Greenhouse control system based on edge computing and control method thereof

Abstract

The application provides a greenhouse control system based on edge computing and a control method thereof, and the system comprises an edge computing device, an environment information acquisition module, a soil information acquisition module, a control module and a cloud server; the environment information acquisition module is used for acquiring greenhouse environment information; the soil information acquisition module is used for acquiring soil environment information in the greenhouse; the edge computing device is used for performing data anomaly detection on the greenhouse environment information and the soil environment information, and sending the detected greenhouse environment information and soil environment information to the cloud server; the cloud server is used for generating a target control instruction based on the detected greenhouse environment information and soil environment information; and the control module is used for controlling a target electrical equipment to operate based on the received target control instruction, so as to adjust the greenhouse environment information and the soil environment information of the greenhouse. The application can realize online acquisition and processing of multiple parameters, identify abnormal data generated by the system, and effectively improve the system efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural technology, and in particular to a greenhouse control system based on edge computing and a control method thereof. BACKGROUND

[0002] In recent years, with the large-scale application of Internet of Things technology and the rapid development of information technology, the construction of agricultural greenhouses is also developing in the direction of production scale and management intelligence. At present, remote servers are usually used to realize intelligent control of greenhouses. Due to the need to process and store a large amount of environmental information data, the remote server is under great pressure, resulting in problems such as data transmission delay. In order to solve the above problems, the edge computing technology with low transmission delay advantage is applied to the greenhouse control system in the prior art.

[0003] However, in the existing greenhouse control system with edge computing devices, online monitoring is usually carried out mainly by measuring temperature and humidity, and the parameters are single, and abnormal data cannot be identified, resulting in low efficiency of the system and difficulty in meeting the actual production needs. SUMMARY

[0004] The present application provides a greenhouse control system based on edge computing and a control method thereof, to solve the defects in the prior art that the online monitoring parameters in the greenhouse control system are single, and abnormal data cannot be identified, resulting in low efficiency of the system and difficulty in meeting the actual production needs.

[0005] The present application provides a greenhouse control system based on edge computing, comprising:

[0006] an edge computing device, an environmental information acquisition module, a soil information acquisition module, a control module and a cloud server;

[0007] The environmental information acquisition module is used to acquire greenhouse environmental information; the soil information acquisition module is used to acquire soil environmental information in the greenhouse;

[0008] The edge computing device is used to detect data anomalies of the greenhouse environmental information and the soil environmental information, and send the detected greenhouse environmental information and soil environmental information to the cloud server;

[0009] The cloud server is used to generate target control instructions based on the detected greenhouse environmental information and soil environmental information;

[0010] The control module is used to control the operation of the target electrical equipment based on the received target control instructions, to adjust the greenhouse environmental information and soil environmental information of the greenhouse.

[0011] According to the greenhouse control system based on edge computing provided by the application, the edge computing device is further used for:

[0012] obtaining first data uploaded by a sensor at a current time; the sensor is any sensor in the environment information collection module and the soil information collection module;

[0013] determining each data uploaded by the sensor in a preset sliding time window; the first data is the last data in the preset sliding time window;

[0014] determining the variance and expectation of each data based on each data uploaded by the sensor in the preset sliding time window;

[0015] in a case where it is determined that the second data meets a target data abnormality condition and the second data is the same as third data at a next time, determining that the second data is abnormal data; the second data is any data in the each data, and the target data abnormality condition is determined based on the variance and expectation of the each data.

[0016] According to the greenhouse control system based on edge computing provided by the application, the edge computing device is further used for:

[0017] in a case where it is determined that the second data meets the target data abnormality condition and the second data is the same as third data at a next time, determining a first abnormality probability of data abnormality of the sensor;

[0018] determining the first abnormality probability corresponding to each sensor in the environment information collection module and the soil information collection module, and determining a second abnormality probability based on each first abnormality probability and a preset weight of each sensor; the second abnormality probability is used to represent the probability of data abnormality of the plurality of sensors;

[0019] determining a data detection result of the sensor based on the second abnormality probability.

[0020] According to the greenhouse control system based on edge computing provided by the application, the edge computing device is further used for:

[0021] in a case where it is determined that the absolute value of the difference between the second abnormality probability and a first mean value is less than the variance, determining that the second data is abnormal data and that the sensor has a sensor event; the first mean value is the mean value of each data uploaded by the sensor in the preset sliding time window;

[0022] or,

[0023] In a case where the absolute value of the difference between the second abnormal probability and the first mean value is not less than the variance, the second data is determined as abnormal data, and the second data is deleted.

[0024] According to the greenhouse control system based on edge computing provided by the application, the edge computing device is specifically further used for:

[0025] In a case where the second data does not satisfy the target data abnormal condition, it is determined that the second data is not abnormal;

[0026] Or,

[0027] In a case where it is determined that the second data satisfies the target data abnormal condition, and the second data is not identical to third data at a next time point, it is determined that the second data is not abnormal;

[0028] Or,

[0029] In a case where it is determined that the second data satisfies the target data abnormal condition, and the second data and a target number of data at continuous time points after the second data in the preset sliding time window are all identical, it is determined that the sensor is in a fault state.

[0030] According to the greenhouse control system based on edge computing provided by the application, the system further comprises:

[0031] An electric energy meter;

[0032] The electric energy meter is connected with the edge computing device;

[0033] The electric energy meter is used for monitoring power consumption information of each electrical device in the greenhouse;

[0034] The edge computing device is specifically further used for:

[0035] Obtaining the power consumption information of the electrical devices, and determining working states of the electrical devices based on the power consumption information of the electrical devices.

[0036] According to the greenhouse control system based on edge computing provided by the application, the system further comprises:

[0037] A meteorological information acquisition module and a mobile terminal;

[0038] The meteorological information acquisition module, the mobile terminal and the edge computing device are connected;

[0039] The meteorological information acquisition module is used for monitoring meteorological information outside the greenhouse;

[0040] The edge computing device is specifically further used for:

[0041] Obtain the weather information, perform data anomaly detection on the weather information, and send the detected weather information to the mobile terminal for display.

