Meteorological monitoring method and system for intelligent agricultural planting

Abstract

The invention belongs to the technical field of smart agriculture, and discloses a meteorological monitoring method and system for smart agricultural planting. According to the method, key meteorological parameters in a farmland environment are monitored in real time through a multi-parameter meteorological sensor, sensor data are collected and preliminarily processed, and the collected data are transmitted to a cloud server or a central database through a wireless or wired network; big data and an artificial intelligence technology are used for processing and analyzing the collected data, and meteorological prediction and decision support are provided; a user and an agricultural manager monitor meteorological data and receive early warning information in real time through a mobile application program or computer end software. The method provides a theoretical basis for scientific decision making and precise agricultural production management.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart agriculture, and in particular relates to a meteorological monitoring method and system for smart agricultural planting. Background Art

[0002] Meteorological monitoring in smart agriculture is a key link, which involves the use of various sensors and monitoring equipment to collect meteorological data in the farmland environment in real time. These data include but are not limited to air temperature and humidity, wind speed and direction, precipitation, light intensity, soil temperature and humidity, etc. With these data, farmers and agricultural managers can make scientific decisions and accurately manage agricultural production, optimize resource utilization efficiency, reduce losses caused by natural disasters, and improve crop yields and quality.

[0003] Through the above analysis, the problems and defects of the existing technology are as follows: the accuracy of the existing technology of meteorological monitoring for agricultural planting needs to be further improved, and it cannot effectively provide a theoretical basis for scientific decision-making and precise management of agricultural production. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the disclosed embodiments of the present invention provide a meteorological monitoring method and system for smart agricultural planting.

[0005] The technical solution is as follows: A meteorological monitoring method for smart agricultural planting, comprising:

[0006] S1, real-time monitoring of meteorological parameters in the farmland environment, including: air temperature and humidity, wind speed and direction, precipitation, light intensity, and soil temperature and humidity data;

[0007] S2, collects multi-parameter meteorological sensor data and performs preliminary processing;

[0008] S3, transmits the processed data to a cloud server or central database via wireless or wired networks;

[0009] S4, using big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support;

[0010] S5, monitor weather data and receive warning information in real time through mobile applications or computer software.

[0011] In step S2, multi-parameter meteorological sensor data is collected and preliminarily processed, including:

[0012] (1) Define the language specifications and encoding rules based on the multi-parameter meteorological sensor data encoder;

[0013] (2) Read the script content, parse the coding rules, and translate the script content into a tag dictionary or data analyzer code;

[0014] (3) Submit the tag dictionary or data analyzer code to the data collection terminal engine and the cloud server engine;

[0015] (4) The data acquisition terminal engine interacts with the cloud server engine.

[0016] In step (3), the data acquisition terminal engine and the cloud server engine have built-in libraries, supported by a data analyzer, and the most basic data abstraction in the data acquisition terminal engine is a resilient distributed data set of a data acquisition terminal structured query language.

[0017] In step (4), the data acquisition terminal engine interacts with the cloud server engine, including:

[0018] (1) The script executes the code with the data analyzer;

[0019] (2) The data acquisition terminal process creates a data communication agreement server and returns the communication agreement host and port information;

[0020] (3) The data collection terminal process starts the selection daemon process;

[0021] (4) Select the daemon process to create a communication protocol client;

[0022] (5) The data acquisition terminal process transmits the data communication port, host information, and data analyzer code through the communication protocol;

[0023] (6) Select the daemon process to receive the above information through the communication agreement and create a data communication agreement client;

[0024] (7) Select the data server that connects to the data collection terminal to obtain data;

[0025] (8) Select a daemon process to automatically send the data analyzer processing logic to the cloud server cluster for execution, use the data analyzer to train the data flow graph model, make the data flow graph model into a database function and register it on the data acquisition terminal, and perform distributed execution on the cloud server;

[0026] (9) The execution results of the cloud server cluster are returned to the selection daemon process and then transmitted to the data acquisition terminal process.

