Prediction control method, device, equipment and storage medium for rice huller

Abstract

The present invention discloses a prediction control method, device, equipment and storage medium for a rice huller. The method includes: acquiring the operation data of the rice huller equipment and the output data after hulling; determining a predicted value of the hulling completion rate according to the hulling completion rate prediction model and the operation data; determining a control quantity according to the predicted value of the hulling completion rate and the output data, and controlling the rice huller equipment based on the control quantity. Since the present invention acquires the operation data of the rice huller equipment and the output data after hulling, then predicts the hulling completion rate under the current operation data according to the hulling completion rate prediction model, and then determines the control quantity according to the predicted value and the actual output data, and controls the rice huller equipment according to the control quantity, it can adaptively adjust the control parameters of the rice huller when the key processing parameters of rice hulling change, so that the hulling effect of the rice huller equipment is better.

Description

Technical Field

[0001] The present invention relates to the technical field of paddy hulling, and in particular, to a predictive control method, device, equipment and storage medium for a hulling machine. Background Art

[0002] As the largest food crop in China, paddy has the largest planting area and output among food crops. At the same time, the processed product of paddy, rice, is also the main variety in the production and consumption of food in China. As the main subsequent processing industry for agricultural products to face the market, paddy processing has an important production and economic status in the national economy and plays a crucial role in promoting the industrialization of agricultural products.

[0003] The paddy processing process generally includes: paddy cleaning and grading, hulling and shelling, separation of the under-runner (rice husk, paddy and brown rice), milling and finished product finishing, etc. In these processing technological processes, hulling and shelling is an extremely important key process, which not only affects the output of subsequent processes, but also directly affects the output rate, output and rice quality of the finished product. At present, paddy hulling uses a hulling machine, among which the rubber roll hulling machine is the key equipment most widely used in the hulling process. The rubber roll hulling machine has good characteristics such as high output, high shelling rate and low breakage rate. The main working parameters affecting the performance of the rubber roll hulling machine are: roll pressure, fast roll linear speed, fast and slow roll linear speed difference and flow rate. These parameters directly affect the shelling and output performance of the hulling machine. The performance evaluation indexes generally include shelling rate, breakage rate, rubber consumption and output, etc.

[0004] At present, for the hulling control system, the degree of automation of domestic hulling equipment is relatively low. The existing hulling equipment cannot dynamically adjust multiple key processing parameters of hulling and shelling, and cannot perform real-time monitoring and detection. At the same time, the loss of the roller rubber of the rubber roll hulling machine during hulling and shelling will cause a change in the radius of the rubber roll, resulting in a change in the key processing parameters, so that the shelling effect of the hulling equipment becomes worse and cannot be adaptively adjusted and controlled according to the change of equipment parameters. Summary of the Invention

[0005] The main object of the present invention is to provide a predictive control method, device, equipment and storage medium for a hulling machine, aiming to solve the technical problem that the existing hulling equipment cannot be adaptively adjusted and controlled according to the change of key processing parameters of hulling and shelling, resulting in poor shelling effect of the hulling equipment.

[0006] To achieve the above object, the present invention provides a predictive control method for a hulling machine, and the method includes the following steps:

[0007] Obtain the operation data of the hulling machine equipment and the output data after shelling;

[0008] Determine the predicted value of the shelling integrity rate according to the shelling integrity rate prediction model and the operation data;

[0009] Determine a control quantity based on the predicted hulling integrity rate value and the output data, and control the hulling machine equipment based on the control quantity.

[0010] Optionally, before the step of determining the predicted hulling integrity rate value according to the hulling integrity rate prediction model and the operation data, the method further includes:

[0011] Obtain the historical operation data and historical output data of the hulling machine equipment;

[0012] Determine the historical hulling integrity rate according to the historical output data;

[0013] Generate a hulling integrity rate training set according to the historical hulling integrity rate, and train a support vector machine regression model according to the hulling integrity rate training set to obtain a hulling integrity rate prediction model.

[0014] Optionally, the determining the historical hulling integrity rate according to the historical output data includes:

[0015] Set weight values for the hulling rate and the breakage rate in the historical output data respectively;

[0016] Determine the historical hulling integrity rate according to the hulling rate, the breakage rate and the weight values.

