Agricultural management method and system based on artificial intelligence

Abstract

The present invention provides an agricultural management method and system based on artificial intelligence, the method comprising: obtaining a first environmental parameter of different positions in a garden; inputting the processing parameter into a first prediction model to output a predicted humidity, obtaining the growth stage of vegetation in the garden, and obtaining the growth humidity required for the vegetation based on the growth stage; calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy reasoning structure to obtain a preliminary irrigation time; obtaining a second environmental parameter of the garden, and eliminating some parameters of the second environmental parameter based on an importance attribute sorting result; inputting the second environmental parameter with some parameters eliminated into a second prediction model to output a predicted transpiration amount, and calculating the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time, thereby saving manpower and improving work efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of agricultural data management, and specifically relates to an agricultural management method and system based on artificial intelligence. Background Art

[0002] In order to ensure that crops can grow normally and achieve high and stable yields, it is crucial to provide sufficient water. However, under natural conditions, precipitation often cannot meet the water needs of crops, either because the precipitation is insufficient or unevenly distributed. These factors may result in crops being unable to obtain the required water, thus affecting their growth and yield.

[0003] In arid and semi-arid areas, due to insufficient natural rainfall, artificial irrigation has become a key measure to ensure crop growth. Through artificial irrigation, farmers can supplement the water needed by crops and ensure that crops can get the necessary water supply even in drought conditions.

[0004] Currently, artificial irrigation usually relies on farmers' experience and observation of crop growth to determine the timing and amount of irrigation. This method has the following limitations:

[0005] Reliance on experience: Irrigation decisions rely on the operator’s experience, which may result in under- or over-irrigation.

[0006] Manpower consuming: manual monitoring and operation are required, which increases labor intensity and cost.

[0007] Inefficiency: It is difficult to accurately control irrigation time and water volume, which may result in waste of water resources.

[0008] Lack of intelligence: Unable to make dynamic adjustments based on real-time data and actual crop needs.

[0009] In order to overcome the limitations of traditional irrigation, an agricultural management method and system based on artificial intelligence is provided. Summary of the invention

[0010] In order to solve the above technical problems, the present invention provides an agricultural management method and system based on artificial intelligence, which are used to solve the technical problems in the prior art.

[0011] On the one hand, the invention provides the following technical solution, an agricultural management method based on artificial intelligence, the method comprising:

[0012] Acquire first environmental parameters at different positions of the garden, and pre-process the first environmental parameters at different positions using interpolation technology and averaging technology to obtain processing parameters;

[0013] Inputting the processing parameters into a first prediction model to output predicted humidity, obtaining the growth stage of the vegetation in the garden, and obtaining the growth humidity required for the vegetation based on the growth stage;

[0014] Calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time;

[0015] Acquire a second environmental parameter of the garden, sort the importance attributes of the second environmental parameter based on a gradient boosting model, and remove some parameters of the second environmental parameter based on the importance attribute sorting result;

[0016] The second environmental parameter with some parameters removed is input into the second prediction model to output the predicted transpiration amount, and the actual irrigation time is calculated based on the predicted transpiration amount and the preliminary irrigation time.

[0017] Compared with the prior art, the beneficial effects of the present application are as follows: by inputting the humidity deviation and the rate of change of the humidity deviation into the fuzzy reasoning structure, the accuracy of the preliminary irrigation time can be effectively improved. At the same time, by calculating the actual irrigation time based on the predicted transpiration and the preliminary irrigation time, the accuracy of the irrigation time can be further improved, thereby effectively preventing excessive irrigation rate or overwatering. In addition, by calculating the actual irrigation time, manual irrigation is replaced, the intelligence of agricultural management is improved, manpower is saved, and work efficiency is improved.

[0018] Furthermore, the step of obtaining the first environmental parameters at different locations of the garden includes:

[0019] Acquire initial first environmental parameters of the garden, analyze the initial first environmental parameters using correlation analysis technology to obtain first environmental parameters, and continuously acquire the first environmental parameters at different locations of the garden at first preset time intervals, wherein the initial first environmental parameters include air temperature, air humidity, light intensity, carbon dioxide concentration, soil temperature, soil moisture, soil conductivity, wind speed, and wind direction, and the first environmental parameters include air temperature, air humidity, light intensity, soil temperature and soil moisture.

