A pesticide compound screening method, system and storage medium

Abstract

The present invention relates to the field of pesticide technology, and in particular to a novel pesticide compound screening method, system and storage medium. Various actual efficacy data of a target pesticide compound within a preset application time period are obtained, and various actual efficacy data of the target pesticide compound within the preset application time period are compared and analyzed with standard efficacy data; if the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound; if one or more efficacy data of the target pesticide compound after application are non-up-to-standard efficacy data, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated. The screening efficiency and accuracy are improved by dynamically evaluating the performance changes of pesticide compounds under different environmental conditions and conducting quantitative analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of pesticides, and in particular to a method, system and storage medium for screening pesticide compounds. Background Art

[0002] As the demand for pesticides in agricultural production continues to grow, the development of efficient, safe, and environmentally friendly pesticide compounds has become the focus of current research. Traditional pesticide screening methods usually rely on laboratory simulation experiments, which are difficult to accurately predict the performance of pesticide compounds in actual environments. In addition, traditional methods have long screening cycles, high costs, and lack evaluation of the performance changes of pesticide compounds under different environmental conditions.

[0003] In order to overcome the limitations of traditional pesticide screening methods, pesticide compound screening methods based on big data and artificial intelligence have gradually emerged in recent years. This type of method uses a large amount of experimental data and environmental data to establish a pesticide compound performance prediction model, and combines actual environmental data to simulate the performance changes of pesticide compounds under different environmental conditions, thereby improving screening efficiency and accuracy.

[0004] However, the existing pesticide compound screening methods based on big data and artificial intelligence still have some shortcomings, such as: lack of dynamic evaluation of the performance changes of pesticide compounds in actual environments, resulting in deviations between screening results and actual applications. Lack of quantitative analysis of the performance changes of pesticide compounds under different environmental conditions makes it difficult to accurately evaluate the comprehensive performance of pesticide compounds. In order to solve the above problems, the present invention proposes a pesticide compound screening method, system and storage medium. Summary of the invention

[0005] The present invention overcomes the deficiencies of the prior art and provides a method, system and storage medium for screening pesticide compounds.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0007] The first aspect of the present invention discloses a method for screening pesticide compounds, comprising the following steps:

[0008] Obtaining standard efficacy data of target pesticide compounds under various environmental characteristic conditions, and constructing a pesticide compound performance prediction model based on the obtained standard efficacy data of target pesticide compounds under various environmental characteristic conditions;

[0009] Obtaining continuous environmental change data of the target pesticide compound within a preset application time period, and obtaining various standard efficacy data of the target pesticide compound within the preset application time period based on the continuous environmental change data of the target pesticide compound within the preset application time period and a pesticide compound performance prediction model;

[0010] Acquire various actual efficacy data of the target pesticide compound within a preset application time period, compare and analyze the various actual efficacy data of the target pesticide compound within the preset application time period with the standard efficacy data, and obtain the status of various efficacy data of the target pesticide compound after application;

[0011] If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and a recommended preparation process parameter scheme for the excellent pesticide compound is obtained;

[0012] If the target pesticide compound has one or more efficacy data that do not meet the standard after application, the target pesticide compound will be screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound will be eliminated.

[0013] Preferably, the standard efficacy data of the target pesticide compound under various environmental characteristic conditions are obtained, and a pesticide compound performance prediction model is constructed based on the standard efficacy data of the target pesticide compound under various environmental characteristic conditions, specifically:

[0014] Acquire structural information of the target pesticide compound, wherein the structural information includes a molecular formula, a SMILES string, and a molecular fingerprint, and construct a network search term based on the structural information of the target pesticide compound;

[0015] Searching the big data network according to the network search terms to obtain standard efficacy data of the target pesticide compound under various environmental characteristic conditions; wherein the efficacy data includes water solubility, degradation rate, bioaccumulation and toxicity;

[0016] Introduce the graph attention network, take various environmental features as environmental feature nodes, and take the standard drug efficacy data under various environmental feature conditions as performance change nodes; connect each environmental feature node with the performance change node with a directed edge to generate an attention network topology graph;

[0017] Performing attention analysis on the attention network topology graph based on the attention mechanism, obtaining attention scores between nodes in the attention network topology graph, and constructing an attention score matrix according to the attention scores between nodes in the attention network topology graph;

[0018] A prediction model is constructed based on a deep learning network, and the attention score matrix is ​​imported into the prediction model for back propagation learning until all model parameters meet preset requirements, and then a pesticide compound performance prediction model is output.

