Method and device for predicting threshing and redrying processing parameters and electronic equipment

Abstract

The embodiment of the invention discloses a method and device for predicting threshing and redrying processing parameters and electronic equipment. The method comprises the following steps: firstly, serially connecting obtained threshing and redrying historical data through a time label to obtain an initial data set, and then, carrying out process division on the initial data set to obtain three divided data sets. And further, determining a model training set corresponding to the divided data set based on a feature extraction method, and finally determining target processing parameters corresponding to the to-be-detected target data according to each process parameter prediction model. Through a process parameter prediction model corresponding to each divided process, threshing and redrying total factor information such as tobacco leaf production place grades, tobacco leaf characteristics and processing environment conditions can be fully considered, and threshing and redrying processing parameters most suitable for the current tobacco leaf situation and processing conditions are determined. The purposes of rapidly responding to production fluctuation and predicting threshing and redrying processing parameters are achieved, the processed tobacco meets the high-quality requirement, and the user satisfaction degree is improved.

Description

TECHNICAL FIELD

[0001] The one or more embodiments of the present specification relate to the technical field of tobacco production and processing, and in particular to a prediction method for threshing and redrying processing parameters. BACKGROUND

[0002] With the development of the tobacco industry, new requirements are put forward for cigarette brand construction and production process. As the front end of cigarette production, threshing and redrying is also given new missions and requirements. In the actual processing of tobacco, there are many device processing parameters in the threshing and redrying process, and the processing quality of tobacco corresponding to different device processing parameters will present a large fluctuation. When the processing environment changes and different tobacco raw materials are alternated, the device processing parameters need to be adjusted to meet the quality requirements of tobacco. At present, when the production factors fluctuate, the device processing parameters are mainly adjusted by the operating personnel according to personal experience, which not only consumes a lot of time and wastes cost, but also the processed tobacco cannot meet the high quality requirements. SUMMARY

[0003] The embodiments of the present specification provide a prediction method, device and electronic equipment for threshing and redrying processing parameters, and the technical solutions are as follows:

[0004] In a first aspect, the embodiments of the present specification provide a prediction method for threshing and redrying processing parameters, and the method comprises:

[0005] Based on the time label, the obtained threshing and redrying historical data is concatenated to obtain an initial data set, and the threshing and redrying historical data includes historical tobacco origin data, historical tobacco grade data, historical process quality data, historical environment data and historical device processing parameters;

[0006] The initial data set is divided into three kinds of division data sets, and the division data set includes a pretreatment process data set, a leaf and stem separation process data set and a redrying process data set;

[0007] For any division data set, a model training set corresponding to the division data set is determined based on a feature extraction method;

[0008] The target processing parameters corresponding to the target data to be measured are determined according to the process parameter prediction models obtained from each model training set.

[0009] In a second aspect, a prediction device for threshing and redrying processing parameters is provided, and the device comprises:

[0010] A concatenation module is configured to concatenate the acquired primary processing historical data based on time labels to obtain an initial data set, wherein the primary processing historical data comprises historical tobacco leaf origin data, historical tobacco leaf grade data, historical process quality data, historical environment data and historical equipment processing parameters;

[0011] A division module is configured to divide the initial data set into three divided data sets, i.e., a pretreatment process data set, a leaf stem separation process data set and a redrying process data set;

[0012] A determination module is configured to determine, for any divided data set, a model training set corresponding to the divided data set based on a feature extraction method;

[0013] A prediction module is configured to determine a target processing parameter corresponding to a target data to be measured according to a process parameter prediction model obtained from each model training set.

[0014] In a third aspect, an electronic device is provided, comprising a device processor and a memory;

[0015] The device processor is connected to the memory;

[0016] The memory is configured to store executable program codes;

[0017] The device processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the steps of the method provided in the first aspect or any possible implementation manner of the first aspect.

