Data processing method and device

Abstract

The invention discloses a data processing method and device, relates to the field of data processing, and is used for monitoring a storage environment in real time and realizing intelligent control on the storage environment in time. The method comprises the steps that first environment data of a storage area and a non-storage area in a first time period are obtained, the first environment data comprise one or more of temperature data, humidity data and gas concentration data, the non-storage area is located in a set distance range outside the storage area, and the first environment data of the storage area and the non-storage area are acquired according to the first environment data and a preset big data model; and predicting second environment data of a second time period, and when the second environment data does not meet a preset storage condition, adjusting the current environment data of the storage area. Through the scheme, the storage environment can be intelligently monitored and adjusted.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a data processing method and device. Background Art

[0002] As global food security issues become increasingly prominent, the importance of food storage management continues to increase. When managing food storage, it is necessary to monitor the temperature and humidity of the food. At present, the food temperature and humidity monitoring system uses temperature and humidity sensors to passively collect temperature and humidity data in the food storage environment. However, this solution lacks the ability of real-time monitoring and intelligent control.

[0003] Therefore, how to intelligently monitor and adjust the storage environment of grain is an issue that needs to be urgently addressed. Summary of the invention

[0004] The present invention provides a data processing method and device for real-time monitoring of a storage environment and timely intelligent control of the storage environment.

[0005] In a first aspect, the present invention provides a data processing method, which includes: obtaining first environmental data of a storage area and a non-storage area within a first time period, the first environmental data including one or more of temperature data, humidity data and gas concentration data, the non-storage area is located within a set distance range outside the storage area, and based on the first environmental data and a preset big data model, predicting second environmental data of a second time period, and when the second environmental data does not meet the preset storage conditions, adjusting the current environmental data of the storage area.

[0006] Through the above scheme, real-time monitoring and early warning of the temperature and humidity of the storage environment can be achieved, the safety and stability of grain storage can be improved, and by using big data technology to analyze and process data, situational awareness of the storage environment can be achieved, and potential temperature and humidity problems of grain can be discovered in time, which can provide a scientific basis and decision-making support for grain storage management.

[0007] Optionally, the temperature data includes one or more of the temperature value of the storage area at each time node, the temperature value of the non-storage area at each time node, and the temperature value of a set location in the storage area; the humidity data includes one or more of the humidity value of the storage area at each time node, the humidity value of the non-storage area at each time node, and the humidity value of a set location in the storage area; before predicting the second environmental data of the second time period, it also includes: deleting abnormal values ​​in the first environmental data, the abnormal values ​​include any one of the temperature value less than the minimum temperature threshold in the temperature data, the temperature value greater than the maximum temperature threshold in the temperature data, the relative humidity less than the minimum humidity threshold in the humidity data, and the relative humidity greater than the maximum humidity threshold in the humidity data; if the data of the first area in the first environmental data at the first time node is a null value, historical data in the historical environmental data whose characteristic information is similar to the characteristic information of the null value is obtained, and the data of the first area at the first time node is corrected using the most recent historical data.

[0008] Through the above solution, the accuracy, completeness and reliability of the first environment data can be ensured, providing a high-quality data foundation for subsequent data analysis.

[0009] Optionally, the first environmental data includes temperature data and / or humidity data, and the big data model is a logistic regression model. Based on the first environmental data and a preset big data model, the second environmental data of the second time period is predicted, including: extracting multiple feature information in the first environmental data, calculating the first correlation between each feature information and grain moldy, screening out the first feature information with a first correlation higher than a preset threshold from the multiple feature information, using the first feature information to train the logistic regression model, and adjusting the parameters of the logistic regression model according to the training results, repeating this step until the value of the logistic regression model under the preset indicator reaches the set value, and predicting the second environmental data of the second time period according to the adjusted logistic regression model.

[0010] Through the above scheme, the classification problem is solved by the logistic regression model to determine whether the grain may be moldy. Through continuous data accumulation and algorithm optimization, the generalization ability of the model can be increased to adapt to the special needs of different grain storage environments.

[0011] Optionally, the first environmental data includes temperature data and / or humidity data, and the big data model is a linear regression model. The second environmental data for the second time period is predicted based on the first environmental data and a preset big data model, including: extracting multiple feature information from the first environmental data, calculating the second correlation between each feature information and the pest, screening out second feature information having a second correlation higher than a preset threshold from the multiple feature information, training the linear regression model using the second feature information, and adjusting the parameters of the linear regression model based on the training results, repeating this step until the value under the preset indicator of the linear regression model reaches the set value, and predicting the second environmental data for the second time period based on the adjusted linear regression model.

[0012] Through the above scheme, the regression problem is processed by the linear regression model, which can predict the development trend and degree of pests. Through continuous data accumulation and algorithm optimization, the generalization ability of the model can be increased to adapt to the special needs of different grain storage environments.

[0013] Optionally, the first environmental data is gas concentration data, and predicting the second environmental data for the second time period includes: obtaining a safety concentration threshold corresponding to a set gas, inputting the safety concentration threshold and the first environmental data into a big data model, and obtaining a prediction result output by the big data model, the prediction result including one or more of the following: whether the current gas concentration of the set gas is higher than the safety threshold, the future gas concentration fluctuation of the set gas, the correlation between the gas concentration of the set gas and other factors, and other factors including one or more of temperature, humidity, and wind speed.

