A rice processing whole-process intelligent quality improvement management and control method

Abstract

The application discloses a rice processing whole-process intelligent quality improvement management and control method and relates to the technical field of rice processing.The application detects the rice storage environment parameters through relevant detection devices, and then obtains the optimal drying parameters and the optimal moisture content of wet rice based on the analysis of the storage environment parameters.After the wet rice is once impurity-removed and input into a dryer, the wet rice is dried based on the optimal drying rate and the optimal moisture content, and the to-be-husked rice is obtained; and then, the optimal husking and the optimal rice milling parameters are obtained through analysis, and the husked rice is husked and milled.The rice processing decision model takes the minimum storage deterioration rate as the optimization target to reversely deduce the optimal drying parameters and the optimal moisture content, and the storage mildew risk and the pest risk are pre-posed and integrated into the drying management and control, so that the problem of the increase of the broken rice rate caused by excessive drying can be avoided, and the problem of the deterioration of the husked rice after storage caused by insufficient drying can be eliminated.

Description

Technical Field

[0001] This invention relates to the field of rice processing technology, specifically to an intelligent quality improvement and control method for the entire rice processing process. Background Technology

[0002] Rice is one of my country's main food crops, and the rice processing industry is crucial for ensuring food supply, improving the quality and efficiency of agricultural products, and ensuring safe storage. The existing traditional rice processing flow mainly involves stages such as paddy cleaning, drying, hulling, milling, and storage. However, the industry currently suffers from prominent problems such as independent control of each stage, reliance on manual experience for setting process parameters, and a disconnect between processing and storage. These issues severely restrict the stability of finished rice quality and its shelf life. Therefore, developing a method for improving and controlling the quality of rice processing is essential.

[0003] Existing technologies still have some shortcomings in the quality control of rice processing, specifically in the following aspects: 1. Existing technologies mostly adopt a crude control method of fixed drying temperature and fixed drying time in the rice drying process. They only make rough adjustments based on the ambient temperature and humidity, and lack comprehensive consideration of the impact of multi-dimensional storage factors such as the concentration of harmful compound gases such as ammonia and hydrogen sulfide, airborne mold spores, storage insect eggs and mite particles on the later mold, yellowing, insect infestation and deterioration of rice. At the same time, there is a lack of intelligent models to make reverse optimization decisions on drying parameters and target moisture content, which can easily lead to over-drying, resulting in increased cracked rice and broken rice rate, or under-drying, making rice susceptible to the influence of storage environment parameters after entering the warehouse.

[0004] 2. The existing traditional rice processing stages are isolated from each other. There is no linkage and coupling mechanism for process parameters between drying, hulling and milling. The upstream processing status cannot be transmitted to the downstream stage in real time for parameter adaptive adjustment. There is an island problem in the major rice processing process components. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to provide an intelligent quality improvement and control method for the entire rice processing process.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent quality improvement and control method for the entire process of rice processing, including: Step 1, detecting the parameters of the rice storage environment through relevant detection devices, and then analyzing the optimal drying parameters and optimal moisture content of wet rice based on the storage environment parameters, inputting the wet rice into the dryer after removing impurities once, and drying the wet rice based on the optimal drying rate and optimal moisture content to obtain rice to be hulled.

[0007] Step 2: Hull the rice to be hulled to obtain brown rice.

[0008] Step 3: Mill the unmilled rice to obtain rice to be stored. After removing impurities three times, store the rice in the warehouse.

[0009] The beneficial effects of this invention are as follows: 1. This invention collects multi-dimensional storage environment parameters such as warehouse temperature and humidity, concentration of compound harmful gases, concentration of airborne microorganisms, and concentration of insect egg particles. Combined with rice varieties, it introduces a pre-trained MLP rice processing decision model. With minimizing the storage spoilage rate as the optimization objective, it reverse-engineers and outputs the optimal drying parameters and target moisture content. This breaks through the traditional extensive mode that relies solely on temperature and humidity and manual experience to set the drying process. By integrating the risks of mold and pests into the drying control in advance, it can avoid over-drying, which can cause rice to burst, crack, and increase the rate of broken rice. It can also prevent under-drying, which can lead to moisture absorption, mold growth, yellowing, browning, and insect infestation after the rice is stored. This significantly reduces the overall storage spoilage rate of rice.

