Intelligent analysis method for grain storage process

Abstract

The invention discloses an intelligent analysis method for a grain storage process, and relates to the technical field of intelligent storage, and the method comprises the steps: analyzing the influence degree of environment data on grain quality, recognizing key environment data and secondary environment data, carrying out the abnormal value detection of the key environment data based on a box plot method, whether abnormal value detection is carried out on the secondary environment data or not is judged according to the key environment data abnormal value detection result and the influence degree of the secondary environment data on the grain quality; correcting the abnormal data based on a particle swarm optimization radial basis function neural network algorithm to obtain an abnormal environment factor; the problems that the response to the change of the storage environment is not timely and the grain storage efficiency is low are solved; the grain storage efficiency is improved, the loss is reduced, and the grain quality is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent storage technology, and more specifically, to an intelligent analysis method for grain storage process. Background Art

[0002] Food is the basic material condition for human survival and the fundamental guarantee for the continuous development and progress of human civilization. my country is a populous country. Meeting the food needs of a population of 1.4 billion and ensuring the safety of food storage are not only related to the national economy and people's livelihood, but also have a key significance for the country's modernization. In recent years, with the increasing severity of climate change and environmental pollution, the production and storage of food are facing extremely severe tests. During the storage of food, environmental factors such as temperature, humidity, and CO2 gas can cause food to undergo adverse changes such as mildew and insect pests.

[0003] Deficiencies of existing technologies:

[0004] In the current grain storage process, many storage facilities lack effective temperature and humidity monitoring systems, resulting in grain being stored in unsuitable environmental conditions, which increases the risk of mildew and decay. The response to changes in the storage environment is not timely, resulting in potential problems not being solved in time, increasing the risk of grain loss.

[0005] The setting of storage strategy is difficult to meet the needs of modern management, resulting in low grain storage efficiency. Therefore, the invention provides an intelligent analysis method for grain storage process to solve the problems of untimely response to storage environment changes and low grain storage efficiency; improve grain storage efficiency, reduce losses, and ensure grain quality.

[0006] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent analysis method for a grain storage process, which solves the problems raised in the above-mentioned background technology through intelligent analysis of the grain storage process.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] An intelligent analysis method for a grain storage process comprises the following steps: collecting environmental data, selecting a granary suitable for grain storage according to the environmental data before grain storage; analyzing the influence of environmental data on grain quality, identifying key environmental data and secondary environmental data, performing outlier detection on the key environmental data based on a box plot method, and judging whether to perform outlier detection on the secondary environmental data according to the outlier detection result of the key environmental data combined with the influence of the secondary environmental data on grain quality; correcting the abnormal data based on a particle swarm optimization radial basis function neural network algorithm to obtain an abnormal environmental factor; rotating grain storage based on a first-in-first-out algorithm to obtain storage data, building a storage evaluation model based on a machine learning algorithm in combination with the environmental data, and generating a storage evaluation coefficient, wherein the storage evaluation model is a weighted evaluation model; comparing the storage evaluation coefficient with a preset evaluation safety threshold, analyzing the performance of grain storage, and maintaining the grain storage process.

[0010] In a preferred embodiment, the environmental data includes temperature and humidity, the collecting of environmental data includes collecting temperature and humidity before grain storage and collecting temperature and humidity during grain storage, and the storage data includes storage loss rate.

[0011] In a preferred embodiment, the specific process of selecting a granary suitable for grain storage based on the environmental data before grain storage is as follows: before grain storage, use temperature and humidity sensors to respectively monitor the temperature and humidity in the storage environment, and compare the collected temperature and humidity average values ​​with preset temperature and humidity thresholds; if the temperature and humidity are both lower than the preset thresholds, select a sealed warehouse to store grain; if the temperature is higher than the preset threshold and the humidity is lower than the preset threshold, select an elevated warehouse to store grain; if the temperature is lower than the preset threshold and the humidity is higher than the preset threshold, select a controlled atmosphere warehouse to store grain; if the temperature and humidity are both higher than the preset thresholds, select a controlled atmosphere warehouse to store grain.

