A method and system for monitoring abnormal grain storage environment

Abstract

The present invention relates to the technical field of grain storage management, and in particular to a method and system for monitoring abnormal grain storage environment, the method comprising the following steps: constructing a multimodal sensing network, and collecting grain storage environment index data and grain images in a grain warehouse based on the multimodal sensing network; establishing a dynamic integration model, and then using the grain storage environment index data and the grain images to construct a multi-source environmental data set; identifying the grain storage environment and judging the abnormal type of the grain storage environment according to the multi-source environmental data set, and adjusting the grain storage environment index when the grain storage environment is abnormal; using a grain storage environment prediction model to predict the grain storage environment, and issuing an alarm when the grain storage environment is abnormal. The present invention can accurately monitor and identify grain storage environment anomalies and abnormal types, improve grain storage management efficiency, and provide strong support for optimizing grain storage management strategies.

Description

Technical Field

[0001] The present invention relates to the technical field of grain storage management, and in particular to a method and system for monitoring abnormal grain storage environment. Background Art

[0002] In recent years, with the rapid development of the Internet of Things, big data and artificial intelligence technologies, grain storage environment monitoring has gradually shifted towards automation and intelligence. Sensor technology, as the core of automated monitoring, has been widely used in real-time monitoring of key parameters such as temperature, humidity, and gas concentration in grain warehouses. These sensors can collect data in real time and transmit it to the data processing center through wireless communication technology, providing warehouse managers with a convenient and efficient monitoring method.

[0003] However, despite the significant progress made in sensor monitoring technology, it still faces many challenges in practical application. On the one hand, the environment in the granary is complex and changeable, and the monitoring data of a single sensor is often difficult to fully reflect the actual situation; on the other hand, how to efficiently process and analyze massive monitoring data, and timely discover and warn potential environmental anomalies or types of environmental anomalies, is still an urgent problem to be solved in the current field of grain storage environment monitoring. Summary of the invention

[0004] In view of the defects in the prior art, the present invention provides a method and system for monitoring abnormality in grain storage environment.

[0005] In order to achieve the above-mentioned purpose, in the first aspect, the present invention provides a method for monitoring abnormal grain storage environment, the method comprising the following steps: constructing a multimodal sensor network, and collecting grain storage environment index data and grain images in the grain warehouse based on the multimodal sensor network; establishing a dynamic integration model, and then using the grain storage environment index data and the grain images to construct a multi-source environmental data set; identifying the grain storage environment and judging the abnormal type of the grain storage environment according to the multi-source environmental data set, and adjusting the grain storage environment index when the grain storage environment is abnormal; using the grain storage environment prediction model to predict the grain storage environment, and issuing an alarm when the grain storage environment is abnormal. The present invention can accurately monitor and identify grain storage environment anomalies and abnormal types, improve grain storage management efficiency, and provide strong support for optimizing grain storage management strategies.

[0006] Optionally, the step of constructing a multimodal sensor network and collecting granary environmental index data and grain images in a grain warehouse based on the multimodal sensor network comprises the following steps:

[0007] Deploy multiple types of environmental monitoring sensors in grain warehouses, with multiple environmental monitoring sensors of each type, to build a multimodal sensor network;

[0008] The environmental monitoring sensor is used to collect the granary environmental index data and the grain image in the grain warehouse in real time.

[0009] Furthermore, the use of multimodal sensor networks to collect real-time grain storage environment indicator data and grain images in grain warehouses can provide more comprehensive data support for storage environment monitoring and improve the accuracy of storage environment anomaly identification and grain storage environment anomaly type identification.

[0010] Optionally, the establishing of the dynamic integration model and then using the granary environmental indicator data and the grain image to construct a multi-source environmental data set comprises the following steps:

[0011] Using a dynamic integration model to fuse the granary environmental index data collected by the environmental monitoring sensors of the same type at the same time, to obtain corresponding granary environmental index integrated data;

[0012] A multi-source environmental dataset is constructed using the granary environmental indicator integrated data and the grain images.

[0013] Furthermore, using a dynamic integration model to obtain integrated data of granary environmental indicators can improve the accuracy of the numerical values ​​of the obtained granary environmental indicators, thereby improving the accuracy of identifying storage environment anomalies and providing a reliable data basis for identifying abnormal types of grain storage environments.