[0027] In step S4, the transmitted data is processed and analyzed using big data and artificial intelligence technologies to provide meteorological forecasts and decision support, including:

[0028] In the meteorological monitoring and identification for agricultural planting, the negative log-likelihood value of the exponential Gaussian model of the meteorological monitoring of the identification data is used as the objective function; the objective function is optimized using the improved bee-ant optimization method to obtain the parameters μ, σ, and τ corresponding to the function matching value; the exponential Gaussian model in meteorological monitoring is fitted using the above three parameters to monitor the data during identification in the meteorological monitoring and identification for agricultural planting;

[0029] Calculate the maximum matching value and the average matching value in the population, and calculate the conversion probability P based on the maximum matching value and the average matching value, so that the algorithm can dynamically control the global meteorological forecast and local meteorological forecast behaviors according to the evolutionary period; based on the meteorological factor cross method, apply the excellent meteorological judgment value of the meteorological factor agricultural planting meteorological single factor to the next iterative prediction of the meteorological single factor for agricultural planting to obtain the optimal value in the optimization.

[0030] Furthermore, the objective function is optimized using the improved bee-ant optimization method to obtain the parameters μ, σ, τ corresponding to the optimal function matching value, including:

[0031] Step 1: Initialize the parameters of the improved bee-ant optimization method, including the maximum number of iterations N iter , flight index L, parameters [μ, σ, τ], population N;

[0032] Step 2: Calculate the matching degree of each meteorological single factor for agricultural planting in the population according to the matching degree function, obtain the global optimal meteorological single factor for agricultural planting in the current population, as well as the maximum matching degree value and the average matching degree value of the population, and calculate the conversion probability p;

[0033] Step 3, select the single factors for meteorological use for agricultural planting with matching degree j in the population to form the single factor sequence for meteorological use for agricultural planting of the population;

[0034] Step 4, randomly select 2 excellent agricultural planting meteorological single factors from the excellent agricultural planting meteorological single factor sequence for crossover to obtain a new excellent agricultural planting meteorological single factor, and compare the matching degree with the global optimal agricultural planting meteorological single factor of the current population. If it is less than, the excellent agricultural planting meteorological single factor after crossover enters the population; otherwise, the global optimal agricultural planting meteorological single factor enters various populations for agricultural planting meteorological single factor learning in the population;

[0035] Step 5: Generate a random number in the range of [0, 1] and compare it with the conversion probability corresponding to the population N. If it is less than , perform a global search operation; otherwise, perform a local search operation;

[0036] Step 6: Update the global optimum of N.

[0037] In step 2, the calculation process of the maximum matching value and the average matching value is:

[0038] Step 2.1, calculate the single factor matching value y for agricultural planting and meteorological use of the population, the expression is:

[0039]

[0040] In the formula, is the probability density of the exponential Gaussian distribution in meteorological monitoring, i is the i-th identification time data obtained in a monitoring experiment, and t is the probability density of the t-th generation of the exponential Gaussian distribution in meteorological monitoring;

[0041] The specific calculation method is:

[0042]

[0043] Where pnf() is the normalized cumulative probability of the exponential Gaussian distribution in meteorological monitoring, eps is the floating point relative accuracy, exp() is the floating point relative accuracy calculation function, is the index of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the fixed exponent of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the variance of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the combined node group value of the ith meteorological factor in the population that has evolved to the tth generation;

[0044] Step 2.2: Calculate the maximum matching value f of the population max And the average matching value f mean :

[0045] f max =max(y)

[0046] f mean =mean(y)

[0047] In the formula, max(y) is the maximum value of the single factor matching degree of the meteorological use for agricultural planting of the population, and mean(y) is the average value of the single factor matching degree of the meteorological use for agricultural planting of the population;

[0048] Step 2.3, calculate the transition probability p:

[0049]

[0050] Where arccos() is the conversion function;

[0051] Select j agricultural planting meteorological single factors with excellent matching degree from various populations to form a population of excellent agricultural planting meteorological single factor sequence, and dynamically select excellent agricultural planting meteorological single factor j from the population composed of parameters to form an excellent agricultural planting meteorological single factor sequence m = {m 1 ,m 2 …m j}, where the update rule for j<N is as follows:

[0052]

[0053] In the formula, γ is a constant, j min is the minimum value of the single factor for agricultural planting meteorology, j max is the maximum value of a single factor for agricultural planting and meteorology, N iter It is a single factor iterative value for meteorological use in agricultural planting.