[0014] Through the above scheme, by analyzing the gas concentration data, it is possible to determine whether the gas environment in the storage area is safe, that is, through concentration comparison analysis, various gas indicators are monitored in real time to ensure the safety of the grain storage environment, and through trend analysis, the changing trends of key indicators are tracked to timely discover potential risks. Through correlation analysis, the internal connections between various indicators are deeply studied to optimize the early warning mechanism.

[0015] Optionally, after predicting the second environmental data of the second period, the method further includes: if the second environmental data is higher than a preset value, an alarm message is issued, and abnormal data is displayed in real time. In this way, it can ensure that management personnel can discover and handle various abnormal situations in a timely manner, greatly improving the safety and management efficiency of grain storage.

[0016] In a second aspect, the present invention provides a data processing device, the device comprising:

[0017] An acquisition module, used to acquire first environmental data of a storage area and a non-storage area within a first period of time, the first environmental data including one or more of temperature data, humidity data and gas concentration data, and the non-storage area is located within a set distance outside the storage area;

[0018] A prediction module, used to predict second environmental data in a second period of time based on the first environmental data and a preset big data model;

[0019] The adjustment module is used to adjust the current environment data of the storage area when the second environment data does not meet the preset storage conditions.

[0020] In one possible implementation, the temperature data includes one or more of the temperature value of the storage area at each time node, the temperature value of the non-storage area at each time node, and the temperature value of a set location in the storage area. The humidity data includes one or more of the humidity value of the storage area at each time node, the humidity value of the non-storage area at each time node, and the humidity value of a set location in the storage area. The acquisition module is also used to: delete abnormal values ​​in the first environmental data, and the abnormal values ​​include any one of a temperature value less than a minimum temperature threshold in the temperature data, a temperature value greater than a maximum temperature threshold in the temperature data, a relative humidity less than a minimum humidity threshold in the humidity data, and a relative humidity greater than a maximum humidity threshold in the humidity data. If the data of the first area in the first environmental data at the first time node is a null value, historical data in the historical environmental data whose characteristic information is similar to the characteristic information of the null value is obtained, and the data of the first area at the first time node is corrected using the most recent historical data.

[0021] In one possible implementation, the first environmental data includes temperature data and / or humidity data, the big data model is a logistic regression model, and the prediction module is specifically used to: extract multiple feature information in the first environmental data, calculate the first correlation between each feature information and grain moldy, screen out the first feature information with a first correlation higher than a preset threshold from the multiple feature information, use the first feature information to train the logistic regression model, and adjust the parameters of the logistic regression model according to the training results, repeat this step until the value of the logistic regression model under the preset indicator reaches the set value, and predict the second environmental data of the second time period according to the adjusted logistic regression model.

[0022] In one possible implementation, the first environmental data includes temperature data and / or humidity data, the big data model is a linear regression model, and the prediction module is specifically used to: extract multiple feature information in the first environmental data, calculate the second correlation between each feature information and the pest, screen out second feature information with a second correlation higher than a preset threshold from the multiple feature information, use the second feature information to train the linear regression model, and adjust the parameters of the linear regression model according to the training results, repeat this step until the value under the preset indicator of the linear regression model reaches the set value, and predict the second environmental data of the second time period according to the adjusted linear regression model.

[0023] In one possible implementation, the first environmental data is gas concentration data, and the prediction module is specifically used to: obtain a safety concentration threshold corresponding to a set gas, input the safety concentration threshold and the first environmental data into a big data model, and obtain a prediction result output by the big data model, wherein the prediction result includes one or more of the following: whether the current gas concentration of the set gas is higher than the safety threshold, the future gas concentration fluctuation of the set gas, and the correlation between the gas concentration of the set gas and other factors, and other factors include one or more of temperature, humidity, and wind speed.

[0024] In a possible implementation, the adjustment module is further used to: if the second environmental data is higher than a preset value, issue an alarm message and display the abnormal data in real time.

[0025] In a third aspect, the present invention further provides a data processing device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, it implements the method described in the various possible designs of the first aspect.

[0026] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program or instructions stored thereon. When the computer program or instructions are executed by a processor, the method described in the various possible designs of the first aspect is implemented.

[0027] In a fifth aspect, the present invention further provides a computer program product, which, when executed on a computer, enables the computer to execute any of the methods described in the first aspect above.

[0028] These implementations or other implementations of the present application will be more concise and understandable in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 A flowchart of a data processing method provided by an embodiment of the present invention;

[0031] Figure 2 A schematic diagram of a broken line of temperature in a first time period provided by an embodiment of the present invention;

[0032] Figure 3 A schematic diagram of a process for predicting the mildew condition of grain in a second period of time provided by an embodiment of the present invention;

[0033] Figure 4 A schematic diagram of a process for predicting the pest situation in the second period provided by an embodiment of the present invention;

[0034] Figure 5 A schematic diagram of a flow chart for predicting the gas concentration in the second time period provided by an embodiment of the present invention;

[0035] Figure 6 A schematic diagram of real-time display of data provided by an embodiment of the present invention;

[0036] Figure 7 A schematic diagram of a model for early warning of abnormal data provided by an embodiment of the present invention;

[0037] Figure 8 A schematic diagram of a data processing system provided by an embodiment of the present invention;

[0038] Fig. 9 A schematic diagram of a data processing device provided by an embodiment of the present invention;

[0039] Fig.10 A schematic diagram of another data processing device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and beneficial effects of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0042] The grain temperature and humidity monitoring system currently on the market uses sensing devices such as temperature and humidity sensors to passively collect temperature and humidity data in the grain storage environment. However, this traditional method has many disadvantages: for example: (1) Low frequency of data collection: Usually only periodic data collection is possible, and continuous real-time monitoring cannot be achieved; (2) Imperfect early warning mechanism: Lack of intelligent analysis and prediction capabilities, making it difficult to detect potential problems in a timely manner; (3) Limited control capabilities: Unable to automatically adjust storage conditions according to environmental changes, requiring manual intervention; (4) Low data utilization: The collected data is often not fully utilized, lacking in-depth analysis and decision support. Therefore, how to intelligently monitor and adjust the grain storage environment is an urgent problem that needs to be solved.