[0020] Furthermore, the interpolation technology includes:

[0021] Determine whether there is data missing for the first environmental parameter, if there is data missing for the first environmental parameter;

[0022] Then determine whether the span of the missing data is less than the threshold, if the span of the missing data is less than the threshold;

[0023] Then, the interpolation formula is used to calculate based on the original first environmental parameter to obtain interpolation data, and the interpolation data is added to the missing data to form a completed first environmental parameter;

[0024] Wherein, the interpolation formula includes:

[0025]

[0026] In the formula, Represented as the missing first environmental parameter, Represented as the first known environmental parameter, located at the beginning of the sequence, Represented as another known first environmental parameter, located in the sequence After Locations, The first environmental parameter represented by the missing The number of intervals between Expressed as and The number of intervals between Expressed as and The number of intervals between

[0027] The averaging techniques include:

[0028] ；

[0029] Expressed as The average value of the first environmental parameter, Expressed as the number of the first environmental parameters, Represents the first environmental parameters at different times.

[0030] Furthermore, the step of obtaining the second environmental parameter of the garden includes:

[0031] Continuously acquiring a second environmental parameter of the garden at a second preset time interval, and using an abnormality judgment formula to judge whether there is an abnormal second environmental parameter in the second environmental parameter;

[0032] If there is an abnormal second environmental parameter among the second environmental parameters, deleting the abnormal second environmental parameter;

[0033] The abnormal repair formula is used to calculate based on the non-abnormal second environmental parameter to obtain the repair data;

[0034] The repair data replaces the abnormal second environmental parameters to obtain complete second environmental parameters.

[0035] Furthermore, the abnormality judgment formula includes:

[0036] ;

[0037] In the formula, Represents the first The second environmental parameter, Expressed as the average value of all second environmental parameters, represents the standard deviation of the data set formed for all second environmental parameters, It is expressed as the test coefficient for measuring the second environmental parameter, where: , represent the maximum and minimum values ​​in the data set formed by the second environmental parameter, respectively;

[0038] The anomaly repair formula includes:

[0039] ；

[0040] In the formula, Represents repair data, The location of the second environmental parameter represented as an anomaly in the dataset, It is expressed as the number of adjacent second environmental parameters selected on both sides of the abnormal second environmental parameter position. Represents the index variable in the summation process, Represents the first A second environmental parameter.

[0041] Furthermore, the step of sorting the importance attributes of the second environmental parameters based on the gradient boosting model and eliminating some parameters of the second environmental parameters based on the importance attribute sorting result includes:

[0042] Inputting the second environmental parameter into a gradient boosting model to obtain an importance score of each parameter of the second environmental parameter, and sorting all the parameters according to the importance score;

[0043] Sequentially select different preset thresholds to determine parameters whose importance scores are lower than the selected preset thresholds, and remove parameters whose importance scores are lower than the selected preset thresholds to form multiple sets of feature sets;

[0044] Sequentially dividing the plurality of feature sets into validation sets and a plurality of training sets, and using the validation set and the plurality of training sets of each group as training and validation data for the gradient boosting model;

[0045] The performance index of the gradient boosting model in the verification process is calculated, and an optimal set of features is obtained based on the performance index to form the second environmental parameters with some parameters eliminated.

[0046] Furthermore, after the step of obtaining the growth humidity required for the vegetation based on the growth stage, the method further includes:

[0047] Acquiring actual soil humidity in the garden, and determining whether the actual soil humidity is less than the growth soil humidity;

[0048] If the actual humidity is less than the growth humidity, the nitrogen fertilizer requirement of the vegetation is obtained by using a fertilizer dosage formula based on the growth stage, and the nitrogen fertilizer requirement is integrated with the actual irrigation time to fertilize and irrigate the vegetation;

[0049] The fertilization dosage formula includes:

[0050] ；

[0051] in, Expressed as the yield of vegetation, Expressed as nitrogen fertilizer requirement, It is expressed as a quadratic coefficient, representing the acceleration of vegetation yield change when nitrogen fertilizer demand increases. Expressed as a linear coefficient, it represents the rate at which vegetation yield changes when nitrogen fertilizer demand increases. Expressed as a constant term, it represents the baseline vegetation yield when the nitrogen fertilizer demand is zero.

[0052] In the second aspect, the invention provides the following technical solution: an agricultural management system based on artificial intelligence, the system comprising:

[0053] A processing module, used for obtaining first environmental parameters at different positions of the garden, and preprocessing the first environmental parameters at different positions by using interpolation technology and averaging technology to obtain processing parameters;

[0054] A required module is used to input the processing parameters into a first prediction model to output predicted humidity, obtain the growth stage of the vegetation in the garden, and obtain the growth humidity required for the vegetation based on the growth stage;

[0055] An input module, used for calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time;

[0056] A elimination module is used to obtain a second environmental parameter of the garden, sort the importance attributes of the second environmental parameter based on a gradient boosting model, and eliminate some parameters of the second environmental parameter based on the importance attribute sorting result;

[0057] The calculation module is used to input the second environmental parameter with some parameters removed into the second prediction model to output the predicted transpiration amount, and calculate the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time.

