Intelligent voice interaction system and method for agricultural meteorological consultation

Abstract

The invention discloses an intelligent voice interaction system and method for agrometeorological consultation, relates to the technical field of intelligent agriculture, and solves the problem that the reply made by a current voice interaction system applied to crops is only simple reply in terms of language and text. Therefore, the technical problem of lack of pertinence and scientificity of decisions given by an interaction system is solved. Real-time weather data, real-time crop data and voice information of a user are collected through a data sensor; analyzing the growth period of the crops according to the real-time crop data to obtain crop demand data; analyzing the growth environment of the crops according to the real-time weather data to obtain environment data; analyzing according to the environment data and the crop demand quantity to obtain farming operation data; recommending the farming operation data to the user; dividing consultation types of the users according to the voice information of the users; analyzing according to the consultation type of the user to obtain answer data; the influence of environment on crops is reduced, and the yield of crops is improved.

Description

Technical Field

[0001] The present invention belongs to the field of smart agriculture and relates to intelligent interaction technology for agricultural meteorological consultation. Specifically, it is an intelligent voice interaction system and method for agricultural meteorological consultation. Background Art

[0002] Agricultural meteorological consultation is a professional service that combines meteorological science and agricultural production needs. It aims to help agricultural producers avoid meteorological risks and optimize production management by providing accurate meteorological data analysis and farming decision-making suggestions. Intelligent voice interaction technology allows users to quickly obtain agricultural meteorological information through voice commands, eliminating the need for manual text input or web searching, thus simplifying the information acquisition process. An intelligent voice interaction system can recognize the voice characteristics of users, understand their intentions and needs, and thus provide more personalized agricultural meteorological services to help users better cope with meteorological changes. Intelligent voice interaction technology is an important part of smart agriculture. Combining it with technologies such as intelligent agricultural equipment and the Internet of Things can achieve real-time monitoring, analysis, and early warning of agricultural meteorological information, providing a scientific basis for agriculture, helping farmers more accurately grasp meteorological conditions, reasonably arrange farming activities, and improve the efficiency and benefits of agricultural production.

[0003] The prior art (a patent application with publication number CN119271765A) discloses an agricultural large model system based on multi-modal fusion technology, including a dialogue management module, a task scheduling engine, an agricultural knowledge base, a model and application library, and a large voice model base. The dialogue management module includes an intelligent interaction terminal, a voice converter, an image processor, a dialogue wrapper, and a prompt template. The task scheduling engine includes a task template, a task scheduler, a task executor, and a task monitor. The data sources of the agricultural knowledge base include an agricultural knowledge graph, an object database, a relational database, a text, and a graph database. The model and application library includes a crop recognition model, a pest and disease recognition model, a spectral recognition model, and several business systems. When the existing voice interaction system replies to the questions and voice commands excluded by users, due to the lack of analysis data on crops, the current applications and the replies made by the voice interaction system for crops are only simple replies in terms of language and text, and the simple replies lack theoretical basis in terms of data, resulting in the decisions given by the interaction system lacking pertinence and scientificity.

[0004] The present invention provides an intelligent voice interaction system and method for agricultural meteorological consultation to solve the above technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present invention proposes an intelligent voice interaction system and method for agricultural meteorological consultation, which is used to solve the problem that when the existing voice interaction system replies to the questions and voice commands eliminated by the user, due to the lack of analysis data on crops, the current application and the reply made by the voice interaction system for crops are only simple replies in terms of language and text, and the simple replies lack theoretical basis in terms of data, resulting in the decision given by the interaction system lacking pertinence and scientificity.

[0006] To achieve the above object, a first aspect of the present invention provides an intelligent voice interaction system for agricultural meteorological consultation, including: an intelligent advice module, and an information collection module and an intelligent voice module connected thereto;

[0007] The information collection module: is used to collect real-time weather data, real-time crop data and the voice information of the user through data sensors;

[0008] The intelligent advice module: is used to analyze the growth period of the crop according to the real-time crop data to obtain crop demand data; analyze the growth environment of the crop according to the real-time weather data to obtain environmental data; analyze and obtain farming operation data according to the environmental data and crop demand; recommend the farming operation data to the user;

[0009] The intelligent voice module: is used to classify the consultation type of the user according to the voice information of the user; analyze and obtain answer data according to the consultation type of the user.