[0043] Based on this, the present invention provides a data processing method, which predicts future environmental data in a storage area through a preset big data model, and adjusts current environmental data in the storage area according to the predicted future environmental data, so as to realize intelligent monitoring and adjustment of the storage environment.

[0044] The present application scheme is described in detail below in conjunction with the accompanying drawings.

[0045] See also Figure 1 , a flow chart of a data processing method is shown, which specifically includes the following steps:

[0046] Step 110: Acquire first environmental data of the storage area and the non-storage area in a first period of time.

[0047] Here, the first environmental data includes one or more of temperature data, humidity data, and gas concentration data, and the non-storage area is located within a set distance range outside the storage area.

[0048] Optionally, various sensors can be used to collect the first environmental data, and the collected first environmental data can be cleaned. For example, by deploying various advanced sensors, such as high-precision temperature and humidity sensors, gas concentration detectors, electromagnetic wave detectors, etc., multi-dimensional environmental data such as temperature, humidity, gas concentration, and pest activity inside the granary can be collected in real time, and the collected raw data can be preliminarily cleaned and preprocessed to remove obvious outliers and noise, thereby improving the accuracy and efficiency of subsequent data analysis. The following are the steps for cleaning different data:

[0049] (1) For bad pixel value.

[0050] It can be understood that bad pixel values ​​usually refer to obviously erroneous data caused by equipment failure, transmission errors, etc., and therefore need to be deleted or repaired.

[0051] Optionally, the first environment data can be screened using data quality assessment indicators to obtain bad pixel value data that does not meet the data quality assessment indicators, and the screened out bad pixel value data can be deleted. Alternatively, the screened out bad pixel value data can be manually reviewed based on the experience of professionals in this field. If it is confirmed to be bad pixel value data after review, the bad pixel value data will be deleted.

[0052] (2) For null values.

[0053] It can be understood that the null value means that the data of the first area at the first time node in the first environment data is null. If there is null value data in the first environment data, historical data with feature information similar to the feature information of the null value in the historical environment data is obtained, and the data of the first area at the first time node is corrected using the most recent historical data.

[0054] Here, the characteristic information includes one or more of time characteristics, area characteristics, and data characteristics, and the data characteristics include one or more of the temperature of the storage area, the temperature of the non-storage area, the humidity of the storage area, the humidity of the non-storage area, and the temperature of a set location in the storage area.

[0055] Optionally, after the most recent historical data is treated as null value data, the rationality of the filled data can be verified.

[0056] (3) Targeting outliers.

[0057] As you can understand, outliers are data that, while the data itself may be real, is outside the normal range and therefore needs to be marked and analyzed.

[0058] Optionally, the abnormal values ​​in the first environmental data can be counted, and an abnormal value processing log recording system can be established to process the abnormal values. Among them, the abnormal value includes any one of a temperature value less than the minimum temperature threshold in the temperature data, a temperature value greater than the maximum temperature threshold in the temperature data, a relative humidity less than the minimum humidity threshold in the humidity data, and a relative humidity greater than the maximum humidity threshold in the humidity data. For example, the minimum temperature threshold is -25°C, the maximum temperature threshold is 45°C, the minimum humidity threshold is 1%, and the maximum humidity threshold is 100%. If the temperature in the first environmental data is between -25°C and 45°C, and the relative humidity is between 1% and 100%, then the data is normal data; conversely, if the temperature in the first environmental data is less than -25°C or greater than 45°C, and the relative humidity is less than 1% or greater than 100%, then the data is abnormal data and needs to be recorded and processed.

[0059] Through the above solution, the accuracy, completeness and reliability of the first environment data can be ensured, providing a high-quality data foundation for subsequent data analysis.

[0060] Optionally, the first environment data can be stored in a distributed manner, for example, the data can be stored in a distributed manner according to time, region or data type. Since the first environment data has a large amount of data and needs to support high-frequency access operations, the first environment data can be stored in an efficient, flexible and reliable distributed system, so that the speed and efficiency of data access can be improved.

[0061] Optionally, the first environment data can be uploaded to the cloud in a variety of network forms, including wired networks (such as Ethernet, optical fiber networks) and wireless networks (such as Wi-Fi, 4G / 5G, LoRa and other low-power wide area network technologies) to ensure real-time and reliable data transmission. In this way, it can adapt to different granary environments and network conditions to ensure the continuity and stability of data transmission.

[0062] Through the above solution, the first environment data is stored in the cloud, which can reduce the burden on the server and support the storage and efficient access of large amounts of data.

[0063] Optionally, multiple backup mechanisms and data encryption technologies, such as an asymmetric encryption algorithm (Rivest-Shamir-Adleman, RSA), may be used to encrypt the first environment data to ensure data security and reliability.

[0064] Step 120: Predict second environmental data for a second period of time based on the first environmental data and a preset big data model.

