Rapeseed oil content prediction method and system based on big data
By analyzing the weights of evaluation indicators in rapeseed growth data and constructing a dynamic prediction model, the problems of unintelligent rapeseed planting management and large fluctuations in oil content were solved, thus improving the accuracy and reliability of rapeseed oil content prediction.
Patent Information
- Application Number
- CN202411141591.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-20
AI Technical Summary
Existing methods for predicting rapeseed oil content suffer from several problems: insufficient intelligence in rapeseed growers' field management, low management efficiency and accuracy, large fluctuations in rapeseed oil content over time, difficulty in flexibly analyzing and predicting under different growth stages and environmental conditions, difficulty in fitting actual fluctuation characteristics with predicted data, and low accuracy and reliability.
By analyzing the weights of evaluation indicators in rapeseed growth data, a dynamic prediction model for oil content was constructed. Dynamic gray factor and similarity measure were introduced, and the model was defined using linear ordinary differential equations with regularization terms set. The prediction model parameters were optimized, and the model was adjusted to fit the fluctuation characteristics of actual oil content data.
It improves the accuracy and management efficiency of rapeseed oil content prediction, enabling flexible analysis and prediction under different growth stages and environmental conditions, thus enhancing the accuracy and reliability of prediction.
Smart Images

Figure CN118657262B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of smart agriculture, and specifically relates to a rapeseed oil content prediction method and system based on big data. Background Art
[0002] The rapeseed oil content prediction method is a method that uses data analysis technology to accurately estimate the oil content in rapeseed. By comprehensively analyzing and modeling the rapeseed growth environment data, rapeseed variety characteristics data, and various data in the planting management process, the final oil content of rapeseed is estimated, aiming to better guide rapeseed planting, variety selection, processing technology optimization and market planning, and improve the economic benefits and sustainable development capabilities of the rapeseed industry. However, the existing rapeseed oil content prediction method has technical problems such as rapeseed growers are not smart enough for field management, the indicators affecting oil content are unclear, management efficiency is low, and accuracy is not high; there are technical problems such as rapeseed oil content fluctuates greatly over time, rapeseed growth environment is complex and changeable, and it is difficult to flexibly analyze and predict under different growth stages and different environmental conditions; there are technical problems such as rapeseed oil content prediction data is difficult to fit the fluctuation characteristics of actual oil content data, and the accuracy and reliability of oil content prediction are not high. Summary of the invention
[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a rapeseed oil content prediction method and system based on big data, which aims at the technical problems that rapeseed growers are not smart enough in field management, the indicators affecting oil content are unclear, the management efficiency is low, and the accuracy is not high. The weights of the evaluation indicators of rapeseed growth data are analyzed to determine the importance ranking of the impact of each evaluation indicator in the rapeseed growth data on the oil content, preset the initial weight value, and calculate the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indicators, and then calculate the specific evaluation indicator weights; in view of the large fluctuation of rapeseed oil content over time, the complex and changeable rapeseed growth environment, and the In order to solve the technical problem that it is difficult to flexibly analyze and predict under different environmental conditions at the same growth stage, a dynamic prediction model for oil content is constructed, and a dynamic grayscale factor is introduced to process complex data. The model is defined by linear ordinary differential equations, and regularization terms are set to constrain the model to evaluate the prediction performance of the model. In order to solve the technical problem that the rapeseed oil content prediction data is difficult to fit the fluctuation characteristics of the actual oil content data, and the accuracy and reliability of the oil content prediction are not high, a similarity measure is introduced to measure the similarity of fluctuations between different rapeseed oil content sequences. According to the results of the similarity measure, the parameters of the dynamic prediction model of oil content are further adjusted to better fit the fluctuation characteristics of the actual oil content data.
[0004] The technical solution adopted by the present invention is as follows: The rapeseed oil content prediction method based on big data provided by the present invention comprises the following steps:
[0005] Step S1: data collection, collecting a large amount of data related to rapeseed oil content prediction, including environmental data, rapeseed growth data, planting management data and sample testing data;
[0006] Step S2: Data analysis, used to analyze the weights of the evaluation indicators of rapeseed growth data, determine the importance ranking of the impact of each evaluation indicator on the oil content in the rapeseed growth data, preset the initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indicators, and then calculate the specific evaluation indicator weights;
[0007] Step S3: construct a dynamic prediction model for oil content, which is used to flexibly analyze and predict rapeseed oil content data. The dynamic prediction model for oil content is established and a dynamic grayscale factor is introduced to process complex data. The model is defined by a linear ordinary differential equation, and a regularization term is set to constrain the model, and the prediction performance of the model is evaluated.