[0014] Optionally, the dynamic integration model satisfies the following relationship:

[0015]

[0016] in, is the integrated data of the granary environmental indicators corresponding to the i-th type of environmental monitoring sensor, n is the number of the i-th type of environmental monitoring sensors, is the granary environmental index data collected by the kth environmental monitoring sensor in the i-th type environmental monitoring sensor, is the integration coefficient of the kth environmental monitoring sensor in the i-th type of environmental monitoring sensor, is the set of integration coefficients corresponding to the i-th type of environmental monitoring sensor, is the maximum integration coefficient in the integration coefficient set corresponding to the i-th type of environmental monitoring sensor, is the minimum integration coefficient in the integration coefficient set corresponding to the i-th type environmental monitoring sensor, m is the number of times the i-th type environmental monitoring sensor has collected the granary environmental index data in history, It is the granary environmental index data collected for the jth time in history by the kth environmental monitoring sensor among the i-th type environmental monitoring sensors.

[0017] Furthermore, the dynamic integrated model of the present invention is simple to calculate, and while ensuring high accuracy in monitoring abnormalities in the grain storage environment, it can also ensure the speed of identifying abnormalities in the grain storage environment.

[0018] Optionally, the grain storage environment abnormality type includes a first abnormality type and a second abnormality type, wherein the first abnormality type is the grain storage environment abnormality caused by changes in the grain storage environmental indicators, and the second abnormality type is the grain storage environment abnormality caused by factors other than changes in the grain storage environmental indicators.

[0019] Furthermore, dividing the abnormal types of grain storage environment into the first abnormal type and the second abnormal type makes it easier for relevant personnel to understand the causes of abnormal grain warehouse environment in a timely manner, and then take targeted measures to solve the problems, which is conducive to improving the efficiency of grain storage management and reducing the cost of grain storage management.

[0020] Optionally, the identifying the grain storage environment and determining the abnormal type of the grain storage environment according to the multi-source environmental data set, and adjusting the grain storage environment index when the grain storage environment is abnormal, comprises the following steps:

[0021] Determine whether the integrated data of the grain storage environment index is within the optimal value range for grain storage, and identify whether there is grain anomaly in the grain image, and determine whether there is anomaly in the grain storage environment according to the determination result and the identification result;

[0022] When the grain storage environment is abnormal, the grain storage time in the abnormal environment is predicted according to the integrated data of the grain storage environment index, and the grain storage environment index is adjusted to suppress the abnormality;

[0023] According to the storage time of grain in the abnormal environment, the classification prediction value is calculated using the abnormal type prediction model;

[0024] If the classification prediction value is greater than the classification threshold, the grain storage environment abnormality type is the first abnormality type, otherwise it is the second abnormality type.

[0025] Furthermore, using the abnormal type prediction model to identify the abnormal type of grain storage environment can enable relevant personnel to timely understand the causes of abnormal grain warehouse environment, and then take targeted measures to solve the problem and prevent further deterioration of grain quality, which is conducive to improving the efficiency of grain storage management and reducing the cost of grain storage management.

[0026] Optionally, the anomaly type prediction model satisfies the following relationship:

[0027]

[0028] in, is the classification prediction value, R is the number of times the granary environmental index data is not within the optimal value range since the grain was stored. is the duration that the granary environmental index data is not within the optimal value range for the rth time, How long can grain be stored in the best grain storage environment? It is the abnormal environment grain storage time corresponding to the rth time when the granary environmental index data is not within the optimal value range.

[0029] Furthermore, the abnormal type prediction model is simple to calculate and has high reliability. Based on the abnormal type prediction model, the abnormal type of the grain storage environment can be quickly and accurately obtained.

[0030] Optionally, the method of using the grain storage environment prediction model to predict the grain storage environment and issuing an alarm when an abnormality occurs in the grain storage environment comprises the following steps:

[0031] Constructing a grain storage environment prediction model, and using the grain storage environment prediction model to obtain prediction values ​​of various grain storage environment indicators;

[0032] If the predicted value of the grain storage environment index is not within the optimal value range, it is determined that the grain storage environment is about to be abnormal;

[0033] An alarm is issued when an abnormality has occurred or is about to occur in the grain storage environment.

[0034] Furthermore, the grain storage environment prediction model is used to predict the grain storage environment and issue an alarm when the grain storage environment may be abnormal, so that relevant personnel can take timely measures to respond to changes in the grain storage environment, prevent deterioration of grain quality, and reduce economic losses.

