flow rate, and limestone powder instantaneous flow rate) and their corresponding limestone slurry densities. The mathematical relationship between these four variables and the limestone slurry density is modeled using the libsvm toolbox in MATLAB. A particle swarm optimization algorithm is then used, treating each particle as a potential solution and iteratively updating its position and velocity to approximate the optimal solution. Deploying the trained model to the actual system allows for real-time prediction of limestone slurry density, improving prediction accuracy and helping to optimize the operation of the desulfurization system.

[0063] Optionally, the above-mentioned model training based on the target dataset using the particle swarm optimization algorithm to obtain the support vector machine model may specifically include: using the particle swarm optimization algorithm to iteratively optimize the regularization parameters and kernel function parameters according to the regularization parameters and kernel function parameters to obtain the optimal regularization parameters and optimal kernel function parameters; and training the model based on the optimal regularization parameters and optimal kernel function parameters according to the target dataset to obtain the support vector machine model.

[0064] In some examples, commonly used kernel functions for Support Vector Machine (SVM) models include linear kernels, polynomial kernels, and RBF kernels. The main parameters include a penalty parameter C that controls the model's fit to the training data and a kernel function parameter σ that controls the smoothness of the decision boundary. The model's performance largely depends on the choice of these parameters, which directly affect the model's complexity and generalization ability. Inappropriate parameter settings can lead to overfitting or underfitting, thus affecting prediction accuracy. Therefore, it is necessary to use the PSO toolbox algorithm in the MATLAB environment to optimize the parameters (C, σ). By using the Particle Swarm Optimization (PSO) algorithm to optimize these parameters, the model's prediction accuracy and generalization ability can be significantly improved. The automated parameter optimization process not only reduces the time and effort of manual trial and error but also automatically finds the optimal parameter combination, improving model performance.

[0065] Optionally, the above-mentioned optimization using a particle swarm optimization algorithm based on regularization parameters and kernel function parameters can be performed iteratively to obtain the optimal regularization parameters and optimal kernel function parameters. Specifically, this may include: initializing various parameters in the particle swarm optimization algorithm based on regularization parameters and kernel function parameters; iteratively updating various parameters based on the initial settings to obtain the global optimal value of particles in the particle swarm; and determining the optimal regularization parameters and optimal kernel function parameters based on the optimization parameters corresponding to the global optimal value.

[0066] In some examples, as shown in Figure 3, a set of initial solutions (particles) is randomly generated, with each particle representing a set of SVM parameters. The particle's velocity and position are initialized. For each particle, the SVM model is trained using the corresponding parameters, updating the individual optimal position and global optimal position of each particle. The particle's velocity and position are then updated based on the global optimal position and individual optimal position. These steps are repeated until the maximum number of iterations is reached or the convergence condition is met. That is, by iteratively updating the particle's position and velocity, the optimal parameter combination for model performance is found. The optimized parameters corresponding to the global optimal value are then the optimal parameters, including the optimal regularization parameter and the optimal kernel function parameter.

[0067] Optionally, the above-mentioned iterative updates of various parameters based on the initial settings to obtain the global optimal value of the particles in the particle swarm include: in each iteration update, evaluating the parameters corresponding to the particles through absolute residuals, mean squared error (MSE), and / or model coefficient of determination to obtain the fitness value of each particle; updating the position and velocity of individual particles in the particle swarm and calculating the updated fitness value; and updating the global optimal value of the particles based on the fitness value.

[0068] For example, as shown in Figure 3, the fitness function value of a particle is calculated. If the current fitness function value is better than the individual optimal value, the individual optimal value and position are updated; if the current fitness function value is better than the global optimal value, the global optimal value and position are updated. The fitness function can be used to evaluate the performance of a support vector machine model. Particle swarm optimization support vector machine (PSO-SVM) models typically use the absolute residual δ, mean squared error (MSE), and / or model determination coefficient R. 2 As a criterion, the particle's velocity and position are updated based on its current position, the individual's optimal position, and the person's optimal position.