[0010] 2. This invention achieves full-process linkage and adaptive optimization of drying quality and rice hulling and milling process parameters, breaking the technical drawbacks of independent parameter adjustment and lack of correlation between different stages of traditional rice processing. It effectively solves the problems of incomplete hulling, high back-hulling rate, over-milling and under-milling, and excessive bran residue caused by traditional manual parameter adjustment. It significantly improves the hulling efficiency of paddy rice and the uniformity of rice milling, while simultaneously reducing the overall broken rice rate and improving the rice yield and finished product consistency.

[0011] 3. This invention relies on a storage environment data collection method that uses equidistant grid multi-point average sampling and periodic time-series average correction to ensure accurate and representative environmental parameter detection. At the same time, it retains the foundation of traditional rice processing technology while realizing intelligent, standardized, and closed-loop management of the entire process, improving the purity, quality stability, and storage and preservation period of finished rice products, and meeting the needs of large-scale, standardized refined rice processing industry for quality improvement and efficiency enhancement. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Reference Figure 1 As shown, the present invention provides an intelligent quality improvement and control method for the entire rice processing process, including: Step 1, detecting the rice storage environment parameters through relevant detection devices, and then analyzing the optimal drying parameters and optimal moisture content of wet rice based on the storage environment parameters, inputting the wet rice into a dryer after removing impurities once, and drying the wet rice based on the optimal drying rate and optimal moisture content to obtain rice to be hulled.

[0016] It should be noted that the primary impurity removal process includes removing large, irregularly shaped impurities such as straw, stalks, weeds, mud, large stones, dust, and light debris from the wet rice using a vibrating screen and a blower.

[0017] It should also be noted that the drying parameters include the drying time for each of the three drying stages, the drying hot air temperature, the tempering time, and the drying air speed.

[0018] In a specific example, the storage environment parameters include temperature and humidity data, complex gas concentration, airborne bacteria concentration, and particulate matter concentration.

[0019] It should be noted that the compound gas includes volatile harmful gases such as ammonia and hydrogen sulfide; airborne bacteria include microorganisms such as mold spores suspended in the air; and particulate matter includes insect eggs, mite spores, and insect debris floating in the air.

[0020] In a specific example, the process of detecting rice storage environment parameters using relevant detection devices is as follows: Sampling points are set up inside the rice storage warehouse using an equidistant grid method. First, temperature and humidity sensors are placed at each sampling point to collect temperature and humidity data. The average temperature and humidity data from each sampling point is used as the warehouse environment temperature and humidity data. A portable composite gas detector is placed at each sampling point to collect the composite gas concentration. The average composite gas concentration from each sampling point is used as the warehouse composite gas concentration. An airborne bacteria sampler is placed at each sampling point to collect the airborne bacteria concentration. The average airborne bacteria concentration from each sampling point is used as the warehouse airborne bacteria concentration. A dust concentration monitor is placed at each sampling point to collect the particulate matter concentration.

[0021] It should be noted that sampling points were set up inside the rice storage warehouse using an equidistant grid method, with the grid spacing set by the relevant staff, for example, both the horizontal and vertical spacing were set to 6 meters.

[0022] It should also be noted that after placing the airborne bacteria sampler at each sampling point, an equal amount of air sample is extracted at each sampling point, and the air sample from each sampling point is attached to a petri dish. Then, the concentration of airborne bacteria in the air is counted through the petri dish.

[0023] Several sampling time points are set within a preset period. The average particulate matter concentration of each sampling point collected within the preset period is taken as the particulate matter concentration in the warehouse. The average values ​​of temperature and humidity data, compound gas concentration, airborne bacteria concentration and particulate matter concentration in the warehouse are collectively recorded as the storage environment parameters.

[0024] It should be noted that the specific length of the preset period is determined by the relevant staff within one to three weeks.

[0025] In a specific example, the process of obtaining the optimal drying parameters and optimal moisture content of wet rice based on the analysis of storage environment parameters is as follows: input the rice storage environment parameters and rice variety into a pre-trained rice processing decision model, and output the optimal drying parameters and optimal moisture content of wet rice through the rice processing decision model.

[0026] In a specific example, the training process of the rice processing decision model is as follows: the training data set of the rice processing decision model is obtained through testing, and the training data set is preprocessed and divided into training set, test set and validation set according to a preset ratio.

