Agricultural park intelligent inspection system and method based on digital twinning
Through edge computing and data fusion technology, we ensure the synchronization and consistency of data sources in digital twin agricultural parks, dynamically adjust the inspection strategy, solve the model prediction error problem caused by inconsistent data updates, and realize efficient monitoring and control of agricultural parks Resource optimization has improved crop yield and quality.
Patent Information
- Application Number
- CN202411062065.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-05
AI Technical Summary
The data update time of different data sources in the digital twin agricultural park is inconsistent, which leads to difficulty in data synchronization and model prediction errors, and the inability to accurately identify crop problems, which in turn affects early remediation and resource optimization.
Using edge computing technology to process sensor data in real time, generate a three-dimensional digital twin model through data fusion algorithm, ensure the consistency and synchronization of data source update frequency, dynamically adjust the inspection strategy, and use Bayesian fusion method and wavelet transform to analyze the data update frequency and Synchronization, evaluate the accuracy of model prediction results.
The comprehensive monitoring and refined management of agricultural parks have been achieved, the accuracy of model prediction has been improved, resource allocation has been optimized, crop yield and quality has been improved, manual intervention and operation costs have been reduced, and the reliability and practicality of the system have been enhanced.
Smart Images

Figure CN119006202A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of agricultural park management, and in particular to an intelligent inspection system and method for agricultural parks based on digital twins. Background Art
[0002] With the development of smart agriculture, digital twin technology is gradually being applied to agricultural production to improve the yield and quality of crops. Digital twin technology creates a virtual model to reflect the status of the physical agricultural park in real time, thereby achieving comprehensive monitoring and management of the agricultural production process. However, when the geographical data, crop data, and environmental data collected by sensors are input into the digital twin platform to generate a three-dimensional model of the agricultural park, if the data update time of different data sources is inconsistent, it may make it difficult to synchronize the data, resulting in errors in model predictions. And if there are errors in model predictions, the latest crop problems may not be accurately identified. At the same time, if crop problems are not discovered and handled in a timely manner, it may lead to the deterioration of crop health and the inability to take remedial measures at an early stage, resulting in greater losses. Summary of the invention
[0003] The purpose of the present invention is to provide an agricultural park intelligent inspection system and method based on digital twins to address the shortcomings of the background technology.
[0004] In order to achieve the above object, the present invention provides the following technical solution: an intelligent inspection method for agricultural parks based on digital twins, comprising the following steps:
[0005] S1: Before building the 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform;
[0006] S2: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model’s prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy.
[0007] S3: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of the update frequency of different data sources is analyzed. The sensors are automatically calibrated based on the analysis results.
[0008] S4: Integrate the data between the edge computing device and the central data server, analyze the real-time synchronization of the integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of the integrated data;
[0009] S5: Comprehensively analyze the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of prediction results of the 3D digital twin model;
[0010] S6: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
[0011] Preferably, in S1, the edge computing device includes an edge server, a gateway device, and an intelligent router; the sensor includes a soil moisture sensor, a temperature sensor, a light sensor, and a meteorological sensor.
[0012] Preferably, a three-dimensional digital twin model is generated by a Bayesian fusion method, specifically:
[0013] Collect real-time sensor observation data, and construct a likelihood function based on sensor characteristics and observation data, that is, the probability distribution of data under given observation values; apply Bayesian theorem to calculate the posterior distribution based on the prior distribution and likelihood function. Bayesian formula: Among them, P(θ|D) is the posterior distribution, P(D|θ) is the likelihood function, P(θ) is the prior distribution, and P(D) is the evidence; through the posterior distribution, the optimal estimate of the data is obtained, and a comprehensive processed data set is generated; the fused data is 3D modeled using geographic information system software, and a 3D modeling software is used to generate a 3D model of the agricultural park, reflecting the terrain, crops and environmental information.
[0014] Preferably, in S3, the data update frequency of each sensor is analyzed to ensure that the acquisition frequencies of different data sources are consistent, the update frequency of each sensor is calculated, and the difference between it and the preset frequency is compared, the frequency difference change of the sensor within a fixed time is analyzed, and the overall update frequency deviation index is generated. The method for obtaining the overall update frequency deviation index is:
[0015] Get the data timestamp of each sensor in the s time period, record the time point of each data update, convert the timestamp into time series data, form the data update sequence of each sensor, select the wavelet function to perform discrete wavelet transform on the sensor data update sequence, extract the multi-scale frequency components, and the transformation expression is: Among them, W j (k) is the wavelet coefficient of the jth layer, x(n) is the original time series, ψ j,k (n) is a wavelet function with a scale of j and a translation of k. The main frequency component is determined by energy density calculation. The specific calculation expression is: E j =∑ k |W j (k)| 2 ; Among them, E j is the wavelet energy of the jth layer. According to the main frequency component, the actual update frequency of each sensor is calculated. Where factual is the actual update frequency, Δd is the time interval corresponding to the main frequency component, the actual update frequency of each sensor is compared with the preset frequency, and the frequency deviation is calculated, Δfi = |factual(i)-fpreset|; where Δfi is the frequency deviation of the i-th sensor, factual(i) is the actual update frequency of the i-th sensor, and fpreset is the preset update frequency; the frequency deviations of all sensors are calculated, and the sum of them is calculated to obtain the overall update frequency deviation index.