[0018] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and the computer readable storage medium stores instructions, which, when executed on a computer or a device processor, cause the computer or the device processor to execute the method provided in the first aspect or any possible implementation manner of the first aspect.

[0019] The technical solutions provided by some embodiments of the present specification have at least the following beneficial effects:

[0020] In one or more embodiments of the present specification, by concatenating the acquired primary processing data based on time labels first, an initial data set is obtained, then, the initial data set is divided into three divided data sets. Further, for any divided data set, a model training set corresponding to the divided data set is determined based on a feature extraction method, and finally, a target processing parameter corresponding to the target data to be measured is determined according to a process parameter prediction model obtained from each model training set. Through the process parameter prediction model corresponding to each divided process, the full-factor information of the primary processing of tobacco leaves such as tobacco origin grade, tobacco characteristics and processing environmental conditions can be fully considered to determine the most suitable primary processing parameter for the current tobacco status and processing conditions. Not only the purpose of quickly responding to production fluctuations and predicting primary processing parameters is achieved, but also the processed tobacco meets the high quality requirements and improves the user satisfaction. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0022] Figure 1 A flow chart of a primary processing parameter prediction method provided by an embodiment of the present specification;

[0023] Figure 2 A structural schematic diagram of a primary processing parameter prediction device provided by an embodiment of the present specification;

[0024] Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the present specification. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application.

[0026] In the present specification, the terms "first", "second", "third", etc. in the description and claims and the above drawings are used to distinguish different objects, and are not used to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed or can optionally include other steps or units inherent to the process, method, product or device.

[0027] The following description provides examples, and is not intended to limit the scope, applicability or examples set forth in the claims. Alterations and further modifications of the described elements are possible without deviating from the scope of the present description. Various examples can omit, substitute, or add various procedures or components as appropriate. For instance, the methods described can be performed in an order different than described, and various steps can be added, omitted, or combined. Also, features described with respect to some examples can be combined in other examples.

[0028] Referring now to the drawings, and more particularly to Figure 1 , Figure 1 A whole flow chart of a prediction method of processing parameters of threshing and redrying provided by the embodiments of the present specification is shown.

[0029] As Figure 1 shown, the prediction method of processing parameters of threshing and redrying can at least include the following steps:

[0030] Step 101, concatenating the obtained threshing and redrying historical data based on time labels to obtain an initial data set.

[0031] The threshing and redrying historical data includes historical tobacco origin data, historical tobacco grade data, historical process quality data, historical environment data and historical equipment processing parameters.

[0032] In the embodiments of the present specification, since there are a large number of equipment processing parameters in the threshing and redrying process of the actual tobacco processing process, and the processing quality of tobacco corresponding to different equipment processing parameters will present a large fluctuation, when the processing environment changes and different tobacco raw materials alternate, etc., the equipment processing parameters need to be adjusted to meet the requirements of tobacco quality. Specifically, there is a complex relationship between tobacco processing quality and tobacco moisture, leaf conditioning temperature, moisture, and threshing process conditions, etc. In order to predict the target processing parameter through the target data to be measured including each tobacco information and environment information, so that the tobacco processing quality result produced under the target processing parameter setting is optimal, it is necessary to obtain the threshing and redrying historical data in the production record of optimal quality result for many years. The threshing and redrying historical data includes historical tobacco origin data, historical tobacco grade data, historical process quality data, historical environment data and historical equipment processing parameters.

[0033] Then, in order to facilitate subsequent unified formatting processing and model training of the threshing and redrying historical data, it is necessary to concatenate the historical tobacco origin data, historical tobacco grade data, historical process quality data, historical environment data and historical equipment processing parameters in the threshing and redrying historical data according to the time labels to obtain an initial data set.

[0034] The historical environment data includes but is not limited to the environment temperature and humidity data during processing of each main process, and the historical process quality data includes but is not limited to the quality index data such as tobacco moisture content, leaf structure, and stem content in leaf, and the process quality data such as tobacco moisture content before and after moisture recovery, tobacco moisture content before and after tobacco moistening, and tobacco temperature.