[0035] Optionally, the step of constructing a grain storage environment prediction model and using the grain storage environment prediction model to obtain prediction values ​​of various grain storage environment indicators comprises the following steps:

[0036] The whale algorithm is used to optimize the weights and thresholds of the BP neural network, and then the BP neural network is used to build an initial prediction model;

[0037] Sequentially taking different granary environmental indicators as outputs of the initial prediction model, taking other granary environmental indicators except the outputs of the initial prediction model as inputs, using the integrated data of granary environmental indicators in the multi-source environmental data set to complete the training and verification of the initial prediction model, and obtaining a grain storage environment prediction model that predicts each of the granary environmental indicators;

[0038] The grain storage environment prediction model of each of the grain storage environment indicators is used to obtain the prediction value of the corresponding grain storage environment indicator.

[0039] Furthermore, using the whale algorithm to optimize the weights and thresholds of the BP neural network can solve the problem that the weights and thresholds of the BP neural network randomly affect the network structure, improve the prediction accuracy of the grain storage environment prediction model, and help prevent grain quality deterioration and reduce economic losses.

[0040] In a second aspect, the present invention further provides a grain storage environment abnormality monitoring system, the grain storage environment abnormality monitoring system uses a grain storage environment abnormality monitoring method provided by the present invention, the system comprises: a data acquisition module, the data acquisition module is used to construct a multimodal sensor network, and based on the multimodal sensor network, collects grain storage environment index data and grain images in the grain warehouse; an anomaly detection module, the anomaly detection module is used to establish a dynamic integration model, and then use the grain storage environment index data and the grain image to construct a multi-source environmental data set; identify the grain storage environment according to the multi-source environmental data set and judge the grain storage The system is used to detect the abnormal type of grain storage environment, and adjust the grain storage environment index when the grain storage environment is abnormal; a prediction alarm module, which is used to predict the grain storage environment using a grain storage environment prediction model, and issue an alarm when the grain storage environment is abnormal; a data storage module, which is used to store the grain storage environment index data, the grain image, the abnormal type of grain storage environment, and the prediction result of the grain storage environment; a data output module, which is used to display the grain storage environment index data, the grain image, the abnormal type of grain storage environment, and the prediction result of the grain storage environment. The system can stably execute the method provided by the present invention, which can not only improve the efficiency of monitoring the abnormal grain storage environment, but also improve the practicality of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 A schematic diagram of a method for monitoring abnormal grain storage environment according to an embodiment of the present invention;

[0043] Figure 2 A method schematic diagram of a grain storage environment abnormality monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0045] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0046] It should be noted in advance that, in an optional embodiment, except for independent explanations, the same symbols or letters appearing in all formulas have the same meanings and values.

[0047] In an alternative embodiment, see Figure 1 The present invention provides a method for monitoring abnormal grain storage environment, the method comprising the following steps:

[0048] S1. Construct a multimodal sensor network, and collect granary environmental index data and grain images in the grain warehouse based on the multimodal sensor network.

[0049] Among them, S1 specifically includes the following steps:

[0050] S11. Deploy various types of environmental monitoring sensors in grain warehouses, with multiple environmental monitoring sensors of each type, to build a multimodal sensor network.

[0051] Specifically, in this embodiment, the granary environmental indicators include temperature, humidity, oxygen concentration and carbon dioxide concentration, and the environmental monitoring sensors include four types of environmental monitoring sensors, namely temperature sensors, humidity sensors, gas concentration sensors and cameras. At least four of each type of environmental monitoring sensors need to be deployed in the grain warehouse to form a multimodal sensing network.

[0052] Furthermore, specific models of various types of environmental monitoring sensors can be selected according to actual conditions and are not limited here.

[0053] S12. Use the environmental monitoring sensor to collect the granary environmental index data and the grain image in the grain warehouse in real time.

[0054] Specifically, in this embodiment, a temperature sensor is used to collect the temperature in the grain warehouse, a humidity sensor is used to collect the humidity in the grain warehouse, a gas concentration sensor is used to collect the oxygen concentration and carbon dioxide concentration in the grain warehouse, and a camera is used to collect grain images in the grain warehouse. Using a multimodal sensor network to collect grain warehouse environmental index data and grain images in real time in the grain warehouse can provide more comprehensive data support for storage environment monitoring and improve the accuracy of subsequent storage environment anomaly recognition and grain storage environment anomaly type recognition.