[0019] Preferably, continuous environmental change data of the target pesticide compound within a preset application time period is obtained, and various standard efficacy data of the target pesticide compound within the preset application time period are obtained according to the continuous environmental change data of the target pesticide compound within the preset application time period and the pesticide compound performance prediction model, specifically:

[0020] Obtaining a preset application time period of the target pesticide compound, and collecting continuous environmental change data of the application site of the target pesticide compound within the preset application time period;

[0021] Performing data cleaning processing on the continuous environmental change data within the preset application time period, and arranging the continuous environmental change data within the preset application time period in chronological order to form time series data;

[0022] Converting the time series data into a feature vector at each time point and loading the pesticide compound performance prediction model;

[0023] Using the characteristic vector at each time point as the input of the pesticide compound performance prediction model, and outputting various standard efficacy data of the target pesticide compound within a preset application time period according to the pesticide compound performance prediction model;

[0024] The environmental change data include temperature, humidity, rainfall, wind speed, soil type, water type and pH value.

[0025] Preferably, various actual efficacy data of the target pesticide compound within the preset application time period are compared and analyzed with the standard efficacy data to obtain the status of various efficacy data of the target pesticide compound after application, specifically:

[0026] Calculating the difference between various actual efficacy data and standard efficacy data of the target pesticide compound within the preset application time period to obtain various efficacy data differences of the target pesticide compound within the preset application time period;

[0027] Preset various difference thresholds of efficacy data, and compare the difference values ​​of various efficacy data of the target pesticide compound within the preset application time period with the corresponding difference thresholds;

[0028] If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is not greater than the corresponding difference threshold, the status of the efficacy data is marked as qualified efficacy data;

[0029] If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is greater than the corresponding difference threshold, the status of the efficacy data is marked as non-standard efficacy data.

[0030] Preferably, if the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and the recommended preparation process parameter scheme of the excellent pesticide compound is obtained, specifically:

[0031] If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound;

[0032] If the target pesticide compound is screened as a high-quality pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the high-quality pesticide compound are extracted according to the operation log;

[0033] Generate an actual preparation process parameter plan for the high-quality pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the period corresponding to the preparation of the low-quality pesticide compound;

[0034] Obtaining a preset preparation process parameter scheme for the excellent pesticide compound; introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme for the excellent pesticide compound and the preset preparation process parameter scheme;

[0035] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to a preset terminal for display;

[0036] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, the actual preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to the preset terminal for display.

[0037] Preferably, if the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated, specifically:

[0038] If the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound;

[0039] If the target pesticide compound is screened as an inferior pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the inferior pesticide compound are extracted according to the operation log;

[0040] Generate an actual preparation process parameter plan for the inferior pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the period corresponding to the preparation of the inferior pesticide compound;

[0041] Obtaining a preset preparation process parameter scheme corresponding to the inferior pesticide compound, and introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme;

[0042] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme for preparing the inferior pesticide compound will be eliminated.

[0043] The following steps are also included:

[0044] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, then various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound are extracted from the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme;

[0045] Calculate the process parameter differences between various actual preparation process parameters and preset preparation process parameters when the preparation production line prepares the inferior pesticide compound;

[0046] Comparing the process parameter differences between various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound with the preset parameter difference threshold;

[0047] Only the preparation process parameters corresponding to the process parameter differences greater than the preset parameter difference threshold are extracted and marked as drift preparation process parameters. The drift preparation process parameters are adjusted according to the corresponding process parameter differences, and then the preparation production line is made to re-prepare the target pesticide compound and then the target pesticide compound is evaluated and screened.

[0048] The second aspect of the present invention discloses a pesticide compound screening system, which includes a memory and a processor. The memory stores a pesticide compound screening method program. When the pesticide compound screening method program is executed by the processor, any one of the pesticide compound screening method steps is implemented.

[0049] The third aspect of the present invention discloses a computer-readable storage medium, which stores a pesticide compound screening method program. When the pesticide compound screening method program is executed by a processor, any one of the pesticide compound screening method steps is implemented.