[0017] Optionally, the generating a hulling integrity rate training set according to the historical hulling integrity rate, and training a support vector machine regression model according to the hulling integrity rate training set to obtain a hulling integrity rate prediction model includes:

[0018] Generate a hulling integrity rate training set according to the historical hulling integrity rate, and train a support vector machine regression model according to the hulling integrity rate training set to obtain a hulling integrity rate support vector machine regression model;

[0019] Perform linear regression processing on the hulling integrity rate support vector machine regression model to obtain a hulling integrity rate prediction model.

[0020] Optionally, the performing linear regression processing on the hulling integrity rate support vector machine regression model to obtain a hulling integrity rate prediction model includes:

[0021] Perform linear regression processing on the hulling integrity rate support vector machine regression model to obtain a hulling integrity rate linear regression model;

[0022] Construct a hyperplane model according to the hulling integrity rate linear regression model, the historical hulling integrity rate and the operation data;

[0023] Solve for the optimal hyperplane of the hyperplane model to obtain a hulling integrity rate prediction model.

[0024] Optionally, performing optimal hyperplane solution on the hyperplane model to obtain a hulling completeness rate prediction model, including:

[0025] Constructing an input sample vector according to the operation data and the historical operation data;

[0026] Constructing an output sample vector according to the historical hulling completeness rate and the real-time hulling completeness rate;

[0027] Training the hyperplane model based on the input sample vector and the output sample vector to obtain a hulling completeness rate prediction model.

[0028] Optionally, determining a control quantity according to the hulling completeness rate prediction value and the output data, and controlling the husker equipment based on the control quantity, including:

[0029] Determining the actual hulling completeness rate according to the output data;

[0030] Determining a feedback control quantity according to the hulling completeness rate prediction value and the actual hulling completeness rate;

[0031] Determining a control increment according to the feedback control quantity and the optimal control parameters;

[0032] Controlling the husker equipment based on the control increment.

[0033] In addition, to achieve the above object, the present invention further provides a husker predictive control device, and the husker predictive control device includes:

[0034] A data acquisition module, configured to acquire the operation data of the husker equipment and the output data after hulling;

[0035] A data prediction module, configured to determine a hulling completeness rate prediction value according to the hulling completeness rate prediction model and the operation data;

[0036] An equipment control module, configured to determine a control quantity according to the hulling completeness rate prediction value and the output data, and control the husker equipment based on the control quantity.

[0037] In addition, to achieve the above object, the present invention further provides a husker predictive control device, and the device includes: a memory, a processor, and a husker predictive control program stored on the memory and executable on the processor, and the husker predictive control program is configured to implement the steps of the husker predictive control method as described above.

[0038] In addition, to achieve the above object, the present invention also provides a storage medium, on which a husker prediction control program is stored. When the husker prediction control program is executed by a processor, the steps of the husker prediction control method as described above are implemented.

[0039] In the present invention, it is disclosed to obtain the operation data of the husker equipment and the output data after shelling; determine the predicted value of the shelling integrity rate according to the shelling integrity rate prediction model and the operation data; determine the control quantity according to the predicted value of the shelling integrity rate and the output data, and control the husker equipment based on the control quantity. Since the present invention obtains the operation data of the husker equipment and the output data after shelling, then predicts the shelling integrity rate under the current operation data according to the shelling integrity rate prediction model, and then determines the control quantity according to the predicted value and the actual output data, and controls the husker equipment according to the control quantity, so that it can adaptively adjust the control parameters of the husker when the key processing parameters of husking and shelling change, and make the shelling effect of the husker equipment better. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic structural diagram of a husker prediction control device in the hardware operating environment related to the embodiment solution of the present invention;

[0041] Figure 2 It is a schematic flowchart of the first embodiment of the husker prediction control method of the present invention;

[0042] Figure 3 It is a schematic flowchart of the second embodiment of the husker prediction control method of the present invention;

[0043] Figure 4 It is a system structure diagram of the husker prediction control method of the present invention;

[0044] Figure 5 It is a prediction control structure diagram based on SVM in the husker prediction control method of the present invention;

[0045] Figure 6 It is a structural block diagram of the first embodiment of the husker prediction control device of the present invention.

[0046] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0048] Refer to Figure 1 , Figure 1 It is a schematic structural diagram of a husker prediction control device in the hardware operating environment related to the embodiment solution of the present invention.