[0058] In the third aspect, the invention provides the following technical solution: a computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the agricultural management method based on artificial intelligence as described above is implemented.

[0059] In a fourth aspect, the invention provides the following technical solution: a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned artificial intelligence-based agricultural management method. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0061] Figure 1 A flowchart of an agricultural management method based on artificial intelligence provided by the first embodiment of the present invention;

[0062] Figure 2 A structural block diagram of an agricultural management system based on artificial intelligence provided by a second embodiment of the present invention;

[0063] Figure 3 A schematic diagram of the hardware structure of a computer provided in the third embodiment of the present invention.

[0064] The embodiments of the present invention will be further described below with reference to the accompanying drawings. DETAILED DESCRIPTION

[0065] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout are the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.

[0066] In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0067] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0068] Embodiment 1

[0069] In the first embodiment of the present invention, see Figure 1 As shown, an agricultural management method based on artificial intelligence includes the following steps S01 to S05:

[0070] S01, obtaining first environmental parameters at different locations of a garden, and preprocessing the first environmental parameters at different locations using an interpolation technique and an averaging technique to obtain processing parameters;

[0071] Optionally, the first environmental parameter may be a parameter of different gardens, so the garden may be a flower bed, a lawn area, a shrub area, a greenhouse area and an orchard. The following is an illustration of a grapefruit orchard.

[0072] Specifically, the step of obtaining the first environmental parameters at different locations of the garden includes:

[0073] Acquire initial first environmental parameters of the garden, analyze the initial first environmental parameters using correlation analysis technology to obtain first environmental parameters, and continuously acquire the first environmental parameters at different locations of the garden at first preset time intervals, wherein the initial first environmental parameters include air temperature, air humidity, light intensity, carbon dioxide concentration, soil temperature, soil moisture, soil conductivity, wind speed, and wind direction, and the first environmental parameters include air temperature, air humidity, light intensity, soil temperature and soil moisture.

[0074] In this embodiment, the Spearman's rank correlation analysis selected in the correlation analysis technology is also called the Spearman's rank correlation coefficient, which is a non-parametric statistical measurement method used to evaluate the correlation or dependence between two variables. The weighted fusion technology is the average technology and the normalization processing technology.

[0075] Specific;

[0076] The interpolation techniques include:

[0077] S11, determining whether there is data missing for the first environmental parameter, if there is data missing for the first environmental parameter;

[0078] S12, determining whether the span of the missing data is smaller than a threshold value, if the span of the missing data is smaller than the threshold value;

[0079] S13, using an interpolation formula to calculate based on the original first environmental parameter to obtain interpolation data, and filling the missing data with the interpolation data to form a completed first environmental parameter;

[0080] After the step of determining whether the span of the missing first environmental parameter is less than a threshold, the method further includes:

[0081] S14, if the span of the missing data is greater than a threshold, interpolating the missing data using data with the same weather conditions as the day when the first environmental parameter was collected;

[0082] Wherein, the interpolation formula includes:

[0083]

[0084] In the formula, Represented as the missing first environmental parameter, Represented as the first known environmental parameter, located at the beginning of the sequence, Represented as another known first environmental parameter, located in the sequence After Locations, The first environmental parameter represented by the missing The number of intervals between Expressed as and The number of intervals between Expressed as and The number of intervals between

[0085] The averaging techniques include:

[0086] ；

[0087] Expressed as The average value of the first environmental parameter, Expressed as the number of the first environmental parameters, Indicates the first environmental parameters at different times. In this embodiment, The value of 48 is the environmental parameter for one day, that is, the environmental parameters are collected every half an hour.

[0088] It is worth noting that sensor data loss is a common problem in industry and scientific research, especially when network transmission and device stability are poor. This data loss has a significant impact on the accuracy of model predictions.

[0089] Optionally, the step of preprocessing the first environmental parameters at different locations further includes:

[0090] The first environmental parameters at different locations are fused using an adaptive weighting algorithm to obtain more comprehensive and reliable orchard environmental information. The data from different sensors are fused and denoised using an adaptive weighting-based algorithm to improve the quality and effect of data processing. The algorithm automatically adjusts the weight according to the importance or reliability of each sensor data, thereby optimizing the results of data fusion.

