Vegetable production index measuring and calculating method and system

Abstract

The invention discloses a vegetable production index measuring and calculating method and system, and the method comprises the steps: collecting production data and environment data of vegetables, and carrying out the preprocessing of the production data and the environment data; the production data comprises planting data and production data; performing time sequence analysis on the planting data according to the periodic yield to obtain a change rate, and correcting the change rate by adopting the production data to obtain an actual change rate; performing variable influence analysis on the environmental data to obtain an attenuation coefficient, and constructing a vegetable production index measurement and calculation model according to the actual change rate and the attenuation coefficient; and optimizing the vegetable production index measuring and calculating model according to the estimation error, inputting data to be measured and calculated into the vegetable production index measuring and calculating model, and outputting an estimation result. The method not only can improve the precision of vegetable production index measurement and calculation, but also has good interpretability, and can be directly applied to a vegetable production index measurement and calculation system.

Description

Technical Field

[0001] The present invention relates to the field of intelligent measurement, and in particular to a vegetable production index measurement method and system. Background Art

[0002] In modern agriculture, efficient management and accurate prediction of vegetable production are crucial to increasing yields, optimizing resource allocation, and meeting market demand. Traditional vegetable production management mainly relies on empirical judgment and simple statistics, which is difficult to accurately reflect the dynamic changes and complex factors in the vegetable production process. With the rapid development of the Internet of Things, big data, and artificial intelligence technologies, it has become possible to accurately calculate the vegetable production index by collecting and analyzing various data in the vegetable production process.

[0003] At present, there are many types of data involved in the vegetable production process, mainly including production data and environmental data. Production data usually covers planting data and output data. These data reflect the entire process of vegetable planting to sales and are an important basis for evaluating vegetable production efficiency and market feedback.

[0004] However, there are many challenges in directly using these raw data to calculate the production index. On the one hand, the raw data may contain noise, outliers or missing values, and need to be preprocessed to improve data quality and availability. On the other hand, there are complex interactions between production data and environmental data. How to accurately capture these relationships and build an effective calculation model is an urgent problem to be solved.

[0005] Therefore, it is urgent to design a vegetable production index calculation method to improve the accuracy, real-time and completeness of vegetable production index calculation. Summary of the invention

[0006] The purpose of the invention is to provide a method for calculating vegetable production index.

[0007] To achieve the above object, the present invention is implemented according to the following technical solutions:

[0008] The present invention comprises the following steps:

[0009] Collecting production data and environmental data of vegetables, and preprocessing the production data and environmental data; the production data includes planting data and output data;

[0010] Performing time series analysis on the planting data according to periodic output to obtain a change rate, and using the production data to correct the change rate to obtain an actual change rate;

[0011] Performing a variable impact analysis on the environmental data to obtain an attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient;

[0012] The vegetable production index calculation model is optimized according to the estimation error, the data to be calculated is input into the vegetable production index calculation model, and the estimation result is output.

[0013] Furthermore, the method of performing time series analysis on the planting data according to the periodic yield to obtain the change rate includes:

[0014] The planting data are arranged in chronological order to obtain a time-series planting data set, and the planting data are divided into regions according to weather conditions and vegetable types;

[0015] Calculate the vegetable yield of the time series planting dataset:

[0016]

[0017] The yield of vegetables of the zth type in the bth region in period x is M x,b,z , the output of the zth type and the bth region at the sth time is M s,z,b , the starting time is T o The end time is T s ;

[0018] Calculate the vegetable growth rate per unit area:

[0019]

[0020] The output of vegetables of type z in region b in the same period last year was MR z,b , the number of vegetable types is n z , the number of areas where vegetables are grown is n b , the statistical error of the yield of the zth type of vegetable in the bth region within period x is σ x,b,z , the growth rate of the zth type of vegetables in the bth area within period x is

[0021] Calculate the average change in vegetable growth rate:

[0022]

[0023] The average change of the vegetable growth rate in the bth area within the period x is θ x,b , step function δ|·|, the vegetable species in the bth area is n1;

[0024] Based on the vegetable areas with weather consistency greater than 80% as the area set, the change rate is calculated:

[0025]

[0026] The rate of change of the i-th region set within period x is γ x,i, the number of regions in the region set is n i .