[0016] Preferably, in S4, a data synchronization real-time index is generated according to the synchronization degree of different data sources in terms of timestamp and geographic location, and the method for obtaining the data synchronization real-time index is:
[0017] Collect the data timestamp sequence of each sensor to form multiple time series T1, T2, …, TN, organize the data timestamp sequence of each sensor into a unified format to ensure the integrity of the time series; for two time series Ti = {ti1, ti2, …, tim} and Tj = {tj1, tj2, …, tjq}, construct a distance matrix D, where D[k, h] represents the distance between timestamps tik and tjh, and D[k, h] = |tik-tjh|, where tik and tjh are the timestamps in sequences Ti and Tj respectively;
[0018] Construct a cumulative distance matrix C, where C[k,h] represents the minimum cumulative distance from the starting point to the position (k,h). The specific calculation expression is: C[k,h]=D[k,h]+min(C[k-1,h],C[k,h-1],C[k-1,h-1]); where the initial condition is C[1,1]=D[1,1]; by backtracking the cumulative distance matrix, find the optimal alignment path W to minimize the total alignment distance of the two time series, where W={(k1,h1),(k2,h2),…,(kp,hp)}; calculate the average alignment distance on the optimal alignment path as the DTW distance between the two time series. The specific calculation expression is: p is the total number of paths;
[0019] Perform DTW calculation on all sensor time series pairs and obtain the average DTW distance of all time series pairs. N is the total number of sensors; the difference between the actual position and the expected position of each data point is calculated, and the data synchronization real-time index is calculated. The specific calculation expression is: DS = αDTW-+βΔs; where α and β are the weight coefficients of time and space synchronization, and the value range of α and β is (0,1); Δs is the difference between the actual position and the expected position of each data point, and DS is the data synchronization real-time index.
[0020] Preferably, the acquired data synchronization real-time index is compared with a preset data synchronization real-time index reference threshold. If the data synchronization real-time index is greater than or equal to the preset data synchronization real-time index reference threshold, the synchronization real-time of different data sources in timestamp and geolocation is high, and the integrated data synchronization is in a normal state, and a normal synchronization signal is generated; if the data synchronization real-time index is less than the preset data synchronization real-time index reference threshold, the synchronization real-time of different data sources in timestamp and geolocation is low, and the integrated data synchronization is in an abnormal state, and a synchronization abnormal signal is generated.
[0021] Preferably, in S5, the overall update frequency deviation index and the data synchronization real-time index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses the accuracy value label of the prediction result of the three-dimensional digital twin model predicted by each group of first eigenvectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy value labels of the prediction results of all three-dimensional digital twin models as the training target. The machine learning model is trained until the sum of the prediction errors converges, and the model training is stopped. The accuracy value of the prediction result of the three-dimensional digital twin model is determined according to the model output results.
[0022] Preferably, the accuracy values of all three-dimensional digital twin model prediction results obtained within a fixed time period are collected, and the corresponding data combination is established, the mean and standard deviation of the data set are calculated, the mean and standard deviation of the accuracy values of the three-dimensional digital twin model prediction results are analyzed, and the accuracy of the model prediction results is divided into accurate predictions and inaccurate predictions according to the analysis results.
[0023] Preferably, if the mean of the accuracy values in the data set is greater than or equal to the reference threshold of the mean of the accuracy values, and the standard deviation of the accuracy values is less than the reference threshold of the standard deviation of the accuracy values, the accuracy of the prediction result is high and stable, and it is classified as an accuracy prediction. At this time, no warning signal is generated, and there is no need to frequently adjust the inspection strategy;
[0024] If the mean of the accuracy values is greater than or equal to the reference threshold of the mean of the accuracy values, and the standard deviation of the accuracy values is greater than or equal to the reference threshold of the standard deviation of the accuracy values, the prediction result is accurate but unstable, and it is classified as an inaccurate prediction. At this time, a third-level warning signal is generated, and the inspection strategy needs to be adjusted to pay attention to the unstable prediction area;
[0025] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low and unstable, and it is classified as an inaccurate prediction. At this time, a first-level warning signal is generated, and the model needs to be re-evaluated and corrected, and the inspection strategy needs to be significantly adjusted, with a focus on inaccurate areas;
[0026] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low but stable, and it is classified as an accuracy prediction. At this time, a secondary warning signal is generated, and the model needs to be adjusted and optimized, and the overall inspection strategy needs to be improved to improve the prediction accuracy.