[0049] As Figure 1 shown, the rice huller predictive control device may include: a processor 1001, such as a Central Processing Unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen and an input unit such as a keyboard. Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wireless Fidelity (WI-FI) interface). The memory 1005 may be a high-speed Random Access Memory (RAM) or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0050] Those skilled in the art can understand that Figure 1 the structure shown in does not constitute a limitation on the rice huller predictive control device, and it may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0051] As Figure 1 shown, the memory 1005, as a storage medium, may include an operating system, a network communication module, a user interface module, and a rice huller predictive control program.

[0052] In Figure 1 the rice huller predictive control device shown, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the rice huller predictive control device of the present invention may be arranged in the rice huller predictive control device. The rice huller predictive control device calls the rice huller predictive control program stored in the memory 1005 through the processor 1001 and executes the rice huller predictive control method provided by the embodiments of the present invention.

[0053] The embodiments of the present invention provide a rice huller predictive control method. Referring to Figure 2 , Figure 2 is a schematic flowchart of the first embodiment of the rice huller predictive control method of the present invention.

[0054] In this embodiment, the rice huller predictive control method includes the following steps:

[0055] Step S10: Obtain the operating data of the rice hulling machine equipment and the output data after hulling.

[0056] It should be noted that the execution subject of the method in this embodiment can be a rice hulling machine control device with functions of predictive control, network communication, and program operation; it can also be a control device with the same or similar functions, or a rice hulling machine equipped with this control device. This embodiment and the following embodiments will be described by taking the rice hulling machine control device as an example.

[0057] It can be understood that currently, the degree of automation of rice hulling equipment is relatively low. Existing rice hulling equipment cannot dynamically adjust the key processing parameters of rice hulling, and cannot perform real-time monitoring and detection. At the same time, the loss of the roller rubber of the rubber roll rice hulling machine during rice hulling will cause changes in the radius of the rubber roll, resulting in changes in key processing parameters, causing the hulling effect of the rice hulling equipment to deteriorate and unable to adaptively adjust and control with changes in equipment parameters. In addition, the rice hulling machine control system has characteristics such as high nonlinearity, time-varying, and coupling, and is a typical complex industrial process. Therefore, adopting advanced intelligent control technology for the rice hulling machine control system can increase the output of rice hulling and improve the hulling quality.

[0058] It should be understood that the operating data of the rice hulling machine equipment (such as, inter-roll pressure, fast roll linear speed, feeding flow rate, linear speed difference, moisture content, etc.) and the output data after hulling (such as, hulling rate, breakage rate, output, and degumming, etc.) can be collected through sensors, and then these data can be recorded on the storage medium in real time.

[0059] Step S20: Determine the predicted value of the hulling integrity rate according to the hulling integrity rate prediction model and the operating data.

[0060] It can be understood that the output of the rice hulling machine can be predicted based on the pre-set hulling integrity rate prediction model and the operating data of the rice hulling machine equipment, and the hulling integrity rate in the output situation of the rice hulling machine can be predicted, so as to adaptively adjust the parameters of the rice hulling machine equipment.

[0061] Step S30: Determine the control quantity according to the predicted value of the hulling integrity rate and the output data, and control the rice hulling machine equipment based on the control quantity.

[0062] It should be understood that since there may be a deviation between the predicted hulling integrity rate based on the operating data and the actual output data, the control quantity of each parameter of the rice hulling machine equipment can be determined according to the existing deviation, and then the rice hulling machine equipment can be controlled based on the control quantity, so that the hulling effect of the rice hulling equipment can be better.

[0063] Further, since there is a deviation between the predicted husking integrity rate and the actual husking integrity rate, the control quantity can be determined according to the ideal input parameter values of the rice huller and the above deviation, and the air velocity, flow rate, pressure, etc. of the rice huller can be controlled according to the control quantity, so that the rice huller reaches the optimal state of the husking integrity rate. Therefore, the step S30 includes: determining the actual husking integrity rate according to the output data; determining the feedback control quantity according to the predicted value of the husking integrity rate and the actual husking integrity rate; determining the control increment according to the feedback control quantity and the optimal control parameters; and controlling the rice huller equipment based on the control increment.

[0064] It should be understood that the actual husking integrity rate can be determined according to the husking rate and the breakage rate in the output data of the rice huller, and the actual operating parameters can be obtained. The predicted value of the husking integrity rate is compared with the actual husking integrity rate to obtain a deviation, and the deviation is added to the predicted value of the husking integrity rate as the feedback control quantity for feedback control. Then, according to the optimal control parameters (ideal input values) and the feedback control quantity, the control increment of the control parameters is obtained, and the control increment is transmitted to the rice huller controller (PLC, industrial computer, etc.) through the data interface to adjust the operating states (parameters such as speed, pressure, and flow rate) of each device of the rice huller, so that the husking integrity rate reaches the ideal value.