[0065] Optionally, before predicting the second environmental data of the second period, the regular characteristics of the first environmental data may be obtained first. Specifically, the first environmental data in step 110 may be plotted into a line graph to show the changes in the first environmental data in different time periods within the first period, thereby identifying the seasonal regularity and periodic fluctuation regularity of the first environmental data. For example, see Figure 2 , showing a line graph of the temperature in the first time period. The first time period in the example in the figure is from June 2024 to October 2024. Line A is the highest temperature in the first time period, line C is the lowest temperature in the first time period, and line B is the average temperature in the first time period. It can be seen from the figure that the temperature in the storage area drops in early October 2024.

[0066] Optionally, time series analysis can be performed first, and then cluster analysis can be performed (for example, the first environmental data can be classified into categories according to storage area temperature, storage area humidity, non-storage area temperature, non-storage area humidity, and point temperature) to obtain the similarities of temperature, humidity, and gas concentration data in different time periods, different areas, or different conditions. In this way, the seasonality and periodic fluctuation patterns of the first environmental data can be obtained more deeply.

[0067] Optionally, the preset big data model can be a comprehensive model based on time series analysis, cluster analysis, and regression analysis. The big data model can be trained and verified based on the first environmental data in the first time period, and predict the second environmental data in the second time period, wherein the first time period can be a past time period, that is, training and verification of historical data, and the second time period can be a future time period, that is, prediction of future environmental data. In this way, the prediction results of the big data model can provide a scientific basis for decisions such as environmental control and equipment maintenance.

[0068] Specifically, step 120 may include the following three situations:

[0069] (I) Case 1: The first environmental data is temperature data and / or humidity data, and the grain moldiness in the second time period is predicted.

[0070] In this case, the big data model is a logistic regression model, see Figure 3, showing a schematic diagram of a process for predicting the mildew condition of grain in the second period, step 120 specifically includes:

[0071] Step 310: Extract multiple feature information from the first environment data.

[0072] Here, characteristic information about grain mold can be extracted based on the temperature data and humidity data in the first environmental data, such as temperature-related characteristics: maximum temperature, minimum temperature, average temperature, temperature fluctuation range; or humidity-related characteristics: average humidity, humidity fluctuation range; or combined characteristics of temperature and humidity: temperature and humidity product, temperature and humidity ratio, etc.

[0073] Step 320: Calculate the first correlation between each feature information and grain mildew.

[0074] Here, correlation analysis or recursive feature elimination method can be used to calculate the first correlation between each feature information and grain mildew, which indicates the predictive ability of each feature for grain mildew, that is, the higher the correlation, the closer the association between the feature information and grain mildew.

[0075] Step 330: Filter out first feature information having a first correlation higher than a preset threshold from the plurality of feature information.

[0076] Here, features with a first correlation higher than a preset threshold and low redundancy may be selected, and these selected features may be used to construct a feature subset.

[0077] Step 340: Use the first feature information to train a logistic regression model, and adjust the parameters of the logistic regression model according to the training results, and repeat this step until the value of the logistic regression model under the preset indicator reaches the set value.

[0078] Here, since judging whether grain is moldy is a classification problem, a logistic regression model is selected. The historical temperature and humidity data and the known grain moldy conditions are used as training data sets, and the training data sets are divided into training sets and validation sets, where the training set is used to train the logistic regression model, and the validation set is used to verify whether the prediction results of the logistic regression model are correct. If the prediction results of the logistic regression model deviate from the actual results, find out the reasons for the misjudgment of the logistic regression model (such as improper feature selection or unreasonable model parameter settings, etc.), and finally adjust the parameters of the logistic regression model (such as learning rate, regularization coefficient and other parameters) and optimize the algorithm to improve the prediction ability of the model until the value of the logistic regression model under the preset indicators reaches the set value, where the preset indicators can be one or more of the accuracy, recall rate, F1 score, and AUC-ROC curve.

[0079] Optionally, methods such as cross-validation can be used to further verify the stability and generalization ability of the model.

[0080] Step 350: Predict the second environmental data of the second period according to the adjusted logistic regression model.

[0081] Finally, the trained model is used to predict future temperature and humidity data, and the prediction results are used to determine whether there is a risk of food mold in the future.

[0082] Through the above scheme, the classification problem is solved by the logistic regression model to determine whether the grain may be moldy. Through continuous data accumulation and algorithm optimization, the generalization ability of the model can be increased to adapt to the special needs of different grain storage environments.

[0083] (II) Case 2: Based on the first environmental data being temperature data and / or humidity data, the pest situation in the second period is predicted.

[0084] In this case, the large data model is a linear regression model, see Figure 4 , showing a schematic diagram of a process for predicting the pest situation in the second period, step 120 specifically includes:

[0085] Step 410: Extract multiple feature information from the first environment data.

[0086] Here, characteristic information on insect pests can be extracted based on the temperature data and humidity data in the first environmental data, such as temperature-related characteristics: maximum temperature, minimum temperature, average temperature, temperature fluctuation range; or humidity-related characteristics: average humidity, humidity fluctuation range; or combined characteristics of temperature and humidity: temperature and humidity product, temperature and humidity ratio, etc.

[0087] Step 420: Calculate the second correlation between each feature information and the pest.

[0088] Here, correlation analysis or recursive feature elimination method can be used to calculate the second correlation between each feature information and the pest, and the second correlation indicates the predictive ability of each feature for the pest, that is, the higher the correlation, the closer the association between the feature information and the pest.