[0027] Furthermore, the method of using the production data to correct the change rate to obtain the actual change rate includes:

[0028] Calculate the deviation between the actual rate of change and the rate of change:

[0029]

[0030] The adjustment weight is ζ, and the rate of change of the i-th region set in period x is γ x,i , the vegetable production of the i-th region set in period x is M x,i , the number of vegetable types in the bth area is n1, and the number of vegetable types is n z , the production data of the bth area in period x is MU x,b , the number of regions in the region set is n1, and the change rate deviation value of the i-th region set in period x is Δγ x,i , the adjustment constant is ε;

[0031] The calculation uses production data to obtain the actual rate of change:

[0032]

[0033] The actual change rate of the i-th region set in period x is

[0034] Introduce particle population, take the minimum deviation value as the search strategy, take the particle position with the minimum deviation value as the target position, and search for the control weight according to the search strategy;

[0035] Calculate the position of the particle:

[0036]

[0037] The position of the ath particle in the yth dimension is The target location is The step factor is η and the maximum dimension is max y , the minimum dimension is min y , the number of dimensions is n y , the walk parameter of the yth dimension is λ y , the initial position of the particle is

[0038] Use the inertia weight to update the particle position to obtain the inertial position. The expression is:

[0039]

[0040] The inertial position of the ath particle in the yth dimension at the t+1th iteration is The position of the ath particle in the yth dimension at the tth iteration is The position of the particle in the y dimension at the tth iteration is Z y (t), the inertia weight of the tth iteration is The random numbers from 0 to 1 are r1 and r2 respectively, and the maximum number of iterations is ty max , the current number of iterations is t;

[0041] Update the inertia weight, the expression is:

[0042]

[0043] The inertia weight of the t+1th iteration is The maximum value of inertia weight is The minimum inertia weight is

[0044] Iterate continuously until the minimum deviation is reached and output the actual rate of change, otherwise update the inertia weight.

[0045] Furthermore, the method of performing variable impact analysis on the environmental data to obtain the attenuation coefficient includes:

[0046] Input the environmental data into the time-varying impact analysis model to extract the time pattern information of the environmental data;

[0047] The temporal pattern information is fed into the attention mechanism to calculate the sensitivity of vegetables to changes in environmental data:

[0048]

[0049] The initial value of the vegetable normalized difference vegetation index is Q0, and the weight of the cth environmental data is K c , the cth environmental data is u c , the step length of environmental data change is Δu c , wavelength is j, environmental data is u c The initial reflectivity of the vegetable leaves at wavelength j is The initial reflectivity of the vegetable leaves when the environmental data changes step length wavelength j is The sensitivity of vegetables to the step length of environmental data changes is The normalized vegetation index of vegetables after the environmental data change step is

[0050] Taking environmental data with a sensitivity greater than 1 as influencing factors, calculate the contribution of influencing factors to vegetable yields:

[0051]

[0052] The impact factor u c The contribution to vegetable production is φ(u c ), the number of impact factors is Impact Factor c The probability of change is p(Δu c ), the inhibition factor is μ;

[0053] Calculate the attenuation coefficient based on the contribution and impact factor:

[0054]

[0055] The attenuation coefficient is The control coefficient is χ and the error constant is ω.

[0056] Furthermore, a method for constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient includes:

[0057] Construct a measurement function based on the actual rate of change and attenuation coefficient. The expression is:

[0058]

[0059] The attenuation coefficient of the kth vegetable is The actual change rate of the kth vegetable is The base period yield of the kth vegetable in the dth area is UL k.d , the base period sales price of the kth vegetable is fL k , the number of base options for the kth vegetable is ρL k , the yield of the kth vegetable in the dth area is U k.d , the sales price of the kth vegetable is f k , the weight of the kth vegetable is ρ k , the production index of the kth vegetable is

[0060] The predicted vegetable production index is obtained according to the estimation function, and the minimum value of the loss function of the predicted vegetable production index and the actual vegetable production index is used as the objective function of the vegetable production index estimation model;

[0061] The vegetable production index calculation model includes hyperspectral imaging technology, time series analysis algorithm, and machine learning algorithm;

[0062] Hyperspectral imaging technology extracts spectral features related to the growth status of vegetables based on the reflection or transmission spectral information of vegetables at different wavelengths;

[0063] The time series analysis algorithm analyzes the inherent laws of spectral features based on the historical time series data of spectral features to obtain spectral feature change data;

[0064] The machine learning algorithm analyzes the vegetable growth status information reflected in the spectral feature change data and obtains the vegetable production index based on the correlation between the spectral features and the vegetable production index.