[0008] Step S4: prediction model optimization, used to optimize the prediction model, introduce similarity measurement, measure the similarity of fluctuations between different rapeseed oil content sequences, and further adjust the model parameters according to the results of the similarity measurement to better fit the fluctuation characteristics of the actual oil content data;
[0009] Step S5: Optimize the oil content improvement. According to the oil content dynamic prediction model, predict the change trend of the oil content, timely discover the factors that may affect the oil content and make targeted adjustments.
[0010] Furthermore, in step S1, the data collection includes the following steps:
[0011] Step S11: Environmental data collection, collecting environmental data such as temperature, humidity, light intensity and soil conditions;
[0012] Step S12: collecting rapeseed growth data, including plant height, number of branches, leaf area, flowering time and pod setting;
[0013] Step S13: Planting management data collection, including sowing time, fertilizer type and amount, and irrigation conditions;
[0014] Step S14: Sample detection data collection, performing oil content index detection on the harvested rapeseed samples, and collecting rapeseed oil content data in historical data.
[0015] Furthermore, in step S2, the data analysis includes the following steps:
[0016] Step S21: sorting by importance, defining that the rapeseed growth data has n evaluation indicators, and the importance order of the n evaluation indicators of rapeseed growth data is expressed as follows:
[0017] ;
[0018] The evaluation indicators of rapeseed growth data include plant height, number of branches, leaf area, flowering time and pod setting, etc.
[0019] Step S22: Preset initial weight values and invite experts in related fields to directly assign initial weight values to evaluation indicators based on their experience;
[0020] Step S23: The relative importance of the evaluation indicators is used to determine the quantitative relationship between the relative importance of each evaluation indicator, calculate the ratio of the weights of different evaluation indicators, and clarify the relative proportion of the importance of two adjacent evaluation indicators, so as to further determine the weight of each evaluation indicator in the overall evaluation system. The formula used is as follows:
[0021] ;
[0022] In the formula, u represents the index of the evaluation index, which is used to distinguish the order of different evaluation indicators. is the initial weight value of the u-th evaluation index, is the initial weight value of the u-1th evaluation index, is the importance ratio, which reflects the ratio of relative importance;
[0023] Step S24: Calculate the specific evaluation index weights, using the following formula:
[0024] ;
[0025] In the formula, Indicates the specific evaluation index weight, Indicates An importance ratio is used to measure the relative importance ratios between different evaluation indicators.
[0026] Further, in step S3, the construction of the oil content dynamic prediction model includes the following steps:
[0027] Step S31: Establish a dynamic prediction model for oil content, introduce a dynamic grayscale factor, better express the periodicity and volatility characteristics of rapeseed growth data and environmental data, more efficiently process and predict data with complex attributes, and be more flexible and accurate in predicting variable and complex data;
[0028] Step S32: Define a dynamic prediction model, and use a linear ordinary differential equation to define the model. The formula used is as follows:
[0029] ;
[0030] In the formula, is the dynamic grayscale factor, represents the predicted value of rapeseed oil content changing over time, It represents the derivation of the predicted value of rapeseed oil content changing with time. are the linear parameters of the model, is the fluctuation coefficient of rapeseed growth data, and are coefficients related to trigonometric terms of different frequencies, used to characterize periodic characteristics, w is the frequency coefficient, is the offset parameter, i represents the sequence number of the trigonometric function terms of different frequencies, and n represents the total number of discrete time points;
[0031] Step S33: Determine the model parameters, and optimize the model parameters according to the difference between the actual value and the expected value of the rapeseed oil content. The formula used is as follows:
[0032] ;
[0033] In the formula, Represents the optimized model parameters, argmin represents the corresponding parameter combination when the minimum value is taken, It is used to measure the difference between the actual value and the expected value. represents the true value of rapeseed oil content, represents the predicted value of rapeseed oil content, k represents the discrete time point index;
[0034] Step S34: Constrain the model. In order to achieve the goals of improving model robustness, reducing sensitivity to noise in training data, and preventing overfitting, regularization is performed at the expense of increasing model complexity. The regularization parameter is set to 0.0001. The formula used is as follows:
[0035] ;
[0036] ;
[0037] In the formula, Represents the parameter group after adding the regularization term, represents the model parameters after constraints;
[0038] Step S35: Model prediction performance evaluation, the formula used is as follows:
[0039] ;
[0040] Where MAPE is the mean absolute percentage error.