[0035] In an implementation manner, the historical data of the threshing and redrying obtained based on the time label is concatenated to obtain an initial data set, including:

[0036] The historical origin coding data corresponding to the historical tobacco origin data and the historical registration coding data corresponding to the historical tobacco grade data are determined according to the one-hot coding database respectively;

[0037] The historical process quality data, the historical environment data, and the historical equipment processing parameter are sorted based on the time label to obtain sorting data;

[0038] The historical origin coding data, the historical registration coding data, and the sorting data are concatenated to obtain the initial data set.

[0039] In the embodiments of the present specification, when the historical data of the threshing and redrying obtained based on the time label is concatenated, the text information corresponding to the historical tobacco origin data and the historical tobacco grade data in the historical data of the threshing and redrying is first converted into a numerical value in a one-hot coding manner through a one-hot coding database. Specifically, the historical tobacco origin data needs to query the origin one-hot coding database corresponding thereto, and the historical tobacco grade data needs to query the grade one-hot coding database corresponding thereto. In each one-hot coding database, an N-bit state register is used to code N states, each state has its independent register bit, and at any time, only one bit is valid. That is, only one bit is 1, and the rest are zero values.

[0040] As an example, a certain constructed origin one-hot coding database is as follows:

[0041]

[0042] A certain constructed grade one-hot coding database is as follows:

[0043]

[0044] Then, the time label corresponding to the historical process quality data, the historical environment data, and the historical equipment processing parameter is determined, and each data is sorted according to the time label corresponding thereto to obtain sorting data. Finally, the time labels corresponding to the historical origin coding data and the historical registration coding data are fixedly arranged at the head of the data set, and then concatenated with the sorting data to obtain the initial data set. As an example, a certain initial data set obtained after concatenation is as follows:

[0045]

[0046] In an implementation manner, the historical process quality data, the historical environment data and the historical equipment processing parameter are sorted based on the time label to obtain sorting data, including:

[0047] The time granularity corresponding to the historical process quality data, the historical environment data and the historical equipment processing parameter is respectively determined based on the time label;

[0048] The historical process quality data, the historical environment data and the historical equipment processing parameter are sorted according to the comparison result of each time granularity to obtain sorting data.

[0049] In the embodiments of the present specification, when the historical process quality data, the historical environment data and the historical equipment processing parameter are sorted by the time label, the time granularity corresponding to the historical process quality data, the historical environment data and the historical equipment processing parameter can be determined first. Then, each time granularity is compared, when the comparison result represents that each time granularity is consistent, the sorting can be directly performed according to the time label, when the comparison result represents that each time granularity is inconsistent, the maximum time granularity is determined, and then the average value of other data is calculated based on the maximum time granularity as a basic unit, and finally the sorting is performed according to the calculated data average value to obtain sorting data.

[0050] Step 102, process division is performed on the initial data set to obtain three divided data sets.

[0051] The divided data set includes a pretreatment process data set, a leaf and stem separation process data set and a redrying process data set.

[0052] In the embodiments of the present specification, after the initial data set is obtained by concatenation, each data in the initial data set can be abnormally cleaned, and the abnormal working condition processing data such as shutdown and material breakage is deleted to obtain the initial data set corresponding to the steady state. Then, the initial data set is separated according to the process characteristics into a pretreatment process, a leaf and stem separation process and a redrying process to obtain three divided data sets, which can include a pretreatment process data set, a leaf and stem separation process data set and a redrying process data set.

[0053] When the initial data set corresponding to the steady state is obtained by abnormally cleaning the initial data set, each abnormal condition judgment rule can be used. As an example, the abnormal condition judgment rule of the first leaf moistening process is as follows:

[0054]

[0055] The abnormal condition judgment rule of the second leaf moistening process is as follows:

[0056]

[0057] The abnormal condition determination rule of the leaf stem separation process is as follows:

[0058] The abnormal condition determination rule of the redrying process is as follows:

[0059]

[0060] Step 103, for any divided data set, determining a model training set corresponding to the divided data set based on a feature extraction method.