[0091] Specifically, when determining whether to perform step S02, the method further includes:

[0092] Determine whether it rains on the current day or in the next day. If it rains on the current day or in the next day, continue to execute step S01.

[0093] S02, inputting the processing parameters into a first prediction model to output predicted humidity, obtaining the growth stage of the vegetation in the garden, and obtaining the growth humidity required for the vegetation based on the growth stage;

[0094] In this embodiment, the first prediction model is the SSA-Elman model, which is a hybrid model that combines the Sparrow Search Algorithm (SSA) and the Elman neural network, and is used to solve dynamic optimization problems and time series prediction.

[0095] Explanation of how the SSA-Elman model works:

[0096] 1. Initialization:

[0097] In the SSA-Elman model, initialization involves generating a group of sparrows, each of which represents a set of parameters in the Elman neural network. These parameters may include the network's weights, biases, etc.

[0098] These parameters are randomly generated and cover a wide region of the parameter space to ensure that the search process can explore the global optimal solution.

[0099] 2. Network training:

[0100] Each sparrow (i.e. each set of parameters) is used to configure an Elman neural network, and then the network is trained using the training dataset.

[0101] The Elman network is able to capture dynamic features in time series data through its recursive connections, thereby learning patterns in the data.

[0102] 3. Fitness evaluation:

[0103] After training, the performance of each network on the validation set is used to evaluate its fitness. The fitness function is usually based on the prediction error, such as mean squared error (MSE) or root mean squared error (RMSE).

[0104] The purpose of fitness evaluation is to distinguish which network configurations (sparrows) have better performance and are more likely to be close to the global optimal solution.

[0105] 4. Update the sparrow position:

[0106] In SSA, each sparrow updates its position (i.e., network parameters) based on its fitness and the fitness of other sparrows.

[0107] Excellent sparrows (solutions with higher fitness) will attract other sparrows to move to their location, which is achieved by simulating the social behavior of sparrows.

[0108] At the same time, in order to maintain the diversity of the population and avoid converging to the local optimal solution too early, sparrows also perform random exploration.

[0109] 5. Iterative optimization:

[0110] By repeating the training, evaluation, and position update processes, the model is continuously iteratively optimized.

[0111] In each iteration, the positions of all sparrows (network parameters) are updated according to the rules of SSA.

[0112] The iterative process continues until the termination condition is met, such as reaching the maximum number of iterations, the fitness no longer improves significantly, or reaching a predetermined fitness threshold.

[0113] 6. Output the optimal solution:

[0114] After the iteration process is over, the network parameters represented by the sparrow with the highest fitness are selected as the optimal solution.

[0115] This optimal solution provides the best network configuration, which enables the Elman neural network to achieve the best performance on specific tasks.

[0116] In this embodiment, the growth stages of the pomelo trees in the pomelo orchard include budding and leafing stage, flowering and fruiting stage, fruit expansion stage and maturity stage, wherein the optimal growth humidity of the pomelo trees is different in each growth stage.

[0117] S03, calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time;

[0118] In this embodiment, the fuzzy reasoning structure is Mamdani type fuzzy logic reasoning.

[0119] By using the humidity deviation and the rate of change of the humidity deviation as the input of the fuzzy inference structure, the accuracy of the initial irrigation time can be effectively improved.

[0120] S04, obtaining a second environmental parameter of the garden, sorting the importance attributes of the second environmental parameter based on a gradient boosting model, and eliminating some parameters of the second environmental parameter based on the importance attribute sorting result;

[0121] Specifically, the step of obtaining the second environmental parameter of the garden includes:

[0122] S41, continuously acquiring a second environmental parameter of the garden at a second preset time interval, and determining whether the second environmental parameter has an abnormal second environmental parameter by using an abnormality determination formula;

[0123] S42, if there is an abnormal second environmental parameter among the second environmental parameters, deleting the abnormal second environmental parameter;

[0124] S43, using an abnormal repair formula to calculate based on a non-abnormal second environmental parameter to obtain repair data;

[0125] S44, replacing the abnormal second environmental parameter with the repair data to obtain a complete second environmental parameter.