[0041] Further, in step S4, the prediction model optimization includes the following steps:
[0042] Step S41: Introduce similarity measurement. The currently proposed index cannot accurately measure the similarity of fluctuations between different sequences. A new similarity form is introduced. The different sequences refer to the sequences formed by the oil content data of different rapeseed varieties at different time points. The formula used is as follows:
[0043] ;
[0044] In the formula, sgn represents the sign function, which is used to determine the positive or negative value. represents a similarity measure used to determine how similar different data series are in terms of fluctuations. Represents the difference between two adjacent data points in the actual rapeseed oil content data series. It represents the difference between two adjacent data points in the rapeseed oil content data series predicted by the model;
[0045] Step S42: Adjust the parameters. If the similarity measure is low, it indicates that the actual data sequence and the predicted data sequence have large differences in volatility. The parameters related to data volatility in the model should be adjusted. If the similarity measure result is high, fine-tuning of minor parameters is sufficient. By performing multiple tests and comparing the similarity measure results under different parameter combinations, the optimal parameter settings can be found to better fit the volatility characteristics of the actual oil content data.
[0046] Furthermore, in step S5, the oil content is improved and optimized, specifically, according to the dynamic prediction model of oil content, the changing trend of oil content is predicted, the factors that may affect the oil content are discovered in time and targeted adjustments are made, such as more suitable sowing time according to model feedback, precise control of fertilizer types and quantities, and reasonable arrangement of irrigation to create conditions that are more conducive to rapeseed growth and increased oil content.
[0047] The rapeseed oil content prediction system based on big data provided by the present invention includes a data collection module, a data analysis module, a module for building a dynamic prediction model for oil content, a prediction model optimization module and an oil content improvement optimization module;
[0048] The data collection module specifically collects a large amount of data related to rapeseed oil content prediction, including environmental data, rapeseed growth data, planting management data and sample detection data;
[0049] The data analysis module is specifically used to determine the importance ranking of the impact of each evaluation index on the oil content in the rapeseed growth data, preset an initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indexes, and then calculate the specific evaluation index weights;
[0050] The construction of the oil content dynamic prediction model module specifically includes establishing the oil content dynamic prediction model and introducing a dynamic grayscale factor to process complex data, defining the model with a linear ordinary differential equation, setting a regularization term to constrain the model, and evaluating the model prediction performance;
[0051] The prediction model optimization module specifically introduces a similarity metric to measure the similarity of fluctuations between different rapeseed oil content sequences, and further adjusts the model parameters according to the results of the similarity metric to better fit the fluctuation characteristics of the actual oil content data;
[0052] The oil content improvement optimization module specifically predicts the changing trend of the oil content based on the oil content dynamic prediction model, timely discovers the factors that may affect the oil content and makes targeted adjustments.