[0065] Furthermore, the method for optimizing the vegetable production index calculation model according to the estimation error includes:

[0066] Introduce the particle population and calculate the fitness of the particles:

[0067]

[0068] The fitness of the wth particle is The measured function value of the wth particle is The actual measured function value of the w-th particle is

[0069] The particle with the largest fitness is taken as the optimal individual, and the fitness of the population is calculated to obtain the optimal population;

[0070] Calculate the transition coefficient:

[0071]

[0072] The transition coefficient is The random number between 0 and 1 is r1, the current number of iterations is t, and the maximum number of iterations is t max ;

[0073] Update the particle's position, the expression is:

[0074]

[0075] The number of particles is N, the random numbers from 1 to N-1 are e1, e2, and e3 respectively, and the candidate optimal populations randomly selected in the tth iteration are The cooperation factor is β, and the updated position of the wth particle in the t+1th iteration is The random number is r;

[0076] Update the particle position to obtain the enhanced position. The expression is:

[0077]

[0078] The reinforcement position of the wth particle in the t+1th iteration is The updated position of the wth particle in the tth iteration is

[0079] Update the particle position to obtain the avoidance position. The expression is:

[0080]

[0081] The avoiding position of the wth particle in the t+1th iteration is Update particles to obtain enhanced particles. The expression is:

[0082]

[0083] The w-th enhanced particle in the t+1th iteration is Continue iterating until the maximum number of iterations is reached, otherwise update the particles.

[0084] In a second aspect, a vegetable production index calculation system includes:

[0085] Data collection module: used to collect production data and environmental data of vegetables, and pre-process the production data and environmental data; the production data includes planting data and output data;

[0086] Analysis and correction module: used for performing time series analysis on the planting data according to periodic output to obtain a change rate, and correcting the change rate using the production data to obtain an actual change rate;

[0087] Analysis and construction module: used for performing variable impact analysis on the environmental data to obtain the attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient;

[0088] Optimization output module: used to optimize the vegetable production index calculation model according to the estimation error, input the data to be measured into the vegetable production index calculation model, and output the estimation result.

[0089] The beneficial effects of the present invention are:

[0090] The present invention is a vegetable production index calculation method and system. Compared with the prior art, the present invention has the following technical effects:

[0091] The present invention can improve the accuracy of vegetable production index calculation through preprocessing, time series analysis, change rate correction, variable impact analysis, model construction and model optimization steps, thereby improving the precision of vegetable production index calculation, optimizing vegetable production index calculation, greatly saving resources, and improving work efficiency, and can realize automatic calculation of vegetable production index, and perform multi-data fusion and change rate correction on vegetable production index calculation in real time, which is of great significance to vegetable production index calculation, can adapt to vegetable production index calculation of different standards and different vegetable production index calculation requirements, and has a certain universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Figure 1 The present invention is a flow chart of the steps of a method for calculating vegetable production index. DETAILED DESCRIPTION

[0093] The present invention is further described below by means of specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention but are not intended to limit the present invention.

[0094] A vegetable production index calculation method and system of the present invention comprises the following steps:

[0095] like Figure 1 As shown, in this embodiment, the following steps are included:

[0096] Collecting production data and environmental data of vegetables, and preprocessing the production data and environmental data; the production data includes planting data and output data;

[0097] In the actual assessment, environmental data include temperature, light, humidity, spectral data, precipitation, soil type, soil fertility, soil pH, soil moisture, air quality, geographical location, and surrounding ecological environment; planting data include planting area, planting density, planting cycle, seed type, fertilizer use, reflectivity, and irrigation conditions; production data include yield, quality, and pest and disease conditions;