[0126] More specifically, the abnormality judgment formula includes:

[0127] ;

[0128] In the formula, Represents the first The second environmental parameter, Expressed as the average value of all second environmental parameters, represents the standard deviation of the data set formed for all second environmental parameters, It is expressed as the test coefficient for measuring the second environmental parameter, where: , represent the maximum and minimum values ​​in the data set formed by the second environmental parameter, respectively;

[0129] The anomaly repair formula includes:

[0130] ；

[0131] In the formula, Represents repair data, The location of the second environmental parameter represented as an anomaly in the dataset, It is expressed as the number of adjacent second environmental parameters selected on both sides of the abnormal second environmental parameter position. Represents the index variable in the summation process, Represents the first A second environmental parameter.

[0132] In this embodiment, the second environmental parameters include air temperature and humidity, soil temperature and humidity, pH, light, CO2 concentration, atmospheric pressure and dew point temperature.

[0133] Specifically, the step of sorting the importance attributes of the second environmental parameters based on the gradient boosting model and eliminating some parameters of the second environmental parameters based on the importance attribute sorting result includes:

[0134] S45, inputting the second environmental parameter into a gradient boosting model to obtain an importance score of each parameter of the second environmental parameter, and sorting all the parameters according to the importance score;

[0135] S46, sequentially selecting different preset thresholds to determine parameters whose importance scores are lower than the selected preset thresholds, and removing parameters whose importance scores are lower than the selected preset thresholds to form multiple sets of feature sets;

[0136] S47, dividing the plurality of groups of feature sets into validation sets and a plurality of training sets in turn, and using the validation set and the plurality of training sets of each group as training and validation data for the gradient boosting model;

[0137] S48, calculating the performance index of the gradient boosting model during the verification process, obtaining an optimal set of features based on the performance index, and forming the second environmental parameters by eliminating some parameters.

[0138] In this embodiment, the gradient boosting model is a LightGBM model. The performance indicator is the root mean square error (RMSE).

[0139] Among them, step S46 can be understood as based on the importance score of the parameter, we can construct feature sets of different sizes. The feature sets include:

[0140] Minimal feature set: may contain only a few features with the highest scores.

[0141] Medium-sized feature sets: May include a progressively larger number of high-scoring features.

[0142] Maximum feature set: This may include all features, or all features except for the lowest-scoring features. In practice, this may mean removing the lowest-scoring features from all features and then using the remaining features for model training and validation. For example:

[0143] The features with the lowest scores of 10% (one of different preset thresholds) are removed, and the remaining features are used for training and validation.

[0144] Remove the features with the lowest 20% scores (one of the different preset thresholds) and train and validate again.

[0145] And so on, until the optimal feature set is found.

[0146] Among them, step S47 can be understood as:

[0147] Divide the dataset: Divide the training dataset into multiple smaller datasets, usually called "folds". A common division method is k-fold cross-validation, where k is a preset number, indicating that the dataset is divided into k parts.

[0148] Iterative training and validation: For each fold, use that fold as the validation set and the remaining k-1 folds as the training set. In this process, the model is trained using the training set and then the validation set is used to evaluate the performance of the model.

[0149] In this embodiment, the second environmental parameters excluding some parameters, that is, the optimal set of features includes: air temperature, light intensity, air humidity, soil humidity and soil temperature.

[0150] S05, inputting the second environmental parameters with some parameters removed into a second prediction model to output a predicted transpiration amount, and calculating the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time.

[0151] In this embodiment, the second prediction model is a DA-LSTM (Dilated Attention Long Short-Term Memory) model, which is a deep learning model that combines a dilated attention mechanism and a long short-term memory network.

[0152] In this embodiment, the step of calculating the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time includes:

[0153] The predicted transpiration amount is calculated based on the time calculation formula to obtain the transpiration amount predicted irrigation time, and the average value of the preliminary irrigation time and the transpiration amount predicted irrigation time is calculated to obtain the actual irrigation time.

[0154] The time calculation formula includes:

[0155] ;

[0156] In the formula, It is expressed as the predicted irrigation time for transpiration, Expressed as the amount of water required by plants, Expressed as the water supply rate of the irrigation system, Expressed as the efficiency of the irrigation system;

[0157] ;

[0158] in, Expressed as the predicted transpiration, The amount of water used by plants for growth and metabolism, Expressed as loss during irrigation.

[0159] By setting the step of calculating the actual irrigation time based on the predicted transpiration and the preliminary irrigation time, the accuracy of the irrigation time can be effectively improved. On the other hand, it replaces manual irrigation, improves the intelligence of agricultural management, saves manpower and improves work efficiency.