[0054] In step 4, two single factors for meteorological purposes for excellent agricultural planting are randomly selected from the single factor sequence for meteorological purposes for excellent agricultural planting and crossover is performed, including:

[0055] Step 4.1: Randomly select two different single factors m for excellent agricultural planting meteorology from the single factor sequence m for excellent agricultural planting meteorology. up ,m down ;

[0056] Step 4.2: Cross the two excellent agricultural planting meteorological factors with single factors, and the expression is:

[0057] c * =rand×m up +(1-rand)×m down

[0058] In the formula, c * is the meteorological single factor crossover value, m up It is the single factor upward value for meteorological use in excellent agricultural planting, m down It is the single factor downward value for meteorological use in excellent agricultural planting, and rand is a random value;

[0059] Step 4.3: Use the single factor crossover value c for the new excellent agricultural planting meteorology * Compare the calculated matching degree with the global optimal matching degree of the population. If it is better, then c * Enter the population for single factor learning of meteorology for agricultural planting in the population.

[0060] In step 5, a random number in the range of [0,1] is generated and compared with the transition probability corresponding to the population N. If it is less than, a global search operation is performed, otherwise a local search operation is performed; including:

[0061] Step 5.1: If the random number is less than the conversion probability corresponding to the population N, then perform a global search operation:

[0062]

[0063] In the formula, g * It is the single factor for global optimal agricultural planting meteorology. are the combined node groups of the ith meteorological factor in the population evolving to the tth generation and the t+1th generation respectively; L( ) is the global search function;

[0064] Step 5.2: If the random number is greater than the conversion probability corresponding to the population N, then perform a local search operation:

[0065]

[0066] In the formula, The distribution is the combined node group of the jth and kth meteorological factors in the population that evolve to the tth generation.

[0067] Another object of the present invention is to provide a meteorological monitoring system for smart agricultural planting, which is implemented by the meteorological monitoring method for smart agricultural planting, and the system includes:

[0068] Multi-parameter meteorological sensor: used to monitor meteorological parameters in farmland environment in real time;

[0069] Data acquisition unit: responsible for collecting multi-parameter meteorological sensor data and performing preliminary processing;

[0070] Communication module: transmits the processed data to the cloud server or central database via wireless or wired network;

[0071] Data processing and analysis platform: Use big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support;

[0072] User interface: used to monitor weather data and receive warning information in real time through mobile applications or computer software.

[0073] In combination with all the above technical solutions, the beneficial effects of the present invention are as follows: the present invention monitors key meteorological parameters in the farmland environment in real time through multi-parameter meteorological sensors, collects sensor data and performs preliminary processing, and transmits the collected data to a cloud server or a central database through a wireless or wired network; uses big data and artificial intelligence technologies to process and analyze the collected data to provide meteorological forecasts and decision support; users and agricultural managers monitor meteorological data and receive early warning information in real time through mobile applications or computer software, providing a theoretical basis for scientific decision-making and precise management of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings herein are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description, serve to explain the principles of the present disclosure;

[0075] Figure 1 It is a schematic diagram of a meteorological monitoring system for smart agricultural planting provided by an embodiment of the present invention;

[0076] Figure 2 This is a flow chart of a meteorological monitoring method for smart agricultural planting provided by an embodiment of the present invention;

[0077] In the figure: 1. Multi-parameter meteorological sensor; 2. Data acquisition unit; 3. Communication module; 4. Data processing and analysis platform; 5. User interface. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific implementation disclosed below.