[0012] In a preferred embodiment, the specific process of identifying key environmental data and secondary environmental data is as follows: obtain the temperature and humidity values ​​of each historical period within n historical storage periods, and standardize the temperature and humidity values; and measure the mildew rate within n historical storage periods respectively; calculate the Pearson correlation coefficient between the environmental data and the mildew rate respectively, and compare the correlation coefficient between the temperature and the mildew rate and the correlation coefficient between the humidity and the mildew rate, wherein the environmental data includes temperature and humidity; if the absolute value of the correlation coefficient between the temperature and the mildew rate is greater than the absolute value of the correlation coefficient between the humidity and the mildew rate, then the temperature is the key environmental data and the humidity is the secondary environmental data; if the absolute value of the correlation coefficient between the temperature and the mildew rate is less than the absolute value of the correlation coefficient between the humidity and the mildew rate, then the humidity is the key environmental data and the temperature is the secondary environmental data.

[0013] In a preferred embodiment, the specific process of outlier detection on key environmental data based on the box plot method is as follows: collect key environmental data within n historical storage periods, arrange the data in ascending order, find the first quartile and the third quartile, subtract the third quartile from the first quartile to obtain the interquartile range; combine the interquartile range with the absolute value of the Pearson correlation coefficient, and the first quartile and the third quartile to determine the lower limit and upper limit of the outlier, check each point in the data to determine whether it is outside the lower limit and the upper limit, and if it is outside the lower limit and the upper limit, mark it as an outlier.

[0014] In a preferred embodiment, the specific process of judging whether to perform outlier detection on the secondary environmental data based on the outlier detection results of the key environmental data in combination with the degree of influence of the secondary environmental data on the grain quality is as follows: obtaining the number of outliers in the key environmental data, dividing the number of outliers by the total number to obtain the outlier ratio of the key environmental data; defining the outlier ratio of the key environmental data and the absolute value of the Pearson correlation coefficient of the secondary environmental data as input variables, and dividing them into different fuzzy sets respectively; defining whether outlier detection is needed for the secondary environment as an output variable, and dividing them into fuzzy sets; formulating fuzzy rules to describe the influence of the definition of the outlier ratio of the key environmental data and the absolute value of the Pearson correlation coefficient of the secondary environmental data on whether outlier detection is needed for the secondary environment; performing fuzzy reasoning according to the fuzzy rules to judge whether to perform outlier detection on the secondary environmental data.

[0015] In a preferred embodiment, the specific process of correcting abnormal data based on the particle swarm optimization radial basis function neural network algorithm to obtain abnormal environmental factors is as follows: inputting key environmental data as environmental factors into the radial basis function neural network, using the radial basis function as the activation function in the hidden layer, and outputting the corrected environmental factors; using the k-means clustering algorithm to determine the hidden layer center and initialize the radial basis function; updating the weight and bias combination of the radial basis function neural network based on the particle swarm optimization algorithm, and using the weight and bias to train the radial basis function neural network until the maximum number of iterations; inputting the detected abnormal data into the trained radial basis function neural network to obtain the corrected environmental factors.

[0016] In a preferred embodiment, the storage assessment coefficient is compared with a preset assessment safety threshold, the performance of grain storage is analyzed, and the grain storage process is maintained. The specific process is as follows: the storage assessment coefficient is compared with the preset assessment safety threshold. When the storage assessment coefficient is greater than the preset assessment safety threshold, the grain storage performance is poor, and it is necessary to check the storage environment, strengthen inventory management, analyze the causes of loss, and take improvement and maintenance measures; when the storage assessment coefficient is less than the preset assessment safety threshold, the grain storage process performance is good, the freshness and quality of the grain are guaranteed, the existing storage strategy is maintained, and regular monitoring and maintenance are performed to improve storage efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a schematic diagram of the structure of an intelligent analysis method for grain storage process. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Embodiment 1, Figure 1 The invention provides an intelligent analysis method for grain storage process.

[0020] S10, collecting environmental data, and selecting a granary suitable for grain storage according to the environmental data before grain storage;

[0021] The environmental data includes temperature and humidity, and collecting environmental data at different stages includes collecting temperature and humidity before grain storage and collecting temperature and humidity during grain storage;

[0022] First, before grain is stored, the temperature and humidity in the storage environment are monitored using temperature and humidity sensors. The collection frequency is set to once an hour, and the average value of the collected temperature and humidity within the preset period is calculated, and the collected average value of temperature and humidity is compared with the preset temperature and humidity thresholds.

[0023] If the temperature and humidity are both below the preset threshold, it is suitable to use a sealed warehouse, which can effectively prevent mildew and insect pests;

[0024] If the temperature is greater than the preset threshold and the humidity is less than the preset threshold, choose a well-ventilated high-bay warehouse to keep it dry;

[0025] If the temperature is lower than the preset threshold and the humidity is higher than the preset threshold, consider adjusting the air in the warehouse to prevent moisture from increasing and causing mildew;

[0026] If both the temperature and humidity are greater than the preset thresholds, a controlled atmosphere is required to maintain a stable storage environment.