[0160] Specifically, after the step of obtaining the growth humidity required for the vegetation based on the growth stage, the method further includes:

[0161] Acquire the actual humidity of the soil in the garden, and determine whether the actual humidity is less than the growth humidity;

[0162] If the actual humidity is less than the growth humidity, the nitrogen fertilizer requirement of the vegetation is obtained by using a fertilizer dosage formula based on the growth stage, and the nitrogen fertilizer requirement is integrated with the actual irrigation time to fertilize and irrigate the vegetation;

[0163] The fertilization dosage formula includes:

[0164] ；

[0165] in, Expressed as the yield of vegetation, Expressed as nitrogen fertilizer requirement, It is expressed as a quadratic coefficient, representing the acceleration of vegetation yield change when nitrogen fertilizer demand increases. Expressed as a linear coefficient, it represents the rate at which vegetation yield changes when nitrogen fertilizer demand increases. Expressed as a constant term, it represents the baseline vegetation yield when the nitrogen fertilizer demand is zero.

[0166] It is understandable that if irrigation water is needed, we can obtain the nitrogen fertilizer needed by the vegetation at this time, and then add the required amount of nitrogen fertilizer to the water so that the water and nitrogen fertilizer can be irrigated together.

[0167] In this embodiment, when the grapefruit tree is in the budding and leafing stage, , and They are -1.5645, 299.8 and -13653 respectively, that is, the nitrogen fertilizer requirement X=95.8kg / hm 2 The yield of grapefruit trees reaches its maximum when the grapefruit trees are in the flowering and fruiting period. , and They are -0.7251, 208.01 and -14207 respectively, so the nitrogen fertilizer requirement is X=95.8kg / hm 2 The yield of grapefruit trees reaches its maximum when the fruit is in the expansion period. , and They are -0.4206, 160.68 and -14638 respectively, so the nitrogen fertilizer requirement X=191.1 kg / hm 2 When the grapefruit tree is in its mature stage, , and They are -7.4673, 711.35 and -16233 respectively, so the nitrogen fertilizer requirement X = 47.6kg / hm 2 The yield of grapefruit trees reaches its maximum when

[0168] In summary, an agricultural management method based on artificial intelligence has the following effects:

[0169] By inputting the humidity deviation and the rate of change of the humidity deviation into the fuzzy reasoning structure, the accuracy of the preliminary irrigation time can be effectively improved. At the same time, by calculating the actual irrigation time based on the predicted transpiration and the preliminary irrigation time, the accuracy of the irrigation time can be further improved, thereby effectively preventing excessive irrigation rate or overwatering. In addition, the step of calculating the actual irrigation time replaces manual irrigation, improves the intelligence of agricultural management, saves manpower, and improves work efficiency.

[0170] Embodiment 2

[0171] like Figure 2 As shown, in a second embodiment of the present invention, an agricultural management system based on artificial intelligence is provided, and the system comprises:

[0172] The processing module 10 is used to obtain the first environmental parameters at different positions of the garden, and pre-process the first environmental parameters at different positions by using interpolation technology and averaging technology to obtain processing parameters;

[0173] The required module 20 is used to input the processing parameters into the first prediction model to output the predicted humidity, obtain the growth stage of the vegetation in the garden, and obtain the growth humidity required for the vegetation based on the growth stage;

[0174] An input module 30, for calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time;

[0175] A elimination module 40 is used to obtain a second environmental parameter of the garden, sort the importance attributes of the second environmental parameter based on a gradient boosting model, and eliminate some parameters of the second environmental parameter based on the importance attribute sorting result;

[0176] The calculation module 50 is used to input the second environmental parameter with some parameters removed into the second prediction model to output the predicted transpiration amount, and calculate the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time.

[0177] The artificial intelligence-based agricultural management system provided in the embodiment of the present invention has the same implementation principle and technical effects as those in the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0178] Embodiment 3

[0179] like Figure 3 As shown, in the third embodiment of the present invention, the embodiment of the present invention provides the following technical solution, a computer, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201, and when the processor 201 executes the computer program, the agricultural management method based on artificial intelligence as described above is implemented.

[0180] Specifically, the processor 201 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0181] Among them, the memory 202 may include a large-capacity memory for data or instructions. For example, but not limitation, the memory 202 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 202 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 202 may be inside or outside the data processing device. In a specific embodiment, the memory 202 is a non-volatile memory. In a specific embodiment, the memory 202 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0182] The memory 202 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 201 .

[0183] The processor 201 implements the above-mentioned agricultural management method based on artificial intelligence by reading and executing computer program instructions stored in the memory 202.