[0089] Step 430: Filter out second feature information having a second correlation higher than a preset threshold from the plurality of feature information.

[0090] Here, features with a second relevance higher than a preset threshold and low redundancy may be selected, and these selected features may be used to construct a feature subset.

[0091] Step 440: Use the second feature information to train a linear regression model, and adjust the parameters of the linear regression model according to the training results, and repeat this step until the value of the preset indicator of the linear regression model reaches the set value.

[0092] Here, since predicting the trend of grain pests is a regression problem, a linear regression model is selected. The historical temperature and humidity data and the known pest conditions are used as training data sets, and the training data sets are divided into training sets and validation sets, where the training set is used to train the linear regression model, and the validation set is used to verify whether the prediction results of the linear regression model are correct. If the prediction results of the linear regression model deviate from the actual results, find out the reasons for the misjudgment of the linear regression model (such as improper feature selection or unreasonable model parameter settings, etc.), and finally adjust the parameters of the linear regression model (such as learning rate, regularization coefficient and other parameters) and optimize the algorithm to improve the prediction ability of the model until the value of the linear regression model under the preset indicators reaches the set value, where the preset indicators can be one or more of the accuracy, recall rate, F1 score, and AUC-ROC curve.

[0093] Optionally, methods such as cross-validation can be used to further verify the stability and generalization ability of the model.

[0094] Step 450: Predict the second environmental data of the second time period according to the adjusted linear regression model.

[0095] Finally, the trained model is used to predict future temperature and humidity data, and the risk of future pest development is determined based on the prediction results.

[0096] Through the above scheme, the regression problem is processed by the linear regression model, which can predict the development trend and degree of pests. Through continuous data accumulation and algorithm optimization, the generalization ability of the model can be increased to adapt to the special needs of different grain storage environments.

[0097] (III) Case 3: The first environmental data is gas concentration data, and the gas concentration in the second time period is predicted.

[0098] See also Figure 5 , showing a schematic flow chart of predicting the gas concentration in the second time period, step 120 specifically includes:

[0099] Step 510: Obtain the safety concentration threshold corresponding to the set gas.

[0100] Optionally, before step 510, key locations that need to be inspected can be determined based on the specific conditions and layout of the storage area, such as the top of the grain pile, the middle of the grain pile, the bottom of the grain pile, the ventilation system outlet, etc., and it can be ensured that the sampling points can represent the gas conditions of the entire storage area to avoid the storage area being affected by local pollution sources.

[0101] Optionally, the set gas can be phosphine, oxygen, nitrogen, carbon monoxide, carbon dioxide, etc., without specific restrictions. And set the safety threshold according to relevant standards and regulations. For example, set the safety concentration threshold of phosphine to 0.2mL / m3 (i.e. 0.2ppm) below, the safe range of carbon monoxide is usually defined between 0-50ppm, and so on.

[0102] Step 520: Input the safety concentration threshold and the first environmental data into the big data model to obtain the prediction result output by the big data model.

[0103] Here, the prediction of gas concentration can select a comprehensive model of concentration comparison, trend analysis, and correlation analysis as a big data model. The real-time monitored gas concentration data is compared with the set safety threshold to determine whether the current gas concentration of the set gas is higher than the safety threshold. If the set gas concentration exceeds the safety threshold, an alarm is issued; or, a time series graph is drawn to analyze the changing trend of the gas concentration to predict the future gas concentration fluctuation of the set gas. For example, if the gas concentration continues to rise or fluctuates greatly, it may indicate a potential safety hazard; or, the correlation between gas concentration and other factors (such as temperature, humidity, wind speed, etc.) is analyzed to understand the possible causes and influencing factors of gas concentration changes, which is conducive to taking corresponding measures for intervention in the future.

[0104] Through the above scheme, one or more prediction results of the following contents can be obtained: whether the gas concentration of the current set gas is higher than the safety threshold, the gas concentration fluctuation of the set gas in the future, and the correlation between the gas concentration of the set gas and other factors. Other factors include one or more of temperature, humidity, and wind speed.

[0105] Through the above scheme, by analyzing the gas concentration data, it is possible to determine whether the gas environment in the storage area is safe, that is, through concentration comparison analysis, various gas indicators are monitored in real time to ensure the safety of the grain storage environment, and through trend analysis, the changing trends of key indicators are tracked to timely discover potential risks. Through correlation analysis, the internal connections between various indicators are deeply studied to optimize the early warning mechanism.

[0106] Step 130: When the second environmental data does not meet the preset storage condition, adjust the current environmental data of the storage area.

[0107] Here, the current environmental data of the storage area can be adjusted according to the second environmental data predicted by the big data model to realize a closed-loop optimization mechanism. For example, when the temperature is too high, the ventilation and cooling system is started, when the temperature is too low, the insulation facilities are turned on, when the humidity is too high, the dehumidification system is activated, when the humidity is too low, the humidification equipment is started, and so on, without specific restrictions. In this way, by jointly controlling with other intelligent devices (such as intelligent ventilation systems, intelligent dehumidification systems, etc.), intelligent management of the grain storage environment can be realized to ensure the safety and quality of grain storage.

[0108] Optionally, if the second environmental data is higher than a preset value, an alarm message is issued and abnormal data is displayed in real time. For example, if the concentration of the set gas exceeds the safety threshold, an alarm message is issued, an emergency response plan is activated, relevant personnel are notified to evacuate quickly, and abnormal data is automatically recorded and an analysis report is generated.