[0027] Choosing the right granary based on the temperature and humidity before grain storage can bring several benefits:

[0028] Prolong shelf life: Suitable temperature and relative humidity can effectively slow down the respiration and biochemical reaction of grains, prolong the shelf life of grains and maintain their freshness;

[0029] Reduce pests: High temperature and high humidity are often breeding grounds for pests. By choosing dry and well-ventilated granaries, the reproduction of pests can be significantly reduced, thus protecting grain from pests.

[0030] Maintain nutritional value: Improper temperature and humidity will lead to the loss of vitamins and minerals in grains. Stable storage conditions can better maintain the nutritional components of grains and ensure that the nutritional value of grains is not damaged;

[0031] Reduce economic losses: An unsuitable storage environment may cause grain to spoil, become moldy or be infested with insects, thus causing direct economic losses. By choosing a suitable granary, losses can be reduced and economic benefits can be improved;

[0032] Improve storage efficiency: Choosing the right granary helps optimize storage space, arrange grain stacking reasonably, improve management efficiency, and reduce waste of manpower and material resources;

[0033] Ensure food safety: A suitable environment can effectively inhibit the growth of harmful microorganisms, reduce food safety risks, and ensure the safety of food throughout the storage process;

[0034] Easy to monitor and manage: Suitable granaries are easy to install temperature and humidity monitoring equipment to monitor storage conditions in real time and adjust storage strategies in time to ensure grain quality;

[0035] To sum up, the rational selection of granaries can not only improve the storage quality of grain, but also reduce losses, ensure food safety, and provide effective support for grain management.

[0036] S20, analyzing the impact of environmental data on food quality, identifying key environmental data and secondary environmental data, performing outlier detection on key environmental data based on the box plot method, and determining whether to perform outlier detection on secondary environmental data based on the outlier detection results of key environmental data and the impact of secondary environmental data on food quality;

[0037] Obtain the temperature and humidity values ​​of each historical period within n historical storage periods. The temperature value is the average temperature value collected within a historical period, and the humidity value is the average humidity value collected within a historical period. Since the units of temperature and humidity are different, the temperature and humidity values ​​need to be standardized;

[0038] And measure the mildew rate in n historical storage cycles respectively;

[0039] The Pearson correlation coefficient is calculated for the environmental data and the mildew rate. The closer the correlation coefficient is to 1 or -1, the greater the influence of the factor on the mildew rate. The environmental data includes temperature and humidity. The calculation formula using the Pearson correlation coefficient is as follows:

[0040]

[0041] In the formula, x represents environmental data, y represents mildew rate, and n is the number of samples;

[0042] The Pearson correlation coefficients between temperature and humidity and the mildew rate were calculated respectively, and the correlation coefficients between temperature and mildew rate and humidity and mildew rate were compared;

[0043] If the absolute value of the correlation coefficient between temperature and mildew rate is greater than the absolute value of the correlation coefficient between humidity and mildew rate, temperature is the key environmental data and humidity is the secondary environmental data;

[0044] If the absolute value of the correlation coefficient between temperature and mildew rate is less than the absolute value of the correlation coefficient between humidity and mildew rate, humidity is the key environmental data and temperature is the secondary environmental data;

[0045] According to the historical key environmental data, the key environmental data is detected based on the box plot method. The specific process is as follows:

[0046] Collect key environmental data within n historical storage cycles, sort the data in ascending order, find the 25% position as the first quartile, find the 75% position as the third quartile, subtract the third quartile from the first quartile to get the interquartile range; combine the interquartile range with the absolute value of the Pearson correlation coefficient, the first quartile and the third quartile to determine the lower and upper limits of the outliers, check each point in the data, determine whether it is outside the lower and upper limits, and mark it as an outlier;

[0047] The calculation formulas for the lower limit and upper limit of the abnormal value are as follows:

[0048]

[0049] Where X is the lower limit of the outlier, S is the upper limit of the outlier, Q1 is the first quartile, Q3 is the third quartile, IQR is the interquartile range, and |r| is the absolute value of the Pearson correlation coefficient;

[0050] Obtain the number of outliers in n key environmental data, and divide the number of outliers by the total number to obtain the outlier ratio of the key environmental data;

[0051] Step C1, define the abnormal value ratio of key environmental data and the absolute value of the Pearson correlation coefficient of secondary environmental data as input variables, and divide them into different fuzzy sets respectively.