[0061] In the embodiments of the present specification, after the initial data set is divided into the pretreatment process data set, the leaf stem separation process data set and the redrying process data set, since the processing environment of each process is different, a corresponding prediction model needs to be constructed for each process in order to accurately predict the equipment parameters corresponding to each process. Wherein, when constructing each equipment parameter prediction model, a large number of model training sets are needed for training, and the obtained divided data sets cannot meet the input requirements of the training sets, therefore, for any divided data set, the model training set corresponding to the divided data set needs to be determined according to the feature extraction method first. Specifically, the feature data corresponding to the divided data set can be determined by the feature extraction method first, and then the historical processing parameter data in the divided data set and the feature data are paired one by one to obtain the model training set.

[0062] Optionally, the feature data obtained by feature extraction can also be multi-modal fusion, or fusion with key field knowledge features, and then the historical processing parameter data is paired to obtain the model training set.

[0063] In an implementable manner, the determination of the model training set corresponding to the divided data set based on the feature extraction method comprises:

[0064] determining the feature data corresponding to the divided data set based on the PCA feature extraction method;

[0065] pairing the feature data and the historical processing parameter data to obtain the model training set.

[0066]

[0067] In the embodiments of the present specification, the PCA feature extraction algorithm can be used to reduce the dimension of the divided data set, calculate the covariance matrix, generate the eigenvector, sort according to the variance contribution rate, retain the first k principal components (such as k = 5), make the cumulative variance contribution rate > 85%, and obtain the feature data. As an example, the 20-dimensional original divided data is compressed to 5-dimensional, reducing the dimension and eliminating redundancy. Then, in order to clarify the mapping relationship between the features and the prediction target, avoid data misplacement, and provide structured input for subsequent prediction models, the extracted feature data can be paired with the historical processing parameter data corresponding to the expected batch to obtain the model training set.

[0068] In an implementation manner, after the PCA feature extraction method determines the feature data corresponding to the divided data set, the method further includes:

[0069] standardizing the feature data to obtain historical standard data;

[0070] The data pairing of the feature data and the historical processing parameter data includes:

[0071] The data pairing of the historical standard data and the historical processing parameter data.

[0072] In the embodiments of the present specification, in order to eliminate the scale difference between the principal components, so as to improve the gradient descent efficiency and accelerate the model convergence in the subsequent process, the obtained feature data can be standardized to obtain historical dimension reduction data. Specifically, the feature data is transformed to transform the data to a range with a mean close to 0 and a standard deviation of 1, and the calculation formula is as follows:

[0073]

[0074] wherein, is the data standardized value, x is the original data value, mean is the average value of the feature data sample, and σ is the standard deviation of the feature data sample.

[0075] Step 104, determining the target processing parameter corresponding to the to-be-tested target data according to the process parameter prediction model obtained from each model training set.

[0076] In the embodiments of the present specification, after obtaining the model training set corresponding to each divided data set, the device parameter prediction model corresponding to each process can be constructed, and each model training set is trained to obtain the trained device parameter prediction model. Further, the target processing parameter corresponding to the to-be-tested target data is determined according to each trained device parameter prediction model.

[0077] The process parameter prediction model corresponding to each process can adopt eight kinds of support multi-output corresponding algorithm models, such as linear regression, lasso regression, elastic network regression, etc.

[0078] In an implementable manner, the process parameter prediction model obtained according to each model training set determines the target processing parameter corresponding to the target data to be measured, comprising:

[0079] Based on the algorithm traversal evaluation of each divided process according to each model training set, the process parameter prediction model corresponding to each target algorithm is obtained;

[0080] According to each process parameter prediction model, the target processing parameter corresponding to the target data to be measured is determined.