[0109] Optionally, a multi-channel alarm mechanism can be adopted, such as: on-site sound and light alarm to ensure that the on-duty personnel can detect it at the first time, SMS and voice call notification, email push detailed abnormality report, mobile APP push real-time warning information, real-time display of abnormal data on the large screen in the monitoring room, etc. Through the multi-level early warning mechanism, it can ensure that managers can discover and deal with various abnormal situations in time, greatly improving the safety and management efficiency of grain storage.

[0110] Optionally, various reports can be automatically generated, including daily, weekly, monthly and annual reports, etc. These reports summarize key environmental data, equipment operating conditions, abnormal event records and other information, and provide in-depth data analysis and trend forecasts, providing important basis for management decisions and long-term planning.

[0111] See also Figure 6 , showing a schematic diagram of real-time display data. The upper left corner is the platform status of the current storage area, the left side is the current first environmental data in storage area A and storage area B, the middle screen is the monitored storage area, the upper right corner is the weather station sensor data, including wind speed data, wind direction data and rainfall data, the right side is the license plate number of the monitored vehicles entering and leaving the storage area, and the lower right corner is the monitoring screen of the storage area. Figure 6 In the middle monitoring storage area, there are monitored storage area 612 and storage area 614, as well as other buildings 611 and other buildings 613. There are sensors 620 and monitoring cameras 630 in storage area 612 and storage area 614 respectively. Sensor 620 is used to measure the temperature and humidity in the storage area (such as the temperature of the set area, the humidity of the entire storage warehouse). If the value measured by sensor 620 exceeds the preset safety threshold range, an early warning is issued on the display interface of the user application layer. For example, 620 turns red in the figure to alert the user that an abnormality has occurred.

[0112] See also Figure 7 , showing a schematic diagram of a model for early warning of abnormal data. The figure takes a storage warehouse as an example. The rectangular block in the storage warehouse is grain. By setting sensors in the storage warehouse, the temperature, humidity and gas concentration data of the storage warehouse can be obtained. The data is displayed to the user in the form of images through the early warning model. For example, Figure 7The arrows indicate the temperature, humidity and gas concentration in each area of ​​the storage bin. If the gas concentration in a certain area exceeds the preset value, the model diagram at the corresponding position will be displayed in red to alert the user of the abnormality.

[0113] Based on the same concept, the present application also provides a data processing system, see Figure 8 The data processing system 800 includes a network perception / control layer 810, a business and data processing layer 820, and an application layer 830. The applications of these three modules are described below:

[0114] (i) Network perception / control layer 810.

[0115] The network perception / control layer 810 is the foundation of the entire system. It is mainly responsible for the collection and upload of internal environmental data of the granary and the control of equipment. It provides a reliable and comprehensive data foundation and flexible control capabilities for the entire grain situation awareness system, laying a solid foundation for the intelligent management of grain storage. The functions of this layer can include the following aspects:

[0116] (1) Data upload: The collected environmental data is uploaded to the business and data processing layer of the cloud or terminal securely and efficiently. During the upload process, data encryption and compression technology are used to ensure the security of data transmission and improve transmission efficiency.

[0117] (2) Receiving commands and controlling equipment: This layer also has the ability to receive control commands from the application layer to achieve remote control of various equipment in the granary. For example, the ventilation system can automatically adjust the ventilation volume and frequency based on temperature and humidity data; the refrigeration system can start the refrigeration equipment to lower the temperature of the granary under high temperature conditions; the dehumidification system can activate the dehumidification device when the humidity is too high to maintain a suitable grain storage environment; the fumigation system can start the automatic fumigation program when necessary based on pest monitoring data.

[0118] (3) Temperature and humidity detection: Real-time control of all sensors and control devices to achieve temperature and humidity detection.

[0119] (ii) Business and data processing layer 820.

[0120] The business and data processing layer 820 is the core of the system. It is mainly responsible for the storage, processing and analysis of environmental data such as temperature, humidity, and gas. It gives full play to the advantages of big data technology and provides powerful data support and decision-making capabilities for the grain situation awareness system, thereby realizing intelligent management and control of the grain storage environment. Its functions may include the following aspects:

[0121] (1) Data storage: Distributed storage technology is used to store a large amount of environmental data collected from the network perception / control layer in the cloud or local big data platform. This storage method can support the rapid reading and writing of massive data, providing a solid foundation for subsequent data processing and analysis.

[0122] (2) Data processing: Based on the specific requirements of grain monitoring tasks, a dedicated big data architecture is constructed, and a variety of data analysis algorithms, such as time series analysis, cluster analysis, and regression analysis, are used to conduct in-depth mining of the environmental data stored in the database.

[0123] (3) Data analysis: Based on historical data and domain knowledge, we build prediction models for granary environmental changes. These models can include temperature change models, humidity change models, gas concentration change models, etc., to predict the changing trend of the granary environment in the future. Finally, we use the data analysis results and prediction models to comprehensively evaluate the current environmental conditions of the granary and deduce possible future development trends.

[0124] (iii) Application layer 830.

[0125] The application layer 830 is the user interface of the system, responsible for data display, decision-making, grain temperature warning, remote control and other functions, and is the main window for users to interact with the system. Specific functions include:

[0126] (1) Data visualization: Through visualization technology, the processed data can be intuitively displayed to users in various forms such as charts, animations, and heat maps. For example, a line chart can be used to display the trend of temperature and humidity changes, a bar chart can be used to compare the environmental parameters of different granaries, or a 3D model can be used to display the temperature distribution inside the grain pile. These intuitive display methods enable users to quickly and accurately understand the real-time situation and historical changes of the granary environment.

