A wireless multi-sensing data monitoring terminal system for smart rice and fish

Abstract

The application discloses a kind of wireless multi-sense data monitoring terminal systems for wisdom rice fish, it is related to data monitoring technical field, including data setting module, set initial monitoring frequency;Environment monitoring module, real-time monitoring rice fish growth condition, soil data and aquatic animal state, timely capture rice fish growth environment change;Data processing module, according to the feature data related to rice fish growth environment change and breeding object anomaly of environment monitoring module extraction, quantize rice fish growth environment change degree;Early warning module, predict growth environment change by machine learning model, when predicting rice fish growth environment change appears anomaly, automatically send early warning signal;Dynamic adjustment module, the monitoring frequency of dynamic adjustment monitoring equipment, to cope with rice fish growth environment change.The application can significantly improve the robustness, timeliness and decision accuracy of the system by adjusting the monitoring frequency to dynamically adapt to the rate of environmental change.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data monitoring, more particularly, the present application relates to a wireless multi-sensing data monitoring terminal system for smart rice-fish. BACKGROUND

[0002] Smart rice-fish integrated breeding is an ecological agricultural mode that combines rice planting with aquaculture (such as fish, shrimp, crab, etc.), and realizes "one water for two uses and one field for double harvest" through resource recycling. This mode has very high requirements for real-time monitoring and accurate regulation of environmental parameters (such as water quality, soil, climate, biological behavior, etc.).

[0003] Disadvantages of the prior art:

[0004] In the prior art, when the environment changes beyond the normal frequency (such as sudden changes in water quality, sudden rise in air temperature, abnormal fluctuations in dissolved oxygen), the key parameters of rice-fish growth (such as dissolved oxygen, pH value, ammonia nitrogen concentration) may change dramatically in a short time; high-frequency monitoring (such as increasing from 1 time / hour to 1 time / minute) can capture minute-level or even second-level fluctuations, while low-frequency monitoring will miss some changes and cause data gaps, and it may not be until the next sampling that the abnormality is found, by which time the ecological damage may have been irreversible. Therefore, when the environment changes beyond the normal frequency, it is necessary to increase the data monitoring frequency to provide higher density data to support decision-making.

[0005] To solve the above problems, the present application provides a solution. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a wireless multi-sensing data monitoring terminal system for smart rice-fish, which adjusts the monitoring frequency to solve the problems raised in the background art.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0008] A wireless multi-sensing data monitoring terminal system for smart rice-fish, comprising a data setting module, an environment monitoring module, a data processing module, a warning module, a dynamic adjustment module, and a connection between the modules:

[0009] The data setting module is used to set the initial monitoring frequency according to the historical breeding experience of rice-fish and the growth environment of rice-fish;

[0010] The environment monitoring module monitors the growth of rice-fish, soil data and the state of aquatic animals in real time, and captures the changes in the growth environment of rice-fish in time;

[0011] The data processing module extracts characteristic data related to changes in the rice-fish growth environment and abnormalities in the aquaculture objects from the environmental monitoring module, and analyzes the characteristic data to quantify the degree of change in the rice-fish growth environment.

[0012] The early warning module inputs the analyzed feature data into a pre-trained machine learning model, which then predicts changes in the growth environment. When an abnormality is predicted in the rice-fish growth environment, an early warning signal is automatically issued.

[0013] The dynamic adjustment module dynamically adjusts the monitoring frequency of the monitoring equipment based on real-time feedback of prediction results and early warning signals in order to cope with changes in the rice-fish growth environment.

[0014] In a preferred embodiment, the process of setting the initial monitoring frequency is as follows:

[0015] Data on growth indicators of rice and fish at different growth stages were collected, and management operations during the aquaculture process were recorded to obtain data on the growth of rice and fish.

[0016] Organize and preprocess historical meteorological and soil data;

[0017] The fluctuation range and trend of various environmental parameters under different categories were analyzed based on the preprocessed historical meteorological data and historical soil data to obtain the fluctuation characteristics of environmental parameters.