[0053] The beneficial results achieved by the present invention using the above scheme are as follows:
[0054] (1) To address the technical problems that rapeseed growers are not smart enough in field management, the indicators that affect oil content are unclear, and the management efficiency and accuracy are low, the weights of the evaluation indicators of rapeseed growth data are analyzed to determine the importance ranking of the impact of each evaluation indicator in the rapeseed growth data on oil content, preset the initial weight value, and calculate the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indicators, and then calculate the specific weight of the evaluation indicator;
[0055] (2) In view of the large fluctuation of rapeseed oil content over time, the complex and changeable rapeseed growth environment, and the difficulty in flexible analysis and prediction under different growth stages and environmental conditions, a dynamic prediction model for oil content was constructed, a dynamic grayscale factor was introduced to process complex data, the model was defined using linear ordinary differential equations, and regularization terms were set to constrain the model, and the prediction performance of the model was evaluated;
[0056] (3) In order to solve the technical problems that the predicted data of rapeseed oil content is difficult to fit the fluctuation characteristics of actual oil content data and the accuracy and reliability of oil content prediction are low, a similarity metric is introduced to measure the similarity of fluctuations between different rapeseed oil content sequences. The model parameters are further adjusted according to the results of the similarity metric to better fit the fluctuation characteristics of actual oil content data. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 A schematic diagram of the process of predicting rapeseed oil content based on big data provided by the present invention;
[0058] Figure 2 A schematic diagram of a rapeseed oil content prediction system based on big data provided by the present invention;
[0059] Figure 3is a schematic flow chart of step S2;
[0060] Figure 4 is a schematic diagram of the process of step S3;
[0061] Figure 5 It is a schematic diagram of the process of step S4.
[0062] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION
[0063] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0064] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.
[0065] Example 1, see Figure 1 The present invention provides a method for predicting rapeseed oil content based on big data, the method comprising the following steps:
[0066] Step S1: data collection, collecting a large amount of data related to rapeseed oil content prediction, including environmental data, rapeseed growth data, planting management data and sample testing data;
[0067] Step S2: Data analysis, used to analyze the weights of the evaluation indicators of rapeseed growth data, determine the importance ranking of the impact of each evaluation indicator on the oil content in the rapeseed growth data, preset the initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indicators, and then calculate the specific evaluation indicator weights;
[0068] Step S3: construct a dynamic prediction model for oil content, which is used to flexibly analyze and predict rapeseed oil content data. The dynamic prediction model for oil content is established and a dynamic grayscale factor is introduced to process complex data. The model is defined by a linear ordinary differential equation, and a regularization term is set to constrain the model, and the prediction performance of the model is evaluated.
[0069] Step S4: prediction model optimization, used to optimize the prediction model, introduce similarity measurement, measure the similarity of fluctuations between different rapeseed oil content sequences, and further adjust the model parameters according to the results of the similarity measurement to better fit the fluctuation characteristics of the actual oil content data;
[0070] Step S5: Optimize the oil content improvement. According to the oil content dynamic prediction model, predict the change trend of the oil content, timely discover the factors that may affect the oil content and make targeted adjustments.
[0071] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the data collection includes the following steps:
[0072] Step S11: Environmental data collection, collecting environmental data such as temperature, humidity, light intensity and soil conditions;
[0073] Step S12: collecting rapeseed growth data, including plant height, number of branches, leaf area, flowering time and pod setting;
[0074] Step S13: Planting management data collection, including sowing time, fertilizer type and amount, and irrigation conditions;
[0075] Step S14: Sample detection data collection, performing oil content index detection on the harvested rapeseed samples, and collecting rapeseed oil content data in historical data.
[0076] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the data analysis includes the following steps:
[0077] Step S21: sorting by importance, defining that the rapeseed growth data has n evaluation indicators, and the importance order of the n evaluation indicators of rapeseed growth data is expressed as follows:
[0078] ;
[0079] The evaluation indicators of rapeseed growth data include plant height, number of branches, leaf area, flowering time and pod setting, etc.
[0080] Step S22: Preset initial weight values and invite experts in related fields to directly assign initial weight values to evaluation indicators based on their experience;
[0081] Step S23: The relative importance of the evaluation indicators is used to determine the quantitative relationship between the relative importance of each evaluation indicator, calculate the ratio of the weights of different evaluation indicators, and clarify the relative proportion of the importance of two adjacent evaluation indicators, so as to further determine the weight of each evaluation indicator in the overall evaluation system. The formula used is as follows:
[0082] ;
[0083] In the formula, u represents the index of the evaluation index, which is used to distinguish the order of different evaluation indicators. is the initial weight value of the u-th evaluation index, is the initial weight value of the u-1th evaluation index, is the importance ratio, which reflects the ratio of relative importance;
[0084] Step S24: Calculate the specific evaluation index weights, using the following formula:
[0085] ;
[0086] In the formula, Indicates the specific evaluation index weight, Indicates An importance ratio is used to measure the relative importance ratios between different evaluation indicators.