[0027] The present invention also provides an intelligent inspection system for agricultural parks based on digital twins, including an edge computing module, a data fusion module, an update frequency analysis module, a data synchronization analysis module, an accuracy assessment module, and an inspection strategy adjustment module;
[0028] Edge computing module: Before building a 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform;
[0029] Data fusion module: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model's prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy.
[0030] Update frequency analysis module: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of update frequency of different data sources is analyzed. The sensors are automatically calibrated according to the analysis results.
[0031] Data synchronization analysis module: Integrate data between edge computing devices and central data servers, analyze the real-time synchronization of integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of integrated data;
[0032] Accuracy evaluation module: Comprehensively analyzes the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of the prediction results of the 3D digital twin model;
[0033] Inspection strategy adjustment module: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
[0034] In the above technical solution, the technical effects and advantages provided by the present invention are:
[0035] 1. The present invention realizes comprehensive monitoring and refined management of agricultural parks through edge computing technology, data fusion algorithm and dynamic adjustment of inspection strategy. Specifically, the system can process and analyze sensor data in real time, generate a three-dimensional digital twin model, and formulate an optimized agricultural management plan based on the model prediction results. By ensuring the synchronization and consistency of data sources, the accuracy of model prediction is effectively improved, thereby optimizing resource allocation and improving crop yield and quality.
[0036] 2. The present invention can timely identify and handle abnormal situations in agricultural parks by automatically calibrating sensors and comprehensively analyzing data update frequency and synchronization, dynamically adjusting inspection strategies. Such a system not only improves the efficiency and benefits of agricultural production, but also reduces the need for manual intervention and reduces operating costs. At the same time, the early warning mechanism and prediction result classification further enhance the reliability and practicality of the system, making agricultural management more intelligent and scientific, thereby bringing significant economic and social benefits to agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0038] Figure 1 The present invention is a flow chart of the method.
[0039] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION
[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0041] Example 1, please refer to Figure 1 As shown, the agricultural park intelligent inspection method based on digital twins described in this embodiment includes the following steps:
[0042] S1: Before building the 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform;
[0043] S2: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model’s prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy.
[0044] S3: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of the update frequency of different data sources is analyzed. The sensors are automatically calibrated based on the analysis results.
[0045] S4: Integrate the data between the edge computing device and the central data server, analyze the real-time synchronization of the integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of the integrated data;
[0046] S5: Comprehensively analyze the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of prediction results of the 3D digital twin model;
[0047] S6: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
[0048] Among them, in S1, before building the 3D model, the sensor data is processed in real time and preliminary analysis is performed through edge computing technology, and the processed data is input into the digital twin platform;
[0049] According to the specific needs of the agricultural park, select appropriate edge computing devices, such as edge servers, gateway devices, or smart routers. Install edge computing devices at key locations in the agricultural park to ensure that the signal range of all sensors can be covered.
[0050] Deploy various sensors, including soil moisture sensors, temperature sensors, light sensors, meteorological sensors, etc. Set the data collection frequency of the sensors to ensure real-time data acquisition.
[0051] Perform preliminary data filtering and cleaning on edge computing devices to remove noise and outliers to ensure data quality. Use simple statistical methods (such as mean, median) or rules (such as threshold filtering) to handle abnormal data.
[0052] Convert sensor data into a unified format to facilitate subsequent data processing and transmission. Standardize data structure to ensure compatibility of data from different sources.
[0053] Use the computing power of edge computing devices to process the collected data in real time. Perform simple data analysis, such as calculating the average, maximum, and minimum values, to gain a preliminary understanding of data characteristics.
[0054] The data is preliminarily analyzed and judged through the set algorithms and rules. For example, the changing trend of soil moisture is monitored to determine whether irrigation is needed. If an abnormal situation is detected (such as low soil moisture, abnormal temperature, etc.), an alarm or corresponding treatment measures are immediately triggered.
[0055] Compress the data after real-time processing to reduce the bandwidth requirements for data transmission. Use data compression algorithms, such as lossless compression or lossy compression, and choose the appropriate method based on the application scenario.
[0056] The processed data is transmitted from the edge computing device to the digital twin platform through wireless communication technologies (such as LoRa, NB-IoT, Wi-Fi, 5G, etc.), ensuring the reliability and real-time performance of data transmission and avoiding data loss and transmission delay.
[0057] S2: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model’s prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy.
[0058] Collect real-time data from sensors such as soil moisture, temperature, weather, light, etc. and perform pre-processing to clean and standardize the data. Obtain high-resolution drone and satellite imagery to provide a comprehensive view of the geography and crop status.
[0059] The pre-processed sensor data, UAV and satellite images are processed by applying data fusion algorithms. Commonly used data fusion algorithms include Kalman filtering, Bayesian fusion method and deep learning method.