[0042] The application further provides a control method applied to the greenhouse control system based on edge computing, comprising:

[0043] The environment information collection module collects greenhouse environment information, and the soil information collection module collects soil environment information in the greenhouse.

[0044] The edge computing device performs data anomaly detection on the greenhouse environment information and the soil environment information, so as to send the detected greenhouse environment information and soil environment information to the control module.

[0045] The control module controls the target electrical equipment to operate based on the detected greenhouse environment information and soil environment information, so as to adjust the greenhouse environment information and soil environment information of the greenhouse.

[0046] According to the control method applied to the greenhouse control system based on edge computing, the method further comprises:

[0047] The edge computing device obtains first data uploaded by a sensor at a current time; the sensor is any sensor in the environment information collection module and the soil information collection module.

[0048] The edge computing device determines each data uploaded by the sensor within a preset sliding time window; the first data is the last data within the preset sliding time window.

[0049] The edge computing device determines the variance and expectation of each data based on each data uploaded by the sensor within the preset sliding time window.

[0050] The edge computing device determines the second data as abnormal data in a case that the second data meets a target data anomaly condition and the second data is the same as third data at a next time; the second data is any data in the each data, and the target data anomaly condition is determined based on the variance and expectation of the each data.

[0051] According to the control method applied to the greenhouse control system based on edge computing, after the edge computing device determines the variance and expectation of each data based on each data uploaded by the sensor within the preset sliding time window, the method further comprises:

[0052] The edge computing device determines a first abnormal probability of the sensor appearing data abnormality in a case that the second data meets the target data abnormal condition and the second data is same as third data at a next time;

[0053] The edge computing device determines the first abnormal probability corresponding to each sensor in the environment information collection module and the soil information collection module, and determines a second abnormal probability based on each first abnormal probability and a preset weight of each sensor; the second abnormal probability is used to represent a probability of multiple sensors appearing data abnormality.

[0054] The edge computing device determines the data detection result of the sensor based on the second abnormal probability.

[0055] The application further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the control method of the greenhouse control system based on edge computing according to any one of the above when executing the program.

[0056] The application further provides a non-transitory computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the control method of the greenhouse control system based on edge computing according to any one of the above.

[0057] The application further provides a computer program product, which includes a computer program, and the computer program is executable on a processor to implement the control method of the greenhouse control system based on edge computing according to any one of the above.

[0058] The greenhouse control system based on edge computing and the control method thereof provided by the application realize online monitoring of multiple modules including an environment information collection module and a soil information collection module, realize online collection and processing of multiple parameters, realize identification of abnormal data generated by the system through abnormal data detection of multiple parameter data by an edge computing device, reduce error data of online measurement, optimize system regulation and control decision, thereby more effectively controlling operation of electrical equipment in the greenhouse, effectively improving system efficiency, improving intelligent level of comprehensive environment control of the greenhouse, and providing a more scientific growth environment for crops in the greenhouse. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0060] Figure 1 Figure 1 is a system structure schematic diagram of a greenhouse control system based on edge computing provided by the present application.

[0061] Figure 2 Figure 2 is one of flow schematic diagrams of a control method of the greenhouse control system based on edge computing provided by the present application.

[0062] Figure 3 Figure 3 is another of flow schematic diagrams of the control method of the greenhouse control system based on edge computing provided by the present application.

[0063] Figure 4 Figure 4 is still another of flow schematic diagrams of the control method of the greenhouse control system based on edge computing provided by the present application. DETAILED DESCRIPTION

[0064] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0065] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connection" and "linking" should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral connection; can be mechanical connection, can also be electrical connection; can be direct connection, can also be indirect connection through an intermediate medium, or can be internal communication of two elements. Those of ordinary skill in the art can understand the specific meaning of the above terms in the present application according to the specific circumstances.

[0066] The present application will be described below with reference to the drawings. Figures 1-4 The present application provides a greenhouse control system based on edge computing and a control method thereof.

[0067] Figure 1 Figure 1 is a system structure schematic diagram of a greenhouse control system based on edge computing provided by the present application, as shown in the figure, the system comprises: an edge computing device 1, an environment information acquisition module 2, a soil information acquisition module 3, a control module 4 and a cloud server 5. Figure 1

[0068] The environment information acquisition module 2 is used for acquiring greenhouse environment information; the soil information acquisition module 3 is used for acquiring soil environment information in the greenhouse.

[0069] The control module 4 is connected with the edge computing device 1, the environment information acquisition module 2, the soil information acquisition module 3 and the cloud server 5, and is used for controlling the greenhouse based on the acquired environment information and soil information.​​​​​​​​​​​​​​​​​​​​​​​​​​​The edge computing device 1 is used for data anomaly detection on greenhouse environment information and soil environment information, and sends the detected greenhouse environment information and soil environment information to a cloud server;

[0070] The cloud server 5 is used for generating a target control instruction based on the detected greenhouse environment information and soil environment information;

[0071] The control module 4 is used for controlling a target electrical device to operate based on the received target control instruction, so as to adjust the greenhouse environment information and soil environment information of the greenhouse.

[0072] Specifically, the greenhouse environment information described in the embodiment of the present application refers to the environmental parameter information of the space environment in the greenhouse, which can specifically include air temperature, air humidity, light intensity, and carbon dioxide concentration, etc.

[0073] The soil environment information described in the embodiment of the present application refers to the environmental parameter information of the soil for planting crops in the greenhouse, which can specifically include soil layer temperature, water content, and salt content of the soil layer, etc.

[0074] The electrical device described in the embodiment of the present application refers to an electrical device arranged in advance in the greenhouse for adjusting the environmental parameters in the greenhouse, for example, a ventilation window device can be used to adjust the air temperature, air humidity, light intensity, and carbon dioxide concentration in the greenhouse; a heating device can be used to increase the air temperature in the greenhouse and reduce the air humidity, etc., and an irrigation device can be used to adjust the water content and salt content of the soil layer, etc.

[0075] The target electrical device described in the embodiment of the present application refers to an electrical device that needs to be regulated and controlled according to the detected greenhouse environment information and soil environment information.

[0076] The target control instruction described in the embodiment of the present application refers to a control instruction generated by the cloud server according to the detected greenhouse environment information and soil environment information, which is used to determine whether each environmental parameter in the greenhouse is at a preset standard level, and is used to control the target electrical device in the greenhouse through the control module on the greenhouse site, so as to adjust the greenhouse environment information and soil environment information in the greenhouse.