[0055] S2. Establish a dynamic integration model, and then use the granary environmental indicator data and the grain image to construct a multi-source environmental data set.

[0056] Among them, S2 specifically includes the following steps:

[0057] S21. Use a dynamic integration model to fuse the granary environmental index data collected by the same type of environmental monitoring sensors at the same time to obtain corresponding granary environmental index integrated data.

[0058] Specifically, in this embodiment, for ease of understanding and explanation, this step is specifically described below by taking the integrated data of the grain warehouse environmental indicators of temperature as an example. Due to the complex and changeable environment in the grain warehouse, the monitoring data of a single temperature sensor is often difficult to reflect the real temperature in the grain warehouse, that is, the integrated data of the grain warehouse environmental indicators of temperature. Therefore, this embodiment uses multiple temperature sensors to monitor the temperature in the grain warehouse. At the same time, considering that even the same model of temperature sensors may have different monitoring accuracy, when using the temperature monitored by multiple temperature sensors to obtain the real temperature in the grain warehouse, this embodiment records the real temperature in the grain warehouse as a nonlinear combination of the temperatures collected by each temperature sensor, and calculates the integration coefficient based on the data collected before each temperature sensor to reflect the proportion that each temperature sensor should occupy when calculating the real temperature, and finally obtains a dynamic integration model, thereby improving the accuracy of the obtained real temperature, thereby ensuring the accuracy of abnormal monitoring of the grain storage environment.

[0059] Furthermore, the dynamic integration model satisfies the following relationship:

[0060]

[0061] in, is the integrated data of granary environmental indicators corresponding to the i-th type of environmental monitoring sensor, n is the number of the i-th type of environmental monitoring sensor, is the granary environmental index data collected by the kth environmental monitoring sensor in the i-th type of environmental monitoring sensor, is the integration coefficient of the kth environmental monitoring sensor in the i-th type of environmental monitoring sensor, is the set of integration coefficients corresponding to the i-th type of environmental monitoring sensor, is the maximum integration coefficient in the set of integration coefficients corresponding to the i-th type of environmental monitoring sensor, is the minimum integration coefficient in the integration coefficient set corresponding to the i-th type of environmental monitoring sensor, m is the number of times the i-th type of environmental monitoring sensor has collected granary environmental indicator data in history, is the grain storage environment index data collected by the kth environmental monitoring sensor in the ith type of environmental monitoring sensor at the jth time in history. The dynamic integration model is simple to calculate, and can ensure the high accuracy of abnormal monitoring of grain storage environment while also ensuring the speed of identifying abnormal grain storage environment.

[0062] Furthermore, the granary environmental index data collected for the m+1th time is usually used as the current granary environmental index data.

[0063] S22. Construct a multi-source environmental data set using the granary environmental indicator integrated data and the grain image.

[0064] Specifically, in this embodiment, since the granary environmental indicator data and grain images are acquired in real time, the granary environmental indicator integrated data and grain images are one-to-one corresponding in time. Therefore, the granary environmental indicator integrated data and grain images can be used to construct a multi-source environmental data set, providing a reliable data basis for identifying abnormal types of grain storage environments.

[0065] S3. Identify the grain storage environment and determine the abnormal type of the grain storage environment based on the multi-source environmental data set, and adjust the grain storage environment indicators when the grain storage environment is abnormal.

[0066] Among them, the grain storage environment abnormality type includes a first abnormality type and a second abnormality type, where the first abnormality type is the grain storage environment abnormality caused by changes in the grain warehouse environmental indicators, and the second abnormality type is the grain storage environment abnormality caused by other factors except changes in the grain warehouse environmental indicators. This embodiment uses an abnormality type prediction model to identify the abnormality type of the grain storage environment, so that relevant personnel can understand the cause of the abnormality of the grain warehouse environment in a timely manner, and then take targeted measures to solve the problem, prevent further deterioration of grain quality, and help improve the efficiency of grain storage management and reduce the cost of grain storage management. S3 specifically includes the following steps:

[0067] S31. Determine whether the integrated data of the grain silo environmental indicators is within the optimal numerical range for grain storage, and identify whether there is grain anomaly in the grain image, and determine whether there is anomaly in the grain storage environment based on the judgment result and the recognition result.