[0050] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: various standard efficacy data of the target pesticide compound in the preset application time period are obtained according to the continuous environmental change data of the target pesticide compound in the preset application time period and the pesticide compound performance prediction model; various actual efficacy data of the target pesticide compound in the preset application time period are obtained, and various actual efficacy data of the target pesticide compound in the preset application time period are compared and analyzed with the standard efficacy data; if the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and the recommended preparation process parameter scheme of the excellent pesticide compound is obtained; if one or more efficacy data of the target pesticide compound after application are non-up-to-standard efficacy data, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated. By constructing a pesticide compound performance prediction model, combining actual environmental data, dynamically evaluating the performance changes of pesticide compounds under different environmental conditions, and conducting quantitative analysis, the screening efficiency and accuracy are improved, providing technical support for the development of efficient, safe, and environmentally friendly pesticide compounds. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, 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, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0052] Figure 1 This is a flow chart of the first method of the pesticide compound screening method;

[0053] Figure 2 This is a flow chart of the second method of the pesticide compound screening method;

[0054] Figure 3 This is the system block diagram of the pesticide compound screening system. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0057] The first aspect of the present invention discloses a method for screening pesticide compounds, such as Figure 1 As shown, the following steps are included:

[0058] S102: Obtaining standard efficacy data of target pesticide compounds under various environmental characteristic conditions, and constructing a pesticide compound performance prediction model based on the obtained standard efficacy data of target pesticide compounds under various environmental characteristic conditions;

[0059] S104: Acquire continuous environmental change data of the target pesticide compound within a preset application time period, and acquire various standard efficacy data of the target pesticide compound within the preset application time period according to the continuous environmental change data of the target pesticide compound within the preset application time period and the pesticide compound performance prediction model;

[0060] S106: Acquire various actual efficacy data of the target pesticide compound within the preset application time period, compare and analyze the various actual efficacy data of the target pesticide compound within the preset application time period with the standard efficacy data, and obtain the status of various efficacy data of the target pesticide compound after application;

[0061] S108: If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and a recommended preparation process parameter scheme for the excellent pesticide compound is obtained;

[0062] S110: If one or more efficacy data of the target pesticide compound after application are non-standard efficacy data, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated.

[0063] The environmental characteristics are environmental change data, including temperature, humidity, rainfall, wind speed, soil type, water type and pH value.

[0064] It should be noted that, first, by collecting and analyzing the standard efficacy data of the target pesticide compound under various specific environmental conditions, a pesticide compound performance prediction model is constructed. This model can predict the performance of pesticide compounds under different environments and provide a theoretical basis for the subsequent screening process. Next, the continuous environmental change data of the target pesticide compound within a preset time period is obtained. Using the constructed performance prediction model, various standard efficacy data of the pesticide compound within the time period can be simulated and predicted to provide a prediction for practical application. After the preset time period, the actual efficacy data of the target pesticide compound is collected and compared with the predicted standard efficacy data. This step helps to understand the performance of the pesticide compound in the actual environment and evaluate whether its performance meets expectations. According to the results of the comparative analysis, if all efficacy data meet the predetermined standards (i.e., qualified efficacy data), the pesticide compound is screened as an excellent pesticide compound, and its recommended preparation process parameter scheme is further obtained to guide large-scale production. On the contrary, if one or more efficacy data do not meet the standards, the pesticide compound is regarded as an inferior pesticide compound and its preset preparation process parameter scheme is eliminated. By constructing a pesticide compound performance prediction model and combining it with actual environmental data, we can dynamically evaluate the performance changes of pesticide compounds under different environmental conditions and conduct quantitative analysis, thereby improving screening efficiency and accuracy and providing technical support for the development of efficient, safe and environmentally friendly pesticide compounds.

[0065] Preferably, the standard efficacy data of the target pesticide compound under various environmental characteristic conditions are obtained, and a pesticide compound performance prediction model is constructed based on the standard efficacy data of the target pesticide compound under various environmental characteristic conditions, such as Figure 2 As shown, specifically:

[0066] S202: Acquire structural information of the target pesticide compound, wherein the structural information includes a molecular formula, a SMILES string, and a molecular fingerprint, and construct a network search term based on the structural information of the target pesticide compound;

[0067] S204: searching the big data network according to the network search term to obtain standard efficacy data of the target pesticide compound under various environmental characteristic conditions; wherein the efficacy data includes water solubility, degradation rate, bioaccumulation and toxicity;

[0068] S206: Introducing a graph attention network, taking various environmental features as environmental feature nodes, and taking standard efficacy data under various environmental feature conditions as performance change nodes; connecting each environmental feature node with the performance change node with a directed edge, and generating an attention network topology graph;

[0069] S208: Performing attention analysis on the attention network topology graph based on the attention mechanism, obtaining attention scores between nodes in the attention network topology graph, and constructing an attention score matrix according to the attention scores between nodes in the attention network topology graph;

[0070] S210: Construct a prediction model based on a deep learning network, import the attention score matrix into the prediction model and perform back propagation learning until all model parameters meet preset requirements, and then output a pesticide compound performance prediction model.