[0010] Preferably, analyzing the growth period of the crop according to the real-time crop data includes:

[0011] Retrieving the real-time crop data; wherein, the real-time crop data includes: crop variety, sowing time and the real-time height of the crop;

[0012] Obtaining the time growth table of the crop; matching the crop variety and sowing time with the time growth table respectively to obtain the growth stage of the crop;

[0013] Obtaining the standard range of the crop height at the growth stage; when the real-time height of the crop is within the standard range, taking the growth stage as the growth period of the crop; otherwise, re-determining the growth period of the crop; wherein, the growth period of the crop includes: germination, growth, flowering, fruiting and maturity;

[0014] Obtaining the growth demand table of the corresponding variety of crop; matching the growth period of the crop with the growth demand table to obtain the corresponding crop demand data; wherein, the crop demand data includes the watering amount, fertilization amount, optimal light intensity and optimal light time.

[0015] It should be noted that the time growth table of crops is set according to the average value of historical data, including the time of each growth period of crops. For example, it takes 1 month for Crop A to germinate after sowing; the growth requirement table of crops is set according to the historical requirement data of the corresponding crops.

[0016] The present invention analyzes the growth period of crops based on the sowing time and variety of crops. However, when the real-time environment is different, the length of the growth period is different; re-determining the growth period of crops according to the growth height of crops can accurately divide the growth period of crops and provide a data basis for subsequent analysis.

[0017] Preferably, analyzing the growth environment of crops according to real-time weather data includes:

[0018] Retrieving real-time weather data within a set time period; integrating the real-time weather data within the set time period in chronological order to obtain an environmental analysis sequence; wherein, the real-time weather data includes: real-time temperature, real-time humidity, real-time precipitation, and real-time light intensity;

[0019] Invoking an environmental analysis model, inputting the environmental analysis sequence into the environmental analysis model to obtain a corresponding environmental label, and matching corresponding environmental data according to the environmental label; wherein, the environmental label is set as a positive integer; the environmental analysis model is constructed based on an artificial intelligence model.

[0020] The present invention comprehensively analyzes environmental data according to real-time weather data within a set time period, uses an artificial intelligence model to analyze the real-time weather data, and obtains environmental data of crops; it can comprehensively analyze the growth environment of crops and provide a data basis for subsequent analysis.

[0021] Preferably, the environmental analysis model is constructed based on an artificial intelligence model, including:

[0022] Obtaining a standard data set; wherein, the standard data set includes standard input data consistent with the content attributes of the environmental analysis sequence and standard output data consistent with the content attributes of the environmental label;

[0023] Dividing the standard data set into a training set, a validation set, and a test set according to a set ratio; training the artificial intelligence model using the training set; adjusting the internal parameters of the artificial intelligence model using the validation set; testing the artificial intelligence model using the test set to obtain test metrics;

[0024] Obtain the index threshold; when the test index is greater than the index threshold, mark the artificial intelligence model as an environmental analysis model; otherwise, retrain the artificial intelligence model; wherein, the artificial intelligence model includes: a convolutional neural network model or a long short-term memory neural network model.

[0025] It should be noted that the index threshold and the division ratio of the standard data set are set by expert evaluation; the test indexes include: accuracy rate, recall rate, and F1 score; when it is necessary to retrain the artificial intelligence model, the division ratio of the standard data set needs to be reset.

[0026] Preferably, the analysis according to environmental data and crop demand data includes:

[0027] Retrieve environmental data and crop demand data; wherein, the environmental data includes: total precipitation, sunshine duration, average sunshine intensity, and day-night temperature difference.

[0028] Obtain the standard temperature difference range; when the day-night temperature difference is greater than the standard temperature difference range, generate a heat preservation signal and send the heat preservation signal to the user; otherwise, calculate the difference between the total precipitation and the watering amount, and re-analyze the watering amount according to the difference.