[0052] For example, "Low", "Medium", "High" are the proportion of outliers for key environmental data, and "Low", "Medium", "High" are the absolute values ​​of Pearson correlation coefficients for secondary environmental data.

[0053] Step C2, whether the secondary environment needs to perform outlier detection is defined as an output variable, which is divided into fuzzy sets, for example, "Yes", "No", for whether the secondary environment needs to perform outlier detection.

[0054] Step C3, formulate a set of fuzzy rules to describe the impact of different input variables on output variables. The definition of rules can be based on professional knowledge or obtained through data analysis and experiments. For example:

[0055] The abnormal value ratio of key environmental data is marked as B, and the absolute value of the Pearson correlation coefficient of secondary environmental data is marked as |r c |, whether the secondary environment needs to perform outlier detection is marked as P and can be defined

[0056] Rule 1:IF(B is Low)AND(|r c |is Low)THEN(P is No)

[0057] Rule 2:IF(B is High)AND(|r c |is High)THEN(P is Yes) ...

[0059] Step C4, perform fuzzy reasoning according to fuzzy rules to determine the secondary environment outlier detection scheme. It should be noted that the division of fuzzy sets can be adjusted according to actual conditions. For example, although this embodiment takes three fuzzy sets as examples, in fact, the outlier ratio of key environmental data, the absolute value of the Pearson correlation coefficient of secondary environmental data, and whether the secondary environment needs outlier detection can be divided into more than three sets to facilitate better and more accurate identification.

[0060] Furthermore, for the judgment of the proportion of abnormal values ​​of key environmental data and the absolute value of the Pearson correlation coefficient of secondary environmental data, the threshold can be set according to the actual situation; when the proportion of abnormal values ​​of key environmental data is higher than 80%, it will be marked as "High", and when the absolute value of the Pearson correlation coefficient of secondary environmental data is higher than 0.7, it will be marked as "High", and so on. I will not go into details here.

[0061] S30, correcting the abnormal data based on the particle swarm optimization radial basis function neural network algorithm to obtain the abnormal environmental factor;

[0062] The specific process of correcting abnormal data based on the particle swarm optimization radial basis function neural network algorithm and obtaining abnormal environmental factors is as follows:

[0063] The key environmental data are input into the radial basis neural network as environmental factors, and the radial basis function is used as the activation function in the hidden layer to output the corrected environmental factors;

[0064] The k-means clustering algorithm is used to determine the hidden layer center and initialize the radial basis function;

[0065] The weight and bias combination of the radial basis neural network is updated based on the particle swarm optimization algorithm, and the radial basis neural network is trained using the weight and bias until the maximum number of iterations;

[0066] The detected abnormal data is input into the trained radial basis neural network to obtain the corrected environmental factors.

[0067] When analyzing grain storage processes, the benefits of outlier monitoring and correction of key environmental factors include:

[0068] Prevent food spoilage: monitor temperature, humidity and other factors to detect abnormalities in time and prevent food from getting damp or spoiling;

[0069] Improve storage efficiency: Through monitoring, optimize storage conditions to ensure that grain is stored in the best environment and reduce losses;

[0070] Ensure food safety: Abnormal value monitoring can help detect harmful substances or pollution in a timely manner and protect consumer health;

[0071] Data-driven decision-making: Real-time monitoring provides important data support to help managers make scientific decisions and improve storage management level;

[0072] Reduce economic losses: Correct abnormal situations in a timely manner, reduce food losses caused by environmental factors, and improve economic benefits.

[0073] S40, rotating the storage of grain based on a first-in-first-out algorithm, acquiring storage data, building a storage evaluation model based on a machine learning algorithm in combination with environmental data, and generating a storage evaluation coefficient, wherein the storage evaluation model is a weighted evaluation model;

[0074] The storage data includes storage loss rate, and the first-in-first-out algorithm ensures that the first stored grain is first out of the warehouse, reducing the risk of corruption and deterioration caused by long-term storage, and reducing the amount of loss and scrap;

[0075] Obtain the amount of food loss due to spoilage, pests, etc., and the amount of food scrapped due to expiration or substandard quality, and add the loss amount and scrap amount to get the loss amount; obtain the total storage volume within a preset period; divide the loss amount by the total storage volume to get the loss rate;

[0076] The benefits of rotating grain storage based on the first-in-first-out (FIFO) algorithm include:

[0077] Reduce the risk of spoilage: ensure that the first stored grain is released first, reducing the risk of spoilage and deterioration caused by long-term storage;