[0068] Specifically, in this embodiment, it is necessary to judge in turn whether the integrated data of the granary environmental indicators of each granary environmental indicator is within the optimal numerical range of grain storage. Taking temperature as an example, it is judged whether the integrated data of the granary environmental indicators of temperature is within the optimal numerical range of grain storage, that is, the optimal temperature range of grain storage. If the integrated data of the granary environmental indicators of temperature is within the optimal temperature range, then there is no abnormality in the grain storage environment, otherwise there is an abnormality.

[0069] Furthermore, it is necessary to identify whether there are any abnormalities in the grain image, such as whether there are insect pests, whether there is mildew, whether there is rot, and the types of insect pests and mildew. If there are no abnormalities in the grain image, then there is no abnormality in the grain storage environment, otherwise there is an abnormality. The process of identifying whether there are any abnormalities in the grain image is a prior art and will not be described in detail here.

[0070] S32. When the grain storage environment is abnormal, the grain storage time in the abnormal environment is predicted based on the integrated data of the grain storage environment indicators, and the grain storage environment indicators are adjusted to suppress the abnormality.

[0071] Specifically, in this embodiment, a grain storage time prediction model is constructed. The construction method can refer to the method of constructing a grain storage environment prediction model in step S4. The input of the grain storage time prediction model is the integrated data of the grain storage environment indicators of each grain storage environment indicator, and the output is the predicted value of grain storage time. The grain storage time in abnormal environment can be predicted based on the integrated data of grain storage environment indicators.

[0072] Furthermore, if the integrated data of the grain warehouse environmental indicators of temperature are not within the optimal temperature range, the temperature in the grain warehouse is adjusted by adjusting the temperature control system in the grain warehouse to keep it within the optimal temperature range; if the integrated data of the grain warehouse environmental indicators of humidity are not within the optimal temperature range, the humidity in the grain warehouse is adjusted by adjusting the humidity control system in the grain warehouse to keep it within the optimal humidity range; if the integrated data of the grain warehouse environmental indicators of oxygen or carbon dioxide are not within the optimal temperature range, nitrogen is filled into the grain warehouse or the oxygen concentration or carbon dioxide concentration in the grain warehouse is adjusted by adjusting the ventilation system in the grain warehouse to keep it within the optimal concentration range.

[0073] S33. Calculate the classification prediction value using the abnormal type prediction model according to the storage time of the grain in the abnormal environment.

[0074] Specifically, in this embodiment, the abnormal type prediction model satisfies the following relationship:

[0075]

[0076] in, is the classification prediction value, R is the number of times the granary environmental index data is not within the optimal value range since the grain was stored. is the duration that the granary environmental index data is not within the optimal value range for the rth time, How long can grain be stored in the best grain storage environment? is the abnormal environment grain storage time corresponding to the time when the grain storage environment index data is not within the optimal value range for the rth time. The abnormal type prediction model is simple to calculate and has high reliability. Based on this abnormal type prediction model, the abnormal type of grain storage environment can be quickly and accurately obtained. The unit of duration is "hour".

[0077] Furthermore, the classification prediction value is actually the loss of storage time caused by the grain storage environment index data exceeding the optimal value range. The length of time that grain can be stored under the optimal grain storage environment is usually a time range, recorded as , and in this embodiment This is the upper limit of the time range. With lower limit The mean of .

[0078] S34. If the classification prediction value is greater than the classification threshold, the grain storage environment abnormality type is the first abnormality type, otherwise it is the second abnormality type.

[0079] Specifically, in this embodiment, the classification threshold Set to , The current storage duration of the grain.

[0080] S4. Use the grain storage environment prediction model to predict the grain storage environment and issue an alarm when the grain storage environment is abnormal.

[0081] The grain storage environment prediction model is used to predict the grain storage environment and issue an alarm when the grain storage environment may be abnormal, so that relevant personnel can take timely measures to respond to changes in the grain storage environment, prevent grain quality deterioration, and reduce economic losses. S4 specifically includes the following steps:

[0082] S41. Construct a grain storage environment prediction model, and use the grain storage environment prediction model to obtain predicted values ​​of various grain storage environmental indicators.

[0083] Among them, S41 specifically includes the following steps:

[0084] S411. Utilize the whale algorithm to optimize the weights and thresholds of the BP neural network, and then use the BP neural network to build an initial prediction model.

[0085] Specifically, in this embodiment, in the BP neural network, the initial weights and thresholds are randomly generated, which will affect the trained network structure and reduce the prediction accuracy of the model. Therefore, this embodiment uses the whale algorithm to optimize the initial weights and thresholds of the BP neural network to improve the prediction accuracy of the grain storage environment prediction model, which is conducive to preventing the deterioration of grain quality and reducing economic losses. This step can refer to the existing technology and will not be repeated here.