[0077] In the embodiment of the present application, the environmental information acquisition module 2 is used for acquiring greenhouse environment information, which can specifically select a greenhouse environment cloud sensing device, for example, a greenhouse doll II, which is a low-power acquisition device designed for sunlight greenhouse, plastic greenhouse, multi-span greenhouse, etc. The device can monitor the air temperature, air humidity, light intensity, carbon dioxide concentration, soil temperature, soil humidity, and other environmental parameters in the greenhouse environment in a time manner.

[0078] In the embodiment of the present application, the soil information acquisition module 3 is used to acquire the soil environment information in the greenhouse, which can be specifically selected as a soil multi-element profile tube monitor (three parameters). The soil multi-element profile tube monitor (three parameters) adopts the frequency domain reflectometry (FDR) principle, integrates the soil moisture, temperature and conductivity profile monitoring technology, can measure the water content and salt content of each soil layer; using a high-precision digital temperature sensor, the temperature of each soil layer can be measured. It can measure the water content, salt content and soil temperature data of up to 16 soil profiles at the same time, and the minimum interval of the soil layer monitoring interval is 10 cm.

[0079] In the embodiment of the present application, the control module 4 can be selected as an extensible greenhouse environment controller, which is a greenhouse environment remote automatic control device designed for sunlight greenhouse and multi-span greenhouse environment regulation and control, which integrates environment acquisition and control. The device can access wired and wireless sensors to collect environmental parameters such as light, air temperature and humidity in the greenhouse, and support up to 48-way environmental parameter input. The device supports switch control and interlocking control, and can realize the control of switch control devices such as fans, spraying, light supplementing lamps and wet curtains, and the control of interlocking devices such as membrane rolling, thermal insulation quilt, internal shading and external shading, and support up to 48-way control signal output. At the same time, it supports automatic control based on timing logic and sensor feedback, and the interlocking device supports opening degree control.

[0080] In the embodiment of the present application, the edge computing device 1 can be composed of a Raspberry Pi, a touch display, a Raspberry Pi special power supply, a LoRa433MHz radio frequency module and a RS485 to TTL module. The edge computing device 1 can be connected to the cloud server 5 through 4G or WIFI, and can also be connected to the control module 4 on site, and adopts LoRa433MHz radio frequency transmission data for greenhouse comprehensive environment regulation and control.

[0081] More specifically, the edge computing device 1 can select a 8G running memory 4B + Raspberry Pi, which has strong data processing performance and Internet of Things communication function, has rich I / O interface as output control, and has good compatibility with Message Queuing Telemetry Transport (MQTT) protocol. By installing and configuring the corresponding support environment package in the system environment in the computing layer, the setting of the related functions can be completed, such as the implementation of the MQTT proxy function by installing the PAHO-MQTT package, and the implementation of the lightweight Internet of Things edge analysis and streaming processing by installing the open source software. The Raspberry Pi gigabit network port is externally connected to a 4G router, and a 4G Internet of Things card is used to complete the transmission task of remote MQTT data. The related pins of the Raspberry Pi are set to serial communication mode, and are connected with a LoRa433MHz radio frequency module through a wired connection. The LoRa433MHz radio frequency module communicates with the environment information acquisition module 2, the soil information acquisition module 3, and the control module 4 through a high-frequency wireless antenna.

[0082] In the embodiment of the present application, the cloud server 5 can build a cloud service, which can be an MQTT server in particular, to provide data storage and decision support for greenhouse comprehensive environment regulation.

[0083] In the embodiment of the present application, by using LoRa433MHz radio frequency transmission data and wireless communication mode networking, the communication between the edge computing device and the environment information acquisition module, the soil information acquisition module, and the control module is realized, the communication distance is extended at a low transmission frequency, and the stability of signal transmission of the edge computing device in the greenhouse is ensured.

[0084] As shown in Figure 1 In the embodiment of the present application, a star-shaped network is built between the edge computing device and the environment information acquisition module, the soil information acquisition module, and the control module, which is easy to diagnose faults. When a connection failure occurs in one device, it will not affect the connection between other devices and the edge computing device, and the network can still operate normally. At the same time, it is easy to expand the network, whether it is to add or delete a node, as long as the edge computing device end is added.

[0085] Further, in the embodiment of the present application, after the greenhouse environment information is collected by the environment information collection module and the soil environment information is collected by the soil information collection module, the environment information collection module sends the greenhouse environment information to the edge computing device, and the soil information collection module sends the soil environment information to the edge computing device. The edge computing device receives the greenhouse environment information and the soil environment information, and can perform data anomaly detection on the greenhouse environment information and the soil environment information, and send the detected greenhouse environment information and soil environment information to the cloud server; at the same time, the sensor number of the sensor that appears data anomaly can also be sent to the cloud server. By performing data anomaly detection on the data collected by each sensor on the edge computing device, removing abnormal data, optimizing decision data, and ensuring that the cloud server can make control decision analysis based on accurate on-site data.

[0086] In the embodiment of the present application, it can be understood that the system as a whole is composed of three parts of the collection control layer, the edge computing layer and the cloud service layer, wherein the collection control layer includes the environment information collection module and the soil information collection module, the edge computing layer includes the edge computing device, and the cloud service layer includes the cloud server.

[0087] The collection control layer acquires the environment parameters through LoRa wireless communication, controls the device switch, and the Raspberry Pi gateway of the edge computing layer is responsible for completing data transmission, and can realize local control decision automation by using a fuzzy control algorithm.

[0088] The edge computing layer as an intermediate layer provides local services for the wireless sensors in the collection control layer, which can improve the quality of service and real-time information feedback. Therefore, the delay caused by remote interaction with the cloud service layer will be reduced, it can also provide initial processing of data and undertake part of the computing tasks of the cloud server to reduce the amount of data uploaded to the cloud server and the bandwidth load of the backbone link. The edge computing device in the edge computing layer plays a crucial role in the above-mentioned anomaly detection method. The edge computing device is used as a relay device to upload the monitoring data collected by the wireless sensor gateway in the collection control layer to the cloud server. At the same time, the collected data is subjected to data anomaly detection. Once the abnormal data is detected, the edge computing device will immediately report to the cloud server and drive the control module on site to execute the emergency response scheme.