[0065] In this embodiment, the operating data of the rice huller equipment and the output data after husking are disclosed; the predicted value of the husking integrity rate is determined according to the husking integrity rate prediction model and the operating data; the control quantity is determined according to the predicted value of the husking integrity rate and the output data, and the rice huller equipment is controlled based on the control quantity. Since this embodiment obtains the operating data of the rice huller equipment and the output data after husking, then predicts the husking integrity rate under the current operating data according to the husking integrity rate prediction model, and then determines the control quantity according to the predicted value and the actual output data, and controls the rice huller equipment according to the control quantity, so that when the key processing parameters of rice hulling and husking change, the control parameters of the rice huller can be adaptively adjusted, and the husking effect of the rice huller equipment is better.

[0066] Reference Figure 3 , Figure 3 is a schematic flowchart of the second embodiment of the predictive control method for the rice huller of the present invention.

[0067] Further, in the predictive control of the rice huller, the historical input data and output data of the rice huller can be obtained. For the key parameters with different attributes and their associated data groups, the prediction model can adopt machine learning methods, and support vector machines can be used. Support vector machines can obtain optimized values whether the samples are few or rich. Using support vector machines, the future output results of the system can be accurately predicted. Therefore, based on the above first embodiment, in this embodiment, the step S20 includes:

[0068] Step S201: Acquire historical operation data and historical output data of the rice huller equipment.

[0069] It is understandable that when predicting and adjusting the rice huller equipment, the real-time operating parameters and output parameters of the rice huller can be obtained, and the historical operating data and historical output data can also be obtained by obtaining the historical test data of the specific rice huller.

[0070] Step S202: determining a historical de-shelling completeness rate based on the historical output data.

[0071] Step S203: generating a deshelling completeness rate training set according to the historical deshelling completeness rate, and training a support vector machine regression model according to the deshelling completeness rate training set to obtain a deshelling completeness rate prediction model.

[0072] It's understandable that Generalized Predictive Control (GPC), a computer control method developed alongside research on adaptive control, has been successfully applied to industrial process control. Support Vector Machine (SVM) regression, based on structural risk minimization, effectively addresses issues such as small sample sizes, nonlinearity, high dimensionality, and local minima, demonstrating strong generalization capabilities.

[0073] It should be noted that the shelling completion rate prediction model can be a pre-configured prediction model based on a support vector machine model, and the output of this model can be a parameter. Since the output data after shelling has multiple parameters, a single parameter can be constructed that significantly reflects the output results after shelling. Previous rice hulling machine shelling trials have shown a strong positive correlation between shelling rate and yield: a high shelling rate indicates a high yield, and vice versa. Therefore, the shelling rate is selected as one of the output parameters here; another parameter is the breakage rate, which reflects the shelling quality. Therefore, the historical shelling completion rate can be determined based on the shelling rate and breakage rate.

[0074] It should be understood that a deshelling completeness rate training set can be generated based on historical deshelling completeness rates, and the deshelling completeness rate training set can be divided into different sub-training sets; the support vector machine model is trained in parallel based on the sub-training sets to obtain a deshelling completeness rate prediction model.

[0075] Furthermore, since there are multiple parameters in the output after hulling, while the output of the regression model of the support vector machine is one parameter, a parameter that can significantly reflect the output result after hulling can be constructed. Also, due to the high hulling rate and large output, the breakage rate reflects the hulling quality. Therefore, the output parameter can be constructed based on the hulling rate and the breakage rate, which reflects both the output and the quality. Therefore, step S202 includes: setting weight values for the hulling rate and the breakage rate in the historical output data respectively; determining the historical hulling integrity rate according to the hulling rate, the breakage rate, and the weight values.

[0076] It can be understood that, in order to construct the output value of a model, weight values can be set for the hulling rate and the breakage rate (for example, each taking 0.5 to obtain the hulling output value), which reflects both the output and the quality. Since the hulling rate is generally much larger than the breakage rate, the information of the breakage rate will be submerged directly, so the integrity rate is used for calculation, that is, the hulling integrity rate in the present invention can be calculated by the following formula:

[0077] y = x1y1 + x2(1 - y2),

[0078] In the formula, x1 is the weight of the hulling rate, y1 is the hulling rate, x2 is the weight of the breakage rate, and y2 is the breakage rate.