[0081] In the embodiments of the present application, since the process parameter prediction model can adopt linear regression algorithm, lasso regression algorithm, elastic network regression algorithm, K nearest neighbor regression algorithm, classification and regression tree algorithm, random forest regression algorithm, single model SVR algorithm and linear model sequence SVR algorithm model, it is necessary to first traverse and evaluate all algorithm models that can be adopted according to a small part of the test set in each model training set according to each divided process, and the model evaluation method adopts the mean and standard deviation of the mean square error to determine the prediction result evaluation of each model corresponding to each divided process. Through the evaluation result, the process parameter prediction model corresponding to each target algorithm is obtained. As an example, for the pre-processing process and the cut tobacco re-drying process, the prediction result evaluation corresponding to the random forest regression algorithm is optimal, and for the leaf and stem separation process, the prediction result evaluation corresponding to the K nearest neighbor regression algorithm is optimal. Further, after the target data to be measured is processed according to the same steps as the leaf re-drying historical data, it is respectively input into each determined process parameter prediction model to obtain the target processing parameter.

[0082] After selecting the process parameter prediction model corresponding to each process, in order to further improve the algorithm prediction accuracy and performance, the K nearest neighbor algorithm can also be selected to adjust the prediction model.

[0083] In an implementable manner, the process parameter prediction model obtained according to each model training set determines the target processing parameter corresponding to the target data to be measured, comprising:

[0084] The target data to be measured is deserialized to obtain target sequence data to be measured;

[0085] Based on each process parameter prediction model, the target sequence data to be measured is determined in the process parameter corresponding to each divided process;

[0086] Integrate each process parameter to obtain the target processing parameter.

[0087] In the embodiments of the present application, when the target processing parameter corresponding to the to-be-tested target data is determined according to the process parameter prediction model, the to-be-tested target data can be first deserialized to obtain to-be-tested target sequence data, so that the data format requirement of subsequent data input into each prediction model is met. Then, the to-be-tested target sequence data is input into each process parameter prediction model according to the sequence, to obtain the process parameters corresponding to each divided process. Finally, the process parameters are integrated to obtain the target processing parameter.

[0088] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order other than that described in the embodiments and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown or sequential order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous or possible.

[0089] Next, please refer to Figure 2 , Figure 2 A structure diagram of a processing parameter prediction device for threshing and redrying provided by an embodiment of the present application is shown. It should be noted that Figure 2 The processing parameter prediction device for threshing and redrying is used to execute the method of the embodiment shown in Figure 1 For ease of illustration, only parts related to the embodiments of the present application are shown, and specific technical details are not disclosed, please refer to the embodiments shown in Figure 1 .

[0090] As Figure 2 shown, the processing parameter prediction device for threshing and redrying can at least include:

[0091] The concatenation module 201 is configured to concatenate the obtained threshing and redrying historical data based on a time label to obtain an initial data set, wherein the threshing and redrying historical data includes historical tobacco leaf origin data, historical tobacco leaf grade data, historical process quality data, historical environment data, and historical equipment processing parameters.

[0092] The division module 202 is configured to divide the initial data set into process division data sets to obtain three kinds of division data sets, wherein the division data sets include a pretreatment process data set, a leaf and stem separation process data set, and a redrying process data set.

[0093] The determination module 203 is configured to determine, for any division data set, a model training set corresponding to the division data set based on a feature extraction method.

[0094] The prediction module 204 is configured to determine a target processing parameter corresponding to to-be-tested target data according to a process parameter prediction model obtained from each model training set.

[0095] In an implementation, the concatenating module 201 is specifically configured to:

[0096] determine, according to the one-hot encoding database, historical origin encoding data corresponding to the historical tobacco origin data and historical registration encoding data corresponding to the historical tobacco grade data, respectively;

[0097] sort the historical process quality data, the historical environment data and the historical equipment processing parameters based on the time label to obtain sorting data;

[0098] concatenate the historical origin encoding data, the historical registration encoding data and the sorting data to obtain an initial data set.