[0089] Further, in the embodiment of the present application, after the cloud server receives the detected greenhouse environment information and soil environment information, it can generate target control instructions based on the detected greenhouse environment information and soil environment information according to the pre-set control decision, and send the target control instructions to the control module on site in the greenhouse. The control module controls the target electrical equipment to operate according to the received target control instructions to adjust the greenhouse environment information and soil environment information of the greenhouse.

[0090] In the embodiment of the present application, the control of the electrical equipment can also be performed in a manual manner through the control module on the greenhouse site, so as to provide the user with a control mode that can be used in an emergency scene.

[0091] For example, taking the control of a ventilation window as an example, the ventilation window is a device that can be controlled in stages and can be set to several static positions. If the cloud server determines that the carbon dioxide concentration index in the greenhouse is too high based on the detected greenhouse environment information, the target control instruction can be generated according to the pre-set control logic to adjust the opening degree of the ventilation window through the control module on the site until the carbon dioxide concentration index in the greenhouse reaches the actual demand level.

[0092] The greenhouse control system based on edge computing provided by the present application realizes online collection and processing of multiple parameters by simultaneously monitoring multiple modules including an environment information collection module and a soil information collection module on the greenhouse, and realizes identification of abnormal data generated by the system, reduces error data of online measurement, and optimizes the control decision of the system, so that the operation of the electrical equipment in the greenhouse can be more effectively controlled, the system efficiency can be effectively improved, the intelligent level of comprehensive environment control of the greenhouse can be improved, and a more scientific growth environment can be provided for crops in the greenhouse.

[0093] Based on the content of the above embodiment, as an optional embodiment, the system further comprises:

[0094] an electric energy meter 6;

[0095] The electric energy meter 6 is connected with the edge computing device 1.

[0096] The electric energy meter 6 is used for monitoring the power consumption information of each electrical equipment in the greenhouse.

[0097] The edge computing device 1 is specifically used for:

[0098] acquiring the power consumption information of each electrical equipment and determining the working state of each electrical equipment based on the power consumption information of each electrical equipment.

[0099] Specifically, as shown in Figure 1 In the greenhouse control system based on edge computing in the embodiment of the present application, an electric energy meter can also be included. The specific pin of the Raspberry Pi in the edge computing device is connected with the electric energy meter through an RS485 to TTL module in a wired manner. The electric energy meter can be used for monitoring the power consumption information of each electrical equipment in the greenhouse, so that the edge computing device can acquire the total energy consumption of each electrical equipment in the greenhouse in a 485 communication mode.

[0100] Meanwhile, the edge computing device can determine the working state of each electrical device according to the total positive useful work electric energy of the electric energy meter based on the power consumption information of each electrical device after obtaining the power consumption information of each electrical device.

[0101] In the embodiment, the electric energy meter can be a DTSU666 electronic guide rail type 380V three-phase meter. The DTSU666 electronic three-phase meter adopts a large-scale integrated circuit and applies a digital sampling technology, and is designed for the power monitoring and electric energy metering requirements of the power system, communication industry, building industry and the like, and mainly measures and displays the parameters such as three-phase voltage, three-phase current, active power, reactive power, frequency, positive and negative electric energy, and four-quadrant electric energy in the electrical circuit in real time.

[0102] In the embodiment of the application, the electrical device needs to be tested for normal working power before use, and the single device working power is recorded to the Raspberry Pi gateway in the edge computing device. Based on the built-in edge computing rules and the instantaneous device total power, the data processing model corresponding to the instantaneous device total power is called from the preset model library to detect and judge the device running condition. For accidental device data anomaly, the fault diagnosis program only gives a warning information and continues to be used. For abnormal power of the electrical device, the fault diagnosis program gives a serious warning information, and lists the electrical device as a disabled device, and at the same time, the actuator control law is reconstructed. For example, if the fan in the greenhouse has a short circuit and other serious faults, the fault diagnosis program will immediately stop the fan and at the same time give an alarm; then, the fan is listed as a disabled device; and the control law is reconstructed, that is, the fan is deleted from the functional devices such as ventilation device and cooling device, and at the same time, the cloud server end is reported to inform the user to repair the device.

[0103] The system of the embodiment of the application can monitor the running state of each electrical device in the greenhouse by adding an electric energy meter and introducing a measurement device power consumption mode, can perform fault diagnosis and emergency treatment on each electrical device, and improves the production safety of greenhouse planting.

[0104] Based on the content of the above embodiment, as an optional embodiment, the system further comprises:

[0105] The meteorological information acquisition module 7 and the mobile terminal 8;

[0106] The meteorological information acquisition module 7 and the mobile terminal 8 are connected with the edge computing device 1.

[0107] The meteorological information acquisition module 7 is used for monitoring the meteorological information outside the greenhouse.

[0108] The edge computing device 1 is specifically further used for:

[0109] The edge computing device 1 is specifically further used for:

[0109] The edge computing device 1 is specifically further used for:

[0110] Specifically, the meteorological information described in the embodiments of the present invention refers to the meteorological information of the environment in which the greenhouse is located, which may include agricultural meteorological information such as air temperature and humidity, atmospheric pressure, radiation intensity, wind speed and direction, and rainfall in the field.

[0111] like Figure 1 As shown, the edge computing-based greenhouse control system of this embodiment may further include a meteorological information acquisition module and a mobile terminal. Similar to the communication method between the edge computing device and the environmental information acquisition module, soil information acquisition module, and control module, communication between the edge computing device and the meteorological information acquisition module and the mobile terminal is achieved by using LoRa 433MHz radio frequency data transmission.

[0112] In embodiments of the present invention, the meteorological information acquisition module can be a fixed remote meteorological station. The fixed remote meteorological station can be a comprehensive remote automatic monitoring device powered by solar energy, capable of automatically acquiring, storing, and remotely transmitting field meteorological information. This device can automatically calculate hourly evapotranspiration (ET) values ​​and acquire image information reflecting crop growth through its onboard camera module. Using communication methods such as mobile networks or wireless data transmission, coupled with USB data export functionality, makes the acquisition of meteorological data more flexible and convenient.