[0079] Furthermore, since the hulling process of the rice huller is a complex industrial process with strong non - linear characteristics, the non - linear support vector machine model can be linearly regressed in this space to improve the effectiveness of model prediction. Therefore, step S203 includes: generating a hulling integrity rate training set according to the historical hulling integrity rate, and training the support vector machine regression model according to the hulling integrity rate training set to obtain a hulling integrity rate support vector machine regression model; performing linear regression processing on the hulling integrity rate support vector machine regression model to obtain a hulling integrity rate prediction model.

[0080] It can be understood that for the regression of the non - linear support vector machine, the data can be mapped to a high - dimensional feature space through a non - linear mapping, and then linear regression is performed in the high - dimensional feature space.

[0081] Further, a hyperplane can be constructed based on historical operation data and real-time operation data of the rice hulling machine. By solving the hyperplane, the optimal hyperplane is used as the prediction model, which can improve the accuracy of model prediction. At the same time, the real-time data generated by the rice hulling machine is added to the historical data to online correct the model, realize automatic compensation and adaptive parameter changes, and realize the intelligence of the rice hulling machine control system. Therefore, the step S203 further includes: performing linear regression processing on the hulling integrity rate support vector machine regression model to obtain a hulling integrity rate linear regression model; constructing a hyperplane model according to the hulling integrity rate linear regression model, the historical hulling integrity rate, and the operation data; solving the optimal hyperplane for the hyperplane model to obtain a hulling integrity rate prediction model.

[0082] It should be noted that a hyperplane model can be constructed according to the hulling integrity rate linear regression model, the historical hulling integrity rate, and the operation data, and it can be constructed according to the following formula:

[0083] w T x - b = 0,

[0084] In the formula, w is the support vector weight, w T is the transpose of the vector weight, x is the support vector on the hyperplane, and b is the historical hulling integrity rate on the hyperplane.

[0085] It can be understood that the specific method for performing linear regression on the model can be achieved through the kernel function to obtain the global optimal solution through quadratic programming. Through derivation, the regression function can be obtained:

[0086]

[0087] In the formula, The sample data corresponding to when it is not equal to zero is the support vector. When b takes the points on the boundary, the average value can be taken.

[0088] Therefore, the support vector model of the rice hulling machine control system is:

[0089]

[0090] In the formula, X i is the support vector, and l is the number of support vectors.

[0091] In specific implementation, the hyperplane model can be solved by the Sequential Minimal Optimization (SMO) algorithm to find the optimal hyperplane and obtain the hulling integrity rate prediction model.

[0092] Furthermore, by training the model with a large amount of operational, historical test data, and output data, and using it as a predictive model, the accuracy of the model's predictions can be improved. Simultaneously, by incorporating real-time data generated by the huller into the historical data, the model can be corrected online, achieving automatic compensation and adaptive parameter changes, thereby realizing an intelligent huller control system. Therefore, step S203 further includes: constructing an input sample vector based on the operational data and the historical operational data; constructing an output sample vector based on the historical and real-time hulling completion rates; and training the hyperplane model based on the input and output sample vectors to obtain a hulling completion rate prediction model.

[0093] It should be understood that the hulling completeness rate is used as the output parameter of the rice huller, so the output y(k) vector of the model can be obtained from the collected or tested hulling completeness rate data. The operating parameters of the rice huller are used as input, so the input x(k) of the model is a vector composed of the collected or tested operating parameters of the rice huller (roller pressure, fast roller speed, feed flow, speed difference, moisture content, etc.). These vectors are used as training samples of the support vector machine, and after training, the support vector X is obtained. i .

[0094] The output is predicted based on the obtained support vector machine model, and the system correction is achieved by feeding back the error between the actual output and the predicted value. The predictive control system structure can be referred to Figure 4 and Figure 5 , Figure 4 This is a system structure diagram of the rice huller predictive control method of the present invention. Figure 5 This is a diagram of the predictive control structure based on SVM in the predictive control method for the rice huller of the present invention. The present invention can also perform rolling optimization on the predictive model. That is, when training and solving the model, not only the historical test data of the rice huller is used, but also the real-time operation data of the rice huller is integrated into the historical test data to optimize the predictive model in real time. It can quantitatively analyze the matching degree between the supply quantity at the equipment end and the terminal load, provide the optimized control logic, and control the fast and slow roller gap adjustment mechanism, the fast and slow roller speed converter, the flow control mechanism and other equipment to achieve optimized control, realize automatic compensation and adaptive parameter changes, and realize the intelligence of the rice huller control system.