[0127] (2) Intelligent decision-making: Based on big data analysis and machine learning algorithms, the system can provide users with intelligent decision-making suggestions. For example, based on current environmental parameters and historical data, the system can recommend the optimal ventilation strategy, dehumidification plan, or fumigation plan.

[0128] (3) Early warning mechanism: The system has set up a multi-level early warning mechanism, including regular warning, emergency warning and critical warning. When environmental parameters approach or exceed the preset threshold, the system will automatically trigger the corresponding level of warning. The early warning information will be sent through various channels, such as system interface pop-up windows, mobile phone text messages, emails, voice calls, etc., to ensure that managers can be informed and deal with potential problems in a timely manner.

[0129] (4) Remote control: Users can remotely control the granary equipment through the application layer interface. For example, start or adjust the ventilation system, turn on the dehumidification equipment, control the fumigation system, etc. These operations can be completed on the computer or mobile terminal (such as smartphones, tablets), greatly improving the flexibility and efficiency of management.

[0130] Through the above scheme, real-time monitoring and early warning of the temperature and humidity of the storage environment can be achieved, the safety and stability of grain storage can be improved, and by using big data technology to analyze and process data, situational awareness of the storage environment can be achieved, and potential temperature and humidity problems of grain can be discovered in time, which can provide a scientific basis and decision-making support for grain storage management.

[0131] Based on the same concept, an embodiment of the present application also provides a data processing device, which can execute the data processing method introduced in the above content.

[0132] See also Fig. 9 , a schematic diagram of the structure of a data processing device provided in an embodiment of the present application is given, such as Fig. 9 As shown, the data processing device includes:

[0133] An acquisition module 901 is used to acquire first environmental data of a storage area and a non-storage area in a first period of time, where the first environmental data includes one or more of temperature data, humidity data, and gas concentration data, and the non-storage area is located within a set distance outside the storage area;

[0134] A prediction module 902, configured to predict second environmental data in a second period of time according to the first environmental data and a preset big data model;

[0135] The adjustment module 903 is used to adjust the current environment data of the storage area when the second environment data does not meet the preset storage condition.

[0136] In one possible implementation, the temperature data includes one or more of the temperature values ​​of the storage area at each time node, the temperature values ​​of the non-storage area at each time node, and the temperature values ​​of the set location in the storage area. The humidity data includes one or more of the humidity values ​​of the storage area at each time node, the humidity values ​​of the non-storage area at each time node, and the humidity values ​​of the set location in the storage area. The acquisition module 901 is also used to: delete abnormal values ​​in the first environmental data, and the abnormal values ​​include any one of the temperature value less than the minimum temperature threshold in the temperature data, the temperature value greater than the maximum temperature threshold in the temperature data, the relative humidity less than the minimum humidity threshold in the humidity data, and the relative humidity greater than the maximum humidity threshold in the humidity data. If the data of the first area in the first environmental data at the first time node is a null value, historical data with characteristic information similar to the characteristic information of the null value in the historical environmental data is obtained, and the data of the first area at the first time node is corrected using the most recent historical data.

[0137] In one possible implementation, the first environmental data includes temperature data and / or humidity data, the big data model is a logistic regression model, and the prediction module 902 is specifically used to: extract multiple feature information in the first environmental data, calculate the first correlation between each feature information and grain mold, screen out the first feature information with a first correlation higher than a preset threshold from the multiple feature information, use the first feature information to train the logistic regression model, and adjust the parameters of the logistic regression model according to the training results, repeat this step until the value of the logistic regression model under the preset indicator reaches the set value, and predict the second environmental data of the second time period according to the adjusted logistic regression model.

[0138] In one possible implementation, the first environmental data includes temperature data and / or humidity data, the big data model is a linear regression model, and the prediction module 902 is specifically used to: extract multiple feature information in the first environmental data, calculate the second correlation between each feature information and the pest, screen out second feature information with a second correlation higher than a preset threshold from the multiple feature information, use the second feature information to train the linear regression model, and adjust the parameters of the linear regression model according to the training results, repeat this step until the value under the preset indicator of the linear regression model reaches the set value, and predict the second environmental data of the second time period according to the adjusted linear regression model.

[0139] In one possible implementation, the first environmental data is gas concentration data, and the prediction module 902 is specifically used to: obtain a safety concentration threshold corresponding to a set gas, input the safety concentration threshold and the first environmental data into a big data model, and obtain a prediction result output by the big data model, wherein the prediction result includes one or more of the following: whether the current gas concentration of the set gas is higher than the safety threshold, the future gas concentration fluctuation of the set gas, and the correlation between the gas concentration of the set gas and other factors, and other factors include one or more of temperature, humidity, and wind speed.

[0140] In a possible implementation, the adjustment module 903 is further configured to: if the second environmental data is higher than a preset value, issue an alarm message and display the abnormal data in real time.

[0141] See also Fig.10 , showing a schematic diagram of the structure of another data processing device provided in an embodiment of the present application, such as Fig.10 As shown, the data processing device includes: a memory 1001 and a processor 1002, and the processor 1002 is coupled to the memory 1001. The memory 1001 is used to store program instructions, and the processor 1002 is used to call the program instructions stored in the memory 1001 and execute the above data processing method according to the obtained program.