[0027] It should be noted that preprocessing includes data cleaning, data normalization, and feature encoding, among which data cleaning, data normalization, and feature encoding are existing technologies and will not be elaborated further.

[0028] It should also be noted that the preset ratio is 7:2:1.

[0029] MLP is adopted as the core architecture of the drying decision model, and a three-layer network framework for the drying decision model is built. The three-layer network framework includes an input layer, a hidden layer and an output layer. The training set, test set and validation set are input into the three-layer network framework for training, testing and validation to obtain the rice processing decision model.

[0030] It should be noted that in the three-layer network framework: the number of neurons in the input layer is the same as the number of input features, and it is responsible for receiving the preprocessed input feature vector and passing it to the hidden layer; there are two hidden layers, with 32 neurons in the first layer and 16 neurons in the second layer, both using the ReLU activation function; the number of neurons in the output layer is 2, using the linear activation function, and it is responsible for outputting the drying parameters and moisture content corresponding to the minimum predicted value of the deterioration rate.

[0031] It should also be noted that the specific steps of the three-layer framework are as follows: Step 1: Install PyTorch, NumPy, Pandas, and other databases, and configure the relevant environment. Step 2: Define the MLP regression model class, build the 3-layer network architecture in the `__init__` method, clarify the parameter dimensions of each layer, and add a parameter backpropagation interface to ensure that after the model training is completed, the input parameters (drying parameters, moisture content) can be solved back by setting the objective function (minimum spoilage rate). Step 3: Instantiate the defined MLP model, initialize the model's internal weights and bias parameters (using default random initialization, which will be optimized iteratively through backpropagation), and set the initial learning rate to 0.001, the number of iterations to 500, and the batch size to 32. Step 4: Select the mean squared error (MSE) as the loss function to calculate the error between the predicted spoilage rate and the actual spoilage rate, and use the Adam optimization algorithm to update the optimizer instead of the basic gradient descent method.

[0032] In a specific example, the training data set for obtaining the rice processing decision model through testing is obtained as follows: equal weights of wet rice samples of each variety are retrieved from the sampling center, and the wet rice samples of each variety are divided equally to obtain samples of each variety of wet rice.

[0033] Any sample of wet rice of any variety is divided into equal amounts into subsamples. Each subsample is further divided into subsample groups, with each subsample group containing the same number of subsamples. Different target moisture contents are then set for each sample group. For each subsample within any subsample group, different drying parameters are applied to each subsample in three stages until the target moisture content is reached. After drying, each subsample within the subsample group is then divided into equal amounts to obtain secondary subsamples. Each secondary subsample is further divided into secondary subsample groups, with each secondary subsample group containing the same number of secondary subsamples. The secondary subsamples within each secondary subsample group are then further divided... The samples were stored in their respective storage compartments. The storage environment parameters of each compartment were adjusted. After a preset storage period, each secondary sub-sample was taken out, and the deterioration rate of each secondary sub-sample under different storage conditions was detected. The samples of each variety of wet rice were tested in the same way. After the test, each variety of wet rice, each sample of each variety of wet rice, each sub-sample group and each sub-sample of each variety of wet rice, as well as each secondary sub-sample group and each secondary sub-sample of each variety of wet rice were numbered. Then, the deterioration rate of each variety of wet rice after drying to each drying rate under each drying parameter and then storing under each storage environment parameter was recorded as a training data set.

[0034] It should be noted that the deterioration rate is detected as follows: A fixed number of grains are extracted from any secondary sub-sample, and each grain is identified as moldy, yellowed, browned, insect-infested, or has an off-odor. The number of deteriorated grains is counted against the total number of grains in the corresponding secondary sub-sample, and the deterioration rate is calculated as the percentage of deteriorated grains to the total number of grains.

[0035] Step 2: Hull the rice to be hulled to obtain brown rice.

[0036] It should be noted that secondary impurity removal includes removing residual stones, fine sand, dust, shriveled grains, broken grains, and small impurities such as metal impurities from the rice through vibrating screens and other means.

[0037] In a specific example, the process of hulling the rice is as follows: a hulling test is performed on the rice to be hulled, and then the optimal hulling parameters are determined based on the hulling test. The rice to be hulled is then input into the rice hulling machine, and the rice is hulled according to the optimal hulling parameters.

