Universe space planting environment regulation and control method and system based on artificial intelligence

Abstract

The invention belongs to the technical field of global space planting, and discloses a global space planting environment regulation and control method and system based on artificial intelligence. The method comprises the following steps: constructing a global space planting environment analysis model and a global space planting environment regulation and control model based on a cloud computing center; acquiring real-time planting environment monitoring data based on a planting environment monitoring device; processing the real-time planting environment monitoring data based on an edge computing gateway; based on the cloud computing center, carrying out global space planting environment analysis; carrying out global space planting environment regulation and control, and sending a real-time regulation and control scheme to a corresponding edge computing gateway; and based on the edge computing gateway, according to the real-time regulation and control scheme, generating a real-time regulation and control instruction, and sending the real-time regulation and control instruction to the planting environment monitoring device and the planting environment regulation and control device. According to the method, the problems of high analysis cost, low efficiency, poor accuracy, unstable regulation and control effect and resource waste in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of global space planting, and specifically relates to a global space planting environment control method and system based on artificial intelligence. Background Art

[0002] Global space planting, also often referred to as vertical farming or vertical agriculture, is an innovative agricultural production method that aims to optimize land use and increase crop yields. The core of this concept is to make full use of spatial resources, including vertical and horizontal space, and control environmental conditions to maximize agricultural production efficiency. The realization of global space planting is a disruptive innovation to traditional agriculture. It provides new ideas and solutions to solve global food security, land resource shortages and environmental sustainability issues.

[0003] In the field of global space planting, precise control of the planting environment has become the key to improving crop yield and quality. In the existing global space planting technology, although the real-time online collection of planting environment monitoring data has been achieved, it still needs to rely on manual analysis of the planting environment monitoring data, which is costly, inefficient and inaccurate. In addition, when formulating the control plan, subjective factors account for a large proportion, and there are large errors and uncertainties, which leads to problems such as unstable control effects and waste of resources. Summary of the invention

[0004] In order to solve the problems of high analysis cost, low efficiency, poor accuracy, unstable regulation effect and waste of resources in the prior art, the present invention aims to provide a global spatial planting environment regulation method and system based on artificial intelligence.

[0005] The technical solution adopted by the present invention is:

[0006] A global spatial planting environment control method based on artificial intelligence comprises the following steps:

[0007] Based on the cloud computing center, a global spatial planting environment analysis model and a global spatial planting environment regulation model are constructed according to the historical planting environment monitoring data of several preset planting crop types;

[0008] Based on the planting environment monitoring device of the current planting area, real-time planting environment monitoring data of the current crop type is collected and sent to the edge computing gateway of the current planting area;

[0009] Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is processed to obtain the processed real-time planting environment monitoring data and upload it to the cloud computing center;

[0010] Based on the cloud computing center, according to the processed real-time planting environment monitoring data, the global spatial planting environment analysis model is used to perform global spatial planting environment analysis and obtain real-time analysis results;

[0011] According to the real-time analysis results, the global spatial planting environment control model is used to control the global spatial planting environment, obtain a real-time control plan, and send the real-time control plan to the corresponding edge computing gateway;

[0012] Based on the edge computing gateway of the current planting area, a real-time control instruction is generated according to the real-time control plan, and the real-time control instruction is sent to the corresponding planting environment monitoring device and planting environment control device.

[0013] Furthermore, based on the cloud computing center, according to the historical planting environment monitoring data of several preset planting crop types, a global spatial planting environment analysis model and a global spatial planting environment regulation model are constructed, including the following steps:

[0014] Based on the cloud computing center, historical planting environment monitoring data of a number of preset planting crop types are collected, and a number of historical planting environment monitoring data are pre-processed to obtain a number of pre-processed historical planting environment monitoring data;

[0015] Performing data dimension reduction on a number of pre-processed historical planting environment monitoring data to obtain a number of reduced-dimensional historical planting environment monitoring data and a set of key monitoring indicators for each preset planting crop type, and sending the set of key monitoring indicators to the edge computing gateway of the corresponding preset planting area;

[0016] Based on several historical planting environment monitoring data after dimensionality reduction, a deep learning algorithm is used to construct a model, obtain a global spatial planting environment analysis model, and generate several historical analysis results;

[0017] Based on several historical analysis results, the reinforcement learning algorithm was used to construct a model to obtain a global spatial planting environment regulation model.

[0018] Furthermore, the global spatial planting environment analysis model is constructed based on the RF-MLP algorithm.

[0019] Furthermore, the global spatial planting environment regulation model is constructed based on the DQN algorithm.