[0086] S412. Different granary environmental indicators are used as outputs of the initial prediction model in turn, and other granary environmental indicators except the output of the initial prediction model are used as inputs. The integrated data of granary environmental indicators in the multi-source environmental data set are used to complete the training and verification of the initial prediction model, so as to obtain a grain storage environment prediction model that predicts each of the granary environmental indicators.

[0087] Specifically, in this embodiment, taking temperature as an example, if a grain storage environment prediction model for predicting temperature is to be constructed, the integrated data of grain silo environmental indicators of temperature is used as the output of the initial prediction model, and the integrated data of grain silo environmental indicators of other grain silo environmental indicators except temperature is used as the input of the initial prediction model. The integrated data of grain silo environmental indicators in the multi-source environmental data set is used to complete the training and verification of the initial prediction model, and finally a grain storage environment prediction model for predicting temperature is obtained.

[0088] S413: using the grain storage environment prediction model of each of the grain storage environment indicators to obtain the prediction value of the corresponding grain storage environment indicator.

[0089] S42. If the predicted value of the grain storage environment index is not within the optimal value range, it is determined that the grain storage environment is about to become abnormal.

[0090] S43. Sending an alarm when an abnormality has occurred or is about to occur in the grain storage environment.

[0091] Specifically, in this embodiment, when an abnormality has occurred or is about to occur in the grain storage environment, a sound alarm is used to prompt relevant personnel to enter the warehouse in time to check the grain and the warehouse environment.

[0092] Furthermore, in other optional embodiments, other alarm methods may be selected, such as flashing lights and email reminders.

[0093] It should be noted that, in some cases, the actions described in the specification can be performed in a different order and still achieve the desired results. In this embodiment, the order of steps given is only to make the embodiment appear clearer and easier to explain, rather than to limit it.

[0094] In an alternative embodiment, see Figure 2 The present invention also provides a grain storage environment abnormality monitoring system, the grain storage environment abnormality monitoring system uses a grain storage environment abnormality monitoring method provided by the present invention, the system can stably execute the method provided by the present invention, not only can improve the efficiency of grain storage environment abnormality monitoring, but also can improve the practicality of the method. The system includes a data acquisition module A1, an abnormality detection module A2, a prediction alarm module A3, a data storage module A4 and a data output module A5.

[0095] The data acquisition module A1 is used to construct a multimodal sensor network, and to collect granary environmental index data and grain images in the grain warehouse based on the multimodal sensor network.

[0096] Specifically, in this embodiment, the data acquisition module A1 includes various types of environmental monitoring sensors, and the data acquisition module A1 is connected to the anomaly detection module A2 and the data storage module A4 respectively through the Internet of Things. The data acquisition module A1 specifically executes the content described in step S1.

[0097] The anomaly detection module A2 is used to establish a dynamic integrated model, and then use the granary environmental indicator data and the grain image to construct a multi-source environmental data set; identify the grain storage environment and determine the type of grain storage environment anomaly based on the multi-source environmental data set, and adjust the granary environmental indicators when an abnormality occurs in the grain storage environment.

[0098] Specifically, in this embodiment, the abnormality detection module A2 is connected to the data acquisition module A1 through the Internet of Things, and is electrically connected to the data storage module A4 and the data output module A5 respectively. The abnormality detection module A2 specifically executes the content described in step S2.

[0099] The prediction alarm module A3 is used to predict the grain storage environment using the grain storage environment prediction model and to issue an alarm when the grain storage environment is abnormal.

[0100] Specifically, in this embodiment, the prediction alarm module A3 is electrically connected to the data storage module A4 and the data output module A5 respectively, and the prediction alarm module A3 specifically executes the content described in step S4.

[0101] The data storage module A4 is used to store the grain storage environment index data, the grain image, the grain storage environment abnormality type and the grain storage environment prediction result.

[0102] Specifically, in this embodiment, the data storage module A4 is electrically connected to the data output module A5.

[0103] The data output module A5 is used to display the grain storage environment index data, the grain image, the grain storage environment abnormality type and the prediction result of the grain storage environment.

[0104] Specifically, in this embodiment, the data output module A5 includes a digital display screen for displaying grain storage environment index data, grain images and grain storage environment abnormality types. In other optional embodiments, relevant personnel can also set the data output module A5 according to actual needs to query and display other data in the data storage module A4.