[0095] In this embodiment, the historical operation data and historical output data of the rice hulling machine equipment are disclosed; the historical hulling integrity rate is determined according to the historical output data; the hulling integrity rate training set is generated according to the historical hulling integrity rate, and the support vector machine regression model is trained according to the hulling integrity rate training set to obtain the hulling integrity rate prediction model. Since the generalized predictive control method based on the support vector machine in this embodiment is based on big data, the intelligent method of machine learning is used to identify the nonlinear system model, and the identified model is used for generalized predictive control. In the process of model identification, a real-time iterative online correction method is adopted, which not only improves the accuracy of the identification system, but also adapts to the changes of parameters in the rice hulling process, reflecting the robustness of the system.

[0096] In addition, an embodiment of the present invention also proposes a storage medium, on which a rice hulling machine predictive control program is stored. When the rice hulling machine predictive control program is executed by a processor, the steps of the rice hulling machine predictive control method as described above are implemented.

[0097] Refer to Figure 6 , Figure 6 which is the structural block diagram of the first embodiment of the rice hulling machine predictive control device of the present invention.

[0098] As Figure 6 shown, the rice hulling machine predictive control device proposed in the embodiment of the present invention includes:

[0099] A data acquisition module 601, configured to acquire the operation data of the rice hulling machine equipment and the output data after hulling;

[0100] A data prediction module 602, configured to determine a hulling integrity rate prediction value according to the hulling integrity rate prediction model and the operation data;

[0101] An equipment control module 603, configured to determine a control amount according to the hulling integrity rate prediction value and the output data, and control the rice hulling machine equipment based on the control amount.

[0102] In this embodiment, the operation data of the rice hulling machine equipment and the output data after hulling are acquired; the hulling integrity rate prediction value is determined according to the hulling integrity rate prediction model and the operation data; the control amount is determined according to the hulling integrity rate prediction value and the output data, and the rice hulling machine equipment is controlled based on the control amount. Since in this embodiment, the operation data of the rice hulling machine equipment and the output data after hulling are acquired, then the hulling integrity rate under the current operation data is predicted according to the hulling integrity rate prediction model, and then the control amount is determined according to the prediction value and the actual output data, and the rice hulling machine equipment is controlled according to the control amount, so that when the key processing parameters of rice hulling change, the control parameters of the rice hulling machine can be adaptively adjusted, and the hulling effect of the rice hulling equipment is better.

[0103] Based on the first embodiment of the rice huller predictive control device of the present invention, a second embodiment of the rice huller predictive control device of the present invention is proposed.

[0104] In this embodiment, the data prediction module 602 is further configured to obtain the historical operation data and historical output data of the rice huller device; determine the historical shelling completeness rate according to the historical output data; generate a shelling completeness rate training set according to the historical shelling completeness rate, and train a support vector machine regression model according to the shelling completeness rate training set to obtain a shelling completeness rate prediction model.

[0105] As an implementation manner, the data prediction module 602 is further configured to set weight values for the shelling rate and the breakage rate in the historical output data respectively; determine the historical shelling completeness rate according to the shelling rate, the breakage rate, and the weight values.

[0106] As an implementation manner, the data prediction module 602 is further configured to generate a shelling completeness rate training set according to the historical shelling completeness rate, and train a support vector machine regression model according to the shelling completeness rate training set to obtain a shelling completeness rate support vector machine regression model; perform linear regression processing on the shelling completeness rate support vector machine regression model to obtain a shelling completeness rate prediction model.

[0107] As an implementation manner, the data prediction module 602 is further configured to perform linear regression processing on the shelling completeness rate support vector machine regression model to obtain a shelling completeness rate linear regression model; construct a hyperplane model according to the shelling completeness rate linear regression model, the historical shelling completeness rate, and the operation data; perform optimal hyperplane solution on the hyperplane model to obtain a shelling completeness rate prediction model.

[0108] As an implementation manner, the data prediction module 602 is further configured to construct an input sample vector according to the operation data and the historical operation data; construct an output sample vector according to the historical shelling completeness rate and the real-time shelling completeness rate; train the hyperplane model based on the input sample vector and the output sample vector to obtain a shelling completeness rate prediction model.