[0029] Fit the light intensity in the environmental data in chronological order to obtain the light curve F(t); mark the optimal light intensity and the optimal light time in the crop demand data as ZS and ZQ respectively; through the formula Calculate the light coefficient of the crops; wherein, α is a proportionality coefficient greater than 0, GY is the lowest effective light intensity; i is the set of start times of the effective light time, and n is the set of end times of the effective light time.

[0030] When the light coefficient is greater than the first-level threshold, generate supplementary light data; when the light coefficient is less than the second-level threshold, generate shading data; integrate the heat preservation signal, watering amount, fertilization amount, supplementary light data, and shading data into agricultural operation data.

[0031] It should be noted that the lowest effective light intensity GY is determined according to the variety and growth period of the crops. The effective light intensities of different crops are different, and the effective light intensities of the same crop in different growth periods are also different. Determining the effective light intensity of the crops according to the variety and growth period of the crops can make the analysis results more accurate; the first-level threshold and the second-level threshold are determined according to the variety and growth period of the crops. The first-level threshold is greater than the second-level threshold. The first-level threshold is the difference between the maximum light received by the crops within the set time period and the optimal light; the second-level threshold is the difference between the minimum light received by the crops within the set time period and the optimal light.

[0032] The present invention analyzes agricultural operation data based on environmental data and crop demand data. When the real-time environment fails to meet the needs of crop growth, it can timely supplement the nutrients required by the crops, which is beneficial to ensuring the normal growth of crops and increasing the crop yield.

[0033] Preferably, dividing the user's consultation types according to the user's voice information includes:

[0034] Retrieve the user's voice information, use speech recognition technology to convert the speech into text to obtain the recognition text; use word segmentation technology to divide the recognition text to obtain a number of divided words; analyze the number of divided words according to the stop word library, remove the stop words in the number of divided words to obtain feature words;

[0035] Obtain the key features of a number of consultation types, and analyze the correlation between the feature words and the key features; extract keywords from the feature words according to the correlation; analyze the matching degree between a number of keywords and the consultation types, and select the type with the highest matching degree as the user's consultation type.

[0036] The present invention converts the user's voice information into a recognition text, extracts keywords from the recognition text, and determines the user's consultation type according to the matching degree between the keywords and the consultation types; it can initially divide the user's needs, laying a data foundation for subsequent answers.

[0037] Preferably, analyzing the obtained answer data according to the user's consultation type includes:

[0038] Retrieve the user's consultation type, and match the corresponding analysis and solution library according to the user's consultation type; analyze the relevance between a number of keywords and the problems in the analysis and solution library;

[0039] Set the weight coefficient of the keyword according to the key degree of the keyword, calculate the product sum of the relevance of the corresponding keyword and the weight coefficient to obtain the matching coefficient; screen out the highest solution measure from the matching coefficients as the answer data, and send the answer data to the user.

[0040] It should be noted that the answer data is presented to the user in two ways: text and voice.

[0041] The present invention matches the corresponding analysis and solution library according to the user's consultation type, calculates the matching coefficient according to the relevance of the keywords and the set weight coefficient, selects the solution measure as the answer data according to the matching coefficient, and sends it to the user; the user only needs to ask questions by voice to obtain the desired answer, which can timely solve the user's problems.

[0042] Preferably, the intelligent suggestion module is also used to analyze abnormal weather according to real-time weather data, including:

[0043] Retrieve the real-time weather data for a set time period, and fit the real-time temperature, real-time precipitation, and real-time light intensity in the real-time weather data into corresponding weather curves; among them, the weather curves include: temperature curve, precipitation curve, and light intensity curve;

[0044] Obtain the peak values of the weather curves; compare the peak values with the corresponding weather threshold ranges; when the peak values exceed the weather threshold ranges, generate an alarm signal and conduct voice broadcast; otherwise, predict abnormal weather based on the weather curves.

[0045] It should be noted that the weather threshold ranges are set according to the varieties, growth periods, and seasons of crops.

[0046] The present invention analyzes abnormal environments based on real-time weather data, can process crops in a timely manner, avoid adverse effects of abnormal environments on the growth of crops, is beneficial to reducing crop losses, and improving crop yields.