[0071] It should be noted that network search terms are constructed by obtaining structural information of target pesticide compounds, such as molecular formula, SMILES string, and molecular fingerprint. These structural information are essential for understanding the chemical properties of compounds. Subsequently, these search terms are used to search in the big data network to obtain standard efficacy data under various environmental characteristics. Efficacy data include but are not limited to key indicators such as water solubility, degradation rate, bioaccumulation, and toxicity, which are important parameters for evaluating the performance of pesticide compounds. The graph attention network is introduced to connect different environmental feature nodes with performance change nodes through directed edges to form an attention network topology. This topology allows the model to pay attention to the importance difference between different environmental features and efficacy data, so as to more accurately predict the performance of pesticide compounds in different environments. Through the attention mechanism, the model can automatically identify and emphasize the relationship between nodes in the topology graph, namely the attention score. These scores reflect the strength of the relationship between each node and are used to construct the attention score matrix. This matrix formally represents the model's understanding and attention to the correlation between different environmental features and efficacy data. Based on the deep learning network, the attention score matrix is ​​used as input for back propagation learning. The model parameters are optimized iteratively until the model performance meets the preset requirements. The final output pesticide compound performance prediction model can predict the performance of new compounds under unknown environmental conditions, providing a scientific basis for the design and preparation of pesticide compounds. Through deep learning and attention mechanisms, the model can capture the subtle correlations between complex data, improve the accuracy and reliability of predictions, and the constructed model can be applied to predictions under various environmental conditions, enhancing the applicability and flexibility of pesticide compounds in different environments.

[0072] Preferably, continuous environmental change data of the target pesticide compound within a preset application time period is obtained, and various standard efficacy data of the target pesticide compound within the preset application time period are obtained according to the continuous environmental change data of the target pesticide compound within the preset application time period and the pesticide compound performance prediction model, specifically:

[0073] Obtaining a preset application time period of the target pesticide compound, and collecting continuous environmental change data of the application site of the target pesticide compound within the preset application time period;

[0074] Performing data cleaning processing on the continuous environmental change data within the preset application time period, and arranging the continuous environmental change data within the preset application time period in chronological order to form time series data;

[0075] Converting the time series data into a feature vector at each time point and loading the pesticide compound performance prediction model;

[0076] Using the characteristic vector at each time point as the input of the pesticide compound performance prediction model, and outputting various standard efficacy data of the target pesticide compound within a preset application time period according to the pesticide compound performance prediction model;

[0077] The environmental change data include temperature, humidity, rainfall, wind speed, soil type, water type and pH value.

[0078] It should be noted that the preset application time period of the target pesticide compound is first clarified, which is the basis for implementing the prediction. During the preset application time period, the environmental change data of the target pesticide compound application site are collected, including but not limited to key factors such as temperature, humidity, rainfall, wind speed, soil type, water type and pH value. The collected environmental change data are cleaned and processed to remove outliers and invalid data to ensure the quality and consistency of the data. Then, the data is arranged in chronological order to form continuous time series data. The time series data is converted into a feature vector at each time point. This step is to convert the continuous data into a form that can be processed by the model. The feature vector is loaded into the pre-built pesticide compound performance prediction model. Through the operation of the model, the feature vector at each time point is used as input to output the predicted various standard efficacy data of the target pesticide compound within the preset application time period. By considering the continuous environmental change data, the model can more accurately predict the performance of pesticides in actual applications, because environmental conditions are one of the key factors affecting efficacy. The predicted efficacy data provides a scientific basis for the development, selection and application of pesticides, and helps to formulate more effective and environmentally friendly agricultural strategies.

[0079] Preferably, various actual efficacy data of the target pesticide compound within the preset application time period are compared and analyzed with the standard efficacy data to obtain the status of various efficacy data of the target pesticide compound after application, specifically:

[0080] Calculating the difference between various actual efficacy data and standard efficacy data of the target pesticide compound within the preset application time period to obtain various efficacy data differences of the target pesticide compound within the preset application time period;

[0081] Preset various difference thresholds of efficacy data, and compare the difference values ​​of various efficacy data of the target pesticide compound within the preset application time period with the corresponding difference thresholds;

[0082] If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is not greater than the corresponding difference threshold, the status of the efficacy data is marked as qualified efficacy data;

[0083] If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is greater than the corresponding difference threshold, the status of the efficacy data is marked as non-standard efficacy data.