[0109] As an implementation manner, the device control module 603 is further configured to determine the actual shelling completeness rate according to the output data; determine a feedback control amount according to the shelling completeness rate prediction value and the actual shelling completeness rate; determine a control increment according to the feedback control amount and the optimal control parameter; control the rice huller device based on the control increment.

[0110] It should be noted that in this document, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, such that a process, method, article or system comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or system comprising such element.

[0111] The serial numbers of the above embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments.

[0112] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium as described above (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to enable a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0113] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structural or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A method for predictive control of a rice huller, characterized in that, The method includes: Obtaining the operation data of the rice hulling machine equipment and the output data after hulling; Determining the predicted value of the hulling completion rate according to the hulling completion rate prediction model and the operation data; Determining the control quantity according to the predicted value of the hulling completion rate and the output data, and controlling the rice hulling machine equipment based on the control quantity; Before the step of determining the predicted value of the hulling completion rate according to the hulling completion rate prediction model and the operation data, it further includes: Obtaining the historical operation data and historical output data of the rice hulling machine equipment; Determining the historical hulling completion rate according to the historical output data; Generating a hulling completion rate training set according to the historical hulling completion rate, and training a support vector machine regression model according to the hulling completion rate training set to obtain a hulling completion rate support vector machine regression model; Performing linear regression processing on the hulling completion rate support vector machine regression model to obtain a hulling completion rate linear regression model; Constructing a hyperplane model according to the hulling completion rate linear regression model, the historical hulling completion rate and the operation data; Solving the optimal hyperplane for the hyperplane model to obtain a hulling completion rate prediction model.

2. The rice huller predictive control method according to claim 1, characterized in that, The determining the historical hulling completion rate according to the historical output data includes: Setting weight values for the hulling rate and the breakage rate in the historical output data respectively; Determining the historical hulling completion rate according to the hulling rate, the breakage rate and the weight values.

3. The rice huller predictive control method according to claim 1, characterized in that The solving the optimal hyperplane for the hyperplane model to obtain a hulling completion rate prediction model includes: Constructing an input sample vector according to the operation data and the historical operation data; Constructing an output sample vector according to the historical hulling completion rate and the real-time hulling completion rate; Training the hyperplane model based on the input sample vector and the output sample vector to obtain a hulling completion rate prediction model.

4. The husking machine predictive control method according to any one of claims 1 to 3, characterized in that, The determining the control quantity according to the predicted value of the hulling completion rate and the output data, and controlling the rice hulling machine equipment based on the control quantity includes: Determining the actual hulling completion rate according to the output data; Determining the feedback control quantity according to the predicted value of the hulling completion rate and the actual hulling completion rate; Determining the control increment according to the feedback control quantity and the optimal control parameters; Controlling the rice hulling machine equipment based on the control increment.

5. A prediction control device for a rice huller, characterized in that, The rice hulling machine prediction control device includes: A data acquisition module, configured to acquire the operation data of the rice hulling machine equipment and the output data after hulling; A data prediction module, configured to determine the predicted value of the hulling completion rate according to the hulling completion rate prediction model and the operation data; An equipment control module, configured to determine the control quantity according to the predicted value of the hulling completion rate and the output data, and control the rice hulling machine equipment based on the control quantity; The data prediction module is further configured to acquire the historical operation data and historical output data of the rice hulling machine equipment; determine the historical hulling completion rate according to the historical output data; generate a hulling completion rate training set according to the historical hulling completion rate, and train a support vector machine regression model according to the hulling completion rate training set to obtain a hulling completion rate support vector machine regression model; The data prediction module is further configured to perform linear regression processing on the dehulling integrity rate support vector machine regression model to obtain a dehulling integrity rate linear regression model; construct a hyperplane model according to the dehulling integrity rate linear regression model, the historical dehulling integrity rate, and the operation data; and solve for the optimal hyperplane of the hyperplane model to obtain a dehulling integrity rate prediction model.

6. A rice huller predictive control device, characterized in that, The device includes: a memory, a processor, and a rice huller prediction control program stored on the memory and executable on the processor, the rice huller prediction control program being configured to implement the steps of the rice huller prediction control method according to any one of claims 1 to 4.

7. A storage medium, characterized in that, A rice huller prediction control program is stored on the storage medium, and when the rice huller prediction control program is executed by a processor, the steps of the rice huller prediction control method according to any one of claims 1 to 4 are implemented.

Images