[0060] The three-dimensional digital twin model is generated by the Bayesian fusion method, specifically:
[0061] Define the prior distribution based on historical data and expert knowledge. For example, define the prior distribution based on soil moisture, temperature and other data accumulated over many years.
[0062] Collect real-time sensor observation data, and construct a likelihood function based on sensor characteristics and observation data to represent the probability distribution of the data given the observation value.
[0063] Apply Bayes' theorem to calculate the posterior distribution based on the prior distribution and likelihood function. Bayes' formula: Among them, P(θ|D) is the posterior distribution, P(D|θ) is the likelihood function, P(θ) is the prior distribution, and P(D) is the evidence.
[0064] Through the posterior distribution, the optimal estimate of the data is obtained to generate a comprehensive processed data set. Use geographic information system (GIS) software (such as ArcGIS) to perform 3D modeling on the fused data. Use 3D modeling software (such as AutoCAD, 3ds Max) to generate a 3D model of the agricultural park, reflecting the terrain, crops and environmental information.
[0065] Based on the prediction results of the model, an optimized agricultural management plan is generated, including irrigation plans, fertilization plans, pest and disease control measures, etc. Irrigation management: Generate accurate irrigation plans based on soil moisture and weather forecasts to avoid water waste. Fertilization management: Analyze soil nutrient content and crop needs, formulate scientific fertilization plans, and improve crop yield and quality. Pest and disease control: Use pest and disease prediction results to take preventive measures in advance to reduce pesticide use and environmental impact.
[0066] Based on the 3D model and prediction results, identify areas that require key inspections, such as plots with poor crop health, areas with high risk of pests and diseases, etc. Determine the inspection frequency: Determine the inspection frequency and develop a detailed inspection plan based on the crop growth cycle, environmental changes and prediction results.
[0067] Use 3D models to plan the path of drone inspections to ensure coverage of all key areas. Use path optimization algorithms (such as A* algorithm, Dijkstra algorithm) to generate efficient inspection routes. During the inspection process, monitor the inspection status of drones in real time, and dynamically adjust the inspection strategy based on real-time data to ensure timely detection and handling of abnormalities. According to the 3D model and inspection requirements, optimize the layout of ground sensors to ensure the comprehensiveness and accuracy of data collection. Use automated inspection equipment (such as automated spraying systems, ground robots) for inspections, and cooperate with drones to achieve all-round monitoring.
[0068] S3: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of the update frequency of different data sources is analyzed. The sensors are automatically calibrated based on the analysis results.
[0069] According to the three-dimensional digital twin model, the areas that need to be monitored in the agricultural park are determined, such as key areas for crop growth, areas with high incidence of pests and diseases, etc. Different types of sensors are deployed in the target area, including soil moisture sensors, meteorological sensors, environmental sensors, etc.
[0070] Sensor types include: Soil sensors: used to monitor soil moisture, temperature and nutrient content. Meteorological sensors: used to monitor meteorological conditions such as temperature, humidity, wind speed, wind direction, rainfall and solar radiation. Environmental sensors: used to monitor environmental factors such as light intensity, carbon dioxide concentration and air quality.
[0071] Analyze the data update frequency of each sensor to ensure that the collection frequency of different data sources is consistent, calculate the update frequency of each sensor, and compare the difference between it and the preset frequency, analyze the frequency difference change of the sensor within a fixed time, and generate the overall update frequency deviation index. The method for obtaining the overall update frequency deviation index is:
[0072] Get the data timestamp of each sensor in the s time period, record the time point of each data update, convert the timestamp into time series data, form the data update sequence of each sensor, select the appropriate wavelet function (such as Haar wavelet, Daubechies wavelet), perform discrete wavelet transform on the sensor data update sequence, extract the multi-scale frequency components, and the transformation expression is: Among them, W j (k) is the wavelet coefficient of the jth layer, x(n) is the original time series, ψ j,k (n) is a wavelet function with a scale of j and a translation of k; the main frequency component is determined by analyzing the wavelet coefficients, and the main frequency component is determined by energy density calculation. The specific calculation expression is: E j =∑ k |W j (k)| 2 ; Among them, E j is the wavelet energy of the jth layer. According to the main frequency components, the actual update frequency of each sensor is calculated. Where factual is the actual update frequency, Δd is the time interval corresponding to the main frequency component, the actual update frequency of each sensor is compared with the preset frequency, and the frequency deviation is calculated, Δfi = |factual(i)-fpreset|; where Δfi is the frequency deviation of the i-th sensor, factual(i) is the actual update frequency of the i-th sensor, and fpreset is the preset update frequency; the frequency deviations of all sensors are calculated, and the sum of them is calculated to obtain the overall update frequency deviation index.