[0184] In some embodiments, the computer may further include a communication interface 203 and a bus 200. Figure 3 As shown, the processor 201, the memory 202, and the communication interface 203 are connected via a bus 200 and communicate with each other.

[0185] The communication interface 203 is used to implement communication between the modules, devices, units and / or equipment in the embodiment of the present application. The communication interface 203 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0186] The bus 200 includes hardware, software or both, and couples the components of the computer to each other. The bus 200 includes but is not limited to at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 200 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, bus 200 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0187] Embodiment 4

[0188] In the fourth embodiment of the present invention, in combination with the above-mentioned agricultural management method based on artificial intelligence, the embodiment of the present invention provides the following technical solution, a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned agricultural management method based on artificial intelligence is implemented.

[0189] Those skilled in the art will appreciate that the logic and / or steps described in the flowchart as data or in other ways herein, for example, can be considered as a sequenced data table of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses.

[0190] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0191] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0192] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0193] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. An agricultural management method based on artificial intelligence, characterized in that: The method comprises: Acquire first environmental parameters at different positions of the garden, and pre-process the first environmental parameters at different positions using interpolation technology and averaging technology to obtain processing parameters; Inputting the processing parameters into a first prediction model to output predicted humidity, obtaining the growth stage of the vegetation in the garden, and obtaining the growth humidity required for the vegetation based on the growth stage; Calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time; Acquire a second environmental parameter of the garden, sort the importance attributes of the second environmental parameter based on a gradient boosting model, and remove some parameters of the second environmental parameter based on the importance attribute sorting result; Inputting the second environmental parameter with some parameters removed into the second prediction model to output a predicted transpiration amount, and calculating the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time; The step of sorting the importance attributes of the second environmental parameters based on the gradient boosting model and eliminating some parameters of the second environmental parameters based on the importance attribute sorting result comprises: Inputting the second environmental parameter into a gradient boosting model to obtain an importance score of each parameter of the second environmental parameter, and sorting all the parameters according to the importance score; Sequentially select different preset thresholds to determine parameters whose importance scores are lower than the selected preset thresholds, and remove parameters whose importance scores are lower than the selected preset thresholds to form multiple sets of feature sets; Sequentially dividing the plurality of feature sets into validation sets and a plurality of training sets, and using the validation set and the plurality of training sets of each group as training and validation data for the gradient boosting model; Calculating the performance index of the gradient boosting model during the verification process, obtaining an optimal set of features based on the performance index, and forming the second environment parameters by eliminating some parameters; The step of calculating the actual irrigation time based on the predicted transpiration and the preliminary irrigation time comprises: The predicted transpiration amount is calculated based on the time calculation formula to obtain the transpiration amount predicted irrigation time, and the average value of the preliminary irrigation time and the transpiration amount predicted irrigation time is calculated to obtain the actual irrigation time.

2. The agricultural management method based on artificial intelligence according to claim 1, characterized in that: The step of obtaining the first environmental parameters at different locations of the garden comprises: Acquire initial first environmental parameters of the garden, analyze the initial first environmental parameters using correlation analysis technology to obtain first environmental parameters, and continuously acquire the first environmental parameters at different locations of the garden at first preset time intervals, wherein the initial first environmental parameters include air temperature, air humidity, light intensity, carbon dioxide concentration, soil temperature, soil moisture, soil conductivity, wind speed, and wind direction, and the first environmental parameters include air temperature, air humidity, light intensity, soil temperature and soil moisture.

3. The agricultural management method based on artificial intelligence according to claim 2, characterized in that: The interpolation techniques include: Determine whether there is data missing for the first environmental parameter, if there is data missing for the first environmental parameter; Then determine whether the span of the missing data is less than the threshold, if the span of the missing data is less than the threshold; Then, the interpolation formula is used to calculate based on the original first environmental parameter to obtain interpolation data, and the interpolation data is added to the missing data to form a completed first environmental parameter; Wherein, the interpolation formula includes: In the formula, Represented as the missing first environmental parameter, Represented as the first known environmental parameter, located at the beginning of the sequence, Represented as another known first environmental parameter, located in the sequence After Locations, Expressed as and The number of intervals between Expressed as and The number of intervals between The averaging techniques include: Expressed as The average value of the first environmental parameter, Expressed as the number of the first environmental parameters, Represents the first environmental parameters at different times.