[0084] It should be noted that by calculating the difference between the actual efficacy data and the standard efficacy data of the target pesticide compound within the preset application time period, the specific differences of various efficacy data are obtained. This step is based on the preset standard to quantitatively evaluate the performance of the pesticide compound under specific environmental conditions. The difference thresholds of various efficacy data are preset, which are set according to the performance requirements and environmental adaptability of the pesticide compound. The setting of the threshold needs to comprehensively consider the purpose of use of the pesticide, environmental factors, safety and economy. The calculated efficacy data difference is compared with the preset difference threshold. In this way, it is possible to intuitively judge whether each efficacy data has reached the expected performance standard. According to the comparison results, the status of the efficacy data is marked as up to standard or not up to standard. Up to standard efficacy data means that the pesticide compound performs well in the corresponding efficacy index, while non-up to standard efficacy data indicates that there is a problem of performance not meeting the standard. Through quantitative comparison, the performance of the pesticide compound can be more accurately evaluated, the advantages and disadvantages can be effectively distinguished, and the accuracy of screening can be improved. For non-up to standard efficacy data, the reasons can be further analyzed, the formula can be adjusted or the production process can be improved in a targeted manner, and the performance of the pesticide compound can be optimized.

[0085] Preferably, if the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and the recommended preparation process parameter scheme of the excellent pesticide compound is obtained, specifically:

[0086] If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound;

[0087] If the target pesticide compound is screened as a high-quality pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the high-quality pesticide compound are extracted according to the operation log;

[0088] Generate an actual preparation process parameter plan for the high-quality pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the period corresponding to the preparation of the low-quality pesticide compound;

[0089] Obtaining a preset preparation process parameter scheme for the excellent pesticide compound; introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme for the excellent pesticide compound and the preset preparation process parameter scheme;

[0090] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to a preset terminal for display;

[0091] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, the actual preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to the preset terminal for display.

[0092] It should be noted that, first, by comparing whether the efficacy data of the pesticide compounds after application meet the standards, the pesticide compounds with excellent performance are screened out. This process ensures that only compounds with good performance can enter the subsequent process optimization stage. Once the excellent pesticide compound is determined, the preparation process data of the compound is obtained from the production line, including the actual working parameters and environmental conditions of each equipment. These data constitute the actual preparation process parameters. Based on the collected actual parameters, a preparation process parameter scheme for the excellent pesticide compound is created. This scheme describes in detail the optimal steps and conditions for producing the compound. Next, the actual preparation process parameter scheme is compared with the preset preparation process parameter scheme. The preset parameters are usually based on theoretical design or past experience, while the actual parameters are the actual performance of the production line. The Pierce correlation coefficient algorithm is used to measure the similarity between the two parameter schemes. The closer this coefficient is to 1, the more matched the two are. Based on the comparison of the Pierce correlation coefficient with the preset threshold, it is decided whether to recommend the actual preparation process parameter scheme or the preset parameter scheme. The recommended parameter scheme will be sent to the preset terminal for display so that the production team can follow and execute it. By comparing and optimizing, we can ensure the efficiency and stability of the preparation process, reduce the cost of trial and error, ensure that the efficacy of the final product meets high standards, and improve the market competitiveness of the product. It also allows parameter adjustment according to actual production conditions, enhancing the adaptability and responsiveness of the production line. In general, by comparing the actual preparation process parameters with the preset parameters to determine the most effective preparation scheme, we can not only ensure the performance of the pesticide compound, but also improve production efficiency and reduce costs.

[0093] Preferably, if the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated, specifically:

[0094] If the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound;

[0095] If the target pesticide compound is screened as an inferior pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the inferior pesticide compound are extracted according to the operation log;

[0096] Generate an actual preparation process parameter plan for the inferior pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the period corresponding to the preparation of the inferior pesticide compound;

[0097] Obtaining a preset preparation process parameter scheme corresponding to the inferior pesticide compound, and introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme;

[0098] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme for preparing the inferior pesticide compound will be eliminated.