[0073] The larger the overall update frequency deviation index, the worse the consistency of update frequency of different data sources. This means that there are large differences in the time intervals of data collection for each sensor, which makes it difficult to align the data in time and affects the effect of data fusion. Inconsistent updates of sensors with large frequency deviations will lead to overall data instability, which in turn affects the accuracy and real-time performance of the digital twin model.
[0074] Frequency inconsistency may be caused by a variety of factors, including sensor hardware failure, environmental interference, or communication delays. These factors can cause the sensor to fail to collect data at the preset frequency, resulting in deviations. The larger the deviation index, the more serious these problems are and the more difficult it is to synchronize data.
[0075] In order to improve the consistency of data source update frequency, sensors need to be calibrated and maintained regularly, and advanced algorithms (such as adaptive sliding windows or Bayesian updates) are used to dynamically adjust and optimize the data collection frequency. This will help reduce the overall update frequency deviation index and ensure the accuracy of data fusion and model prediction.
[0076] S4: Integrate the data between the edge computing device and the central data server, analyze the real-time synchronization of the integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of the integrated data.
[0077] Data integration between edge computing devices and central data servers means aggregating and transmitting data scattered on edge computing devices to the central data server. This process includes standardizing data from different sources and in different formats to ensure consistency and integrity during transmission. Real-time data transmission is achieved through wireless communication technologies (such as LoRa, NB-IoT, Wi-Fi, 5G, etc.), so that data can be aggregated to the central server in a timely manner for further analysis and processing.
[0078] Analyzing the synchronization and real-time performance of integrated data in time and space refers to evaluating the synchronization and real-time performance of data in time and geographic location. The synchronization check in time ensures that the timestamps of different data sources are consistent, solving the data deviation caused by time asynchrony; the synchronization check in space ensures the accurate positioning of data in geographic space, ensuring that data from different sources can correctly correspond to the actual geographic location. Through these analyses, potential problems in data synchronization can be identified and corrected, improving the accuracy and real-time performance of data, thereby enhancing the reliability and effectiveness of the digital twin model.
[0079] The data synchronization real-time index is generated according to the synchronization degree of different data sources in timestamp and geographic location. The method for obtaining the data synchronization real-time index is:
[0080] Collect the data timestamp sequence of each sensor to form multiple time series T1, T2, …, TN, and organize the data timestamp sequence of each sensor into a unified format to ensure the integrity of the time series; for two time series Ti = {ti1, ti2, …, tim} and Tj = {tj1, tj2, …, tjn}, construct a distance matrix D, where D[k, h] represents the distance between timestamps tik and tjh, and D[k, h] = |tik-tjh|, where tik and tjh are the timestamps in sequences Ti and Tj respectively;
[0081] Construct a cumulative distance matrix C, where C[k,h] represents the minimum cumulative distance from the starting point to the position (k,h). The specific calculation expression is: C[k,h]=D[k,h]+min(C[k-1,h],C[k,h-1],C[k-1,h-1]); where the initial condition is C[1,1]=D[1,1]; by backtracking the cumulative distance matrix, find the optimal alignment path W to minimize the total alignment distance of the two time series, where W={(k1,h1),(k2,h2),…,(kp,hp)}; calculate the average alignment distance on the optimal alignment path as the DTW distance between the two time series. The specific calculation expression is: p is the total number of paths;
[0082] Perform DTW calculation on all sensor time series pairs and obtain the average DTW distance of all time series pairs. N is the total number of sensors; the difference between the actual position and the expected position of each data point is calculated, and the data synchronization real-time index is calculated. The specific calculation expression is: DS = αDTW-+βΔs; where α and β are the weight coefficients of time and space synchronization, and the value range of α and β is (0,1); Δs is the difference between the actual position and the expected position of each data point, and DS is the data synchronization real-time index.
[0083] The acquired data synchronization real-time index is compared with the preset data synchronization real-time index reference threshold. If the data synchronization real-time index is greater than or equal to the preset data synchronization real-time index reference threshold, it means that the synchronization real-time of different data sources in timestamp and geographic positioning is high, and the integrated data synchronization is in a normal state. At this time, a normal synchronization signal is generated; if the data synchronization real-time index is less than the preset data synchronization real-time index reference threshold, it means that the synchronization real-time of different data sources in timestamp and geographic positioning is low, and the integrated data synchronization is in an abnormal state. At this time, a synchronization abnormality signal is generated.
[0084] S5: Comprehensively analyze the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of the prediction results of the 3D digital twin model.
[0085] The overall update frequency deviation index and the data synchronization real-time index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses the accuracy value label of the prediction result of the three-dimensional digital twin model predicted by each group of first eigenvectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy value labels of the prediction results of all three-dimensional digital twin models as the training target. The machine learning model is trained until the sum of the prediction errors converges, then the model training is stopped, and the accuracy value of the prediction result of the three-dimensional digital twin model is determined according to the model output results.