[0087] By executing the above operations, the weights of the evaluation indicators of rapeseed growth data are analyzed, the importance ranking of the impact of each evaluation indicator in the rapeseed growth data on the oil content is determined, the initial weight value is preset, and the quantitative relationship of relative importance is clarified by calculating the ratio of the weights of different evaluation indicators, and then the specific evaluation indicator weights are calculated, which solves the technical problems of rapeseed growers' insufficient intelligent field management, unclear indicators affecting oil content, low management efficiency and low accuracy.
[0088] Example 4, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S3, the construction of the oil content dynamic prediction model includes the following steps:
[0089] Step S31: Establish a dynamic prediction model for oil content, introduce a dynamic grayscale factor, better express the periodicity and volatility characteristics of rapeseed growth data and environmental data, more efficiently process and predict data with complex attributes, and be more flexible and accurate in predicting variable and complex data;
[0090] Step S32: Define a dynamic prediction model, and use a linear ordinary differential equation to define the model. The formula used is as follows:
[0091] ;
[0092] In the formula, is the dynamic grayscale factor, represents the predicted value of rapeseed oil content changing over time, It represents the derivation of the predicted value of rapeseed oil content changing with time. are the linear parameters of the model, is the fluctuation coefficient of rapeseed growth data, and are coefficients related to trigonometric terms of different frequencies, used to characterize periodic characteristics, w is the frequency coefficient, is the offset parameter, i represents the sequence number of the trigonometric function terms of different frequencies, and n represents the total number of discrete time points;
[0093] Step S33: Determine the model parameters, and optimize the model parameters according to the difference between the actual value and the expected value of the rapeseed oil content. The formula used is as follows:
[0094] ;
[0095] In the formula, Represents the optimized model parameters, argmin represents the corresponding parameter combination when the minimum value is taken, It is used to measure the difference between the actual value and the expected value. represents the true value of rapeseed oil content, represents the predicted value of rapeseed oil content, k represents the discrete time point index;
[0096] Step S34: Constrain the model. In order to achieve the goals of improving model robustness, reducing sensitivity to noise in training data, and preventing overfitting, regularization is performed at the expense of increasing model complexity. The regularization parameter is set to 0.0001. The formula used is as follows:
[0097] ;
[0098] ;
[0099] In the formula, Represents the parameter group after adding the regularization term, represents the model parameters after constraints;
[0100] Step S35: Model prediction performance evaluation, the formula used is as follows:
[0101] ;
[0102] Where MAPE is the mean absolute percentage error.
[0103] By performing the above operations, a dynamic prediction model for oil content is constructed, a dynamic grayscale factor is introduced to process complex data, the model is defined by a linear ordinary differential equation, the regularization term is set to constrain the model, and the model prediction performance is evaluated. The technical problems that the oil content of rapeseed fluctuates greatly over time, the rapeseed growth environment is complex and changeable, and it is difficult to flexibly analyze and predict under different growth stages and environmental conditions are solved.
[0104] Example 5, see Figure 1 and Figure 5 This embodiment is based on the above embodiment. In step S4, the prediction model optimization includes the following steps:
[0105] Step S41: Introduce similarity measurement. The currently proposed index cannot accurately measure the similarity of fluctuations between different sequences. A new similarity form is introduced. The different sequences refer to the sequences formed by the oil content data of different rapeseed varieties at different time points. The formula used is as follows:
[0106] ;
[0107] In the formula, sgn represents the sign function, which is used to determine the positive or negative value. represents a similarity measure used to determine how similar different data series are in terms of fluctuations. Represents the difference between two adjacent data points in the actual rapeseed oil content data series. It represents the difference between two adjacent data points in the rapeseed oil content data series predicted by the model;
[0108] Step S42: Adjust the parameters. If the similarity measure is low, it indicates that the actual data sequence and the predicted data sequence have large differences in volatility. The parameters related to data volatility in the model should be adjusted. If the similarity measure result is high, fine-tuning of minor parameters is sufficient. By performing multiple tests and comparing the similarity measure results under different parameter combinations, the optimal parameter settings can be found to better fit the volatility characteristics of the actual oil content data.