[0020] Furthermore, the method further comprises the following steps:

[0021] Based on the trusted organization, key generation and identity registration are performed for the edge computing gateways in all preset planting areas, and the public-private key pair and signature information of each edge computing gateway are obtained. The private key and signature information of the public-private key pair are sent to the corresponding edge computing gateway, and the public key of the public-private key pair is published to the cloud computing center.

[0022] Furthermore, based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is processed to obtain the processed real-time planting environment monitoring data, and uploaded to the cloud computing center, including the following steps:

[0023] Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is preprocessed to obtain the preprocessed real-time planting environment monitoring data;

[0024] According to the key monitoring indicator set of the current crop type in the current planting area, the pre-processed real-time planting environment monitoring data is subjected to data dimension reduction to obtain the real-time planting environment monitoring data after dimension reduction;

[0025] According to the private key and signature information of the edge computing gateway, the real-time planting environment monitoring data after dimensionality reduction is encrypted and signed to obtain the processed real-time planting environment monitoring data, that is, the encrypted real-time planting environment monitoring data and signature data;

[0026] The encrypted real-time planting environment monitoring data and signature data are uploaded to the cloud computing center.

[0027] Furthermore, based on the cloud computing center, according to the processed real-time planting environment monitoring data, the global spatial planting environment analysis model is used to perform global spatial planting environment analysis to obtain real-time analysis results, including the following steps:

[0028] Call the trusted institution to verify the signature data. After the signature verification is passed, the encrypted real-time planting environment monitoring data is decrypted according to the public key of the edge computing gateway to obtain the decrypted real-time planting environment monitoring data.

[0029] According to the decrypted real-time planting environment monitoring data, the global spatial planting environment analysis model is used to extract key features and obtain several real-time key features;

[0030] A number of real-time key features are fused to obtain real-time fused features, and based on the real-time fused features, the global spatial planting environment analysis is performed to obtain real-time analysis results.

[0031] Furthermore, according to the real-time analysis results, the global space planting environment control model is used to control the global space planting environment, obtain a real-time control plan, and send the real-time control plan to the corresponding edge computing gateway, including the following steps:

[0032] According to the real-time analysis results, the model parameters of the global spatial planting environment control model are updated to obtain updated model parameters;

[0033] Based on the updated model parameters, the global spatial planting environment control model is used to control the global spatial planting environment and obtain a real-time control plan;

[0034] According to the public key of the edge computing gateway, the real-time control plan is encrypted to obtain the encrypted real-time control plan, and the encrypted real-time control plan is sent to the corresponding edge computing gateway.

[0035] Furthermore, based on the edge computing gateway of the current planting area, according to the real-time control scheme, a real-time control instruction is generated, and the real-time control instruction is sent to the corresponding planting environment monitoring device and the planting environment control device, including the following steps:

[0036] Based on the edge computing gateway of the current planting area, the encrypted real-time control plan is decrypted according to the private key to obtain the decrypted real-time control plan;

[0037] According to the decrypted real-time control scheme, control instructions are generated to obtain real-time control instructions, and the real-time control instructions are sent to the corresponding planting environment monitoring device and planting environment control device.

[0038] A global space planting environment control system based on artificial intelligence is used to implement a global space planting environment control method. The system includes a cloud computing center, a trusted institution, an edge computing gateway, a planting environment monitoring device, and a planting environment control device. The cloud computing center and the trusted institution are respectively connected to the edge computing gateways of several preset planting areas, and the cloud computing center is also connected to the trusted institution. The edge computing gateway is connected to the planting environment monitoring device and the planting environment control device in the same preset planting area.

[0039] The cloud computing center is provided with a model building unit, a data analysis unit and an environmental control unit which are connected in sequence. The model building unit, the data analysis unit and the environmental control unit are respectively connected to the edge computing gateways of several preset planting areas for communication.

[0040] The beneficial effects of the present invention are:

[0041] The present invention discloses a global spatial planting environment control method and system based on artificial intelligence, which combines a cloud computing center and an edge computing gateway to realize rapid data processing and distributed transmission, reduce the resource processing pressure of the cloud computing center, and use edge computing technology to realize localized data processing, shorten the time of data transmission and processing, and improve the response speed to environmental changes; through the global spatial planting environment analysis model constructed by a deep learning algorithm and the global spatial planting environment control model constructed by a reinforcement learning algorithm, the intelligent analysis and automatic control of the planting environment reduce the cost of data analysis, improve the analysis efficiency and accuracy, and use real-time planting environment monitoring data to dynamically adjust the real-time control scheme to ensure that the planting environment is always in the best state, effectively reduce resource waste, and improve the utilization efficiency of resources such as water, fertilizer, and energy, thereby improving the growth efficiency and quality of crops, and avoiding the defects of unstable control effect and resource waste; by analyzing historical planting environment monitoring data, customized environmental control schemes can be provided for different types of crops, thereby enhancing the adaptability and universality of the method.