[0086] The method for obtaining the accuracy value of the prediction result of the three-dimensional digital twin model is as follows: from the first eigenvector training data of the trained machine learning model, obtain the corresponding function expression: MT=f1(KP, DS); where f1 is the output function of the model, KP is the overall update frequency deviation index, DS is the data synchronization real-time index, and MT is the accuracy value of the prediction result of the three-dimensional digital twin model.
[0087] S6: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
[0088] The accuracy values of all three-dimensional digital twin model prediction results obtained within a fixed time period are collected, and the corresponding data combination is established to calculate the mean and standard deviation of the data set. The mean and standard deviation of the accuracy values of the three-dimensional digital twin model prediction results are analyzed, and the accuracy of the model prediction results is divided into accurate predictions and inaccurate predictions according to the analysis results.
[0089] If the mean accuracy value in the data set is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the prediction result is highly accurate and stable, and it is classified as an accuracy prediction. At this time, no warning signal is generated, and there is no need to frequently adjust the inspection strategy;
[0090] If the mean of the accuracy values is greater than or equal to the reference threshold of the mean of the accuracy values, and the standard deviation of the accuracy values is greater than or equal to the reference threshold of the standard deviation of the accuracy values, the prediction result is accurate but unstable, and it is classified as an inaccurate prediction. At this time, a third-level warning signal is generated, and the inspection strategy needs to be adjusted to pay attention to the unstable prediction area;
[0091] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low and unstable, and it is classified as an inaccurate prediction. At this time, a first-level warning signal is generated, and the model needs to be re-evaluated and corrected, and the inspection strategy needs to be significantly adjusted, with a focus on inaccurate areas;
[0092] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low but stable, and it is classified as an accuracy prediction. At this time, a secondary warning signal is generated, and the model needs to be adjusted and optimized, and the overall inspection strategy needs to be improved to improve the prediction accuracy.
[0093] It should be noted here that the management level of the first-level warning signal is higher than that of the second-level warning signal, and the management level of the second-level warning signal is higher than that of the third-level warning signal. Relevant personnel can dynamically adjust the agricultural park inspection strategy according to the level of the warning signal.
[0094] According to the division results, the inspection strategy of the agricultural park is dynamically adjusted: for high-accuracy and low-volatility prediction results, the existing inspection frequency and key areas are maintained.
[0095] For forecast results with high accuracy but large fluctuations, increase the inspection frequency and focus on areas with large fluctuations.
[0096] For low accuracy and high volatility forecast results, recalibrate and optimize the model, and significantly increase the inspection frequency to cover more areas.
[0097] For prediction results with low accuracy but small fluctuations, optimize the model and adjust the inspection strategy to improve prediction accuracy.
[0098] In this embodiment, before building a three-dimensional model, the sensor data is processed in real time and preliminarily analyzed by edge computing technology, and the processed data is input into the digital twin platform. The digital twin platform generates a three-dimensional digital twin model using a data fusion algorithm, and preliminarily generates an agricultural management plan based on the prediction results of the model, and determines the park inspection strategy; according to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of the update frequency of different data sources is analyzed, and the sensor is automatically calibrated according to the analysis results; data is integrated between the edge computing device and the central data server, and the synchronization and real-time performance of the integrated data in time and space is analyzed to evaluate whether the synchronization of the integrated data is abnormal; the consistency of the update frequency of different data sources and the synchronization and real-time performance of the integrated data in time and space are comprehensively analyzed to evaluate the accuracy of the prediction results of the three-dimensional digital twin model; according to the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accuracy prediction and inaccuracy prediction, and the park inspection strategy is dynamically adjusted according to the division results. It can more effectively identify and solve problems in agricultural parks, optimize resource allocation, improve crop yield and quality, and ultimately realize the intelligent and refined agricultural management.
[0099] Example 2, please refer to Figure 2 As shown, the agricultural park intelligent inspection system based on digital twins in this embodiment includes an edge computing module, a data fusion module, an update frequency analysis module, a data synchronization analysis module, an accuracy assessment module, and an inspection strategy adjustment module;
[0100] Edge computing module: Before building a 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform;
[0101] Data fusion module: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model's prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy.
[0102] Update frequency analysis module: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of update frequency of different data sources is analyzed. The sensors are automatically calibrated according to the analysis results.
[0103] Data synchronization analysis module: Integrate data between edge computing devices and central data servers, analyze the real-time synchronization of integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of integrated data;
[0104] Accuracy evaluation module: Comprehensively analyzes the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of the prediction results of the 3D digital twin model;
[0105] Inspection strategy adjustment module: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
[0106] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.
[0107] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0108] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.