[0047] Preferably, the predicting abnormal weather based on the weather curves includes:

[0048] Retrieve the weather curves, and use the method of curve trend analysis to predict the real-time weather data within a set time period to obtain a predicted weather curve;

[0049] Integrate the predicted weather curves to obtain corresponding intersection points; number the intersection points in chronological order, and extract the predicted weather data corresponding to the time of the intersection points; among them, the predicted weather data includes: predicted temperature, predicted precipitation, and predicted light intensity;

[0050] Mark the predicted temperature, predicted precipitation, and predicted light intensity in the predicted weather data as YWj, YSj, and YQj respectively; through the formula Calculate the abnormality coefficient corresponding to the intersection points; where β and γ are adjustment coefficients greater than 0; WY, SY, and QY are the temperature threshold, precipitation threshold, and light intensity threshold respectively; DL is the dimension removal coefficient; j is the number of the intersection point, j = 1, 2,..., k; k is a positive integer; when the abnormality coefficient is greater than the coefficient threshold, generate a warning signal and send the warning signal to the user.

[0051] It should be noted that the temperature threshold, precipitation threshold, and light intensity threshold are the median values of the standard ranges, and are adjusted according to seasons, varieties, and growth periods of crops; it can make the calculated abnormality coefficient more in line with the actual situation; as long as there are intersection points in any two of the three curves, they are marked as intersection points; the adjustment coefficient is used to adjust the accuracy of the prediction result, and the dimension removal coefficient DL is used to remove the dimension in the formula, and the value is generally set to 1.

[0052] The present invention predicts the weather within a set time period according to the weather curve, analyzes the intersection points of the weather curve, calculates the abnormal coefficient corresponding to the predicted data of the intersection points according to the intersection points, and generates a warning signal according to the abnormal coefficient and the coefficient threshold, which can timely predict the upcoming abnormal weather, remind users to take measures in advance for the abnormal weather, is beneficial to reducing the influence of the abnormal weather on crops, and improving the yield of crops.

[0053] The second aspect of the present invention provides an intelligent voice interaction method for agricultural meteorological consultation, including:

[0054] Step S1: Collect real-time weather data, real-time crop data and the voice information of users through data sensors;

[0055] Step S2: Analyze the growth period of the crops according to the real-time crop data to obtain crop demand data; analyze the growth environment of the crops according to the real-time weather data to obtain environmental data;

[0056] Step S3: Analyze and obtain farming operation data according to the environmental data and the crop demand; recommend the farming operation data to the users;

[0057] Step S4: Classify the consultation types of the users according to the voice information of the users; analyze and obtain answer data according to the consultation types of the users.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] 1. The present invention analyzes the growth period of the crops according to the sowing time and variety of the crops. However, when the real-time environment is different, the length of the growth period is different; re-determine the growth period of the crops according to the growth height of the crops, which can accurately divide the growth period of the crops and provide a data basis for subsequent analysis; comprehensively analyze the environmental data according to the real-time weather data within a set time period, and analyze the real-time weather data by using an artificial intelligence model to obtain the environmental data of the crops; can comprehensively analyze the growth environment of the crops and provide a data basis for subsequent analysis; analyze the farming operation data according to the environmental data and the crop demand data. When the real-time environment fails to meet the growth requirements of the crops, it can timely supplement the nutrients required by the crops, which is beneficial to ensuring the normal growth of the crops and improving the yield of the crops.

[0060] 2. The present invention converts the user's voice information into recognized text, extracts keywords from the recognized text, and determines the user's consultation type according to the matching degree between the keywords and the consultation type; it can initially classify the user's needs, laying a data foundation for subsequent answers; it matches the corresponding analysis and solution library according to the user's consultation type, calculates the matching coefficient based on the relevance of the keywords and the set weight coefficient, selects the solution measure as the answer data according to the matching coefficient, and sends it to the user; the user only needs to ask questions by voice to obtain the desired answer, which can solve the user's problems in a timely manner; it analyzes the abnormal environment according to the real-time weather data, can process the crops in a timely manner, avoid the adverse effects of the abnormal environment on the growth of the crops, is conducive to reducing the loss of the crops, and improving the yield of the crops; it predicts the weather within a set time period according to the weather curve, analyzes the intersection points of the weather curve, calculates the corresponding intersection point anomaly coefficient according to the predicted data of the intersection points, and generates a warning signal according to the anomaly coefficient and the coefficient threshold, which can predict the upcoming abnormal weather in a timely manner, remind the user to take measures in advance for the abnormal weather, and is conducive to reducing the impact of the abnormal weather on the crops. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0062] Figure 1 It is a schematic diagram of the overall system solution steps of the present invention;