[0099] It should be noted that by comparing whether the efficacy data of the pesticide compounds after application meet the standards, the pesticide compounds with poor performance are screened out. Once the inferior pesticide compounds are determined, the preparation process data of the compound is obtained from the production line, including the actual working parameters and environmental conditions of each equipment. These data constitute the actual preparation process parameters. Based on the collected actual parameters, a preparation process parameter scheme for the inferior pesticide compound is created. This scheme describes in detail the process details and conditions for producing the compound. Then, the actual preparation process parameter scheme is compared with the preset preparation process parameter scheme. The preset parameters are usually based on theoretical design or past experience, while the actual parameters are the actual performance of the production line. The Pierce correlation coefficient algorithm is used to measure the similarity between two parameter schemes. The closer this coefficient is to 1, the more matched the two are. Based on the comparison of the Pierce correlation coefficient with the preset threshold, it is decided whether to eliminate the preset preparation process parameter scheme. If the actual preparation process parameter scheme is significantly inconsistent with the preset scheme in terms of similarity, it means that there may be problems with the preset scheme, so it will be eliminated. Through comparison and optimization, the efficiency and stability of the preparation process are ensured, the trial and error costs are reduced, the efficacy of the final product is ensured to meet high standards, and the market competitiveness of the product is improved. By eliminating mismatched parameter schemes, the production team can find a more optimized preparation method and promote continuous improvement of the process. It avoids the use of invalid preparation parameter schemes, reduces unnecessary resource consumption, and achieves cost savings.

[0100] The following steps are also included:

[0101] If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, then various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound are extracted from the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme;

[0102] Calculate the process parameter differences between various actual preparation process parameters and preset preparation process parameters when the preparation production line prepares the inferior pesticide compound;

[0103] Comparing the process parameter differences between various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound with the preset parameter difference threshold;

[0104] Only the preparation process parameters corresponding to the process parameter differences greater than the preset parameter difference threshold are extracted and marked as drift preparation process parameters. The drift preparation process parameters are adjusted according to the corresponding process parameter differences, and then the preparation production line is made to re-prepare the target pesticide compound and then the target pesticide compound is evaluated and screened.

[0105] It should be noted that, first, the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is calculated. If the correlation between the actual and preset parameters is lower than the preset threshold, it means that there is a large deviation between the actual operation and the expectation. At this time, the specific difference between the actual preparation process parameters and the preset parameters will be further calculated to quantify these deviations. Then, these parameter differences are compared with the preset parameter difference threshold. If the difference of a parameter exceeds the threshold, it means that the parameter has changed significantly in the actual operation, which may be the reason for the decline in product performance. The parameters exceeding the threshold are marked as "drift preparation process parameters". Then, according to the size and direction of these parameter differences, the corresponding process parameters are adjusted. This step is intended to correct operations that deviate from the preset parameters. The adjusted process parameters are reapplied to the preparation production line so that the production line re-prepares the target pesticide compound. Then, the re-prepared compounds are evaluated and screened. By accurately adjusting the process parameters, the problem of product performance deviation caused by the drift of the preparation parameters during the production process can be effectively corrected, thereby improving the reliability of the pesticide compound screening results, strengthening the real-time monitoring and feedback of the production process, and helping to establish a more stable and controllable production process.

[0106] In addition, the method further comprises the following steps:

[0107] Acquiring real-time efficacy data of the target pesticide compound at a preset time point within a preset application time period of the target pesticide compound;

[0108] Introducing a fuzzy evaluation algorithm, using the fuzzy evaluation algorithm to perform fuzzy evaluation on the real-time efficacy data of the target pesticide compound, and obtaining the membership degree of each real-time efficacy data;

[0109] Constructing a decision tree model, and determining a split node of the decision tree model according to each real-time drug efficacy data, and iteratively splitting the split node according to the membership degree of each real-time drug efficacy data to generate a plurality of child nodes;

[0110] Acquire the real-time efficacy data in each child node. When there is no redundant data in the real-time efficacy data in each child node, stop iterative splitting, calculate the Euclidean distance between each child node, and construct a distance matrix based on the Euclidean distance between each child node.

[0111] Perform membership transfer probability calculation on the distance matrix based on hidden Markov chain calculation to obtain the state transfer probability of the target pesticide compound at a preset time node;

[0112] If the state transition probability of the target pesticide compound at the preset time node is greater than the preset probability threshold, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated.