[0042] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flowchart of the global space planting environment control method based on artificial intelligence in the present invention;

[0044] Figure 2 It is a structural block diagram of the global space planting environment control system based on artificial intelligence in the present invention. DETAILED DESCRIPTION

[0045] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0046] Embodiment 1:

[0047] like Figure 1 As shown, this embodiment provides a global spatial planting environment control method based on artificial intelligence, comprising the following steps:

[0048] S1: Based on the cloud computing center, according to the historical planting environment monitoring data of several preset planting crop types, a global spatial planting environment analysis model and a global spatial planting environment regulation model are constructed, including the following steps:

[0049] S1-1: Based on the cloud computing center, historical planting environment monitoring data of a number of preset planting crop types are collected, and the historical planting environment monitoring data are preprocessed to obtain a number of preprocessed historical planting environment monitoring data;

[0050] The monitoring indicators of planting environment monitoring data include temperature, humidity, light, soil nutrients, oxygen content, etc.

[0051] Preprocessing includes data cleaning, missing value processing, normalization, etc., to facilitate the subsequent analysis of deep learning algorithms;

[0052] S1-2: Using the principal component analysis (PCA) method, a plurality of pre-processed historical planting environment monitoring data are subjected to data dimension reduction to obtain a plurality of reduced-dimensional historical planting environment monitoring data and a set of key monitoring indicators for each preset planting crop type, and the set of key monitoring indicators is sent to the edge computing gateway of the corresponding preset planting area, including the following steps:

[0053] S1-2-1: Use the PCA method to transform the data space matrix of some pre-processed historical planting environment monitoring data to obtain the corresponding historical planting environment monitoring data space matrix X = [x1, x2, ... x p ,...,x n ] T , where x p is the row vector of the p-th preprocessed historical planting environment monitoring data, p is the row vector indicator, n is the total number of preprocessed historical planting environment monitoring data, and T is the transposition mark;

[0054] S1-2-2: Standardize the spatial matrix of historical planting environment monitoring data to obtain the corresponding standardized spatial matrix of historical planting environment monitoring data;

[0055] The formula is:

[0056]

[0057] Where X' is the standardized spatial matrix of historical planting environment monitoring data; μ is the mean of the initial spatial matrix of historical planting environment monitoring data; σ is the variance of the initial spatial matrix of historical planting environment monitoring data; X is the initial spatial matrix of historical planting environment monitoring data;

[0058] S1-2-3: Obtain the covariance matrix of the standardized historical planting environment monitoring data space matrix, and obtain the corresponding candidate principal component matrix according to the historical planting environment monitoring data space matrix and the covariance matrix;

[0059] The formula is:

[0060]

[0061] Where D is the covariance matrix of the standardized historical planting environment monitoring data space matrix; Y is the candidate principal component matrix; P is the conversion matrix; E is the unit eigenvector matrix; n is the total number of historical planting environment monitoring data after preprocessing; X' is the standardized historical planting environment monitoring data space matrix;

[0062] Y=PX'

[0063] Where, Y is the candidate principal component matrix; P is the transformation matrix; X' is the standardized historical planting environment monitoring data space matrix;

[0064] S1-2-4: Set the candidate principal component matrix Y = [y1, y2, ...y l ,...,y L ] Several candidate principal components with the top 90% of the cumulative variance contribution rate are used as the corresponding principal components, and a set of key monitoring indicators consisting of the monitoring indicators corresponding to the principal components is obtained, among which y l is the lth candidate principal component column vector, l is the column vector indicator, and L is the total number of candidate principal components;

[0065] The formula is:

[0066]

[0067] In the formula, G is the cumulative contribution rate of variance; l The lth candidate principal component column vector y l The variance of ; l is the indicator of the candidate principal component; L is the total number of candidate principal components; K is the total number of principal components;

[0068] S1-2-5: According to the key monitoring indicator set, the standardized historical planting environment monitoring data space matrix is ​​subjected to data space dimension reduction to obtain the corresponding historical planting environment monitoring data space matrix after dimension reduction Y'=[y'1,y'2,...y' l ,...,y' K ], where y' l is the lth principal component column vector, l is the column vector indicator, and K is the total number of candidate principal components;

[0069] S1-2-6: Performing an inverse transformation on the spatial matrix of the historical planting environment monitoring data after dimensionality reduction to obtain a number of corresponding historical planting environment monitoring data after dimensionality reduction;

[0070] S1-3: Based on some historical planting environment monitoring data after dimensionality reduction, the model is constructed using the Random Forest (RF)-Multilayer Perceptron (MLP) algorithm to obtain a global spatial planting environment analysis model and generate some historical analysis results;

[0071] The RF module uses the internal Classification And Regression Tree (CART) to screen the key features of the input planting environment monitoring data and extract the key features related to the data labels. The MLP network is a fully connected layer, which is characterized by all neurons between layers being interconnected. It trains its own network structure through a large number of key features, predicts new data features, and realizes fast, efficient and accurate prediction of data labels by mining the deep relationship between data features and data labels.