[0099] In an implementation, the concatenating module 201 is specifically configured to:

[0100] determine, based on the time label, time granularities corresponding to the historical process quality data, the historical environment data and the historical equipment processing parameters, respectively;

[0101] sort the historical process quality data, the historical environment data and the historical equipment processing parameters according to the comparison results of the time granularities to obtain sorting data.

[0102] In an implementation, the determining module 203 is specifically configured to:

[0103] determine, based on the PCA feature extraction method, feature data corresponding to the divided data set;

[0104] pair the feature data and historical processing parameter data to obtain a model training set.

[0105] In an implementation, the determining module 203 is specifically configured to:

[0106] standardize the feature data to obtain historical standard data;

[0107] The pairing of the feature data and the historical processing parameter data includes:

[0108] pairing the historical standard data and the historical processing parameter data.

[0109] In an implementation, the predicting module 204 is specifically configured to:

[0110] perform algorithm traversal evaluation on each divided process based on each model training set to obtain a process parameter prediction model corresponding to each target algorithm;

[0111] determine target processing parameters corresponding to to-be-tested target data according to each process parameter prediction model.

[0112] In an implementation, the prediction module 204 is further configured to:

[0113] deserializing the target data to obtain target sequence data;

[0114] determining the process parameters corresponding to each division process based on the process parameter prediction model;

[0115] integrating the process parameters to obtain target processing parameters.

[0116] Those skilled in the art can clearly understand that the technical solutions of the embodiments of the present application can be implemented by means of software and / or hardware. The "unit" and "module" in the specification refer to software and / or hardware that can independently complete or cooperate with other components to complete a specific function, and the hardware may, for example, be a field programmable gate array (FPGA), an integrated circuit (IC), etc.

[0117] The various processing units and / or modules of the embodiments of the present application can be implemented by means of analog circuits that implement the functions described in the embodiments of the present application, or by means of software that implements the functions described in the embodiments of the present application.

[0118] Next, please refer to Figure 3 , Figure 3 a structural schematic diagram of an electronic device provided by an embodiment of the present specification is shown.

[0119] As Figure 3 shown, the electronic device 300 can include at least one device processor 301, at least one network interface 303, a user interface 303, a memory 305, and at least one communication bus 302.

[0120] The communication bus 302 can be used to realize the connection and communication of the above-mentioned components.

[0121] The user interface 303 can include a key, and the optional user interface can further include a standard wired interface, a wireless interface.

[0122] The network interface 304 can include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.

[0123] The device processor 301 can include one or more processing cores. The device processor 301 connects various parts within the entire electronic device 300 by various interfaces and lines, executes various functions of the electronic device 300 and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Alternatively, the device processor 301 can be implemented in at least one of a hardware form of a DSP, an FPGA, and a PLA. The device processor 301 can integrate one or a combination of a CPU, a GPU, and a modem. Among them, the CPU mainly processes operating systems, user interfaces, and application programs; the GPU is responsible for rendering and drawing the content required to be displayed on the display screen; and the modem is used for processing wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the device processor 301, but can be implemented by a separate chip.

[0124] The memory 305 can include a RAM and can also include a ROM. Alternatively, the memory 305 includes a non-transitory computer readable medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 305 can also be at least one storage device located away from the aforementioned device processor 301. As shown, the memory 305 as a computer storage medium can include an operating system, a network communication module, a user interface module, and program instructions. Figure 3

[0125] Specifically, the device processor 301 can be used to call the prediction application program of the primary processing parameter of the leaf tobacco reprocessing stored in the memory 305, and specifically perform the following operations:

[0126] The obtained leaf tobacco reprocessing historical data is concatenated based on a time label to obtain an initial data set, and the leaf tobacco reprocessing historical data includes historical tobacco leaf origin data, historical tobacco leaf grade data, historical process quality data, historical environment data, and historical equipment processing parameters;

[0127] The initial data set is divided into processes to obtain three divided data sets, and the divided data sets include a pretreatment process data set, a leaf stem separation process data set, and a reprocessing process data set;

[0128] For any divided data set, a model training set corresponding to the divided data set is determined based on a feature extraction method; ​

[0129] The process parameter prediction model obtained according to each model training set is used to determine the target processing parameter corresponding to the target data to be measured.