[0063] Figure 2 It is a schematic diagram of the agricultural operation data analysis steps of the present invention;

[0064] Figure 3 It is a schematic diagram of the voice question and answer steps of the present invention;

[0065] Figure 4 It is a schematic diagram of the abnormal weather monitoring and prediction steps of the present invention;

[0066] Figure 5 It is a schematic diagram of the specific steps of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0068] Please refer to Figure 1 , an embodiment of the first aspect of the present invention provides an intelligent voice interaction system and method for agricultural meteorological consultation, including: an intelligent recommendation module, and an information collection module and an intelligent voice module connected thereto;

[0069] The information collection module: is used to collect real-time weather data, real-time crop data and the voice information of users through data sensors;

[0070] The intelligent recommendation module: is used to analyze the growth period of crops according to the real-time crop data to obtain crop demand data; analyze the growth environment of crops according to the real-time weather data to obtain environmental data; analyze and obtain farming operation data according to the environmental data and crop demand; recommend the farming operation data to users;

[0071] The intelligent voice module: is used to classify the consultation types of users according to the voice information of users; analyze and obtain answer data according to the consultation types of users.

[0072] Please refer to Figure 2 , collect real-time crop data through data sensors; wherein, the real-time crop data includes: crop variety, sowing time and the real-time height of crops; obtain the time growth table of crops; match the crop variety and sowing time with the time growth table respectively to obtain the growth stage of crops; obtain the standard range of the height of crops in the growth stage; when the real-time height of crops is within the standard range, then use the growth stage as the growth period of crops; otherwise, re-determine the growth period of crops; wherein, the growth period of crops includes: germination, growth, flowering, fruiting and maturity; obtain the growth demand table of corresponding variety crops; match the growth period of crops with the growth demand table to obtain the corresponding crop demand data; wherein, the crop demand data includes watering amount, fertilization amount, optimal light intensity and optimal light time;

[0073] Collect real-time weather data through a data sensor; integrate the real-time weather data within a set time period in chronological order to obtain an environmental analysis sequence; where the real-time weather data includes: real-time temperature, real-time humidity, real-time precipitation, and real-time light intensity; call an environmental analysis model, input the environmental analysis sequence into the environmental analysis model to obtain corresponding environmental labels, and match corresponding environmental data according to the environmental labels; where the environmental labels are set as positive integers; the environmental analysis model is constructed based on an artificial intelligence model.

[0074] Retrieve environmental data and crop requirement data; where the environmental data includes: total precipitation, sunlight duration, average light intensity, and day-night temperature difference; obtain the standard range of the temperature difference; when the day-night temperature difference is greater than the standard range of the temperature difference, generate a heat preservation signal and send the heat preservation signal to the user; otherwise, calculate the difference between the total precipitation and the watering amount, and re-analyze the watering amount according to the difference; fit the light intensity in the environmental data in chronological order to obtain a light curve F(t); mark the optimal light intensity and the optimal sunlight duration in the crop requirement data as ZS and ZQ respectively; through the formula Calculate the light coefficient of the crops; where α is a proportionality coefficient greater than 0, GY is the lowest effective light intensity; i is the set of start times of the effective light duration, and n is the set of end times of the effective light duration; when the light coefficient is greater than the first-level threshold, generate supplementary light data; when the light coefficient is less than the second-level threshold, generate shading data; integrate the heat preservation signal, watering amount, fertilization amount, supplementary light data, and shading data into farm operation data.

[0075] It should be noted that represents the total sum of the effective light intensity within a set time period, and ZS×ZQ represents the comprehensive optimal light intensity; calculating the light coefficient through the formula can analyze and judge the light intensity received by the crops. When the light intensity is too high, shading treatment can be performed on the crops; when the light intensity is too low, supplementary light treatment can be performed on the crops.