[0072] S1-4: Based on several historical analysis results, the Deep Q Network (DQN) algorithm is used to construct a model to obtain a global spatial planting environment regulation model, including the following steps:

[0073] S1-4-1: Generate a global spatial planting environment control scheme as a simulation environment for the DQN algorithm to build an intelligent agent and experience replay pool;

[0074] S1-4-2: Define the state space of the DQN algorithm according to each analysis type corresponding to the historical analysis results, and the parameters of the state space correspond to each analysis type;

[0075] S1-4-3: Define the action space of the DQN algorithm according to the actions that need to be output for global spatial planting environment control; including actions such as turning on or off ventilation devices, turning on or off irrigation devices, and adjusting the frequency or period of monitoring data collection;

[0076] S1-4-4: Based on the possible impact of each action in the action space, define the reward function of the DQN algorithm to evaluate the quality or impact of the action;

[0077] S1-4-5: Construct the input layer, several hidden layers, and output layer of the deep Q network, connect the input layer to the state space, and connect the output layer to the action space;

[0078] S1-4-6: Based on the state space, action space and reward function, according to several historical analysis results, the deep Q network and the agent are optimized and trained to build a global spatial planting environment control model, and the generated environmental control experiences are stored in the experience playback pool;

[0079] S1.0: Based on the trusted institution, key generation and identity registration are performed for the edge computing gateways in all preset planting areas, and the public-private key pair and signature information of each edge computing gateway are obtained. The private key and signature information of the public-private key pair are sent to the corresponding edge computing gateway, and the public key of the public-private key pair is published to the cloud computing center, including the following steps:

[0080] S1.0-1: Based on the trusted institution, perform key initialization to generate public parameters GP, master key MSK and initial key PK. The formula is:

[0081] GP={g,g1,g a ,e(g,g) a , H1, H2, H3, H4, H5, H6}

[0082] PK={g,g1,g a ,e(g,g) a ,H u}

[0083] MSK = {g a ,a}

[0084] Where GP is the public parameter; MSK is the master key; PK is the initial key; a is the integer domain Z p Random numbers; H1, H2, H3, H4, H5, H6, H u All are target hash functions; g, g1, g a are all random numbers that are generators of the cyclic group G; e(g,g) a is a bilinear map of random numbers g;

[0085] S1.0-2: Based on the public parameters GP, master key MSK, initial key PK and attribute information V of the edge computing gateway u , generate the corresponding public-private key pair of the edge computing gateway, the formula is:

[0086] SK u = {MSK,V u ,K u =g a g ab ,L u =g b ,(K=H3(V u ) b )}

[0087]

[0088] In the formula, SK u is the private key of the edge computing gateway u; b is the integer domain Z p Random number; L u , K u is the private key parameter of the edge computing gateway u; H3 is the target hash function of the public parameter GP; u is the edge computing gateway indicator; MSK is the master key; PK is the initial key; PK u is the public key of the edge computing gateway u; g b , ga , g ab is the random number of the generator of the cyclic group G; V u is the attribute information of the edge computing gateway u;

[0089] S1.0-3: According to the private key in the public-private key pair and the corresponding device ID, perform identity registration and obtain the signature information of the corresponding edge computing gateway. The formula is:

[0090]

[0091] In the formula, k" is a random number; K u Registration parameters for edge computing gateway u; KID u KID is the signature ID of the edge computing gateway u; u and the corresponding K u The signature information {K u ,KID u}; H1 is the target hash function; ID u SK is the device ID of the edge computing gateway u; u is the private key of edge computing gateway u; is the prime order; P′ is the base point of the prime domain;

[0092] S1.0-4: Send the private key and signature information of the public-private key pair to the corresponding edge computing gateway, and publish the public key of the public-private key pair to the cloud computing center;

[0093] Combining asymmetric encryption algorithms and digital identity authentication technology improves the security of data transmission and avoids data leakage;

[0094] S2: Based on the planting environment monitoring device of the current planting area, collect the real-time planting environment monitoring data of the current planting crop type and send it to the edge computing gateway of the current planting area;

[0095] S3: Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is processed to obtain the processed real-time planting environment monitoring data, and uploaded to the cloud computing center, including the following steps:

[0096] S3-1: Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is preprocessed to obtain the preprocessed real-time planting environment monitoring data;

[0097] Preprocessing includes standardization processing, format conversion, etc., which are carried out in sequence to improve data quality and convert it into a unified data format that can be recognized by the cloud computing center;

[0098] S3-2: According to the key monitoring indicator set of the current crop type in the current planting area, the pre-processed real-time planting environment monitoring data is subjected to data dimension reduction to obtain the real-time planting environment monitoring data after dimension reduction;

[0099] For example, for watermelon, key monitoring indicators include light, humidity, etc., while for wood ear, key monitoring indicators include temperature, humidity, soil nutrients, etc. Different crop types in different planting areas require different data dimensionality reduction, which improves the accuracy of data analysis.