[0105] In summary, firstly, the method uses a multimodal sensor network to collect grain warehouse environmental index data and grain images in real time in the grain warehouse, and uses a dynamic integration model to fuse the grain warehouse environmental index data collected from the same type of environmental monitoring sensors at the same time, and obtains the corresponding grain warehouse environmental index integrated data, thereby improving the accuracy of storage environment anomaly identification and providing a reliable data basis for the identification of abnormal types of grain storage environment; secondly, after identifying the abnormality of the grain storage environment, the method issues an alarm and automatically adjusts the grain warehouse environmental index to suppress the anomaly, and uses an abnormal type prediction model to identify the type of abnormal grain storage environment, so that relevant personnel can understand the cause of the abnormal grain warehouse environment in a timely manner, and then take targeted measures to solve the problem in a timely manner to prevent further deterioration of grain quality, which is conducive to ensuring grain quality, improving the efficiency of grain storage management and reducing the cost of grain storage management; finally, the method uses a grain storage environment prediction model to predict the grain storage environment, and issues an alarm when the grain storage environment may be abnormal, prompting relevant personnel to take timely measures to respond to changes in the grain storage environment, prevent grain quality deterioration, and reduce economic losses. In addition, the system can stably execute the method provided by the present invention, which can not only improve the efficiency of abnormal monitoring of grain storage environment, but also improve the practicality of the method.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for monitoring abnormal grain storage environment, characterized in that: The steps include: Deploy multiple types of environmental monitoring sensors in grain warehouses, with multiple environmental monitoring sensors of each type, to build a multimodal sensor network; Using the environmental monitoring sensor to collect granary environmental index data and grain images in the grain warehouse in real time; The dynamic integration model is used to fuse the granary environmental index data collected by the environmental monitoring sensors of the same type at the same time to obtain the corresponding granary environmental index integrated data. The dynamic integration model satisfies the following relationship: ， in, is the integrated data of the granary environmental indicators corresponding to the i-th type of environmental monitoring sensor, n is the number of the i-th type of environmental monitoring sensors, is the granary environmental index data collected by the kth environmental monitoring sensor in the i-th type environmental monitoring sensor, is the integration coefficient of the kth environmental monitoring sensor in the i-th type of environmental monitoring sensor, is the set of integration coefficients corresponding to the i-th type of environmental monitoring sensor, is the maximum integration coefficient in the integration coefficient set corresponding to the i-th type of environmental monitoring sensor, is the minimum integration coefficient in the integration coefficient set corresponding to the i-th type environmental monitoring sensor, m is the number of times the i-th type environmental monitoring sensor has collected the granary environmental index data in history, The granary environmental index data collected at the jth time in history by the kth environmental monitoring sensor among the i-th type environmental monitoring sensors; Construct a multi-source environmental data set using the granary environmental indicator integrated data and the grain image; Determine whether the integrated data of the grain storage environment index is within the optimal value range for grain storage, and identify whether there is grain anomaly in the grain image, and determine whether there is anomaly in the grain storage environment according to the determination result and the identification result; When the grain storage environment is abnormal, the grain storage time in the abnormal environment is predicted according to the integrated data of the grain storage environment index, and the grain storage environment index is adjusted to suppress the abnormality; According to the storage time of grain in the abnormal environment, the classification prediction value is calculated using the abnormal type prediction model; If the classification prediction value is greater than the classification threshold, the grain storage environment abnormality type is the first abnormality type, otherwise it is the second abnormality type; The grain storage environment abnormality type includes the first abnormality type and the second abnormality type, the first abnormality type is the grain storage environment abnormality caused by the change of the grain storage environment index, and the second abnormality type is the grain storage environment abnormality caused by other factors except the change of the grain storage environment index; Use the grain storage environment prediction model to predict the grain storage environment and issue an alarm when the grain storage environment is abnormal.

2. A method for monitoring abnormal grain storage environment according to claim 1, characterized in that: The anomaly type prediction model satisfies the following relationship: ， in, is the classification prediction value, R is the number of times the granary environmental index data is not within the optimal value range since the grain was stored. is the duration that the granary environmental index data is not within the optimal value range for the rth time, How long can grain be stored in the best grain storage environment? It is the abnormal environment grain storage time corresponding to the rth time when the granary environmental index data is not within the optimal value range.