[0098] Different vegetables from 12 growing areas in XXX County were selected as research objects, and production and environmental data were collected on an annual basis;

[0099] Performing time series analysis on the planting data according to periodic output to obtain a change rate, and using the production data to correct the change rate to obtain an actual change rate;

[0100] In the actual evaluation, the rate of change was 0.0603; the actual rate of change was 0.0625;

[0101] Performing a variable impact analysis on the environmental data to obtain an attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient;

[0102] In the actual evaluation, the attenuation coefficient is 0.817;

[0103] The vegetable production index calculation model is optimized according to the estimation error, the data to be calculated is input into the vegetable production index calculation model, and the estimation result is output.

[0104] In this embodiment, the method of performing time series analysis on the planting data according to periodic yield to obtain the change rate includes:

[0105] The planting data are arranged in chronological order to obtain a time-series planting data set, and the planting data are divided into regions according to weather conditions and vegetable types;

[0106] Calculate the vegetable yield of the time series planting dataset:

[0107]

[0108] The yield of vegetables of the zth type in the bth region in period x is M x,b,z , the output of the zth type and the bth region at the sth time is M s,z,b , the starting time is T o The end time is T s ;

[0109] Calculate the vegetable growth rate per unit area:

[0110]

[0111] The output of vegetables of type z in region b in the same period last year was MR z,b , the number of vegetable types is n z , the number of areas where vegetables are grown is n b , the statistical error of the yield of the zth type of vegetable in the bth region within period x is σ x,b,z , the growth rate of the zth type of vegetables in the bth area within period x is

[0112] Calculate the average change in vegetable growth rate:

[0113]

[0114] The average change of the vegetable growth rate in the bth area within the period x is θ x,b , step function δ|·|, the vegetable species in the bth area is n1;

[0115] Based on the vegetable areas with weather consistency greater than 80% as the area set, the change rate is calculated:

[0116]

[0117] The rate of change of the i-th region set within period x is γ x,i , the number of regions in the region set is n i .

[0118] In this embodiment, the method of using the production data to correct the change rate to obtain the actual change rate includes:

[0119] Calculate the deviation between the actual rate of change and the rate of change:

[0120]

[0121] The adjustment weight is ζ, and the rate of change of the i-th region set in period x is γ x,i, the vegetable production of the i-th region set in period x is M x,i , the number of vegetable types in the bth area is n1, and the number of vegetable types is n z , the production data of the bth area in period x is MU x,b , the number of regions in the region set is n1, and the change rate deviation value of the i-th region set in period x is Δγ x,i , the adjustment constant is ε;

[0122] The calculation uses production data to obtain the actual rate of change:

[0123]

[0124] The actual change rate of the i-th region set in period x is

[0125] Introduce particle population, take the minimum deviation value as the search strategy, take the particle position with the minimum deviation value as the target position, and search for the control weight according to the search strategy;

[0126] Calculate the position of the particle:

[0127]

[0128] The position of the ath particle in the yth dimension is The target location is The step factor is η and the maximum dimension is max y , the minimum dimension is min y , the number of dimensions is n y , the walk parameter of the yth dimension is λ y , the initial position of the particle is

[0129] Use the inertia weight to update the particle position to obtain the inertial position. The expression is:

[0130]

[0131] The inertial position of the ath particle in the yth dimension at the t+1th iteration is The position of the ath particle in the yth dimension at the tth iteration is The position of the particle in the y dimension at the tth iteration is Z y (t), the inertia weight of the tth iteration is The random numbers from 0 to 1 are r1 and r2 respectively, and the maximum number of iterations is ty max , the current number of iterations is t;

[0132] Update the inertia weight, the expression is:

[0133]

[0134] The inertia weight of the t+1th iteration is The maximum value of inertia weight is The minimum inertia weight is

[0135] Iterate continuously until the minimum deviation is reached and output the actual rate of change, otherwise update the inertia weight.