[0113] In embodiments of the present invention, the meteorological information acquisition module can monitor agricultural meteorological parameters such as air temperature, humidity, light intensity, atmospheric pressure, wind speed, and wind direction outside the greenhouse in real time, and send the collected data to an edge computing device. The edge computing device can detect data anomalies in the acquired meteorological information to ensure the accuracy of the data uploaded by the meteorological information acquisition module. Simultaneously, the edge computing device sends the detected accurate meteorological information to a mobile terminal, such as a mobile phone, for display, allowing users to intuitively view various meteorological monitoring information in the field. In the event of severe weather, the meteorological information acquisition module can promptly report the situation. Leveraging the low transmission latency of the edge computing device, users can be promptly alerted via mobile terminals, ensuring that users can respond promptly to sudden weather conditions, take preventative measures to protect crops, and minimize disaster losses.

[0114] The system of this invention, by adding a meteorological information acquisition module and a mobile terminal, can monitor the meteorological environment of the greenhouse in real time and provide real-time reminders to users through the mobile terminal, thereby improving the meteorological disaster early warning capability of the greenhouse control system and providing a guarantee for stable and high crop yields.

[0115] Based on the above embodiments, as an optional embodiment, the edge computing device is further used for:

[0116] obtaining first data uploaded by a sensor at a current time; the sensor is any one of an environment information collection module and a soil information collection module;

[0117] determining each data uploaded by the sensor within a preset sliding time window; the first data is the last data within the preset sliding time window;

[0118] determining a variance and an expectation of each data based on each data uploaded by the sensor within the preset sliding time window;

[0119] in a case where it is determined that the second data meets a target data abnormality condition and the second data is the same as third data at a next time, determining that the second data is abnormal data; the second data is any one of the data, and the target data abnormality condition is determined based on the variance and the expectation of each data.

[0120] Specifically, the sensor described in the embodiments of the present application refers to any one of the environment information collection module and the soil information collection module.

[0121] The first data described in the embodiments of the present application refers to the last data within the preset sliding time window, and the second data is any one of the data within the preset sliding time window. It can be understood that the second data can also include the first data.

[0122] It should be noted that in the embodiments of the present application, the data of the sensor is transmitted to the edge computing device in the form of a data stream, so the data structure thereof can be defined as:

[0123]

[0124] wherein, D Tm represents a set of n sensor data received in m periods, S i represents a set of n sensor data received in the i th period, that is, S i ={r1(t i ),r2(t i ),r3(t i ),…,r n (t i )}, r i (t i ) represents data transmitted by the i th sensor to the edge computing device at t i .

[0125] Since the storage capacity of the edge computing device is limited, and the data uploaded by the sensor is constantly accumulated and increased, the memory capacity of the edge computing device will eventually reach saturation, so the edge computing device uses a sliding time window to manage the data uploaded by the sensor.

[0126] The sensor data with a window size of |w| is intercepted from the data stream uploaded by the sensor in the preset sliding time window, and divided into n data blocks, i.e., (block1, block2, block3, … blockn), each data block has a length of m, when a new data block blockn+1 is uploaded by the sensor, the earliest received data block block1 in the sliding time window will be replaced. n new oldest

[0127] In the embodiments of the present application, the sliding time window is used to process the data streams uploaded by different sensors connected to the same edge computing device.

[0128] Suppose that before time t q , q data from a certain sensor are loaded into the preset sliding time window with a window size of |w| = q, i.e., (r j (t1), r j (t2), r j (t3), …, r j (t q )), then the variance of the data in the preset sliding window is:

[0129]

[0130] wherein, represents the mean of the data in the preset sliding time window, r j (t i ) represents any data in the preset sliding time window, i.e., the second data, r j (t q ) represents the last data in the preset sliding time window, i.e., the first data.

[0131] When the sliding time window slides forward, i.e., the new first data r j (t q+1 ) is received by the edge computing device and replaces r j (t1), then the variance is:

[0132]

[0133] It should be noted that, under normal circumstances, the data received at the next moment has a small change fluctuation with the data in the current data window, and when the data fluctuation is large, it means that the data is largely abnormal data.

[0134] ​​​In the embodiments of the present application, the target data anomaly condition is determined based on the variance and expectation of each data in a preset sliding time window, and the target data anomaly condition can be expressed as:

[0135]

[0136] wherein, r j (t i ) represents the data transmitted by the jth sensor to the edge computing device at t i , i.e. the second data; E ej (t) represents the mathematical expectation of each data uploaded by the jth sensor in the preset sliding time window in which r j (t q ) is located; E nj (t) represents the mathematical expectation of each data uploaded by the jth sensor in the preset sliding time window in which r j (t q+1 ) is located; δ 2 represents the variance of each data uploaded by the jth sensor in the preset sliding time window in which r j (t q ) is located.

[0137] In addition, when a certain sensor itself fails, the data uploaded by the sensor may be the same in a period of time, i.e. the second data is the same as the third data at the next time, and there is:

[0138] r j (t i ) = r j (t i+1 ).

[0139] Therefore, this can be used as a basis for judging whether the sensor itself uploads abnormal data.

[0140] When the edge computing device determines that the second data satisfies the above target data anomaly condition, and the second data is the same as the third data at the next time, it can be determined that the second data is abnormal data.

[0141] The system of the embodiments of the present application considers that the storage data capacity of the edge computing device is limited, so that the edge computing device uses a sliding time window to manage the data uploaded by the sensor, and performs data anomaly detection on the data in the sliding time window, realizes the identification of abnormal data generated by the system, reduces the error data of online measurement, and is beneficial to optimize the control decision of the system.

[0142] Optionally, the edge computing device is further used for:

[0143] In a case where it is determined that the second data meets the target data anomaly condition and the second data is the same as third data at a next moment, a first anomaly probability of the sensor appearing data anomaly is determined;

[0144] The first anomaly probability corresponding to each sensor in the environment information collection module and the soil information collection module is determined, and a second anomaly probability is determined based on each first anomaly probability and a preset weight of each sensor; the second anomaly probability is used to represent a probability of multiple sensors appearing data anomaly;

[0145] Based on the second anomaly probability, a data detection result of the sensor is determined.