[0109] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0110] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0111] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. The intelligent inspection method of agricultural parks based on digital twins is characterized by: The steps include: S1: Before building the 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform; S2: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model’s prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy. S3: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of the update frequency of different data sources is analyzed. The sensors are automatically calibrated based on the analysis results. S4: Integrate the data between the edge computing device and the central data server, analyze the real-time synchronization of the integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of the integrated data; S5: Comprehensively analyze the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of prediction results of the 3D digital twin model; S6: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
2. The agricultural park intelligent inspection method based on digital twin according to claim 1 is characterized in that: In S1, edge computing devices include edge servers, gateway devices, and smart routers; sensors include soil moisture sensors, temperature sensors, light sensors, and meteorological sensors.
3. The agricultural park intelligent inspection method based on digital twin according to claim 1 is characterized in that: The three-dimensional digital twin model is generated by the Bayesian fusion method, specifically: Collect real-time sensor observation data, and construct a likelihood function based on sensor characteristics and observation data, that is, the probability distribution of data under given observation values; apply Bayesian theorem to calculate the posterior distribution based on the prior distribution and likelihood function. Bayesian formula: Among them, P(θ|D) is the posterior distribution, P(D|θ) is the likelihood function, P(θ) is the prior distribution, and P(D) is the evidence; through the posterior distribution, the optimal estimate of the data is obtained, and a comprehensive processed data set is generated; the fused data is 3D modeled using geographic information system software, and a 3D modeling software is used to generate a 3D model of the agricultural park, reflecting the terrain, crops and environmental information.
4. The agricultural park intelligent inspection method based on digital twin according to claim 1 is characterized in that: In S3, the data update frequency of each sensor is analyzed to ensure that the acquisition frequencies of different data sources are consistent. The update frequency of each sensor is calculated and compared with the preset frequency. The frequency difference change of the sensor within a fixed time is analyzed and the overall update frequency deviation index is generated. The overall update frequency deviation index is obtained as follows: Get the data timestamp of each sensor in the s time period, record the time point of each data update, convert the timestamp into time series data, form the data update sequence of each sensor, select the wavelet function to perform discrete wavelet transform on the sensor data update sequence, extract the multi-scale frequency components, and the transformation expression is: Among them, W j (k) is the wavelet coefficient of the jth layer, x(n) is the original time series, ψ j,k (n) is a wavelet function with a scale of j and a translation of k. The main frequency component is determined by energy density calculation. The specific calculation expression is: E j =∑ k |W j (k)| 2 ; Among them, E j is the wavelet energy of the jth layer. According to the main frequency component, the actual update frequency of each sensor is calculated. Where factual is the actual update frequency, Δd is the time interval corresponding to the main frequency component, the actual update frequency of each sensor is compared with the preset frequency, and the frequency deviation is calculated, Δfi = |factual(i)-fpreset|; where Δfi is the frequency deviation of the i-th sensor, factual(i) is the actual update frequency of the i-th sensor, and fpreset is the preset update frequency; the frequency deviations of all sensors are calculated, and the sum of them is calculated to obtain the overall update frequency deviation index.
5. The method for intelligent inspection of agricultural parks based on digital twins according to claim 4 is characterized in that: In S4, a data synchronization real-time index is generated according to the synchronization degree of different data sources in timestamp and geographic location. The method for obtaining the data synchronization real-time index is: Collect the data timestamp sequence of each sensor to form multiple time series T1, T2, …, TN, organize the data timestamp sequence of each sensor into a unified format to ensure the integrity of the time series; for two time series Ti = {ti1, ti2, …, tim} and Tj = {tj1, tj2, …, tjq}, construct a distance matrix D, where D[k, h] represents the distance between timestamps tik and tjh, and D[k, h] = |tik-tjh|, where tik and tjh are the timestamps in sequences Ti and Tj respectively; Construct a cumulative distance matrix C, where C[k,h] represents the minimum cumulative distance from the starting point to the position (k,h). The specific calculation expression is: C[k,h]=D[k,h]+min(C[k-1,h],C[k,h-1],C[k-1,h-1]); where the initial condition is C[1,1]=D[1,1]; by backtracking the cumulative distance matrix, find the optimal alignment path W to minimize the total alignment distance of the two time series, where W={(k1,h1),(k2,h2),…,(kp,hp)}; calculate the average alignment distance on the optimal alignment path as the DTW distance between the two time series. The specific calculation expression is: p is the total number of paths; Perform DTW calculation on all sensor time series pairs and obtain the average DTW distance of all time series pairs. N is the total number of sensors; the difference between the actual position and the expected position of each data point is calculated, and the data synchronization real-time index is calculated. The specific calculation expression is: DS = αDTW-+βΔs; where α and β are the weight coefficients of time and space synchronization, and the value range of α and β is (0,1); Δs is the difference between the actual position and the expected position of each data point, and DS is the data synchronization real-time index.