[0100] S3-3: According to the private key and signature information of the edge computing gateway, the real-time planting environment monitoring data after dimensionality reduction is encrypted and signed to obtain the processed real-time planting environment monitoring data, that is, the encrypted real-time planting environment monitoring data and signature data;

[0101] The formula is:

[0102] M u =E(SK u ,m u )

[0103] Where M u is the encrypted real-time planting environment monitoring data of the edge computing gateway u; E(*) is the asymmetric encryption function; m u Real-time planting environment monitoring data for edge computing gateway u; SK u is the private key of the edge computing gateway u; u is the edge computing gateway indicator;

[0104]

[0105] In the formula, r' is a random number; H2 is the target hash function; K u Registration parameters for edge computing gateway u; KID u is the signature ID of the edge computing gateway u; M u It is the encrypted real-time planting environment monitoring data of the edge computing gateway u; ID u is the device ID of the edge computing gateway u; the signature information is {ID u ,M u ,γ'={K u ,R u ,B u}}; R u ,B u ,γ' are the signature parameters of edge computing gateway u;

[0106] S3-4: Upload the encrypted real-time planting environment monitoring data and signature data to the cloud computing center;

[0107] S4: Based on the cloud computing center, according to the processed real-time planting environment monitoring data, the global space planting environment analysis model is used to perform global space planting environment analysis to obtain real-time analysis results, including the following steps:

[0108] S4-1: Call the trusted institution to perform signature verification on the signature data. After the signature verification is passed, the encrypted real-time planting environment monitoring data is decrypted according to the public key of the edge computing gateway to obtain the decrypted real-time planting environment monitoring data;

[0109] The formula is:

[0110] β u B u P=β u H2(R u ,M u ,ID u ,K u )R u +β u K u +β u H1(ID u ,K u )PK u

[0111] In the formula, β u PK is the signature verification parameter of the edge computing gateway u; u is the public key of the edge computing gateway u; if the left formula is equal to the right formula, the signature verification is successful;

[0112] m u =E - (PK u ,M u )

[0113] In the formula, m u The decrypted real-time planting environment monitoring data of the edge computing gateway u; E - (*) is an asymmetric decryption function; PK u is the public key of edge computing gateway u; M u It is the encrypted real-time planting environment monitoring data of the edge computing gateway u;

[0114] S4-2: Based on the decrypted real-time planting environment monitoring data, the global spatial planting environment analysis model is used to extract key features and obtain several real-time key features, including the following steps:

[0115] S4-2-1: Use the RF structure trained in the global spatial planting environment analysis model to extract the feature contribution of several real-time candidate features in the decrypted real-time planting environment monitoring data;

[0116] The formula is:

[0117]

[0118] In the formula, is the feature contribution of the jth real-time candidate feature; is the feature contribution of the jth real-time candidate feature in the i-th tree of the random forest; i is the CART tree indicator; j is the real-time candidate feature indicator; n' is the total number of CARTs;

[0119]

[0120] In the formula, GI m GI l GI r is the Gini index of CART tree node m, node l and node r of random forest; p mk is the proportion of category k in CART tree node m; K is the total number of categories; m, l, r are node indicators; k is the category indicator;

[0121] S4-2-2: normalizing the feature contributions of several real-time candidate features to obtain corresponding normalized feature contributions;

[0122] The formula is:

[0123]

[0124] Where VIM j is the contribution of the feature after normalization; J is the total number of real-time candidate features;

[0125] S4-2-3: Generate feature selection standard values ​​for several real-time candidate features based on the normalized feature contribution;

[0126] The formula is:

[0127]

[0128] In the formula, CFC j VIM is the feature selection standard value for the jth real-time candidate feature; j' is the normalized feature contribution of the j'th real-time candidate feature; j' is the indicator of the candidate feature;

[0129] S4-2-4: sorting the real-time candidate features in descending order according to the feature selection standard value, and selecting the first N real-time candidate features as real-time key features to obtain N real-time key features, where N is the total number of real-time key features;

[0130] S4-3: Fusing several real-time key features to obtain real-time fusion features, and performing global spatial planting environment analysis based on the real-time fusion features to obtain real-time analysis results;