[0136] In this embodiment, the method of performing variable impact analysis on the environmental data to obtain the attenuation coefficient includes:

[0137] Input the environmental data into the time-varying impact analysis model to extract the time pattern information of the environmental data;

[0138] The temporal pattern information is fed into the attention mechanism to calculate the sensitivity of vegetables to changes in environmental data:

[0139]

[0140] The initial value of the vegetable normalized difference vegetation index is Q0, and the weight of the cth environmental data is K c , the cth environmental data is u c , the step length of environmental data change is Δu c , wavelength is j, environmental data is u c The initial reflectivity of the vegetable leaves at wavelength j is The initial reflectivity of the vegetable leaves when the environmental data changes step length wavelength j is The sensitivity of vegetables to the step length of environmental data changes is The normalized vegetation index of vegetables after the environmental data change step is

[0141] Taking environmental data with a sensitivity greater than 1 as influencing factors, calculate the contribution of influencing factors to vegetable yields:

[0142]

[0143] The impact factor u c The contribution to vegetable production is φ(u c ), the number of impact factors is Impact Factor c The probability of change is p(Δu c ), the inhibition factor is μ;

[0144] Calculate the attenuation coefficient based on the contribution and impact factor:

[0145]

[0146] The attenuation coefficient is The control coefficient is χ and the error constant is ω.

[0147] In this embodiment, the method for constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient includes:

[0148] Construct a measurement function based on the actual rate of change and attenuation coefficient. The expression is:

[0149]

[0150] The attenuation coefficient of the kth vegetable is The actual change rate of the kth vegetable is The base period yield of the kth vegetable in the dth area is UL k.d , the base period sales price of the kth vegetable is fL k , the number of base options for the kth vegetable is ρL k , the yield of the kth vegetable in the dth area is U k.d , the sales price of the kth vegetable is f k , the weight of the kth vegetable is ρ k , the production index of the kth vegetable is

[0151] The predicted vegetable production index is obtained according to the estimation function, and the minimum value of the loss function of the predicted vegetable production index and the actual vegetable production index is used as the objective function of the vegetable production index estimation model;

[0152] The vegetable production index calculation model includes hyperspectral imaging technology, time series analysis algorithm, and machine learning algorithm;

[0153] Hyperspectral imaging technology extracts spectral features related to the growth status of vegetables based on the reflection or transmission spectral information of vegetables at different wavelengths;

[0154] The time series analysis algorithm analyzes the inherent laws of spectral features based on the historical time series data of spectral features to obtain spectral feature change data;

[0155] The machine learning algorithm analyzes the vegetable growth status information reflected in the spectral feature change data and obtains the vegetable production index based on the correlation between the spectral features and the vegetable production index.

[0156] In this embodiment, the method for optimizing the vegetable production index calculation model according to the estimation error includes:

[0157] Introduce the particle population and calculate the fitness of the particles:

[0158]

[0159] The fitness of the wth particle is The measured function value of the wth particle is The actual measured function value of the w-th particle is

[0160] The particle with the largest fitness is taken as the optimal individual, and the fitness of the population is calculated to obtain the optimal population;

[0161] Calculate the transition coefficient:

[0162]

[0163] The transition coefficient is The random number between 0 and 1 is r1, the current number of iterations is t, and the maximum number of iterations is t max ;

[0164] Update the particle's position, the expression is:

[0165]

[0166] The number of particles is N, the random numbers from 1 to N-1 are e1, e2, and e3 respectively, and the candidate optimal populations randomly selected in the tth iteration are The cooperation factor is β, and the updated position of the wth particle in the t+1th iteration is The random number is r;

[0167] Update the particle position to obtain the enhanced position. The expression is:

[0168]

[0169] The reinforcement position of the wth particle in the t+1th iteration is The updated position of the wth particle in the tth iteration is

[0170] Update the particle position to obtain the avoidance position. The expression is:

[0171]

[0172] The avoiding position of the wth particle in the t+1th iteration is Update particles to obtain enhanced particles. The expression is:

[0173]

[0174]

[0175] The w-th enhanced particle in the t+1th iteration is Continue iterating until the maximum number of iterations is reached, otherwise update the particles.