[0079] Embodiment 1, as Figure 1 As shown, the meteorological monitoring system for smart agricultural planting provided by the embodiment of the present invention includes:

[0080] Multi-parameter meteorological sensor 1: used to monitor meteorological parameters in farmland environment in real time;

[0081] Data acquisition unit 2: responsible for collecting multi-parameter meteorological sensor data and performing preliminary processing;

[0082] Communication module 3: transmits the processed data to the cloud server or central database via wireless or wired network;

[0083] Data processing and analysis platform 4: Use big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support;

[0084] User interface 5: used to monitor weather data and receive warning information in real time through mobile applications or computer software

[0085] Embodiment 2, as Figure 2 As shown, the meteorological monitoring method for smart agricultural planting provided by the embodiment of the present invention includes:

[0086] S1, real-time monitoring of meteorological parameters in the farmland environment, including: air temperature and humidity, wind speed and direction, precipitation, light intensity, and soil temperature and humidity data;

[0087] S2, collects multi-parameter meteorological sensor data and performs preliminary processing;

[0088] S3, transmits the processed data to a cloud server or central database via wireless or wired networks;

[0089] S4, using big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support;

[0090] S5, monitor weather data and receive warning information in real time through mobile applications or computer software.

[0091] Embodiment 3, as a preferred embodiment, in step S2, collecting multi-parameter meteorological sensor data and performing preliminary processing includes:

[0092] S201, defining a language specification and encoding rules according to a multi-parameter meteorological sensor data encoder;

[0093] S202, reading the script content, parsing the coding rules, and translating the script content into a tag dictionary or a data analyzer code;

[0094] S203, submitting the tag dictionary or the code of the data analyzer to the data collection terminal engine and the cloud server engine;

[0095] S204, the data collection terminal engine interacts with the cloud server engine.

[0096] In step S203-step S204, the data acquisition terminal engine and the cloud server engine have built-in libraries, are supported by the data analyzer, and have out-of-the-box functions;

[0097] The data acquisition terminal engine and the cloud server engine are both distributed, and the connection is distributed;

[0098] In step S204, the data interaction between the data acquisition terminal cluster and the cloud server cluster specifically includes:

[0099] (1) The script executes the code with the data analyzer;

[0100] (2) The data acquisition terminal process creates a data communication agreement server and returns the communication agreement host and port information;

[0101] (3) The data collection terminal process starts the selection daemon process;

[0102] (4) Select the daemon process to create a communication protocol client;

[0103] (5) The data acquisition terminal process transmits the data communication port, host information, and data analyzer code through the communication protocol;

[0104] (6) Select the daemon process to receive the above information through the communication agreement and create a data communication agreement client;

[0105] (7) Select the data server that connects to the data collection terminal to obtain data;

[0106] (8) Select a daemon process to automatically send the data analyzer processing logic to the cloud server cluster for execution;

[0107] (9) The cloud server cluster execution results are returned to the selection daemon process and then transmitted to the data collection terminal process;

[0108] The data analyzer processing logic specifically includes training a data flow graph model with the data analyzer, making the data flow graph model into a database function and registering it on the data acquisition terminal, and then performing distributed execution on the cloud server;

[0109] The data analyzer processing logic is automatically sent to the cloud server cluster for execution to address the support of multi-parameter meteorological sensor data environment.

[0110] The most basic data abstraction in the data acquisition terminal engine is the elastic distributed data set of the data acquisition terminal structured query language, which represents an immutable, partitionable set whose elements can be calculated in parallel;

[0111] The cloud server engine is a high-performance distributed execution engine and an open source artificial intelligence framework. One of its goals is to allow developers to use a prototype algorithm running on a laptop and easily convert it into a high-performance distributed application suitable for running on a computer cluster or a single multi-core computer by just adding a few lines of code.

[0112] Embodiment 4, as a preferred embodiment, in step S4, using big data and artificial intelligence technology to process and analyze the collected data to provide weather forecast and decision support includes:

[0113] In the meteorological monitoring and identification for agricultural planting, the negative log-likelihood value of the exponential Gaussian model of the meteorological monitoring of the identification data is used as the objective function; the improved bee-ant optimization method is used to optimize the objective function, and the parameters μ, σ, τ corresponding to the function matching value are obtained; the above three parameters are used to fit the exponential Gaussian model in meteorological monitoring, and the data is monitored during the identification in the meteorological monitoring and identification for agricultural planting;

[0114] Calculate the maximum matching value and the average matching value in the population, and calculate the conversion probability P based on the maximum matching value and the average matching value, so that the algorithm can dynamically control the global meteorological forecast and local meteorological forecast behaviors according to the evolutionary period; based on the meteorological factor cross method, apply the excellent meteorological judgment value of the meteorological factor agricultural planting meteorological single factor to the next iterative prediction of the meteorological single factor for agricultural planting to obtain the optimal value in the optimization.