[0131] S5: According to the real-time analysis results, the global space planting environment control model is used to control the global space planting environment, obtain a real-time control plan, and send the real-time control plan to the corresponding edge computing gateway, including the following steps:

[0132] S5-1: According to the real-time analysis results, the model parameters of the global spatial planting environment control model are updated to obtain updated model parameters, including the following steps:

[0133] S5-1-1: According to the real-time analysis results, extract several matching environmental control experiences from the experience playback pool of the mine car control optimization model;

[0134] S5-1-2: Based on some environmental control experience, update the action space of the global spatial planting environment control model to obtain the updated action space A'=[a'1,...,a' j" ,...,a' I ], where a' j" is the updated j-th "action value", j" is the action indicator; I is the total number of action space dimensions;

[0135] S5-1-3: According to the real-time analysis results, the state space of the global spatial planting environment control model is updated to obtain the updated state space S'=[s'1,...,s' i" ,...,s' I' ], where s' i" is the updated i'th state value, i' is the state indicator; I' is the total number of state space dimensions;

[0136] S5-2: Based on the updated model parameters, i.e. A' = [a'1, ..., a' j" ,...,a' I ] and S'=[s'1,...,s' i" ,...,s' I' ], using the global space planting environment control model, the global space planting environment control is carried out to obtain a real-time control scheme, including the following steps:

[0137] S5-2-1: Update the state space S' = [s'1, ..., s' i' ,...,s' I ] as the input of the global spatial planting environment control model, using the deep Q network to generate an updated action space A'=[a'1,...,a' j' ,...,a'I ] is the Q value of each possible action;

[0138] S5-2-2: Use the reward function to obtain the reward value of each possible action in the updated action space, and update the Q value of the possible action according to the reward value to obtain the updated Q value of the possible action;

[0139] The formula is:

[0140] Q(s' p' ,a' p' )=(1-α")·Q(s p' ,a p' )+α"·(R(s p' ,a p' ,s' p' )+γ·Q max (s p' ,a p' ))

[0141] In the formula, Q(s' p' ,a' p' ) is the updated state value s' p' and the updated action value a' p' The corresponding updated Q value; Q(s p' ,a p' ) is the state value s p' and action value a p' The corresponding predicted Q value; α" is the learning rate; Q max (s p' ,a p' ) is the highest predicted Q value;

[0142] S5-2-3: Repeat the above steps until the iteration threshold is reached, use the greedy strategy, take the possible action corresponding to the highest updated Q value as the execution action, and output the execution action as a real-time control plan; including shutting down the ventilation device, shutting down the irrigation device, shortening the monitoring data collection frequency, etc.;

[0143] S5-3: Encrypt the real-time control plan according to the public key of the edge computing gateway to obtain the encrypted real-time control plan, and send the encrypted real-time control plan to the corresponding edge computing gateway;

[0144] The formula is:

[0145] M' u =E(PK u ,QS)

[0146] In the formula, M' uis the encrypted real-time control scheme sent to the edge computing gateway u; E(*) is the asymmetric encryption function; QS is the real-time control scheme; PK u is the public key of the edge computing gateway u; u is the edge computing gateway indicator;

[0147] S6: Based on the edge computing gateway of the current planting area, according to the real-time control scheme, a real-time control instruction is generated, and the real-time control instruction is sent to the corresponding planting environment monitoring device and planting environment control device, including the following steps:

[0148] S6-1: Based on the edge computing gateway of the current planting area, the encrypted real-time control plan is decrypted according to the private key to obtain the decrypted real-time control plan;

[0149] The formula is:

[0150] QS'=E - (SK u ,M' u )

[0151] Where, QS' is the real-time control scheme after decryption; E - (*) is an asymmetric decryption function; M' u is the encrypted real-time control scheme sent to the edge computing gateway u; SK u is the private key of the edge computing gateway u; u is the edge computing gateway indicator;

[0152] S6-2: Generate a control instruction according to the decrypted real-time control scheme to obtain the real-time control instruction, and send the real-time control instruction to the corresponding planting environment monitoring device and planting environment control device;

[0153] The decrypted real-time control scheme of this embodiment is to turn off the ventilation device, turn off the irrigation device, shorten the monitoring data collection frequency, etc. The generated real-time control instructions include the ventilation device shutdown instruction in the planting environment control device, the irrigation device shutdown instruction and the frequency shortening instruction of the planting environment monitoring device.