4. The agricultural management method based on artificial intelligence according to claim 1, characterized in that: The step of obtaining the second environmental parameter of the garden comprises: Continuously acquiring a second environmental parameter of the garden at a second preset time interval, and using an abnormality judgment formula to judge whether there is an abnormal second environmental parameter in the second environmental parameter; If there is an abnormal second environmental parameter among the second environmental parameters, deleting the abnormal second environmental parameter; The abnormal repair formula is used to calculate based on the non-abnormal second environmental parameter to obtain the repair data; The repair data replaces the abnormal second environmental parameters to obtain complete second environmental parameters.

5. The agricultural management method based on artificial intelligence according to claim 4, characterized in that: The abnormality judgment formula includes: In the formula, Represents the first The second environmental parameter, Expressed as the average value of all second environmental parameters, represents the standard deviation of the data set formed for all second environmental parameters, It is expressed as the test coefficient for measuring the second environmental parameter, where: , represent the maximum and minimum values ​​in the data set formed by the second environmental parameter, respectively; The anomaly repair formula includes: In the formula, Represents repair data, The location of the second environmental parameter represented as an anomaly in the dataset, It is expressed as the number of adjacent second environmental parameters selected on both sides of the abnormal second environmental parameter position. Represents the index variable in the summation process, Represents the first A second environmental parameter.

6. The agricultural management method based on artificial intelligence according to claim 1, characterized in that: After the step of obtaining the growth humidity required for the vegetation based on the growth stage, the method further includes: Acquiring actual soil humidity in the garden, and determining whether the actual soil humidity is less than the growth soil humidity; If the actual humidity is less than the growth humidity, the nitrogen fertilizer requirement of the vegetation is obtained by using a fertilizer dosage formula based on the growth stage, and the nitrogen fertilizer requirement is integrated with the actual irrigation time to fertilize and irrigate the vegetation; The fertilization dosage formula includes: in, Expressed as the yield of vegetation, Expressed as nitrogen fertilizer requirement, It is expressed as a quadratic coefficient, representing the acceleration of vegetation yield change when nitrogen fertilizer demand increases. Expressed as a linear coefficient, it represents the rate at which vegetation yield changes when nitrogen fertilizer demand increases. Expressed as a constant term, it represents the baseline vegetation yield when the nitrogen fertilizer demand is zero.

7. An agricultural management system based on artificial intelligence, characterized in that: The system comprises: A processing module, used for obtaining first environmental parameters at different positions of the garden, and preprocessing the first environmental parameters at different positions by using interpolation technology and averaging technology to obtain processing parameters; A required module is used to input the processing parameters into a first prediction model to output predicted humidity, obtain the growth stage of the vegetation in the garden, and obtain the growth humidity required for the vegetation based on the growth stage; An input module, used for calculating a humidity deviation based on the predicted humidity and the growth humidity, and inputting the humidity deviation and the rate of change of the humidity deviation into a fuzzy inference structure to obtain a preliminary irrigation time; A elimination module is used to obtain a second environmental parameter of the garden, sort the importance attributes of the second environmental parameter based on a gradient boosting model, and eliminate some parameters of the second environmental parameter based on the importance attribute sorting result; A calculation module, used for inputting the second environmental parameter with some parameters removed into a second prediction model to output a predicted transpiration amount, and calculating the actual irrigation time based on the predicted transpiration amount and the preliminary irrigation time; The step of sorting the importance attributes of the second environmental parameters based on the gradient boosting model and eliminating some parameters of the second environmental parameters based on the importance attribute sorting result comprises: Inputting the second environmental parameter into a gradient boosting model to obtain an importance score of each parameter of the second environmental parameter, and sorting all the parameters according to the importance score; Sequentially select different preset thresholds to determine parameters whose importance scores are lower than the selected preset thresholds, and remove parameters whose importance scores are lower than the selected preset thresholds to form multiple sets of feature sets; Sequentially dividing the plurality of feature sets into validation sets and a plurality of training sets, and using the validation set and the plurality of training sets of each group as training and validation data for the gradient boosting model; Calculating the performance index of the gradient boosting model during the verification process, obtaining an optimal set of features based on the performance index, and forming the second environment parameters by eliminating some parameters; The step of calculating the actual irrigation time based on the predicted transpiration and the preliminary irrigation time comprises: The predicted transpiration amount is calculated based on the time calculation formula to obtain the transpiration amount predicted irrigation time, and the average value of the preliminary irrigation time and the transpiration amount predicted irrigation time is calculated to obtain the actual irrigation time.

8. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the artificial intelligence-based agricultural management method as described in any one of claims 1 to 6 is implemented.

9. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the artificial intelligence-based agricultural management method as described in any one of claims 1 to 6.

Images