[0038] It should be noted that the optimal parameters for rice hulling include the clearance between the rubber rollers, the roller linear speed, the roller speed difference, the parallelism and coaxiality of the rubber rollers, etc.

[0039] In a specific example, the dehulling test of the rice to be hulled is carried out as follows: A preset mass of rice samples to be hulled is obtained from the rice to be hulled. The preset mass of rice samples to be hulled is divided equally into dehulling test samples. The hulling parameter range of the rice hulling machine is obtained from the relevant technical manual of the rice hulling machine. The hulling parameter range is divided according to a preset gradient to obtain each dehulling test parameter. Each dehulling test sample is input into the rice hulling machine in sequence, and dehulling is carried out according to each dehulling test parameter. The dehulling rate and broken rice rate of each test sample after dehulling are detected. The dehulling rate and broken rice rate are normalized and then weighted and summed to obtain the comprehensive dehulling coefficient of the rice to be hulled. The dehulling test parameter corresponding to the maximum comprehensive dehulling coefficient is taken as the optimal rice hulling parameter.

[0040] It should be noted that for any hulled test sample after hulling, a quartering method is used to reduce the sample size. Brown rice and unhulled paddy are manually separated grain by grain. The number of brown rice grains and the total number of grains are counted. The hulling rate is calculated as the percentage of brown rice grains to the total number of grains. Similarly, for any hulled test sample after hulling, a quartering method is used to reduce the sample size. Whole rice and broken rice are separated using a standard grading sieve. The weight of whole rice and broken rice is weighed separately. The broken rice rate is calculated as the percentage of broken rice weight to the total weight of the sample. Based on this, the hulling rate and broken rice rate of each hulled test sample can be detected.

[0041] It should be noted that the hulling rate and broken rice rate are normalized and then weighted and summed, with the weights of the normalized hulling rate and broken rice rate both being 0.5.

[0042] Step 3: Mill the unmilled rice to obtain rice to be stored. After removing impurities three times, store the rice in the warehouse.

[0043] It should be noted that the three-stage impurity removal process includes removing impurities such as fine bran, broken rice, and dust through a blower air separator.

[0044] In a specific example, the process of milling the brown rice to be milled is as follows: a milling test is performed on the brown rice to be milled, and then the optimal milling parameters for the brown rice to be milled are determined based on the milling test. The brown rice to be milled is then input into the rice milling machine, and the brown rice to be milled is processed according to the optimal milling parameters.

[0045] It should be noted that the rice milling parameters include the feed flow rate of the rice milling machine, the rotation speed of the milling rollers, the linear speed, the whitening pressure, the rice knife advance and retreat gap, the air volume, the air pressure, and the air speed.

[0046] In a specific example, the process of conducting a rice milling test on the unmilled brown rice is as follows: A predetermined mass of unmilled brown rice sample is obtained from the unmilled brown rice. This predetermined mass of unmilled brown rice sample is then divided equally into several test samples. The rice milling parameter range of the rice milling machine is obtained from the relevant technical specifications of the rice milling machine. This range is then divided according to a predetermined gradient to obtain the specific rice milling test parameters. Each test sample is sequentially input into the rice milling machine, and the rice is milled according to its respective test parameters. The dehulling quality of each test sample after milling is then tested. The dehulling rate, broken rice rate, and moisture content were compared with the moisture content of each test sample after milling after the rice processing decision model was lowered by 2%. When the moisture content of each test sample after milling was the same as the optimal moisture content output by the rice processing decision model (lowered by 2%), the dehulling rate and broken rice rate of each milled rice test sample were obtained. After normalizing the dehulling rate and broken rice rate of each milled rice test sample, a weighted sum was performed to obtain the comprehensive milling coefficient of the brown rice to be milled. The milling parameter corresponding to the maximum comprehensive milling coefficient was taken as the optimal milling parameter.

[0047] It should be noted that any milled rice test sample is divided into quarters using the quartering method. Whole rice and broken rice are separated using a standard grading sieve. The weight of whole rice and broken rice is weighed separately. The broken rice rate is calculated as the percentage of broken rice weight to the total weight of the sample. Similarly, the milled rice sample to be tested is divided into quarters using the quartering method. The total mass of the sample is weighed, and the free bran and bran powder adhering to the surface of the rice grains are blown off. All the bran is collected and weighed. The bran removal rate is calculated as the percentage of the mass of the bran removed to the total mass of the milled rice test sample. Based on this, the broken rice rate and bran removal rate of each milled rice test sample are obtained.