[0154] Embodiment 2:

[0155] like Figure 2 As shown, this embodiment provides a global space planting environment control system based on artificial intelligence, which is used to implement a global space planting environment control method. The system includes a cloud computing center, a trusted institution, an edge computing gateway, a planting environment monitoring device, and a planting environment control device. The cloud computing center and the trusted institution are respectively connected to the edge computing gateways of several preset planting areas, and the cloud computing center is also connected to the trusted institution. The edge computing gateway is connected to the planting environment monitoring device and the planting environment control device of the same preset planting area;

[0156] The trusted institution generates keys and registers identities for the edge computing gateways in all preset planting areas, obtains the public-private key pair and signature information of each edge computing gateway, sends the private key and signature information of the public-private key pair to the corresponding edge computing gateway, and publishes the public key of the public-private key pair to the cloud computing center;

[0157] The edge computing gateway is used to process the real-time planting environment monitoring data, obtain the processed real-time planting environment monitoring data, and upload it to the cloud computing center; generate real-time control instructions according to the real-time control plan, and send the real-time control instructions to the corresponding planting environment monitoring device and planting environment control device;

[0158] Planting environment monitoring device, used to collect real-time planting environment monitoring data of the current crop type and send it to the edge computing gateway of the current planting area;

[0159] A planting environment control device is used to control the planting environment control device of the current planting area according to real-time control instructions to meet the remote control of real-time planting environment monitoring data;

[0160] The cloud computing center is provided with a model building unit, a data analysis unit and an environmental control unit which are connected in sequence, and the model building unit, the data analysis unit and the environmental control unit are respectively connected to the edge computing gateways of several preset planting areas for communication;

[0161] A model building unit, used to build a global spatial planting environment analysis model and a global spatial planting environment regulation model according to historical planting environment monitoring data of several preset planting crop types sent by the edge computing gateway;

[0162] A data analysis unit is used to perform global spatial planting environment analysis based on the processed real-time planting environment monitoring data sent by the edge computing gateway and obtain real-time analysis results using a global spatial planting environment analysis model;

[0163] The environmental control unit is used to control the global spatial planting environment according to the real-time analysis results and use the global spatial planting environment control model to obtain a real-time control plan, and send the real-time control plan to the corresponding edge computing gateway.

[0164] The present invention discloses a global spatial planting environment control method and system based on artificial intelligence, which combines a cloud computing center and an edge computing gateway to realize rapid data processing and distributed transmission, reduce the resource processing pressure of the cloud computing center, and use edge computing technology to realize localized data processing, shorten the time of data transmission and processing, and improve the response speed to environmental changes; through the global spatial planting environment analysis model constructed by a deep learning algorithm and the global spatial planting environment control model constructed by a reinforcement learning algorithm, the intelligent analysis and automatic control of the planting environment reduce the cost of data analysis, improve the analysis efficiency and accuracy, and use real-time planting environment monitoring data to dynamically adjust the real-time control scheme to ensure that the planting environment is always in the best state, effectively reduce resource waste, and improve the utilization efficiency of resources such as water, fertilizer, and energy, thereby improving the growth efficiency and quality of crops, and avoiding the defects of unstable control effect and resource waste; by analyzing historical planting environment monitoring data, customized environmental control schemes can be provided for different types of crops, thereby enhancing the adaptability and universality of the method.

[0165] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. A global spatial planting environment control method based on artificial intelligence, characterized in that: The steps include: Based on the cloud computing center, a global spatial planting environment analysis model and a global spatial planting environment regulation model are constructed according to the historical planting environment monitoring data of several preset planting crop types; Based on the planting environment monitoring device of the current planting area, real-time planting environment monitoring data of the current crop type is collected and sent to the edge computing gateway of the current planting area; Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is processed to obtain the processed real-time planting environment monitoring data and upload it to the cloud computing center; Based on the cloud computing center, according to the processed real-time planting environment monitoring data, the global spatial planting environment analysis model is used to perform global spatial planting environment analysis and obtain real-time analysis results; According to the real-time analysis results, the global spatial planting environment control model is used to control the global spatial planting environment, obtain a real-time control plan, and send the real-time control plan to the corresponding edge computing gateway; Based on the edge computing gateway of the current planting area, a real-time control instruction is generated according to the real-time control plan, and the real-time control instruction is sent to the corresponding planting environment monitoring device and planting environment control device.

2. The method for controlling the global spatial planting environment based on artificial intelligence according to claim 1, characterized in that: Based on the cloud computing center, according to the historical planting environment monitoring data of several preset planting crop types, a global spatial planting environment analysis model and a global spatial planting environment regulation model are constructed, including the following steps: Based on the cloud computing center, historical planting environment monitoring data of a number of preset planting crop types are collected, and a number of historical planting environment monitoring data are pre-processed to obtain a number of pre-processed historical planting environment monitoring data; Performing data dimension reduction on a number of pre-processed historical planting environment monitoring data to obtain a number of reduced-dimensional historical planting environment monitoring data and a set of key monitoring indicators for each preset planting crop type, and sending the set of key monitoring indicators to the edge computing gateway of the corresponding preset planting area; Based on several historical planting environment monitoring data after dimensionality reduction, a deep learning algorithm is used to construct a model, obtain a global spatial planting environment analysis model, and generate several historical analysis results; Based on several historical analysis results, the reinforcement learning algorithm was used to construct a model to obtain a global spatial planting environment regulation model.