[0078] Ensure freshness: Ensure consumers have access to fresh food and improve product quality through storage rotation;

[0079] Effective inventory management: FIFO helps to better manage inventory, avoid expiration and backlog, and improve storage efficiency;

[0080] Reduce economic losses: timely handle food that is about to expire to reduce the economic burden caused by losses;

[0081] Comply with regulatory requirements: Some areas have legal requirements for food storage, and FIFO can ensure compliance with relevant standards;

[0082] Combine storage data with environmental data to build a storage assessment model based on machine learning algorithms to generate storage assessment coefficients;

[0083] The specific calculation formula of the storage evaluation coefficient is as follows:

[0084] PG=μ1|T-T0|+μ2|D-D0|+μ3H;

[0085] Wherein, PG is the storage evaluation coefficient, T is the storage environment temperature, T0 is the environment temperature reference value, D is the storage environment humidity, D0 is the environment humidity reference value, H is the storage loss rate; μ1 is the temperature weight coefficient, μ2 is the humidity weight coefficient, and μ3 is the storage loss rate weight coefficient.

[0086] S50, comparing the storage assessment coefficient with a preset assessment safety threshold, analyzing the performance of grain storage, and maintaining the grain storage process.

[0087] The storage assessment coefficient is compared with the preset assessment safety threshold. When the storage assessment coefficient is greater than the preset assessment safety threshold, the grain storage performance is poor, and it is necessary to check the storage environment, strengthen inventory management, analyze the causes of losses, and take improvement and maintenance measures; when the storage assessment coefficient is less than the preset assessment safety threshold, the grain storage process performance is good, the freshness and quality of the grain are guaranteed, the existing storage strategy is maintained, and regular monitoring and maintenance are carried out to improve storage efficiency.

[0088] The storage assessment coefficient is compared with the preset assessment safety threshold, and the performance of grain storage is analyzed based on the comparison results. The functions of maintaining the grain storage process are as follows:

[0089] Ensure food safety: Through regular inspection and monitoring, ensure that the storage process meets safety standards and reduce food safety risks;

[0090] Improve storage efficiency: Improve inventory turnover and reduce loss rate by optimizing storage processes and management strategies;

[0091] Enhance customer satisfaction: Provide fresh and high-quality food products to improve consumer trust and satisfaction;

[0092] Reduce economic losses: timely identify problems and take measures to reduce economic losses caused by loss or expiration.

[0093] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0094] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0095] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0096] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0097] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0098] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent analysis method for grain storage process, characterized in that: The following steps are involved: Collect environmental data and select granaries suitable for grain storage based on the environmental data before grain storage; Analyze the impact of environmental data on food quality, identify key environmental data and secondary environmental data, perform outlier detection on key environmental data based on the box plot method, and determine whether to perform outlier detection on secondary environmental data based on the outlier detection results of key environmental data and the impact of secondary environmental data on food quality; Based on the particle swarm optimization radial basis function neural network algorithm, the abnormal data is corrected to obtain the abnormal environmental factors; The grain is rotated and stored based on the first-in-first-out algorithm, the storage data is obtained, and a storage evaluation model is constructed based on a machine learning algorithm in combination with environmental data to generate a storage evaluation coefficient, wherein the storage evaluation model is a weighted evaluation model; The storage assessment coefficient is compared with the preset assessment safety threshold, the performance of grain storage is analyzed, and the grain storage process is maintained.

2. The intelligent analysis method for grain storage process according to claim 1, characterized in that: The environmental data includes temperature and humidity, the collecting of environmental data includes collecting temperature and humidity before grain storage and collecting temperature and humidity during grain storage, and the storage data includes storage loss rate.

3. The intelligent analysis method for grain storage process according to claim 2, characterized in that: The specific process of selecting a granary suitable for grain storage based on the environmental data before grain storage is as follows: Before grain is stored, the temperature and humidity in the storage environment are monitored using temperature and humidity sensors, and the collected average temperature and humidity values ​​are compared with the preset temperature and humidity thresholds; If the temperature and humidity are both lower than the preset threshold, choose to store grain in a sealed warehouse; If the temperature is greater than the preset threshold and the humidity is less than the preset threshold, choose the high-bay warehouse to store the grain; If the temperature is lower than the preset threshold and the humidity is higher than the preset threshold, the gas-controlled storage is selected to store the grain; If the temperature and humidity are both greater than the preset thresholds, choose to store grain in a controlled atmosphere warehouse.