[0146] Specifically, in the embodiment of the present application, the edge computing device can further consider the correlation between different sensors in the data anomaly detection process.

[0147] The first anomaly probability described in the embodiment of the present application refers to a probability of a single sensor appearing data anomaly.

[0148] The second anomaly probability described in the embodiment of the present application refers to a probability of multiple sensors appearing data anomaly, which integrates the correlation between sensors.

[0149] Further, in the embodiment of the present application, in a case where it is determined that the second data meets the target data anomaly condition and the second data is the same as third data at a next moment, a first anomaly probability of the sensor appearing data anomaly is determined.

[0150] In the embodiment of the present application, the probability of a single sensor appearing data anomaly, i.e., the first anomaly probability, can be defined as:

[0151] P J (t i+1 )=P J (t i )+c·k 2 ;

[0152] Wherein, the initial P J (t0) can be set to 0, k represents the serial number of the second data in the data stream within the preset sliding time window; and c represents a preset constant.

[0153] When r j (t i )=r j (t i+1 ), P J (t i+1 ) is calculated; when r j (t i )≠r j (t i+1 ), PJ (t i+1 ) clear.

[0154] Therefore, the edge computing device can determine the first abnormal probability corresponding to each sensor in the environment information collection module and the soil information collection module, and determine the second abnormal probability based on each first abnormal probability and the preset weight of each sensor, that is:

[0155]

[0156] wherein P T (t i ) represents the second abnormal probability for representing that multiple sensors appear data abnormality, P j (t i ) represents the first abnormal probability corresponding to each sensor; λ j represents the weight coefficient of the preset weight of each sensor, and has

[0157] Considering that λ j is related to the fluctuation amplitude of the data, it can be obtained by the variance of the data stream corresponding to each sensor, that is:

[0158] λ1:λ2:λ3…λ j =δ1:δ2:δ3…δ j ;

[0159] Further, after obtaining the second abnormal probability, the data detection result of the sensor can be determined according to the second abnormal probability.

[0160] The system of the embodiment of the application, in the process of data abnormality detection of the edge computing device, by further considering the correlation between different sensors, each data in the preset sliding time window is detected more finely, which is beneficial to improve the accuracy of the data abnormality detection result.

[0161] Optionally, the edge computing device in the embodiment of the application is specifically further used for:

[0162] In the case where the absolute value of the difference between the second abnormal probability and the first mean value is less than the variance, the second data is determined as abnormal data, and the sensor event occurs; the first mean value is the mean value of each data uploaded by the sensor in the preset sliding time window.

[0163] Specifically, the sensor event described in the embodiment of the application refers to a program failure event occurring at the software level of the sensor system.

[0164] In the embodiment of the present application, according to the second abnormal probability calculated according to the foregoing calculation, the first mean value and the variance of each data uploaded by the sensor in the preset sliding time window are calculated, and in the case that the absolute value of the difference between the second abnormal probability and the first mean value is less than the variance, the second abnormal probability satisfies |P T (t i )-μ|<δ 2 , it can be determined that the second data is abnormal data, and the sensor event occurs. At the same time, the sensor event can be processed according to the preset processing scheme. It can be understood that the sensor event causes the data abnormality of the second data.

[0165] Wherein, μ represents the mean value of each data uploaded by the sensor in the preset sliding time window.

[0166] Optionally, the edge computing device in the embodiment of the present application is specifically further used for:

[0167] In the case that the absolute value of the difference between the second abnormal probability and the first mean value is not less than the variance, it is determined that the second data is abnormal data, and the second data is deleted.

[0168] That is, in the present embodiment, if the second abnormal probability satisfies:

[0169] |P T (t i )-μ|≥δ 2 ;

[0170] , it can be determined that the second data is abnormal data, and the second data is deleted. It can be understood that the abnormality of the second data at this time is not caused by the sensor event.

[0171] The system of the embodiment of the present application, in the process of data abnormality detection by the edge computing device, judges the fault type of the abnormality of the data uploaded by the sensor by calculating the relationship between the second abnormal probability and the mean value and the variance of each data in the preset sliding time window, and further improves the effectiveness of the abnormal data detection by the edge computing device.

[0172] Based on the content of the above embodiment, as an optional embodiment, the edge computing device is specifically further used for:

[0173] In the case that the second data does not satisfy the target data abnormality condition, it is determined that the second data is not abnormal;

[0174] In the case that it is determined that the second data satisfies the target data abnormality condition, and the second data is not identical to the third data at the next moment, it is determined that the second data is not abnormal;

[0175] In a case where it is determined that the second data meets the target data abnormality condition and the target number of data at continuous time points after the second data is the same, it is determined that the sensor is in a fault state.

[0176] Specifically, the target number described in the embodiment of the present application refers to a preset number threshold, which is used to determine whether the sensor is faulty.

[0177] In the embodiment of the present application, in a case where it is determined that the second data meets the target data abnormality condition and the second data is not the same as the third data at the next time point, it is determined that the second data is not abnormal, that is, when the condition of and r j (t i )≠r j (t i+1 ) is met, it can be further determined that the second data r j (t i ) is not abnormal.

[0178] In the embodiment of the present application, if it is determined that the second data meets the target data abnormality condition and the target number of data in the preset sliding time window and at continuous time points after the second data is the same, assuming that the target number can be set to 4, that is, when the condition of r j (t i )=r j (t i+1 )=r j (t i+2 )=r j (t i+3 )=r j (t i+4 ) is met, it can be determined that the sensor has failed and the sensor is in a fault state.

[0179] The system of the embodiment of the present application determines that the second data is not abnormal when it is determined that the second data does not meet the target data abnormality condition, and further determines the data at the time point after the second data in a case where it is determined that the second data meets the target data abnormality condition, to determine different detection results, thereby improving the precision of abnormality detection of the data reported by the sensor.

[0180] The control method of the greenhouse control system based on edge computing provided by the present application is described below, and the control method of the greenhouse control system based on edge computing described below can be correspondingly referred to the greenhouse control system based on edge computing described above.