[0113] It should be noted that the real-time efficacy data of the pesticide is obtained within the preset time period of the application of the pesticide compound. The fuzzy evaluation algorithm is introduced to analyze the collected real-time efficacy data. This step is mainly to quantify the uncertainty of the efficacy data. The algorithm will give each data point a "membership degree" according to the characteristics of the data, that is, the degree to which the data belongs to a certain evaluation level, thereby forming a more comprehensive and flexible evaluation system. Based on the above-mentioned membership information, a decision tree model is constructed. The decision tree generates multiple child nodes by continuously splitting nodes (i.e., classifying according to the characteristics of the efficacy data). Each step of splitting is based on the optimal division of the data under the current node to better reflect the distribution law of the efficacy data. This process is carried out through iteration until the data in all child nodes does not contain redundant information. After the decision tree is constructed, the Euclidean distance between the real-time efficacy data in each child node is calculated to form a distance matrix. The Euclidean distance is used here to measure the degree of difference between different states (represented by child nodes). The distance matrix is ​​analyzed using the hidden Markov chain model to calculate the transition probability between different states (i.e., the states of the child nodes). Hidden Markov chains can effectively capture the dynamic changes of state sequences. Through the probability transition matrix, the changing trend of the state of pesticide compounds over time can be predicted. Finally, if the state transition probability of a pesticide compound exceeds the set threshold at the preset time node, it means that the compound is unstable or ineffective during use (such as easy decomposition, rapid decline in efficacy, etc.), then it will be identified as an inferior pesticide compound and its preparation process parameter scheme will be eliminated. This process helps to optimize pesticide formulations, improve agricultural production efficiency and environmental protection levels. Overall, this method integrates real-time data collection, fuzzy logic evaluation, decision tree analysis, distance calculation and probability model, realizes intelligent and refined screening and evaluation of pesticide compounds, and provides a scientific basis and efficient tools for pesticide research and development.

[0114] The second aspect of the present invention discloses a pesticide compound screening system, such as Figure 3As shown, the pesticide compound screening system includes a memory 50 and a processor 80. The memory 50 stores a pesticide compound screening method program. When the pesticide compound screening method program is executed by the processor 80, any one of the pesticide compound screening method steps is implemented.

[0115] The third aspect of the present invention discloses a computer-readable storage medium, which stores a pesticide compound screening method program. When the pesticide compound screening method program is executed by a processor, any one of the pesticide compound screening method steps is implemented.