[0176] In a second aspect, a vegetable production index calculation system includes:

[0177] Data collection module: used to collect production data and environmental data of vegetables, and pre-process the production data and environmental data; the production data includes planting data and output data;

[0178] Analysis and correction module: used for performing time series analysis on the planting data according to periodic output to obtain a change rate, and correcting the change rate using the production data to obtain an actual change rate;

[0179] Analysis and construction module: used for performing variable impact analysis on the environmental data to obtain the attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient;

[0180] Optimization output module: used to optimize the vegetable production index calculation model according to the estimation error, input the data to be measured into the vegetable production index calculation model, and output the estimation result.

[0181] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating vegetable production index, characterized in that: The following steps are involved: Collecting production data and environmental data of vegetables, and preprocessing the production data and environmental data; the production data includes planting data and output data; Performing time series analysis on the planting data according to periodic output to obtain a change rate, and using the production data to correct the change rate to obtain an actual change rate; Performing a variable impact analysis on the environmental data to obtain an attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient; The vegetable production index calculation model is optimized according to the estimation error, the data to be calculated is input into the vegetable production index calculation model, and the estimation result is output.

2. A vegetable production index calculation method according to claim 1, characterized in that: The method of performing time series analysis on the planting data according to periodic yield to obtain the change rate includes: The planting data are arranged in chronological order to obtain a time-series planting data set, and the planting data are divided into regions according to weather conditions and vegetable types; Calculate the vegetable yield of the time series planting dataset: The yield of vegetables of the zth type in the bth region in period x is M x,b,z , the output of the zth type and the bth region at the sth time is M s,z,b , the starting time is T o , the end time is T s ; Calculate the vegetable growth rate per unit area: The output of vegetables of type z in region b in the same period last year was MR z,b , the number of vegetable types is n z , the number of areas where vegetables are grown is n b , the statistical error of the yield of the zth type of vegetable in the bth region within period x is σ x,b,z , the growth rate of the zth type of vegetables in the bth area within period x is Calculate the average change in vegetable growth rate: The average change of the vegetable growth rate in the bth area within the period x is θ x,b , step function δ|·|, the vegetable species in the bth area is n1; Based on the vegetable areas with weather consistency greater than 80% as the area set, the change rate is calculated: The rate of change of the i-th region set within period x is γ x,i , the number of regions in the region set is n i .

3. The vegetable production index calculation method according to claim 1, characterized in that: The method of using the production data to correct the change rate to obtain the actual change rate includes: Calculate the deviation between the actual rate of change and the rate of change: The adjustment weight is ζ, and the rate of change of the i-th region set in period x is γ x,i , the vegetable production of the i-th region set in period x is M x,i , the number of vegetable types in the bth area is n1, and the number of vegetable types is n z , the production data of the bth area in period x is MU x,b , the number of regions in the region set is n1, and the change rate deviation value of the i-th region set in period x is Δγ x,i , the adjustment constant is ε; The calculation uses production data to obtain the actual rate of change: The actual change rate of the i-th region set in period x is Introduce particle population, take the minimum deviation value as the search strategy, take the particle position with the minimum deviation value as the target position, and search for the control weight according to the search strategy; Calculate the position of the particle: The position of the ath particle in the yth dimension is The target location is The step factor is η and the maximum dimension is max y , the minimum dimension is min y , the number of dimensions is n y , the walk parameter of the yth dimension is λ y , the initial position of the particle is Use the inertia weight to update the particle position to obtain the inertial position. The expression is: The inertial position of the ath particle in the yth dimension at the t+1th iteration is The position of the ath particle in the yth dimension at the tth iteration is The position of the particle in the y dimension at the tth iteration is Z y (t), the inertia weight of the tth iteration is The random numbers from 0 to 1 are r1 and r2 respectively, and the maximum number of iterations is ty max , the current number of iterations is t; Update the inertia weight, the expression is: The inertia weight of the t+1th iteration is The maximum value of inertia weight is The minimum inertia weight is Iterate continuously until the minimum deviation is reached and output the actual rate of change, otherwise update the inertia weight.