[0181] Figure 2 is one of the flowcharts of the control method of the greenhouse control system based on edge computing provided by the present application, asFigure 2 As shown in the figure, the method is applied to the aforementioned greenhouse control system based on edge computing, and the method comprises:

[0182] In step 210, the environment information collection module collects the greenhouse environment information, and the soil information collection module collects the soil environment information in the greenhouse.

[0183] In step 220, the edge computing device performs data anomaly detection on the greenhouse environment information and the soil environment information, so as to send the detected greenhouse environment information and soil environment information to the control module.

[0184] In step 230, the control module controls the target electrical equipment to operate based on the detected greenhouse environment information and soil environment information, so as to adjust the greenhouse environment information and soil environment information of the greenhouse.

[0185] The control method of the greenhouse control system based on edge computing described in the embodiment can be applied to the above-mentioned embodiment of the greenhouse control system based on edge computing, and has similar principles and technical effects, which will not be described here.

[0186] The control method of the greenhouse control system based on edge computing provided by the application can realize online collection and processing of multiple parameters by simultaneously online monitoring of multiple modules including the environment information collection module and the soil information collection module in the greenhouse, and can realize identification of abnormal data generated by the system, reduce error data of online measurement, and optimize the regulation and control decision of the system, so that the operation of the electrical equipment in the greenhouse can be more effectively controlled, the system efficiency can be effectively improved, the intelligent level of comprehensive environment control of the greenhouse can be improved, and a more scientific growth environment can be provided for crops in the greenhouse.

[0187] Figure 3 is a flowchart of the control method of the greenhouse control system based on edge computing provided by the application, and as shown in the figure, Figure 3 The method further comprises:

[0188] In step 310, the edge computing device acquires first data uploaded by the sensor at the current time; the sensor is any sensor in the environment information collection module and the soil information collection module.

[0189] In step 320, the edge computing device determines each data uploaded by the sensor within a preset sliding time window; the first data is the last data within the preset sliding time window.

[0190] In step 330, the edge computing device determines the variance and expectation of each data based on each data uploaded by the sensor within the preset sliding time window.

[0191] In step 340, the edge computing device determines that the second data is abnormal data in a case where it is determined that the second data meets the target data abnormality condition and the second data is the same as third data at a next time; the second data is any one of the respective data, and the target data abnormality condition is determined based on the variance and the expectation of the respective data.

[0192] Based on the content of the above embodiment, as an optional embodiment, after the edge computing device determines the variance and the expectation of the respective data based on the respective data uploaded by the sensor within the preset sliding time window, the method further comprises:

[0193] The edge computing device determines a first abnormal probability of the sensor in a case where it is determined that the second data meets the target data abnormality condition and the second data is the same as third data at a next time;

[0194] The edge computing device determines the first abnormal probability corresponding to each sensor in the environment information acquisition module and the soil information acquisition module, and determines a second abnormal probability based on the respective first abnormal probability and a preset weight of each sensor; the second abnormal probability is used to represent the probability of data abnormality of the plurality of sensors;

[0195] The edge computing device determines the data detection result of the sensor based on the second abnormal probability.

[0196] Based on the content of the above embodiment, as an optional embodiment, the edge computing device determines the data detection result of the sensor based on the second abnormal probability, which comprises:

[0197] The edge computing device determines that the second data is abnormal data and that the sensor event occurs in a case where it is determined that the absolute value of the difference between the second abnormal probability and the first mean value is less than the variance; the first mean value is the mean value of the respective data uploaded by the sensor within the preset sliding time window;

[0198] Or,

[0199] The edge computing device determines that the second data is abnormal data and deletes the second data in a case where it is determined that the absolute value of the difference between the second abnormal probability and the first mean value is not less than the variance.

[0200] Based on the content of the above embodiment, as an optional embodiment, after the edge computing device determines the variance and the expectation of the respective data based on the respective data uploaded by the sensor within the preset sliding time window, the method further comprises:

[0201] The edge computing device determines that the second data is not abnormal in a case where the second data does not meet the target data abnormality condition;

[0202] Or,

[0203] In a case where it is determined that the second data meets the target data abnormality condition and the second data is not identical to third data at a next time point, it is determined that the second data is not abnormal.

[0204] Or,

[0205] In a case where the edge computing device determines that the second data meets the target data abnormality condition and the second data is identical to target data at a plurality of continuous time points after the second data within the preset sliding time window, it is determined that the sensor is in a fault state.

[0206] Based on the content of the above embodiment, as an optional embodiment, the method further comprises:

[0207] The electric energy meter monitors the power consumption information of each electrical device in the greenhouse.

[0208] The edge computing device acquires the power consumption information of each electrical device and determines the working state of each electrical device based on the power consumption information of each electrical device.

[0209] Based on the content of the above embodiment, as an optional embodiment, the method further comprises:

[0210] The meteorological information acquisition module monitors meteorological information outside the greenhouse.

[0211] The edge computing device acquires the meteorological information, performs data abnormality detection on the meteorological information, and sends the detected meteorological information to a mobile terminal for display.

[0212] Figure 4 is a flowchart of a control method of a greenhouse control system based on edge computing provided by the application, as shown in Figure 4 The method further comprises:

[0213] Step 400, start;

[0214] Step 410, receive data: acquire first data uploaded by the sensor at a current time point and determine each data uploaded by the sensor within a preset sliding time window;

[0215] Step 420, process data and calculate variance and mathematical expectation: based on each data uploaded by the sensor within the preset sliding time window, calculate the variance and expectation of each data;

[0216] Step 430, determine whether any data, i.e., second data, in each data meets a target data abnormality condition: if yes, execute step 440, and if no, it is determined that the sensor is normal, the second data is not abnormal, and jump to step 470;

[0217] Step 440, determine whether r j (t i) = r j (t i+1 ) is equal to r j (t i ) is equal to r j (t i+1 ), it is determined that the sensor is normal, the second data is normal, and the step 470 is jumped to;

[0218] In step 450, it is judged whether the second data is equal to the third data at the next time, if yes, step 460 is executed; if no, r j (t i ) is equal to r j (t i+1 ), it is determined that the sensor is normal, the second data is normal, and the step 470 is jumped to;

[0219] In step 460, it is judged whether the absolute value of the difference between the second abnormal probability and the first mean value is less than the variance of each data uploaded by the sensor in the preset sliding time window, if yes, it is determined that the sensor event occurs, the sensor event is processed, and the step 470 is jumped to; if no, it is determined that the second data is abnormal data, and the second data is discarded;

[0220] In step 470, the method ends.