[0130] As an option of the embodiment of the present specification, the historical data of threshing and redrying obtained based on the time label is concatenated to obtain an initial data set, including:

[0131] According to the one-hot encoding database, historical origin coding data corresponding to the historical tobacco origin data and historical registration coding data corresponding to the historical tobacco grade data are determined respectively;

[0132] The historical process quality data, historical environment data and historical equipment processing parameters are sorted based on the time label to obtain sorting data;

[0133] The historical origin coding data, historical registration coding data and sorting data are concatenated to obtain the initial data set.

[0134] As an option of the embodiment of the present specification, the historical process quality data, historical environment data and historical equipment processing parameters are sorted based on the time label to obtain sorting data, including:

[0135] The time granularity corresponding to the historical process quality data, historical environment data and historical equipment processing parameters is determined based on the time label respectively;

[0136] The historical process quality data, historical environment data and historical equipment processing parameters are sorted according to the comparison results of each time granularity to obtain sorting data.

[0137] As an option of the embodiment of the present specification, the model training set corresponding to the divided data set is determined based on the feature extraction method, including:

[0138] The feature data corresponding to the divided data set is determined based on the PCA feature extraction method;

[0139] The feature data and historical processing parameter data are paired to obtain the model training set.

[0140] As an option of the embodiment of the present specification, the feature data is standardized to obtain historical standard data, including:

[0141] The feature data is standardized to obtain historical standard data;

[0142] The feature data and historical processing parameter data are paired, including:

[0143] The historical standard data and historical processing parameter data are paired.

[0144] As an option of the embodiment of the present specification, the procedure parameter prediction model obtained according to each of the model training sets determines the target machining parameter corresponding to the target data to be measured, comprising:

[0145] Each target algorithm corresponding to the procedure parameter prediction model is obtained by algorithm traversal evaluation of each of the model training sets on each divided procedure;

[0146] The target machining parameter corresponding to the target data to be measured is determined according to each of the procedure parameter prediction models.

[0147] As an option of the embodiment of the present specification, the procedure parameter prediction model obtained according to each of the model training sets determines the target machining parameter corresponding to the target data to be measured, comprising:

[0148] The target data to be measured is deserialized to obtain target sequence data to be measured;

[0149] The procedure parameter corresponding to each divided procedure is determined based on each of the procedure parameter prediction models and the target sequence data to be measured;

[0150] The target machining parameter is obtained by integrating each of the procedure parameters.

[0151] The embodiment of the present specification also provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the steps of the above method. The computer readable storage medium can include but is not limited to any type of disk, including floppy disk, optical disk, DVD, CD-ROM, micro drive, and magneto-optical disk, ROM, RAM, EPROM, EEPROM, DRAM, VRAM, flash memory device, magnetic card or optical card, nanosystem (including molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0152] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action order described, because according to the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0153] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0154] In several embodiments provided in the present application, it should be understood that the disclosed apparatus can be implemented in other manners. For example, the division of the apparatus embodiments described above is merely illustrative, and the division of the units can be changed according to actual needs. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0155] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0156] In addition, each functional unit in the various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0157] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that makes a contribution to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0158] Those of ordinary skill in the art can understand that all or part of the steps of the various methods in the above embodiments can be completed by a program instructing relevant hardware, and the program can be stored in a computer readable storage medium, which can include a flash disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, etc.

[0159] The above described embodiments of the present description have been described. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In certain implementations, multitasking and parallel processing can be advantageous.