4. The vegetable production index calculation method according to claim 1, characterized in that: The method of performing variable impact analysis on the environmental data to obtain the attenuation coefficient comprises: Input the environmental data into the time-varying impact analysis model to extract the time pattern information of the environmental data; The temporal pattern information is fed into the attention mechanism to calculate the sensitivity of vegetables to changes in environmental data: The initial value of the vegetable normalized difference vegetation index is Q0, and the weight of the cth environmental data is K c , the cth environmental data is u c , the step length of environmental data change is Δu c , wavelength is j, environmental data is u c The initial reflectivity of the vegetable leaves at wavelength j is The initial reflectivity of the vegetable leaves when the environmental data changes step length wavelength j is The sensitivity of vegetables to the step length of environmental data changes is The normalized vegetation index of vegetables after the environmental data change step is Taking environmental data with a sensitivity greater than 1 as influencing factors, calculate the contribution of influencing factors to vegetable yields: The impact factor u c The contribution to vegetable production is φ(u c ), the number of impact factors is Impact Factor c The probability of change is p(Δu c ), the inhibition factor is μ; Calculate the attenuation coefficient based on the contribution and impact factor: The attenuation coefficient is The control coefficient is χ and the error constant is ω.

5. The vegetable production index calculation method according to claim 1, characterized in that: The method for constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient includes: Construct a measurement function based on the actual rate of change and attenuation coefficient. The expression is: The attenuation coefficient of the kth vegetable is The actual change rate of the kth vegetable is The base period yield of the kth vegetable in the dth area is UL k.d , the base period sales price of the kth vegetable is fL k , the number of base options for the kth vegetable is ρL k , the yield of the kth vegetable in the dth area is U k.d , the sales price of the kth vegetable is f k , the weight of the kth vegetable is ρ k , the production index of the kth vegetable is The predicted vegetable production index is obtained according to the estimation function, and the minimum value of the loss function of the predicted vegetable production index and the actual vegetable production index is used as the objective function of the vegetable production index estimation model; The vegetable production index calculation model includes hyperspectral imaging technology, time series analysis algorithm, and machine learning algorithm; Hyperspectral imaging technology extracts spectral features related to the growth status of vegetables based on the reflection or transmission spectral information of vegetables at different wavelengths; The time series analysis algorithm analyzes the inherent laws of spectral features based on the historical time series data of spectral features to obtain spectral feature change data; The machine learning algorithm analyzes the vegetable growth status information reflected in the spectral feature change data and obtains the vegetable production index based on the correlation between the spectral features and the vegetable production index.

6. The method for calculating vegetable production index according to claim 1, characterized in that: The method for optimizing the vegetable production index calculation model according to the estimation error comprises: Introduce the particle population and calculate the fitness of the particles: The fitness of the wth particle is The measured function value of the wth particle is The actual measured function value of the w-th particle is The particle with the largest fitness is taken as the optimal individual, and the fitness of the population is calculated to obtain the optimal population; Calculate the transition coefficient: The transition coefficient is The random number between 0 and 1 is r1, the current number of iterations is t, and the maximum number of iterations is t max ; Update the particle's position, the expression is: The number of particles is N, the random numbers from 1 to N-1 are e1, e2, and e3 respectively, and the candidate optimal populations randomly selected in the tth iteration are The cooperation factor is β, and the updated position of the wth particle in the t+1th iteration is The random number is r; Update the particle position to obtain the enhanced position. The expression is: The reinforcement position of the wth particle in the t+1th iteration is The updated position of the wth particle in the tth iteration is Update the particle position to obtain the avoidance position. The expression is: The avoiding position of the wth particle in the t+1th iteration is Update particles to obtain enhanced particles. The expression is: The w-th enhanced particle in the t+1th iteration is Continue iterating until the maximum number of iterations is reached, otherwise update the particles.

7. A vegetable production index calculation system, used to execute the method described in any one of claims 1 to 6, characterized in that: include: Data collection module: used to collect production data and environmental data of vegetables, and pre-process the production data and environmental data; the production data includes planting data and output data; Analysis and correction module: used for performing time series analysis on the planting data according to periodic output to obtain a change rate, and correcting the change rate using the production data to obtain an actual change rate; Analysis and construction module: used for performing variable impact analysis on the environmental data to obtain the attenuation coefficient, and constructing a vegetable production index calculation model according to the actual change rate and the attenuation coefficient; Optimization output module: used to optimize the vegetable production index calculation model according to the estimation error, input the data to be measured into the vegetable production index calculation model, and output the estimation result.

Images