[0221] The method of the embodiment of the application reduces the error data of the online measurement of the multiple sensors, realizes the abnormal data detection and the logic control at the edge, and greatly reduces the safety risk of the greenhouse production.

[0222] The device embodiments described above are only schematic, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or distributed on a plurality of network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the embodiment scheme. Those skilled in the art can understand and implement it without creative labor.

[0223] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0224] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An edge-computing based greenhouse control system, characterized in that, The edge computing device, the environment information collection module, the soil information collection module, the control module and the cloud server are included. The environment information collection module is used for collecting greenhouse environment information; the soil information collection module is used for collecting soil environment information in the greenhouse. The edge computing device is used for performing data anomaly detection on the greenhouse environment information and the soil environment information, and sending the detected greenhouse environment information and soil environment information to the cloud server. The cloud server is used for generating a target control instruction based on the detected greenhouse environment information and soil environment information. The control module is used for controlling a target electrical equipment to operate based on the received target control instruction, so as to adjust the greenhouse environment information and soil environment information of the greenhouse. The edge computing device is specifically further used for: obtaining first data uploaded by a sensor at a current time; the sensor is any sensor in the environment information collection module and the soil information collection module; determining each data uploaded by the sensor within a preset sliding time window; the first data is the last data within the preset sliding time window; based on each data uploaded by the sensor within the preset sliding time window, determining the variance and expectation of the each data; in a case where it is determined that second data meets a target data anomaly condition and the second data is the same as third data at a next time, determining that the second data is abnormal data; the second data is any data in the each data, and the target data anomaly condition is determined based on the variance and expectation of the each data; The edge computing device is specifically further used for: in a case where it is determined that the second data meets the target data anomaly condition and the second data is the same as the third data at the next time, determining a first anomaly probability of data anomaly of the sensor; determining the first anomaly probability corresponding to each sensor in the environment information collection module and the soil information collection module, and determining a second anomaly probability based on each first anomaly probability and a preset weight of each sensor; the second anomaly probability is used to represent the probability of data anomaly of a plurality of sensors; based on the second anomaly probability, determining a data detection result of the sensor. The edge computing device is specifically further used for:

2. The edge-computing-based greenhouse control system according to claim 1, characterized in that, in a case where it is determined that the absolute value of the difference between the second anomaly probability and a first mean value is less than the variance, determining that the second data is abnormal data and that a sensor event occurs to the sensor; the first mean value is a mean value of each data uploaded by the sensor within the preset sliding time window; or, in a case where it is determined that the absolute value of the difference between the second anomaly probability and the first mean value is not less than the variance, determining that the second data is abnormal data and deleting the second data. The edge computing device is specifically further used for:

3. The edge-computing-based greenhouse control system according to claim 1, characterized in that, in a case where the second data does not meet the target data anomaly condition, determining that the second data is not abnormal; or, ​ In a case where it is determined that the second data meets a target data abnormality condition and the second data is not identical to third data at a next time point, it is determined that the second data is not abnormal. Or, In a case where it is determined that the second data meets a target data abnormality condition and the second data is identical to a target number of data at continuous time points after the second data within the preset sliding time window, it is determined that the sensor is in a fault state.

4. The edge-computing based greenhouse control system according to any one of claims 1-3, characterized in that, The system further comprises: an electric energy meter; the electric energy meter is connected with the edge computing device; the electric energy meter is configured to monitor power consumption information of each electrical device in the greenhouse; the edge computing device is specifically further configured to: obtain the power consumption information of the electrical devices, and determine working states of the electrical devices based on the power consumption information of the electrical devices.

5. The edge-computing based greenhouse control system according to any one of claims 1-3, characterized in that, The system further comprises: a meteorological information acquisition module and a mobile terminal; the meteorological information acquisition module, the mobile terminal, and the edge computing device are connected; the meteorological information acquisition module is configured to monitor meteorological information outside the greenhouse; the edge computing device is specifically further configured to: obtain the meteorological information, perform data abnormality detection on the meteorological information, and send the detected meteorological information to the mobile terminal for display.

6. A control method applied to the edge-computing based greenhouse control system according to any one of claims 1-5, characterized in that, The system further comprises: the environmental information acquisition module acquires greenhouse environmental information, and the soil information acquisition module acquires soil environmental information in the greenhouse; the edge computing device performs data abnormality detection on the greenhouse environmental information and the soil environmental information, so as to send the detected greenhouse environmental information and soil environmental information to the control module; the control module controls target electrical devices to operate based on the detected greenhouse environmental information and soil environmental information, so as to adjust the greenhouse environmental information and soil environmental information of the greenhouse. The method further comprises: the edge computing device obtains first data uploaded by a sensor at a current time point; the sensor is any one of the environmental information acquisition module and the soil information acquisition module; the edge computing device determines each data uploaded by the sensor within a preset sliding time window; the first data is the last data within the preset sliding time window; the edge computing device determines variances and expectations of the data based on each data uploaded by the sensor within the preset sliding time window; in a case where it is determined that second data meets a target data abnormality condition and the second data is identical to third data at a next time point, the edge computing device determines that the second data is abnormal data; the second data is any one of the data, and the target data abnormality condition is determined based on the variances and expectations of the data; after the edge computing device determines the variances and expectations of the data based on each data uploaded by the sensor within the preset sliding time window, the method further comprises: in a case where it is determined that the second data meets the target data abnormality condition and the second data is identical to third data at a next time point, the edge computing device determines a first abnormal probability of data abnormality of the sensor; The edge computing device determines the first anomaly probability corresponding to each sensor in the environment information collection module and the soil information collection module, and determines a second anomaly probability based on the first anomaly probability of each sensor and a preset weight of each sensor; the second anomaly probability is used to represent the probability of data anomaly of the plurality of sensors; The edge computing device determines the data detection result of the sensor based on the second anomaly probability.

Images