[0115] The objective function is optimized using the improved bee-ant optimization method, and the parameters μ, σ, τ corresponding to the optimal function matching value are obtained, including:

[0116] Step 1: Initialize the parameters of the improved bee-ant optimization method, including the maximum number of iterations N iter , flight index L, parameters [μ, σ, τ], population N;

[0117] Step 2: Calculate the matching degree of each meteorological single factor for agricultural planting in the population according to the matching degree function, obtain the global optimal meteorological single factor for agricultural planting in the current population, as well as the maximum matching degree value and the average matching degree value of the population, and calculate the conversion probability p;

[0118] Step 3, select the single factors for meteorological use for agricultural planting with matching degree j in the population to form the single factor sequence for meteorological use for agricultural planting of the population;

[0119] Step 4, randomly select 2 excellent agricultural planting meteorological single factors from the excellent agricultural planting meteorological single factor sequence for crossover to obtain a new excellent agricultural planting meteorological single factor, and compare the matching degree with the global optimal agricultural planting meteorological single factor of the current population. If it is less than, the excellent agricultural planting meteorological single factor after crossover enters the population; otherwise, the global optimal agricultural planting meteorological single factor enters various populations for agricultural planting meteorological single factor learning in the population;

[0120] Step 5: Generate a random number in the range of [0,1] and compare it with the conversion probability corresponding to the population N. If it is less than , perform a global search operation; otherwise, perform a local search operation;

[0121] Step 6: Update the global optimum of N.

[0122] In step 2, the calculation process of the maximum matching value and the average matching value is:

[0123] Step 2.1, calculate the single factor matching value y for agricultural planting and meteorological use of the population, the expression is:

[0124]

[0125] In the formula, is the probability density of the exponential Gaussian distribution in meteorological monitoring, i is the i-th identification time data obtained in a monitoring experiment, and t is the probability density of the t-th generation of the exponential Gaussian distribution in meteorological monitoring;

[0126] The specific calculation method is:

[0127]

[0128] Where pnf() is the normalized cumulative probability of the exponential Gaussian distribution in meteorological monitoring, eps is the floating point relative accuracy, exp() is the floating point relative accuracy calculation function, is the index of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the fixed exponent of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the variance of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the combined node group value of the ith meteorological factor in the population that has evolved to the tth generation;

[0129] Step 2.2: Calculate the maximum matching value f of the population max And the average matching value f mean :

[0130] f max =max(y)

[0131] f mean =mean(y)

[0132] In the formula, max(y) is the maximum value of the single factor matching degree of the meteorological use for agricultural planting of the population, and mean(y) is the average value of the single factor matching degree of the meteorological use for agricultural planting of the population;

[0133] Step 2.3, calculate the transition probability p:

[0134]

[0135] Where arccos() is the conversion function;

[0136] Select j agricultural planting meteorological single factors with excellent matching degree from various populations to form a population of excellent agricultural planting meteorological single factor sequence, and dynamically select excellent agricultural planting meteorological single factor j from the population composed of parameters to form an excellent agricultural planting meteorological single factor sequence m = {m 1 , m 2 …m j}, where the update rule for j<N is as follows:

[0137]

[0138] In the formula, γ is a constant, j min is the minimum value of the single factor for agricultural planting meteorology, j max is the maximum value of a single factor for agricultural planting and meteorology, N iter It is a single factor iterative value for meteorological use in agricultural planting.