Claims

1. A method for predicting parameters of threshing and redrying processing, characterized in that: The method comprises: The acquired threshing and redrying historical data are concatenated based on the time tags to obtain an initial data set, wherein the threshing and redrying historical data includes historical tobacco leaf origin data, historical tobacco leaf grade data, historical process quality data, historical environmental data, and historical equipment processing parameters; Dividing the initial data set into three divided data sets, wherein the divided data sets include a pre-processing process data set, a leaf stem separation process data set, and a redrying process data set; For any divided data set, determining a model training set corresponding to the divided data set based on a feature extraction method; The target processing parameters corresponding to the target data to be measured are determined according to the process parameter prediction model obtained from each of the model training sets.

2. The method according to claim 1, characterized in that The obtained leaf threshing and redrying historical data are concatenated based on the time tags to obtain an initial data set, including: Determining the historical origin coding data corresponding to the historical tobacco origin data and the historical registration coding data corresponding to the historical tobacco grade data respectively according to a one-hot coding database; Sorting the historical process quality data, historical environment data, and historical equipment processing parameters based on time tags to obtain sorted data; The historical origin coding data, historical registration coding data and sorting data are concatenated to obtain an initial data set.

3. The method according to claim 2, characterized in that The historical process quality data, historical environment data, and historical equipment processing parameters are sorted based on the time tags to obtain sorted data, including: Determining the time granularity corresponding to the historical process quality data, the historical environmental data, and the historical equipment processing parameters respectively based on the time tags; The historical process quality data, historical environment data and historical equipment processing parameters are sorted according to the comparison results of each of the time granularities to obtain sorted data.

4. The method according to claim 1, wherein The determining of the model training set corresponding to the divided data set based on the feature extraction method includes: Determine feature data corresponding to the divided data set based on the PCA feature extraction method; The feature data and historical processing parameter data are paired to obtain a model training set.

5. The method according to claim 4, characterized in that After determining the feature data corresponding to the divided data set based on the PCA feature extraction method, the method further includes: performing standardization processing on the characteristic data to obtain historical standard data; The data pairing of the characteristic data and the historical processing parameter data includes: The historical standard data and the historical processing parameter data are paired.

6. The method according to claim 1, characterized in that The process parameter prediction model obtained from each model training set determines the target processing parameter corresponding to the target data to be measured, including: Perform algorithm traversal evaluation on each divided process based on each model training set to obtain a process parameter prediction model corresponding to each target algorithm; The target processing parameters corresponding to the target data to be measured are determined according to the process parameter prediction models.

7. The method according to claim 6, characterized in that Determining the target processing parameters corresponding to the target data to be measured according to each of the process parameter prediction models includes: Deserialize the target data to be tested to obtain the target sequence data to be tested; Determine the process parameters corresponding to each divided process of the target sequence data to be tested based on each process parameter prediction model; The process parameters are integrated to obtain target processing parameters.

8. A device for predicting parameters of leaf threshing and redrying processing, characterized in that: The device comprises: A concatenation module is used to concatenate the acquired leaf threshing and redrying historical data based on time tags to obtain an initial data set, wherein the leaf threshing and redrying historical data includes historical tobacco leaf origin data, historical tobacco leaf grade data, historical process quality data, historical environmental data, and historical equipment processing parameters; a partitioning module, configured to partition the initial data set into three partitioned data sets, wherein the partitioned data sets include a pre-processing data set, a leaf stem separation data set, and a redrying data set; A determination module, configured to determine, for any partitioned data set, a model training set corresponding to the partitioned data set based on a feature extraction method; The prediction module is used to determine the target processing parameters corresponding to the target data to be measured based on the process parameter prediction model obtained from each of the model training sets.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, wherein the computer-readable storage medium stores instructions, which, when the instructions are executed on a computer or a processor, cause the computer or processor to execute the steps of the method according to any one of claims 1 to 7.

Images