[0076] It should be explained that the growth period of the crops is analyzed based on the crop data, and the farm operation data is analyzed based on the environmental data. When the growth environment of the crops is not conducive to the growth of the crops, human intervention can be carried out, which is beneficial to making the growth trend of the crops better.

[0077] It is worth noting that the environmental analysis model is constructed based on an artificial intelligence model, including:

[0078] Obtain a standard data set; where the standard data set includes standard input data consistent with the content attributes of the environmental analysis sequence and standard output data consistent with the content attributes of the environmental labels.

[0079] Divide the standard data set into a training set, a validation set, and a test set according to a set ratio; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the artificial intelligence model; use the test set to test the artificial intelligence model to obtain test metrics;

[0080] Obtain the metric threshold; when the test metric is greater than the metric threshold, mark the artificial intelligence model as an environmental analysis model; otherwise, retrain the artificial intelligence model; wherein, the artificial intelligence model includes: a convolutional neural network model or a long short-term memory neural network model.

[0081] Please refer to Figure 3 , collect the user's voice information through a data sensor; use speech recognition technology to convert the speech into text to obtain the recognition text; use word segmentation technology to divide the recognition text to obtain a number of divided words; analyze the number of divided words according to the stop word library, remove the stop words in the number of divided words to obtain feature words; obtain the key features of several consultation types, and analyze the correlation between the feature words and the key features; extract keywords from the feature words according to the correlation; analyze the matching degree between several keywords and the consultation types, and select the type with the highest matching degree as the user's consultation type;

[0082] Match the corresponding analysis and solution library according to the user's consultation type; analyze the relevance between several keywords and the problems in the analysis and solution library; set the weight coefficient of the keywords according to the key degree of the keywords, calculate the product sum of the relevance of the corresponding keywords and the weight coefficient to obtain the matching coefficient; screen out the highest solution measure from the matching coefficients as the answer data and send the answer data to the user.

[0083] It should be noted that performing voice interaction based on the user's voice information can timely solve the problems existing in the user and timely answer the user, and through intelligent voice answering, the user does not need to input, improving the efficiency of answering questions.

[0084] Please refer to Figure 4 , retrieve the real-time weather data for a set time period, and fit the real-time temperature, real-time precipitation, and real-time light intensity in the real-time weather data into corresponding weather curves; wherein, the weather curves include: temperature curves, precipitation curves, and illumination intensity curves; obtain the peaks of the weather curves; compare the peaks with the corresponding weather threshold ranges; when the peaks exceed the weather threshold ranges, generate an alarm signal and perform voice broadcast; otherwise, predict abnormal weather according to the weather curves;

[0085] Predict the real-time weather data within a set time period using the method of curve trend analysis to obtain a predicted weather curve; integrate the predicted weather curve to obtain corresponding intersection points; number the intersection points in chronological order and extract the predicted weather data corresponding to the time of the intersection points; among them, the predicted weather data includes: predicted temperature, predicted precipitation, and predicted light intensity;

[0086] Mark the predicted temperature, predicted precipitation, and predicted light intensity in the predicted weather data as YWj, YSj, and YQj respectively; through the formula Calculate the anomaly coefficient of the corresponding intersection point; where β and γ are adjustment coefficients greater than 0; WY, SY, and QY are the temperature threshold, precipitation threshold, and light intensity threshold respectively; DL is the dimension removal coefficient; j is the number of the intersection point, j = 1, 2,..., k; k is a positive integer; when the anomaly coefficient is greater than the coefficient threshold, an early warning signal is generated and sent to the user.

[0087] It should be noted that the greater the difference between the predicted data and the environmental threshold, the greater the anomaly coefficient; which indicates that the environment has a greater impact on the crops. Timely early warning can ensure that users can intervene in the growth environment of the crops in a timely manner, which is beneficial to ensuring the growth trend of the crops and increasing the crop yield.

[0088] Please refer to Figure 5 For the second aspect of the invention, an intelligent voice interaction method for agricultural meteorological consultation is provided, including:

[0089] Step S1: Collect real-time weather data, real-time crop data, and the user's voice information through data sensors;

[0090] Step S2: Analyze the growth period of the crops based on the real-time crop data to obtain crop demand data; analyze the growth environment of the crops based on the real-time weather data to obtain environmental data;

[0091] Step S3: Analyze and obtain farming operation data based on the environmental data and crop demand; recommend the farming operation data to the user;

[0092] Step S4: Classify the user's consultation type based on the user's voice information; analyze and obtain answer data based on the user's consultation type.