[0048] It should also be noted that the dehulling rate and broken rice rate of each rice milling test sample were normalized and then weighted and summed, with the weight of the normalized dehulling rate and broken rice rate of each test sample being 0.5.

[0049] It should be noted that since the heat generated by the rice milling machine during the rice milling process has a significant impact on the moisture content of the rice, in order to ensure that the change in the moisture content of the rice does not affect the subsequent storage of the rice, it is necessary to select the optimal moisture content output by the rice processing decision model and lower it by 2% so that it is the same as the moisture content of each rice sample to be tested after milling, and then analyze the corresponding rice test samples.

[0050] This invention uses relevant detection devices to monitor rice storage environment parameters, and then analyzes these parameters to obtain the optimal drying parameters and optimal moisture content for wet paddy. After initial impurity removal, the wet paddy is fed into a dryer and dried based on the optimal drying rate and optimal moisture content, yielding paddy ready for hulling. Further analysis yields the optimal hulling and milling parameters, which are then used for hulling and milling. This invention utilizes a rice processing decision model, with the goal of minimizing storage spoilage, to reverse-engineer and output the optimal drying parameters and optimal moisture content. It incorporates the risks of mold and pests into the drying control process upfront, preventing over-drying that increases broken rice rates and under-drying that leads to spoilage after storage.

[0051] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0052] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for intelligent quality improvement and control of the entire rice processing process, characterized in that, Includes the following steps: Step 1: Detect the storage environment parameters of rice using relevant testing devices, and then analyze the optimal drying parameters and optimal moisture content of wet rice based on the storage environment parameters. After removing impurities once, the wet rice is fed into the dryer and dried based on the optimal drying rate and optimal moisture content to obtain rice to be hulled. Step 2: Hull the rice to be hulled to obtain brown rice to be milled; Step 3: Mill the unmilled rice to obtain rice to be stored. After removing impurities three times, store the rice in the warehouse.

2. The intelligent quality improvement and control method for the entire rice processing process according to claim 1, characterized in that, The storage environment parameters include temperature and humidity data, concentration of complex gases, concentration of airborne bacteria, and concentration of particulate matter.

3. The intelligent quality improvement and control method for the entire rice processing process according to claim 2, characterized in that, The specific process of detecting rice storage environment parameters using relevant detection devices is as follows: Sampling points were set up inside the rice storage warehouse using an equidistant grid method. First, temperature and humidity sensors were placed at each sampling point to collect temperature and humidity data. The average temperature and humidity data from all sampling points was used as the warehouse's environmental temperature and humidity data. Next, a portable composite gas detector was placed at each sampling point to collect composite gas concentrations. The average composite gas concentration from all sampling points was used as the warehouse's composite gas concentration. Finally, an airborne bacteria sampler was placed at each sampling point to collect airborne bacteria concentrations. The average airborne bacteria concentration from all sampling points was used as the warehouse's airborne bacteria concentration. Finally, a dust concentration monitor was placed at each sampling point to collect particulate matter concentrations. Several sampling time points are set within a preset period. The average particulate matter concentration of each sampling point collected within the preset period is taken as the particulate matter concentration in the warehouse. The average values ​​of temperature and humidity data, compound gas concentration, airborne bacteria concentration and particulate matter concentration in the warehouse are collectively recorded as the storage environment parameters.

4. The intelligent quality improvement and control method for the entire rice processing process according to claim 3, characterized in that, The optimal drying parameters and optimal moisture content of wet rice were obtained based on the analysis of storage environment parameters. The specific process is as follows: The rice storage environment parameters and rice variety are input into a pre-trained rice processing decision model, which then outputs the optimal drying parameters and optimal moisture content of the wet paddy.

5. The intelligent quality improvement and control method for the entire rice processing process according to claim 4, characterized in that, The training process of the rice processing decision model is as follows: The training dataset for the rice processing decision model was obtained through testing. After preprocessing, the training dataset was divided into training set, test set and validation set according to a preset ratio. MLP is adopted as the core architecture of the drying decision model, and a three-layer network framework for the drying decision model is built. The three-layer network framework includes an input layer, a hidden layer and an output layer. The training set, test set and validation set are input into the three-layer network framework for training, testing and validation to obtain the rice processing decision model.