4. The intelligent analysis method for grain storage process according to claim 3 is characterized in that: The specific process of identifying key environmental data and secondary environmental data is as follows: Obtain the temperature and humidity values ​​of each historical period within n historical storage periods, and perform standardization on the temperature and humidity values; And measure the mildew rate in n historical storage cycles respectively; Calculating the Pearson correlation coefficient between environmental data and mildew rate respectively, and comparing the correlation coefficient between temperature and mildew rate and the correlation coefficient between humidity and mildew rate, wherein the environmental data includes temperature and humidity; If the absolute value of the correlation coefficient between temperature and mildew rate is greater than the absolute value of the correlation coefficient between humidity and mildew rate, temperature is the key environmental data and humidity is the secondary environmental data; If the absolute value of the correlation coefficient between temperature and mildew rate is smaller than the absolute value of the correlation coefficient between humidity and mildew rate, humidity is the key environmental data and temperature is the secondary environmental data.

5. The intelligent analysis method for grain storage process according to claim 4, characterized in that: The specific process of detecting outliers on key environmental data based on the box plot method is as follows: Collect key environmental data within n historical storage cycles, sort the data in ascending order, find the first quartile and the third quartile, and subtract the third quartile from the first quartile to obtain the interquartile range; The interquartile range and the absolute value of the Pearson correlation coefficient are combined with the first quartile and the third quartile to determine the lower and upper limits of the outliers. Each point in the data is checked to determine whether it is outside the lower and upper limits. If it is outside the lower and upper limits, it is marked as an outlier.

6. The intelligent analysis method for grain storage process according to claim 5, characterized in that: The specific process of judging whether to perform outlier detection on the secondary environmental data according to the outlier detection results of the key environmental data and the degree of influence of the secondary environmental data on the grain quality is as follows: Obtain the number of outliers in the key environmental data, and divide the number of outliers by the total number to obtain the outlier ratio of the key environmental data; The proportion of abnormal values ​​of key environmental data and the absolute value of Pearson correlation coefficient of secondary environmental data are defined as input variables, and they are divided into different fuzzy sets respectively; Whether the secondary environment needs outlier detection is defined as the output variable and divided into fuzzy sets; Formulate fuzzy rules to describe the impact of the proportion of outliers in key environmental data and the absolute value of the Pearson correlation coefficient of secondary environmental data on whether outlier detection is needed in the secondary environment; Fuzzy reasoning is performed based on fuzzy rules to determine whether to perform outlier detection on secondary environmental data.

7. The intelligent analysis method for grain storage process according to claim 6, characterized in that: The specific process of correcting abnormal data based on the particle swarm optimization radial basis function neural network algorithm to obtain abnormal environmental factors is as follows: The key environmental data are input into the radial basis neural network as environmental factors, and the radial basis function is used as the activation function in the hidden layer to output the corrected environmental factors; The k-means clustering algorithm is used to determine the hidden layer center and initialize the radial basis function; The weight and bias combination of the radial basis neural network is updated based on the particle swarm optimization algorithm, and the radial basis neural network is trained using the weight and bias until the maximum number of iterations; The detected abnormal data is input into the trained radial basis neural network to obtain the corrected environmental factors.

8. The intelligent analysis method for grain storage process according to claim 7, characterized in that: The specific process of comparing the storage evaluation coefficient with the preset evaluation safety threshold, analyzing the performance of grain storage, and maintaining the grain storage process is as follows: Compare the storage assessment coefficient with the preset assessment safety threshold. When the storage assessment coefficient is greater than the preset assessment safety threshold, the grain storage performance is poor, and it is necessary to check the storage environment, strengthen inventory management, analyze the causes of loss, and take improvement and maintenance measures; When the storage assessment coefficient is less than the preset assessment safety threshold, the grain storage process performance is good, the freshness and quality of the grain are guaranteed, the existing storage strategy is maintained, and regular monitoring and maintenance are carried out to improve storage efficiency.

9. The intelligent analysis method for grain storage process according to claim 8, characterized in that: The Pearson correlation coefficient calculation formula is as follows: In the formula, x represents environmental data, y represents mildew rate, and n is the number of samples.

10. The intelligent analysis method for grain storage process according to claim 9, characterized in that: The calculation formulas for the lower limit and upper limit of the abnormal value are as follows: Where X is the lower limit of the outlier, S is the upper limit of the outlier, Q1 is the first quartile, Q3 is the third quartile, IQR is the interquartile range, and |r| is the absolute value of the Pearson correlation coefficient.

Images