[0093] Some of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the real situation; the preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0094] Working principle of the present invention: The present invention collects real-time weather data, real-time crop data, and the voice information of users through data sensors; analyzes the growth period of crops based on the real-time crop data to obtain crop demand data; analyzes the growth environment of crops based on the real-time weather data to obtain environmental data; analyzes and obtains farming operation data based on the environmental data and crop demand; recommends the farming operation data to users; classifies the consultation types of users according to the voice information of users; analyzes and obtains answer data according to the consultation types of users.

[0095] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent voice interaction system for agricultural meteorological consultation, characterized in that: include: Intelligent suggestion module, and the information collection module and intelligent voice module connected to it; Information collection module: used to collect real-time weather data, real-time crop data and user voice information through data sensors; Intelligent suggestion module: used to analyze the growth period of crops according to real-time crop data to obtain crop demand data; analyze the growth environment of crops according to real-time weather data to obtain environmental data; obtain agricultural operation data according to environmental data and crop demand analysis; and recommend agricultural operation data to users; Intelligent voice module: used to classify users’ consultation types according to their voice information; Analyze the user's inquiry type to get answer data.

2. The intelligent voice interaction system for agricultural meteorological consultation according to claim 1, characterized in that: Analyze the crop growth period based on real-time crop data, including: Retrieve real-time crop data; real-time crop data includes: crop variety, sowing time and real-time height of crops; Obtain a time growth table for crops; match the crop varieties and sowing times with the time growth table to obtain the growth stages of the crops; Obtaining the standard range of crop heights at the growth stage; when the real-time height of the crop is within the standard range, the growth stage is used as the crop growth period; otherwise, the crop growth period is re-determined; wherein the crop growth period includes: germination, growth, flowering, fruiting and maturity; Obtain a growth requirement table for the corresponding variety of crops; match the growth period of the crops with the growth requirement table to obtain corresponding crop requirement data; wherein the crop requirement data includes watering amount, fertilizer amount, optimal light intensity and optimal light time.

3. The intelligent voice interaction system for agricultural meteorological consultation according to claim 1, characterized in that: The analysis of the growing environment of crops based on real-time weather data includes: Retrieve real-time weather data within a set time period; integrate the real-time weather data within the set time period in chronological order to obtain an environmental analysis sequence; wherein the real-time weather data includes: real-time temperature, real-time humidity, real-time precipitation, and real-time light intensity; Call the environmental analysis model, input the environmental analysis sequence into the environmental analysis model, obtain the corresponding environmental label, and match the corresponding environmental data according to the environmental label; wherein the environmental label is set to a positive integer; the environmental analysis model is constructed based on the artificial intelligence model.

4. The intelligent voice interaction system for agricultural meteorological consultation according to claim 3 is characterized in that: The environmental analysis model is constructed based on an artificial intelligence model and includes: Acquire a standard data set; wherein the standard data set includes standard input data consistent with the content attributes of the environment analysis sequence, and standard output data consistent with the content attributes of the environment label; Divide the standard data set into a training set, a validation set, and a test set according to a set ratio; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the artificial intelligence model; use the test set to test the artificial intelligence model and obtain test indicators; Obtain an indicator threshold; when the test indicator is greater than the indicator threshold, mark the artificial intelligence model as an environmental analysis model; otherwise, retrain the artificial intelligence model; wherein the artificial intelligence model includes: a convolutional neural network model or a long short-term memory neural network model.