[0109] By performing the above operations, the similarity measurement is introduced to measure the similarity of fluctuations between different rapeseed oil content sequences. The model parameters are further adjusted according to the results of the similarity measurement to better fit the fluctuation characteristics of the actual oil content data. This solves the technical problems that the rapeseed oil content prediction data is difficult to fit the fluctuation characteristics of the actual oil content data, and the accuracy and reliability of the oil content prediction are not high.
[0110] Example 6, see Figure 1This embodiment is based on the above embodiment. In step S5, the oil content is improved and optimized, specifically, according to the dynamic prediction model of oil content, the changing trend of oil content is predicted, the factors that may affect the oil content are discovered in time and targeted adjustments are made, such as feedback from the model on a more suitable sowing time, precise control of the type and amount of fertilizer, and reasonable arrangement of irrigation to create conditions that are more conducive to rapeseed growth and increased oil content.
[0111] Embodiment 7, see Figure 2 , this embodiment is based on the above embodiment, and the rapeseed oil content prediction system based on big data provided by the present invention includes a data collection module, a data analysis module, a module for building a dynamic prediction model for oil content, a prediction model optimization module and an oil content improvement optimization module;
[0112] The data collection module specifically collects a large amount of data related to rapeseed oil content prediction, including environmental data, rapeseed growth data, planting management data and sample detection data;
[0113] The data analysis module is specifically used to determine the importance ranking of the impact of each evaluation index on the oil content in the rapeseed growth data, preset an initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indexes, and then calculate the specific evaluation index weights;
[0114] The construction of the oil content dynamic prediction model module specifically includes establishing the oil content dynamic prediction model and introducing a dynamic grayscale factor to process complex data, defining the model with a linear ordinary differential equation, setting a regularization term to constrain the model, and evaluating the model prediction performance;
[0115] The prediction model optimization module specifically introduces a similarity metric to measure the similarity of fluctuations between different rapeseed oil content sequences, and further adjusts the model parameters according to the results of the similarity metric to better fit the fluctuation characteristics of the actual oil content data;
[0116] The oil content improvement optimization module specifically predicts the changing trend of the oil content based on the oil content dynamic prediction model, timely discovers the factors that may affect the oil content and makes targeted adjustments.
[0117] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0118] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.
[0119] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.
Claims
1. A rapeseed oil content prediction method based on big data, characterized by: The method comprises the following steps: Step S1: data collection; Step S2: Data analysis, used to analyze the weights of the evaluation indicators of rapeseed growth data, determine the importance ranking of the impact of each evaluation indicator on the oil content in the rapeseed growth data, preset the initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indicators, and then calculate the specific evaluation indicator weights; Step S3: construct a dynamic prediction model for oil content, which is used to flexibly analyze and predict rapeseed oil content data. The dynamic prediction model for oil content is established and a dynamic grayscale factor is introduced to process complex data. The model is defined by a linear ordinary differential equation, and a regularization term is set to constrain the model, and the prediction performance of the model is evaluated. Step S4: prediction model optimization, used to optimize the prediction model, introduce similarity measurement, measure the similarity of fluctuations between different rapeseed oil content sequences, and further adjust the model parameters according to the results of the similarity measurement; Step S5: oil content improvement and optimization; In step S3, the construction of the oil content dynamic prediction model includes the following steps: Step S31: Establish a dynamic prediction model for oil content and introduce a dynamic grayscale factor; Step S32: define a dynamic prediction model, and use a linear ordinary differential equation to define the model. The formula used is as follows: ; In the formula, is the dynamic grayscale factor, represents the predicted value of rapeseed oil content changing over time, It represents the derivation of the predicted value of rapeseed oil content changing with time. is the linear parameter of the dynamic prediction model, is the fluctuation coefficient of rapeseed growth data, and are coefficients related to trigonometric terms of different frequencies, used to characterize periodic characteristics, w is the frequency coefficient, is the offset parameter, i represents the sequence number of the trigonometric function terms of different frequencies, and n represents the total number of discrete time points; Step S33: Determine the model parameters, and optimize the model parameters according to the difference between the actual value and the expected value of the rapeseed oil content. The formula used is as follows: ; In the formula, Represents the optimized model parameters, argmin represents the corresponding parameter combination when the minimum value is taken, It is used to measure the difference between the actual value and the expected value. represents the true value of rapeseed oil content, represents the predicted value of rapeseed oil content, and k represents the discrete time point index; Step S34: Constrain the model and set the regularization parameter to 0.0001. The formula used is as follows: ; ; In the formula, Represents the parameter group after adding the regularization term, represents the model parameters after constraints; Step S35: Model prediction performance evaluation, the formula used is as follows: ; Where MAPE is the mean absolute percentage error.