[0139] Randomly select two excellent agricultural planting meteorological single factors from each excellent agricultural planting meteorological single factor sequence for crossover to obtain three new excellent agricultural planting meteorological single factors, and compare the matching degree with the global optimal agricultural planting meteorological single factor of the current population. If it is less than, the excellent agricultural planting meteorological single factors after crossover enter their respective populations; otherwise, the global optimal enters each population for learning other agricultural planting meteorological single factors in the population. The specific steps are as follows:

[0140] Step 4.1: Randomly select two different single factors m for excellent agricultural planting meteorology from the single factor sequence m for excellent agricultural planting meteorology. up , m down ;

[0141] Step 4.2: Cross the two excellent agricultural planting meteorological factors with single factors, and the expression is:

[0142] c * =rand×m up +(1-rand)×m down

[0143] In the formula, c * is the meteorological single factor crossover value, m up It is the single factor upward value for meteorological use in excellent agricultural planting, m down It is the single factor downward value for meteorological use in excellent agricultural planting, and rand is a random value;

[0144] Step 4.3: Use the single factor crossover value c for the new excellent agricultural planting meteorology * Compare the calculated matching degree with the global optimal matching degree of the population. If it is better, then c * Enter the population for single factor learning of meteorology for agricultural planting in the population.

[0145] In step 5, a random number in the range of [0,1] is generated and compared with the transition probability corresponding to the population N. If it is less than, a global search operation is performed, otherwise a local search operation is performed; including:

[0146] Step 5.1: If the random number is less than the conversion probability corresponding to the population N, then perform a global search operation:

[0147]

[0148] In the formula, g * It is the single factor for global optimal agricultural planting meteorology. are the combined node groups of the ith meteorological factor in the population evolved to the tth generation and the t+1th generation respectively; L() is the global search function;

[0149] Step 5.2: If the random number is greater than the conversion probability corresponding to the population N, then perform a local search operation:

[0150]

[0151] In the formula, The distribution is the combined node group of the jth and kth meteorological factors in the population that evolve to the tth generation.

[0152] It can be seen from the above multiple embodiments that in order to better reflect the AI ​​ecosystem, two engines are used, one is the data acquisition terminal, and the other is the cloud server. Both are distributed and connected. A data flow graph model is trained with a data analyzer, and this model is made into a database function and registered on the data acquisition terminal, and then distributedly executed on the cloud server. The user himself is unaware and just registers his own database function normally. In this way, batch processing and stream processing can be performed, and even exposed as an API to provide external services. Support for data can update and write data into the data.

[0153] Support for multiple environments. Traditional data collection terminals have a big problem. The data analyzer worker and Java Executor are together. This mixed architecture will have a problem. If you run a large query, the memory occupied by the data analyzer worker will be very large, or the CPU will be fully utilized, and there is no way to isolate resources. The present invention aims to realize that the data analyzer processing logic is automatically sent to the cloud server cluster for execution to solve the support for multiple environments. Support data analyzers, embrace the AI ​​ecosystem, and solve the data connection problem. The present invention can find the appropriate three parameters in the exponential Gaussian model in meteorological monitoring in the data during identification to analyze the top-down and bottom-up monitoring processing and response processes in meteorological monitoring and identification for agricultural planting, and provide empirical data and theoretical hypotheses for related analysis.

[0154] The experimental results show that in the laboratory test using the method of the present invention, the optimal value is among 60 results. The optimal value of the solution result of the exponential Gaussian model parameters in meteorological monitoring is closer to the theoretical optimal value, and better results can be found in the solution of the exponential Gaussian model parameters in meteorological monitoring.

[0155] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered within the protection scope of the present invention.

Claims

1. A meteorological monitoring method for smart agricultural planting, characterized in that: The method includes: S1, real-time monitoring of meteorological parameters in the farmland environment, including: air temperature and humidity, wind speed and direction, precipitation, light intensity, and soil temperature and humidity data; S2, collects multi-parameter meteorological sensor data and performs preliminary processing; S3, transmits the processed data to a cloud server or central database via wireless or wired networks; S4, using big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support; S5, monitor weather data and receive warning information in real time through mobile applications or computer software.

2. The meteorological monitoring method for smart agricultural planting according to claim 1, characterized in that: In step S2, multi-parameter meteorological sensor data is collected and preliminarily processed, including: (1) Define the language specifications and encoding rules based on the multi-parameter meteorological sensor data encoder; (2) Read the script content, parse the coding rules, and translate the script content into a tag dictionary or data analyzer code; (3) Submit the tag dictionary or data analyzer code to the data collection terminal engine and the cloud server engine; (4) The data acquisition terminal engine interacts with the cloud server engine.