3. The method for controlling the global spatial planting environment based on artificial intelligence according to claim 2 is characterized in that: The global spatial planting environment analysis model is constructed based on the RF-MLP algorithm.

4. The method for controlling the global space planting environment based on artificial intelligence according to claim 3 is characterized in that: The global spatial planting environment control model is constructed based on the DQN algorithm.

5. The method for controlling the global space planting environment based on artificial intelligence according to claim 4 is characterized in that: The following steps are also included: Based on the trusted organization, key generation and identity registration are performed for the edge computing gateways in all preset planting areas, and the public-private key pair and signature information of each edge computing gateway are obtained. The private key and signature information of the public-private key pair are sent to the corresponding edge computing gateway, and the public key of the public-private key pair is published to the cloud computing center.

6. The method for controlling the global spatial planting environment based on artificial intelligence according to claim 5 is characterized in that: Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is processed to obtain the processed real-time planting environment monitoring data and upload it to the cloud computing center, including the following steps: Based on the edge computing gateway of the current planting area, the real-time planting environment monitoring data is preprocessed to obtain the preprocessed real-time planting environment monitoring data; According to the key monitoring indicator set of the current crop type in the current planting area, the pre-processed real-time planting environment monitoring data is subjected to data dimension reduction to obtain the real-time planting environment monitoring data after dimension reduction; According to the private key and signature information of the edge computing gateway, the real-time planting environment monitoring data after dimensionality reduction is encrypted and signed to obtain the processed real-time planting environment monitoring data, that is, the encrypted real-time planting environment monitoring data and signature data; The encrypted real-time planting environment monitoring data and signature data are uploaded to the cloud computing center.

7. The method for controlling the global space planting environment based on artificial intelligence according to claim 6 is characterized in that: Based on the cloud computing center, according to the processed real-time planting environment monitoring data, the global space planting environment analysis model is used to perform global space planting environment analysis to obtain real-time analysis results, including the following steps: Call the trusted institution to verify the signature data. After the signature verification is passed, the encrypted real-time planting environment monitoring data is decrypted according to the public key of the edge computing gateway to obtain the decrypted real-time planting environment monitoring data. According to the decrypted real-time planting environment monitoring data, the global spatial planting environment analysis model is used to extract key features and obtain several real-time key features; A number of real-time key features are fused to obtain real-time fused features, and based on the real-time fused features, the global spatial planting environment analysis is performed to obtain real-time analysis results.

8. The method for controlling the global space planting environment based on artificial intelligence according to claim 5 is characterized in that: According to the real-time analysis results, the global space planting environment control model is used to control the global space planting environment, obtain the real-time control plan, and send the real-time control plan to the corresponding edge computing gateway, including the following steps: According to the real-time analysis results, the model parameters of the global spatial planting environment control model are updated to obtain updated model parameters; Based on the updated model parameters, the global spatial planting environment control model is used to control the global spatial planting environment and obtain a real-time control plan; According to the public key of the edge computing gateway, the real-time control plan is encrypted to obtain the encrypted real-time control plan, and the encrypted real-time control plan is sent to the corresponding edge computing gateway.

9. The method for controlling the global space planting environment based on artificial intelligence according to claim 8, characterized in that: Based on the edge computing gateway of the current planting area, a real-time control instruction is generated according to the real-time control scheme, and the real-time control instruction is sent to the corresponding planting environment monitoring device and planting environment control device, including the following steps: Based on the edge computing gateway of the current planting area, the encrypted real-time control plan is decrypted according to the private key to obtain the decrypted real-time control plan; According to the decrypted real-time control scheme, control instructions are generated to obtain real-time control instructions, and the real-time control instructions are sent to the corresponding planting environment monitoring device and planting environment control device.

10. A global space planting environment control system based on artificial intelligence, used to implement the global space planting environment control method as described in any one of claims 1 to 9, characterized in that: The system includes a cloud computing center, a trusted institution, an edge computing gateway, a planting environment monitoring device, and a planting environment control device. The cloud computing center and the trusted institution are respectively connected to the edge computing gateways of several preset planting areas, and the cloud computing center is also connected to the trusted institution. The edge computing gateway is connected to the planting environment monitoring device and the planting environment control device in the same preset planting area. The cloud computing center is provided with a model building unit, a data analysis unit and an environmental control unit which are connected in sequence. The model building unit, the data analysis unit and the environmental control unit are respectively connected to the edge computing gateways of several preset planting areas for communication.

Images