2. The rapeseed oil content prediction method based on big data according to claim 1, characterized in that: In step S2, the data analysis includes the following steps: Step S21: sorting by importance, defining that there are n evaluation indicators for rapeseed growth data, and the order of importance of the evaluation indicators for rapeseed growth data is as follows: ; Step S22: preset initial weight value; Step S23: The relative importance of the evaluation indicators is used to determine the quantitative relationship between the relative importance of each evaluation indicator, calculate the ratio of the weights of different evaluation indicators, and clarify the relative proportion of the importance of two adjacent evaluation indicators. The formula used is as follows: ; In the formula, u represents the index of the evaluation index, which is used to distinguish the order of different evaluation indicators. is the initial weight value of the u-th evaluation index, is the initial weight value of the u-1th evaluation index, is the importance ratio, which reflects the ratio of relative importance; Step S24: Calculate the specific evaluation index weights, using the following formula: ; In the formula, Indicates the specific evaluation index weight, Indicates An importance ratio is used to measure the relative importance ratios between different evaluation indicators.
3. The rapeseed oil content prediction method based on big data according to claim 1, characterized in that: In step S4, the prediction model optimization includes the following steps: Step S41: Introduce similarity measurement, the formula used is as follows: ; In the formula, sgn represents the sign function, which is used to determine the positive or negative value. represents a similarity measure used to determine how similar different data series are in terms of fluctuations. Represents the difference between two adjacent data points in the actual rapeseed oil content data series. It represents the difference between two adjacent data points in the rapeseed oil content data series predicted by the model; Step S42: Adjust parameters, adjust model parameters according to the similarity measurement results, and fit the fluctuation characteristics of actual data.
4. The rapeseed oil content prediction method based on big data according to claim 1, characterized in that: In step S1, the data collection includes the following steps: Step S11: Environmental data collection; Step S12: collecting rapeseed growth data; Step S13: Planting management data collection; Step S14: sample detection data collection.
5. The rapeseed oil content prediction method based on big data according to claim 1, characterized in that: In step S5, the oil content improvement optimization is specifically to predict the change trend of the oil content according to the oil content dynamic prediction model, timely discover the factors affecting the oil content and make targeted adjustments.
6. A rapeseed oil content prediction system based on big data, used to implement the rapeseed oil content prediction method based on big data as described in any one of claims 1 to 5, characterized in that: It includes data collection module, data analysis module, oil content dynamic prediction model building module, prediction model optimization module and oil content improvement optimization module.
7. The rapeseed oil content prediction system based on big data according to claim 6 is characterized by: The data collection module specifically collects a large amount of data related to rapeseed oil content prediction, including environmental data, rapeseed growth data, planting management data and sample detection data; The data analysis module is specifically used to determine the importance ranking of the impact of each evaluation index on the oil content in the rapeseed growth data, preset an initial weight value, and clarify the quantitative relationship of relative importance by calculating the ratio of the weights of different evaluation indexes, and then calculate the specific evaluation index weights; The construction of the oil content dynamic prediction model module specifically includes establishing the oil content dynamic prediction model and introducing a dynamic grayscale factor to process complex data, defining the model with a linear ordinary differential equation, setting a regularization term to constrain the model, and evaluating the model prediction performance; The prediction model optimization module specifically introduces a similarity metric to measure the similarity of fluctuations between different rapeseed oil content sequences, and adjusts the model parameters according to the results of the similarity metric to fit the fluctuation characteristics of the actual data; The oil content improvement optimization module specifically predicts the changing trend of the oil content based on the oil content dynamic prediction model and performs targeted adjustments.