3. The meteorological monitoring method for smart agricultural planting according to claim 2, characterized in that: In step (3), the data acquisition terminal engine and the cloud server engine have built-in libraries, supported by a data analyzer, and the most basic data abstraction in the data acquisition terminal engine is a resilient distributed data set of a data acquisition terminal structured query language.

4. The meteorological monitoring method for smart agricultural planting according to claim 2, characterized in that: In step (4), the data acquisition terminal engine interacts with the cloud server engine, including: (1) The script executes the code with the data analyzer; (2) The data acquisition terminal process creates a data communication agreement server and returns the communication agreement host and port information; (3) The data collection terminal process starts the selection daemon process; (4) Select the daemon process to create a communication protocol client; (5) The data acquisition terminal process transmits the data communication port, host information, and data analyzer code through the communication protocol; (6) Select the daemon process to receive the above information through the communication agreement and create a data communication agreement client; (7) Select the data server that connects to the data collection terminal to obtain data; (8) Select a daemon process to automatically send the data analyzer processing logic to the cloud server cluster for execution, use the data analyzer to train the data flow graph model, make the data flow graph model into a database function and register it on the data acquisition terminal, and perform distributed execution on the cloud server; (9) The execution results of the cloud server cluster are returned to the selection daemon process and then transmitted to the data acquisition terminal process.

5. The meteorological monitoring method for smart agricultural planting according to claim 1, characterized in that: In step S4, the transmitted data is processed and analyzed using big data and artificial intelligence technologies to provide meteorological forecasts and decision support, including: In the meteorological monitoring and identification for agricultural planting, the negative log-likelihood value of the exponential Gaussian model of the meteorological monitoring of the identification data is used as the objective function; the improved bee-ant optimization method is used to optimize the objective function, and the parameters μ, σ, τ corresponding to the function matching value are obtained; the above three parameters are used to fit the exponential Gaussian model in meteorological monitoring, and the data is monitored during the identification in the meteorological monitoring and identification for agricultural planting; Calculate the maximum matching value and the average matching value in the population, and calculate the conversion probability P based on the maximum matching value and the average matching value, so that the algorithm can dynamically control the global meteorological forecast and local meteorological forecast behaviors according to the evolutionary period; based on the meteorological factor cross method, apply the excellent meteorological judgment value of the meteorological factor agricultural planting meteorological single factor to the next iterative prediction of the meteorological single factor for agricultural planting to obtain the optimal value in the optimization.

6. The meteorological monitoring method for smart agricultural planting according to claim 5, characterized in that: The objective function is optimized using the improved bee-ant optimization method, and the parameters μ, σ, τ corresponding to the optimal function matching value are obtained, including: Step 1: Initialize the parameters of the improved bee-ant optimization method, including the maximum number of iterations N iter , flight index L, parameters [μ, σ, τ], population N; Step 2: Calculate the matching degree of each meteorological single factor for agricultural planting in the population according to the matching degree function, obtain the global optimal meteorological single factor for agricultural planting in the current population, as well as the maximum matching degree value and the average matching degree value of the population, and calculate the conversion probability p; Step 3, select the single factors for meteorological use for agricultural planting with matching degree j in the population to form the single factor sequence for meteorological use for agricultural planting of the population; Step 4, randomly select 2 excellent agricultural planting meteorological single factors from the excellent agricultural planting meteorological single factor sequence for crossover to obtain a new excellent agricultural planting meteorological single factor, and compare the matching degree with the global optimal agricultural planting meteorological single factor of the current population. If it is less than, the excellent agricultural planting meteorological single factor after crossover enters the population; otherwise, the global optimal agricultural planting meteorological single factor enters various populations for agricultural planting meteorological single factor learning in the population; Step 5: Generate a random number in the range of [0,1] and compare it with the conversion probability corresponding to the population N. If it is less than , perform a global search operation; otherwise, perform a local search operation; Step 6: Update the global optimum of N.