5. The intelligent voice interaction system for agricultural meteorological consultation according to claim 1, characterized in that: The analysis based on environmental data and crop demand data includes: Retrieve environmental data and crop demand data; environmental data includes: total precipitation, sunshine hours, average light intensity, and day-night temperature difference; Get the standard range of temperature difference; when the temperature difference between day and night is greater than the standard range, generate a heat preservation signal and send the heat preservation signal to the user; otherwise, calculate the difference between the total precipitation and the watering amount, and re-analyze the watering amount based on the difference; The light intensity in the environmental data is fitted in time sequence to obtain the light curve F(t); the optimal light intensity and optimal light time in the crop demand data are marked as ZS and ZQ respectively; through the formula Calculate the light coefficient of crops; where α is a proportional coefficient greater than 0, GY is the minimum effective light intensity; i is the set of effective light time start time, and n is the set of effective light time end time; When the illumination coefficient is greater than the first-level threshold, fill-light data will be generated; when the illumination coefficient is less than the second-level threshold, shading data will be generated; the insulation signal, watering amount, fertilization amount, fill-light data and shading data are integrated into agricultural operation data.

6. The intelligent voice interaction system for agricultural meteorological consultation according to claim 1, characterized in that: The method of classifying the user's consultation type according to the user's voice information includes: Retrieve the user's voice information, convert the voice into text using voice recognition technology, and obtain recognized text; use word segmentation technology to divide the recognized text to obtain a number of divided words; analyze the number of divided words according to the stop word library, remove the stop words in the number of divided words, and obtain characteristic words; Obtain key features of several consultation types, analyze the correlation between feature words and key features; extract keywords from feature words based on the correlation; analyze the matching degree between several keywords and consultation types, and select the type with the highest matching degree as the user's consultation type. The present invention converts the user's voice information into recognized text, extracts keywords from the recognized text, and determines the user's consultation type based on the degree of matching between the keywords and the consultation type; it can first make a preliminary division of the user's needs, laying a data foundation for subsequent answers.

7. The intelligent voice interaction system for agricultural meteorological consultation according to claim 6, characterized in that: The answer data obtained by analyzing the user's inquiry type includes: Retrieve the user's inquiry type and match the corresponding analysis and solution library according to the user's inquiry type; analyze the relevance of several keywords and the problems in the analysis and solution library; The weight coefficient of the keyword is set according to the criticality of the keyword, and the product sum of the relevance of the corresponding keyword and the weight coefficient is calculated to obtain the matching coefficient; the highest solution measure is selected from the matching coefficient as the answer data, and the answer data is sent to the user.

8. The intelligent voice interaction system for agricultural meteorological consultation according to claim 1, characterized in that: The intelligent suggestion module is also used to analyze abnormal weather according to real-time weather data, including: Retrieving real-time weather data for a set time period, and fitting the real-time temperature, real-time precipitation and real-time light intensity in the real-time weather data into a corresponding weather curve; wherein the weather curve includes: a temperature curve, a precipitation curve and a light intensity curve; Get the peak value of the weather curve; compare the peak value with the corresponding weather threshold range; when the peak value exceeds the weather threshold range, generate an alarm signal and make a voice broadcast; otherwise, predict abnormal weather based on the weather curve.

9. The intelligent voice interaction system for agricultural meteorological consultation according to claim 8, characterized in that: The forecasting of abnormal weather according to the weather curve includes: Retrieve the weather curve, use the curve trend analysis method to predict the real-time weather data within the set time period, and obtain the predicted weather curve; Integrate the forecast weather curves to obtain corresponding intersections; number the intersections in chronological order, and extract the forecast weather data at the time corresponding to the intersections; wherein the forecast weather data includes: forecast temperature, forecast precipitation, and forecast light intensity; The anomaly coefficient of the corresponding intersection is calculated by mapping the predicted temperature, predicted precipitation and predicted light intensity with the anomaly coefficient of the intersection; when the anomaly coefficient is greater than the coefficient threshold, an early warning signal is generated and sent to the user.

10. An intelligent voice interaction method for agricultural meteorological consultation, applied to an intelligent voice interaction system for agricultural meteorological consultation as claimed in any one of claims 1 to 9, characterized in that: include: Step S1: Collecting real-time weather data, real-time crop data and user's voice information through data sensors; Step S2: Analyze the growth period of crops according to real-time crop data to obtain crop demand data; analyze the growth environment of crops according to real-time weather data to obtain environmental data; Step S3: obtaining agricultural operation data according to environmental data and crop demand analysis; and recommending the agricultural operation data to the user; Step S4: classifying the user's consultation type according to the user's voice information; Analyze the user's inquiry type to get answer data.

Images