[0116] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for screening pesticide compounds, characterized in that: The following steps are involved: Obtaining standard efficacy data of target pesticide compounds under various environmental characteristic conditions, and constructing a pesticide compound performance prediction model based on the obtained standard efficacy data of target pesticide compounds under various environmental characteristic conditions; Obtaining continuous environmental change data of the target pesticide compound within a preset application time period, and obtaining various standard efficacy data of the target pesticide compound within the preset application time period based on the continuous environmental change data of the target pesticide compound within the preset application time period and a pesticide compound performance prediction model; Acquire various actual efficacy data of the target pesticide compound within a preset application time period, compare and analyze the various actual efficacy data of the target pesticide compound within the preset application time period with the standard efficacy data, and obtain the status of various efficacy data of the target pesticide compound after application; If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and a recommended preparation process parameter scheme for the excellent pesticide compound is obtained; If the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated; Among them, the standard efficacy data of the target pesticide compound under various environmental characteristic conditions are obtained, and the pesticide compound performance prediction model is constructed based on the standard efficacy data of the target pesticide compound under various environmental characteristic conditions, specifically: Acquire structural information of the target pesticide compound, wherein the structural information includes a molecular formula, a SMILES string, and a molecular fingerprint, and construct a network search term based on the structural information of the target pesticide compound; Searching the big data network according to the network search terms to obtain standard efficacy data of the target pesticide compound under various environmental characteristic conditions; wherein the efficacy data includes water solubility, degradation rate, bioaccumulation and toxicity; Introduce the graph attention network, take various environmental features as environmental feature nodes, and take the standard drug efficacy data under various environmental feature conditions as performance change nodes; connect each environmental feature node with the performance change node with a directed edge to generate an attention network topology graph; Performing attention analysis on the attention network topology graph based on the attention mechanism, obtaining attention scores between nodes in the attention network topology graph, and constructing an attention score matrix according to the attention scores between nodes in the attention network topology graph; Constructing a prediction model based on a deep learning network, importing the attention score matrix into the prediction model, and performing back propagation learning until all model parameters meet preset requirements, and then outputting a pesticide compound performance prediction model; Among them, the continuous environmental change data of the target pesticide compound within the preset application time period is obtained, and various standard efficacy data of the target pesticide compound within the preset application time period are obtained according to the continuous environmental change data of the target pesticide compound within the preset application time period and the pesticide compound performance prediction model, specifically: Obtaining a preset application time period of the target pesticide compound, and collecting continuous environmental change data of the application site of the target pesticide compound within the preset application time period; Performing data cleaning processing on the continuous environmental change data within the preset application time period, and arranging the continuous environmental change data within the preset application time period in chronological order to form time series data; Converting the time series data into a feature vector at each time point and loading the pesticide compound performance prediction model; Using the characteristic vector at each time point as the input of the pesticide compound performance prediction model, and outputting various standard efficacy data of the target pesticide compound within a preset application time period according to the pesticide compound performance prediction model; The environmental change data include temperature, humidity, rainfall, wind speed, soil type, water type and pH value; Among them, if the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound, and the recommended preparation process parameter scheme of the excellent pesticide compound is obtained, specifically: If the various efficacy data of the target pesticide compound after application are all up-to-standard efficacy data, the target pesticide compound is screened as an excellent pesticide compound; If the target pesticide compound is screened as a high-quality pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the high-quality pesticide compound are extracted according to the operation log; Generate an actual preparation process parameter plan for the high-quality pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the period corresponding to the preparation of the low-quality pesticide compound; Obtaining a preset preparation process parameter scheme for the excellent pesticide compound; introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme for the excellent pesticide compound and the preset preparation process parameter scheme; If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to a preset terminal for display; If the Pierce correlation coefficient between the actual preparation process parameter scheme of the excellent pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, the actual preparation process parameter scheme is screened as the recommended preparation process parameter scheme of the excellent pesticide compound, and the recommended preparation process parameter scheme is sent to the preset terminal for display; Among them, if the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound, and the preset preparation process parameter scheme for preparing the inferior pesticide compound is eliminated, specifically: If the target pesticide compound has one or more efficacy data that are not up to standard after application, the target pesticide compound is screened as an inferior pesticide compound; If the target pesticide compound is screened as an inferior pesticide compound, an operation log of the preparation production line is obtained, and actual working parameters and actual working environment of each preparation equipment of the preparation production line during the period corresponding to the preparation of the inferior pesticide compound are extracted according to the operation log; Generate an actual preparation process parameter plan for the inferior pesticide compound according to the actual working parameters and actual working environment of each preparation equipment in the preparation production line during the corresponding period of time for preparing the inferior pesticide compound; Obtaining a preset preparation process parameter scheme corresponding to the inferior pesticide compound, and introducing a Pierce correlation coefficient algorithm to calculate the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme; If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is greater than the preset coefficient threshold, the preset preparation process parameter scheme for preparing the inferior pesticide compound will be eliminated; The following steps are also included: If the Pierce correlation coefficient between the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme is not greater than the preset coefficient threshold, then various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound are extracted from the actual preparation process parameter scheme of the inferior pesticide compound and the preset preparation process parameter scheme; Calculate the process parameter differences between various actual preparation process parameters and preset preparation process parameters when the preparation production line prepares the inferior pesticide compound; Comparing the process parameter differences between various actual preparation process parameters and preset preparation process parameters of the preparation production line when preparing the inferior pesticide compound with the preset parameter difference threshold; Only the preparation process parameters corresponding to the process parameter differences greater than the preset parameter difference threshold are extracted and marked as drift preparation process parameters. The drift preparation process parameters are adjusted according to the corresponding process parameter differences, and then the preparation production line is made to re-prepare the target pesticide compound and then the target pesticide compound is evaluated and screened.

2. A method for screening pesticide compounds according to claim 1, characterized in that: The various actual efficacy data of the target pesticide compound within the preset application time period are compared and analyzed with the standard efficacy data to obtain the status of various efficacy data of the target pesticide compound after application, specifically: Calculating the difference between various actual efficacy data and standard efficacy data of the target pesticide compound within the preset application time period to obtain various efficacy data differences of the target pesticide compound within the preset application time period; Preset various difference thresholds of efficacy data, and compare the difference values ​​of various efficacy data of the target pesticide compound within the preset application time period with the corresponding difference thresholds; If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is not greater than the corresponding difference threshold, the status of the efficacy data is marked as qualified efficacy data; If the difference of a certain efficacy data of the target pesticide compound within the preset application time period is greater than the corresponding difference threshold, the status of the efficacy data is marked as non-standard efficacy data.

3. A pesticide compound screening system, characterized in that: The pesticide compound screening system includes a memory and a processor, wherein a pesticide compound screening method program is stored in the memory, and when the pesticide compound screening method program is executed by the processor, the pesticide compound screening method steps as described in any one of claims 1 to 2 are implemented.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a pesticide compound screening method program, and when the pesticide compound screening method program is executed by a processor, the pesticide compound screening method steps as described in any one of claims 1 to 2 are implemented.

Images