3. A method for monitoring abnormal grain storage environment according to claim 1, characterized in that: The method of using the grain storage environment prediction model to predict the grain storage environment and issuing an alarm when the grain storage environment is abnormal includes the following steps: Constructing a grain storage environment prediction model, and using the grain storage environment prediction model to obtain prediction values ​​of various grain storage environment indicators; If the predicted value of the grain storage environment index is not within the optimal value range, it is determined that the grain storage environment is about to be abnormal; An alarm is issued when an abnormality has occurred or is about to occur in the grain storage environment.

4. A method for monitoring abnormal grain storage environment according to claim 3, characterized in that: The step of constructing a grain storage environment prediction model and using the grain storage environment prediction model to obtain prediction values ​​of various grain storage environment indicators comprises the following steps: The whale algorithm is used to optimize the weights and thresholds of the BP neural network, and then the BP neural network is used to build an initial prediction model; Sequentially taking different granary environmental indicators as outputs of the initial prediction model, taking other granary environmental indicators except the outputs of the initial prediction model as inputs, using the integrated data of granary environmental indicators in the multi-source environmental data set to complete the training and verification of the initial prediction model, and obtaining a grain storage environment prediction model that predicts each of the granary environmental indicators; The grain storage environment prediction model of each of the grain storage environment indicators is used to obtain the prediction value of the corresponding grain storage environment indicator.

5. A grain storage environment abnormality monitoring system, the grain storage environment abnormality monitoring system using a grain storage environment abnormality monitoring method according to any one of claims 1 to 4, characterized in that: include: A data acquisition module, wherein the data acquisition module is used to deploy multiple types of environmental monitoring sensors in the grain warehouse, wherein each type of environmental monitoring sensor is multiple, thereby constructing a multimodal sensor network; Using the environmental monitoring sensor to collect granary environmental index data and grain images in the grain warehouse in real time; The anomaly detection module is used to use a dynamic integration model to fuse the granary environmental index data collected by the environmental monitoring sensors of the same type at the same time to obtain the corresponding granary environmental index integrated data. The dynamic integration model satisfies the following relationship: ， in, is the integrated data of the granary environmental indicators corresponding to the i-th type of environmental monitoring sensor, n is the number of the i-th type of environmental monitoring sensors, is the granary environmental index data collected by the kth environmental monitoring sensor in the i-th type environmental monitoring sensor, is the integration coefficient of the kth environmental monitoring sensor in the i-th type of environmental monitoring sensor, is the set of integration coefficients corresponding to the i-th type of environmental monitoring sensor, is the maximum integration coefficient in the integration coefficient set corresponding to the i-th type of environmental monitoring sensor, is the minimum integration coefficient in the integration coefficient set corresponding to the i-th type environmental monitoring sensor, m is the number of times the i-th type environmental monitoring sensor has collected the granary environmental index data in history, The granary environmental index data collected at the jth time in history by the kth environmental monitoring sensor among the i-th type environmental monitoring sensors; Construct a multi-source environmental data set using the granary environmental indicator integrated data and the grain image; Determine whether the integrated data of the grain storage environment index is within the optimal value range for grain storage, and identify whether there is grain anomaly in the grain image, and determine whether there is anomaly in the grain storage environment according to the determination result and the identification result; When the grain storage environment is abnormal, the grain storage time in the abnormal environment is predicted according to the integrated data of the grain storage environment index, and the grain storage environment index is adjusted to suppress the abnormality; According to the storage time of grain in the abnormal environment, the classification prediction value is calculated using the abnormal type prediction model; If the classification prediction value is greater than the classification threshold, the grain storage environment abnormality type is the first abnormality type, otherwise it is the second abnormality type; The grain storage environment abnormality type includes the first abnormality type and the second abnormality type, the first abnormality type is the grain storage environment abnormality caused by the change of the grain storage environment index, and the second abnormality type is the grain storage environment abnormality caused by other factors except the change of the grain storage environment index; A prediction alarm module, which is used to predict the grain storage environment using a grain storage environment prediction model and issue an alarm when the grain storage environment is abnormal; A data storage module, the data storage module is used to store the grain storage environment index data, the grain image, the grain storage environment abnormality type and the grain storage environment prediction result; A data output module, wherein the data output module is used to display the grain storage environment indicator data, the grain image, the abnormal type of the grain storage environment and the prediction result of the grain storage environment.

Images