6. The method for intelligent inspection of agricultural parks based on digital twins according to claim 5 is characterized in that: The acquired data synchronization real-time index is compared with the preset data synchronization real-time index reference threshold. If the data synchronization real-time index is greater than or equal to the preset data synchronization real-time index reference threshold, the synchronization real-time of different data sources in timestamp and geolocation is high, and the integrated data synchronization is in a normal state. At this time, a normal synchronization signal is generated; if the data synchronization real-time index is less than the preset data synchronization real-time index reference threshold, the synchronization real-time of different data sources in timestamp and geolocation is low, and the integrated data synchronization is in an abnormal state. At this time, a synchronization abnormality signal is generated.
7. The method for intelligent inspection of agricultural parks based on digital twins according to claim 6 is characterized in that: In S5, the overall update frequency deviation index and the data synchronization real-time index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses the accuracy value label of the prediction result of the three-dimensional digital twin model predicted by each group of first eigenvectors as the prediction target, and minimizes the sum of the prediction errors of the accuracy value labels of the prediction results of all three-dimensional digital twin models as the training target. The machine learning model is trained until the sum of the prediction errors converges, and the model training is stopped. The accuracy value of the prediction result of the three-dimensional digital twin model is determined according to the model output results.
8. The method for intelligent inspection of agricultural parks based on digital twins according to claim 1 is characterized in that: The accuracy values of all three-dimensional digital twin model prediction results obtained within a fixed time period are collected, and the corresponding data combination is established to calculate the mean and standard deviation of the data set. The mean and standard deviation of the accuracy values of the three-dimensional digital twin model prediction results are analyzed, and the accuracy of the model prediction results is divided into accurate predictions and inaccurate predictions according to the analysis results.
9. The method for intelligent inspection of agricultural parks based on digital twins according to claim 8 is characterized in that: If the mean accuracy value in the data set is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the prediction result is highly accurate and stable, and it is classified as an accuracy prediction. At this time, no warning signal is generated, and there is no need to frequently adjust the inspection strategy; If the mean of the accuracy values is greater than or equal to the reference threshold of the mean of the accuracy values, and the standard deviation of the accuracy values is greater than or equal to the reference threshold of the standard deviation of the accuracy values, the prediction result is accurate but unstable, and it is classified as an inaccurate prediction. At this time, a third-level warning signal is generated, and the inspection strategy needs to be adjusted to pay attention to the unstable prediction area; If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low and unstable, and it is classified as an inaccurate prediction. At this time, a first-level warning signal is generated, and the model needs to be re-evaluated and corrected, and the inspection strategy needs to be significantly adjusted, with a focus on inaccurate areas; If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the accuracy of the prediction result is low but stable, and it is classified as an accuracy prediction. At this time, a secondary warning signal is generated, and the model needs to be adjusted and optimized, and the overall inspection strategy needs to be improved to improve the prediction accuracy.
10. An intelligent inspection system for agricultural parks based on digital twins, used to implement the intelligent inspection method for agricultural parks based on digital twins according to any one of claims 1 to 9, characterized in that: It includes edge computing module, data fusion module, update frequency analysis module, data synchronization analysis module, accuracy assessment module and inspection strategy adjustment module; Edge computing module: Before building a 3D model, edge computing technology is used to process sensor data in real time and conduct preliminary analysis, and the processed data is input into the digital twin platform; Data fusion module: The digital twin platform uses data fusion algorithms to generate a three-dimensional digital twin model. Based on the model's prediction results, it preliminarily generates an agricultural management plan and determines the park inspection strategy. Update frequency analysis module: According to the park inspection strategy, different types of sensors are used to collect soil, meteorological and environmental data in the park in real time, and the consistency of update frequency of different data sources is analyzed. The sensors are automatically calibrated according to the analysis results. Data synchronization analysis module: Integrate data between edge computing devices and central data servers, analyze the real-time synchronization of integrated data in time and space, and evaluate whether there are any abnormalities in the synchronization of integrated data; Accuracy evaluation module: Comprehensively analyzes the consistency of update frequencies of different data sources and the synchronization and real-time performance of integrated data in time and space to evaluate the accuracy of the prediction results of the 3D digital twin model; Inspection strategy adjustment module: Based on the evaluation results, the prediction results of the three-dimensional digital twin model are divided into accurate predictions and inaccurate predictions, and the park inspection strategy is dynamically adjusted based on the division results.
Citation Information
Patent Citations
Vibration data wireless synchronous acquisition method and system and vibration monitoring system
CN115038161A
Park monitoring method based on digital twinning and related device
CN116129366A
Smart park control system based on digital twinning
CN116468287A
Orchard monitoring system based on digital twinning
CN117456363A
Aero-engine blisk digital twin data consistency verification method
CN117745767A
Cited By
Inspection method and system combined with electric power operation scene characteristics
CN119179992A
Intelligent inspection method and system for power grid regulation and control based on digital twinning
CN119891518A
Intelligent agricultural data analysis method and system based on machine learning
CN120123703A
A Machine Learning-Based Smart Agriculture Data Analysis Method and System
CN120123703B
Agricultural ecological environment monitoring method and system based on digital twinborn
CN120338979A