7. The meteorological monitoring method for smart agricultural planting according to claim 6, characterized in that: In step 2, the calculation process of the maximum matching value and the average matching value is: Step 2.1, calculate the single factor matching value y for agricultural planting and meteorological use of the population, the expression is: In the formula, is the probability density of the exponential Gaussian distribution in meteorological monitoring, i is the i-th identification time data obtained in a monitoring experiment, and t is the probability density of the t-th generation of the exponential Gaussian distribution in meteorological monitoring; The specific calculation method is: Where pnf() is the normalized cumulative probability of the exponential Gaussian distribution in meteorological monitoring, eps is the floating point relative accuracy, exp() is the floating point relative accuracy calculation function, is the index of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the fixed exponent of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the variance of the probability density of the tth generation of data in the i-th identification time that obeys the exponential Gaussian distribution in meteorological monitoring, is the combined node group value of the ith meteorological factor in the population that has evolved to the tth generation; Step 2.2: Calculate the maximum matching value f of the population max And the average matching value f mean : f max =max(y) f mean =mean(y) In the formula, max(y) is the maximum value of the single factor matching degree of the meteorological use for agricultural planting of the population, and mean(y) is the average value of the single factor matching degree of the meteorological use for agricultural planting of the population; Step 2.3, calculate the transition probability p: Where arccos( ) is the conversion function; Select j single factors for agricultural planting meteorology with excellent matching degrees from various groups to form a sequence of excellent single factors for agricultural planting meteorology. Dynamically select j excellent single factors for agricultural planting meteorology from the population composed of parameters respectively to form a sequence of excellent single factors for agricultural planting meteorology m = {m1, m2…m j}, where the update rule for j < N is as follows: In the formula, γ is a constant, j min is the minimum value of the single factor for agricultural planting meteorology, j max is the maximum value of a single factor for agricultural planting and meteorology, N iter It is a single factor iterative value for meteorological use in agricultural planting.

8. The meteorological monitoring method for smart agricultural planting according to claim 6, characterized in that: In step 4, two single factors for meteorological purposes for excellent agricultural planting are randomly selected from the single factor sequence for meteorological purposes for excellent agricultural planting and crossover is performed, including: Step 4.1: Randomly select two different single factors m for excellent agricultural planting meteorology from the single factor sequence m for excellent agricultural planting meteorology. up ,m down ; Step 4.2: Cross the two excellent agricultural planting meteorological factors with single factors, and the expression is: c * =row×m up +(1-row)×m down In the formula, c * is the meteorological single factor crossover value, m up It is the single factor upward value for meteorological use in excellent agricultural planting, m down It is the single factor downward value for meteorological use in excellent agricultural planting, and rand is a random value; Step 4.3: Use the single factor crossover value c for the new excellent agricultural planting meteorology * Compare the calculated matching degree with the global optimal matching degree of the population. If it is better, then c * Enter the population for single factor learning of meteorology for agricultural planting in the population.

9. The meteorological monitoring method for smart agricultural planting according to claim 6, characterized in that: In step 5, a random number in the range of [0,1] is generated and compared with the transition probability corresponding to the population N. If it is less than, a global search operation is performed, otherwise a local search operation is performed; including: Step 5.1: If the random number is less than the conversion probability corresponding to the population N, then perform a global search operation: In the formula, g * It is the single factor for global optimal agricultural planting meteorology. are the combined node groups of the ith meteorological factor in the population evolved to the tth generation and the t+1th generation respectively; L() is the global search function; Step 5.2: If the random number is greater than the conversion probability corresponding to the population N, then perform a local search operation: In the formula, The distribution is the combined node group of the jth and kth meteorological factors in the population that evolve to the tth generation.

10. A meteorological monitoring system for smart agricultural planting, characterized in that: The system is implemented by the meteorological monitoring method for smart agricultural planting according to any one of claims 1 to 9, and the system includes: Multi-parameter meteorological sensor (1): used to monitor meteorological parameters in farmland environment in real time; Data acquisition unit (2): responsible for collecting multi-parameter meteorological sensor data and performing preliminary processing; Communication module (3): transmits the processed data to a cloud server or a central database via a wireless or wired network; Data processing and analysis platform (4): uses big data and artificial intelligence technologies to process and analyze transmitted data to provide weather forecasts and decision support; User interface (5): used to monitor meteorological data in real time and receive warning information through mobile applications or computer software.

Images