[0069] In some examples, the results of inputting limestone slurry-related feature data into the original support vector machine model are: Mean squared error = 0.135792 (regression), Squared correlation coefficient = 0.434852 (regression); Mean squared error = 0.565684 (regression), Squared correlation coefficient = 0.00622 (regression). The results of inputting the same feature data into the particle swarm optimization-optimized support vector machine model are: Mean squared error = 0.00323137 (regression), Squared correlation coefficient = 0.490683 (regression); Mean squared error = 0.00391151 (regression); Squared correlation coefficient = 0.288028 (regression). By comparing the results, it is clear that the results of the particle swarm optimization are superior to those of the original support vector machine, demonstrating the effectiveness of the particle swarm optimization-optimized support vector machine model.

[0070] Compared with existing technologies, this embodiment retrieves feature data highly correlated with limestone slurry density from the DCS system. In the MATLAB environment, a mathematical model is constructed using a support vector machine algorithm to establish the relationship between the feature data and the limestone slurry density, enabling real-time prediction of the limestone slurry density. This serves as a supplement and correction to existing limestone slurry density measurement methods. This method avoids the problems of existing measurement methods, such as significant susceptibility to environmental influences, numerous equipment defects, high spare parts consumption, and long measurement cycles. It improves the accuracy of limestone slurry density measurement, providing a basis for production. Furthermore, by controlling the limestone slurry density, the desulfurization efficiency of the unit and the economic efficiency of the desulfurization process can be improved.

[0071] Furthermore, as a specific implementation of the method shown in Figure 1, this embodiment provides a limestone slurry density prediction device, as shown in Figure 4. The device includes: an acquisition module 31 and a prediction module 32.

[0072] The acquisition module 31 is configured to acquire characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current value, limestone slurry instantaneous flow rate and limestone powder instantaneous flow rate.

[0073] The prediction module 32 is configured to predict the density of the limestone slurry based on the feature data using a particle swarm optimization support vector machine model.

[0074] In some examples, factors affecting the density of limestone slurry are determined through Pierce correlation analysis; accordingly, the acquisition module 31 is specifically configured to retrieve the feature data from the DCS based on the factors affecting the density of limestone slurry.

[0075] In some examples, prediction module 32 is specifically configured to retrieve multiple sets of feature data from the DCS under different time periods and different loads to construct a target dataset; based on the target dataset, the model is trained using the particle swarm optimization algorithm to obtain the support vector machine model.

[0076] In some examples, prediction module 32 is further configured to perform iterative optimization using a particle swarm optimization algorithm based on regularization parameters and kernel function parameters to obtain optimal regularization parameters and optimal kernel function parameters; and to train the model based on the target dataset and the optimal regularization parameters and optimal kernel function parameters to obtain the support vector machine model.

[0077] In some examples, the prediction module 32 is further configured to initialize various parameters in the particle swarm optimization algorithm according to the regularization parameter and the kernel function parameter; iteratively update the various parameters based on the initialized settings to obtain the global optimal value of the particles in the particle swarm; and determine the optimal regularization parameter and the optimal kernel function parameter according to the optimization parameters corresponding to the global optimal value.

[0078] In some examples, the prediction module 32 is specifically configured to evaluate the parameters corresponding to the particle by absolute residuals, mean squared errors and / or model determination coefficients in each iteration update to obtain the fitness value of each particle; update the position and velocity of individual particles in the particle swarm and calculate the updated fitness value; and update the global optimum value of the particle based on the fitness value.

[0079] In some examples, the prediction module 32 is further configured to input the feature data and the range corresponding to the feature data into the support vector machine model to obtain the density of the limestone slurry.

[0080] Based on the methods shown in Figures 1 and 2, this embodiment also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the methods shown in Figures 1 and 2.