6. The intelligent quality improvement and control method for the entire rice processing process according to claim 5, characterized in that, The training dataset for the rice processing decision model was obtained through testing, and the specific process is as follows: Samples of wet rice of each variety of the same weight were retrieved from the sample retention center, and the wet rice of each variety was divided into equal portions to obtain samples of each variety of wet rice. Any sample of wet rice of any variety is divided into equal amounts into subsamples. Each subsample is further divided into subsample groups, with each subsample group containing the same number of subsamples. Different target moisture contents are then set for each sample group. For each subsample within any subsample group, different drying parameters are applied to each subsample in three stages until the target moisture content is reached. After drying, each subsample within the subsample group is then divided into equal amounts to obtain secondary subsamples. Each secondary subsample is further divided into secondary subsample groups, with each secondary subsample group containing the same number of secondary subsamples. The secondary subsamples within each secondary subsample group are then further divided... The samples were stored in their respective storage compartments. The storage environment parameters of each compartment were adjusted. After a preset storage period, each secondary sub-sample was taken out, and the deterioration rate of each secondary sub-sample under different storage conditions was detected. The samples of each variety of wet rice were tested in the same way. After the test, each variety of wet rice, each sample of each variety of wet rice, each sub-sample group and each sub-sample of each variety of wet rice, as well as each secondary sub-sample group and each secondary sub-sample of each variety of wet rice were numbered. Then, the deterioration rate of each variety of wet rice after drying to each drying rate under each drying parameter and then storing under each storage environment parameter was recorded as a training data set.

7. The intelligent quality improvement and control method for the entire rice processing process according to claim 6, characterized in that, The specific process for hulling the rice is as follows: The rice to be hulled is tested to determine the optimal hulling parameters. The rice to be hulled is then fed into the rice huller and hulled according to the optimal hulling parameters.

8. The intelligent quality improvement and control method for the entire rice processing process according to claim 7, characterized in that, The specific process for the dehulling test of the rice grains to be hulled is as follows: A predetermined mass of paddy rice samples was obtained from the paddy rice to be hulled. These samples were then divided into equal portions for each hulling test sample. The hulling parameter range for the hulling machine was obtained from its technical specifications. This range was then divided according to a predetermined gradient to obtain the hulling test parameters. Each hulling test sample was sequentially input into the hulling machine, and hulling was performed according to the specified parameters. The hulling rate and broken rice rate of each sample were measured after hulling. The hulling rate and broken rice rate were normalized and then weighted and summed to obtain the comprehensive hulling coefficient of the paddy rice. The hulling test parameters corresponding to the maximum comprehensive hulling coefficient were taken as the optimal hulling parameters.

9. The intelligent quality improvement and control method for the entire rice processing process according to claim 8, characterized in that, The specific process for milling the brown rice is as follows: A rice milling test is conducted on the brown rice to be milled, and the optimal milling parameters are determined based on the test. The brown rice is then fed into the rice milling machine and milled according to the optimal milling parameters.

10. The intelligent quality improvement and control method for the entire rice processing process according to claim 9, characterized in that, The specific process for conducting a rice milling test on the brown rice to be milled is as follows: A predetermined mass of unmilled brown rice samples was obtained from the unmilled rice. These samples were then divided equally into several milling test samples. The milling parameter ranges for the rice milling machine were obtained from its technical specifications. These ranges were then divided according to a predetermined gradient to obtain the milling test parameters. Each test sample was sequentially input into the rice milling machine and milled according to its respective test parameters. The hulling rate, broken rice rate, and moisture content of each milled sample were then measured to determine the rice processing parameters. The optimal moisture content output by the decision model is lowered by 2% and compared with the moisture content of each test sample after milling. When the optimal moisture content output by the rice processing decision model is lowered by 2% and the moisture content of each test sample after milling is the same, the dehulling rate and broken rice rate of each milled rice test sample are taken. After normalizing the dehulling rate and broken rice rate of each milled rice test sample, a weighted sum is taken to obtain the comprehensive milling coefficient of the brown rice to be milled. The milling parameter corresponding to the maximum comprehensive milling coefficient is taken as the optimal milling parameter.

Images