[0142] Optionally, the data processing device may further include an interface circuit 1003, which may be a transceiver or an input / output interface. The input / output interface is used to input and / or output information, where output may be understood as sending and input may be understood as receiving. The processor 1002 may communicate with other devices in the data processing device or other devices other than the data processing device through the interface circuit 1003 to obtain information required for executing the above data processing method.

[0143] When the data processing device 1000 is used to implement Figure 1 When the method shown is performed, the processor 1002 is used to implement the functions of the above-mentioned acquisition module 901, prediction module 902, and adjustment module 903.

[0144] It is understandable that the processor in the embodiments of the present application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. The general-purpose processor may be a microprocessor or any conventional processor.

[0145] The memory in the embodiments of the present application can be a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, a register, a hard disk, a mobile hard disk, a compact disc read-only memory (CD-ROM) or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. The storage medium can also be a component of the processor.

[0146] Based on the same technical concept, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the computer executes the above-mentioned data processing method.

[0147] Based on the same technical concept, an embodiment of the present invention further provides a computer-readable program product, which, when executed, enables a computer to execute the above data processing method.

[0148] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0149] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0150] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0151] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0152] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A data processing method, characterized in that: The method comprises: Acquire first environmental data of a storage area and a non-storage area within a first time period, wherein the first environmental data includes one or more of temperature data, humidity data, and gas concentration data, and the non-storage area is located within a set distance outside the storage area; Predicting second environmental data for a second period of time according to the first environmental data and a preset big data model; When the second environmental data does not meet the preset storage condition, the current environmental data of the storage area is adjusted.

2. The method according to claim 1, characterized in that The temperature data includes one or more of the temperature value of the storage area at each time node, the temperature value of the non-storage area at each time node, and the temperature value of a set location in the storage area; the humidity data includes one or more of the humidity value of the storage area at each time node, the humidity value of the non-storage area at each time node, and the humidity value of a set location in the storage area; Before predicting the second environmental data of the second period, the method further includes: Deleting abnormal values ​​in the first environmental data, wherein the abnormal values ​​include any one of a temperature value in the temperature data that is less than a minimum temperature threshold, a temperature value in the temperature data that is greater than a maximum temperature threshold, a relative humidity in the humidity data that is less than a minimum humidity threshold, and a relative humidity in the humidity data that is greater than a maximum humidity threshold; If the data of the first area in the first environmental data at the first time node is a null value, historical data in the historical environmental data with feature information similar to the feature information of the null value is obtained, and the data of the first area at the first time node is corrected using the most recent historical data.

3. The method according to claim 1, characterized in that The first environmental data includes temperature data and / or humidity data, and the big data model is a logistic regression model; The step of predicting second environmental data for a second period of time according to the first environmental data and a preset big data model includes: Extracting a plurality of feature information from the first environmental data; Calculate the first correlation between each feature information and grain mildew; Filtering out first feature information whose first correlation is higher than a preset threshold from the multiple feature information; Using the first feature information to train the logistic regression model, and adjusting the parameters of the logistic regression model according to the training results, and repeating this step until the value of the logistic regression model under the preset indicator reaches a set value; The second environmental data for the second time period is predicted according to the adjusted logistic regression model.

4. The method according to claim 1, characterized in that The first environmental data includes temperature data and / or humidity data, and the big data model is a linear regression model; The step of predicting second environmental data for a second period of time according to the first environmental data and a preset big data model includes: Extracting a plurality of feature information from the first environmental data; Calculate the second correlation between each feature information and the pest; Filtering out second feature information whose second correlation is higher than a preset threshold from the multiple feature information; Using the second special diagnosis information to train the linear regression model, and adjusting the parameters of the linear regression model according to the training results, and repeating this step until the value of the preset indicator of the linear regression model reaches the set value; The second environmental data of the second time period is predicted according to the adjusted linear regression model.

5. The method according to claim 1, characterized in that The first environmental data is gas concentration data, and the second environmental data for predicting the second time period includes: Get the safety concentration threshold corresponding to the set gas; The safety concentration threshold and the first environmental data are input into the big data model to obtain a prediction result output by the big data model, wherein the prediction result includes one or more of the following: whether the current gas concentration of the set gas is higher than the safety threshold, the future gas concentration fluctuation of the set gas, and the correlation between the gas concentration of the set gas and other factors, wherein the other factors include one or more of temperature, humidity, and wind speed.

6. The method according to any one of claims 1 to 5, characterized in that After predicting the second environmental data of the second time period, the method further includes: If the second environmental data is higher than a preset value, an alarm message is issued and the abnormal data is displayed in real time.

7. A data processing device, characterized in that: The device comprises: An acquisition module, configured to acquire first environmental data of a storage area and a non-storage area within a first time period, wherein the first environmental data includes one or more of temperature data, humidity data, and gas concentration data, and the non-storage area is located within a set distance outside the storage area; A prediction module, used to predict second environmental data in a second period of time according to the first environmental data and a preset big data model; The adjustment module is used to adjust the current environmental data of the storage area when the second environmental data does not meet the preset storage conditions.

8. A data processing device, characterized in that: include: A processor, wherein the processor is coupled to a memory, the memory is used to store a computer program or an instruction, and the processor is used to execute the computer program or the instruction to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: It stores a computer program executable by a computer device, and when the program is run on the computer device, the computer device executes the steps of any method described in claims 1 to 6.

10. A computer program product, characterized in that When the method is executed on a computer, the computer is enabled to execute the method according to any one of claims 1 to 6.

Images