[0081] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0082] Based on the methods shown in Figures 1 and 2, and the virtual device embodiment shown in Figure 4, in order to achieve the above objectives, this application embodiment also provides an electronic device, which may include a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to implement the methods shown in Figures 1 and 2.

[0083] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0084] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0085] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0086] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented using software plus necessary general-purpose hardware platforms, or it can be implemented in hardware. By applying the solution of this embodiment, the problems of existing measurement methods being greatly affected by the usage environment, having many equipment defects, high spare parts consumption, and long measurement cycles are avoided. Based on the characteristic data related to limestone slurry, the support vector machine algorithm with particle swarm optimization is used to predict the density of limestone slurry in real time, which improves the accuracy of measuring the density of limestone slurry, provides a basis for production, and further improves the desulfurization efficiency of the unit and the economy of the desulfurization process by controlling the density of limestone slurry.

[0087] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0088] The above are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method of predicting the density of a limestone slurry, characterized by, include: Acquire characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current value, limestone slurry instantaneous flow rate, and limestone powder instantaneous flow rate; Based on the aforementioned feature data, the density of the limestone slurry is predicted using a particle swarm optimization support vector machine model.

2. The method of claim 1, wherein, Before acquiring characteristic data related to limestone slurry, the method further includes: Pierce correlation analysis was used to determine the factors affecting the density of limestone slurry; The acquisition of characteristic data related to limestone slurry includes: Based on the factors affecting the density of limestone slurry, the characteristic data is retrieved from the distributed control system (DCS).

3. The method of claim 1, wherein, The training process of the support vector machine model includes: Multiple sets of characteristic data under different time periods and loads are retrieved from the DCS to construct the target dataset; Based on the target dataset, the support vector machine model is obtained by training the model using the particle swarm optimization algorithm.

4. The method of claim 3, wherein, The step of training the support vector machine model using the particle swarm optimization algorithm based on the target dataset to obtain the model includes: Based on the regularization parameter and kernel function parameter, the particle swarm optimization algorithm is used for iterative optimization to obtain the optimal regularization parameter and optimal kernel function parameter; Based on the target dataset, the model is trained using the optimal regularization parameters and the optimal kernel function parameters to obtain the support vector machine model.

5. The method of claim 4, wherein, The step of iteratively optimizing the regularization parameters and kernel function parameters using a particle swarm optimization algorithm to obtain the optimal regularization parameters and optimal kernel function parameters includes: Based on the regularization parameter and the kernel function parameter, the various parameters in the particle swarm optimization algorithm are initialized. Based on the various parameters set in the initialization, the global optimal value of the particles in the particle swarm is obtained by iteratively updating them. Based on the optimization parameters corresponding to the global optimum, the optimal regularization parameter and the optimal kernel function parameter are determined.

6. The method of claim 5, wherein, The iterative update of various parameters based on the initial settings to obtain the global optimal value of the particles in the particle swarm includes: In each iteration update, the parameters corresponding to the particle are evaluated by absolute residual, mean square error and / or model determination coefficient to obtain the fitness value of each particle; Update the position and velocity of individual particles in the particle swarm, and calculate the updated fitness value; The global optimum of the particle is updated based on the fitness value.

7. The method of claim 1, wherein, The step of predicting the density of the limestone slurry using a particle swarm optimization support vector machine model based on the feature data includes: The feature data and the range corresponding to the feature data are input into the support vector machine model to obtain the density of the limestone slurry.

8. A device for predicting the density of a limestone slurry, characterized in that, include: The acquisition module is configured to acquire characteristic data related to limestone slurry, including limestone slurry tank level, limestone slurry pump current value, limestone slurry instantaneous flow rate, and limestone powder instantaneous flow rate. The prediction module is configured to predict the density of the limestone slurry based on the feature data using a particle swarm optimization support vector machine model.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, The processor implements the method in any one of claims 1 to 7 when executing the computer program.

Images