[0018] An initial monitoring frequency is set based on the fluctuation characteristics of environmental parameters and the growth of rice and fish.

[0019] In a preferred embodiment, the process for obtaining the soil microbial diversity change index is as follows:

[0020] Obtain the relative abundance of each microorganism in the initial state, and divide the current relative abundance of each microorganism by the relative abundance of each microorganism in the initial state to obtain the relative abundance factor;

[0021] The soil microbial diversity change index is calculated based on the relative abundance factor, using the following formula:

[0022]

[0023] In the formula, H is the soil microbial diversity change index, and S is the total number of detected microbial species. It represents the relative abundance of each type of microorganism. It represents the relative abundance of each microorganism in its initial state.

[0024] In a preferred embodiment, the relative abundance of each microorganism is obtained as follows:

[0025] The detected microorganisms are classified according to their morphological characteristics and gene sequences to obtain the total number of detected microbial species.

[0026] By combining the number of pulse signals corresponding to each microorganism, i.e. the number of individual microorganisms, the relative abundance of each microorganism is calculated.

[0027] In a preferred embodiment, the process for obtaining the number of pulse signals corresponding to each microorganism, i.e., the number of individual microorganisms, is as follows:

[0028] Sampling points were set up at key locations in the rice-fish farming area, and buoy devices equipped with microbial sensors were deployed. The microbial sensors use fluorescent labeling and flow cytometry technology to identify and count different microbial groups.

[0029] The microbial sensor collects soil samples at a set frequency and detects the fluorescent labeling signals of different microorganisms in the samples through the built-in fluorescence detection system. It records the number of pulse signals for each microorganism, and the number of these pulse signals corresponds to the number of individual microorganisms.

[0030] The sensor transmits the raw pulse signal data it collects to the data processing center via the LoRa wireless communication module, where it is stored in a time-series database.

[0031] The data processing center processes the received pulse signal data and uses the sensor's calibration parameters to convert the pulse signal into the actual number of individual microorganisms.

[0032] In a preferred embodiment, the process for acquiring aquatic animal behavioral entropy change data is as follows:

[0033] Acquire video data of aquatic animal activities, and extract one frame from the video data at regular time intervals to obtain an image sequence;

[0034] The extracted image is converted to grayscale, and Gaussian filtering is used to remove noise from the image;

[0035] The preprocessed image is processed using a deep learning object detection algorithm to identify aquatic animals in the image and determine their outlines and position coordinates in the image.

[0036] The activity area of ​​aquatic animals is divided into m equal-sized grids on the image;

[0037] The frequency of each animal appearing in different grids within a certain time period is counted, and the frequency of animal appearance in each grid is calculated. Combined with the frequency of animal appearance in each grid under the initial state, the change in aquatic animal behavioral entropy is calculated. The specific calculation formula is as follows:

[0038]

[0039] In the formula, B represents the data on the change of aquatic animal behavioral entropy. It represents the frequency of animal occurrences within each grid. It represents the frequency of animal occurrences within each grid in the initial state. It is the frequency index in the initial state.

[0040] In a preferred embodiment, the process of predicting changes in the growth environment using a machine learning model is as follows:

[0041] Long Short-Term Memory Network was chosen as the machine learning model. The model was trained using a historical feature dataset. During the training process, the feature data of the time series was used as input and the feature data of a certain future time period was used as output. The model parameters were adjusted through the backpropagation algorithm to minimize the prediction error of the model.

[0042] The feature dataset analyzed by the data processing module is organized in time series format and input into the trained LSTM model. Based on the input feature data, the model outputs a comprehensive ecological disturbance index through internal neuron calculations and state propagation to predict changes in the growth environment.

[0043] In a preferred embodiment, the specific formula for calculating the comprehensive ecological disturbance index is as follows:

[0044]

[0045] In the formula, G is the comprehensive ecological disturbance index, and H is the soil microbial diversity change index. B is the weighting coefficient of the soil microbial diversity change index, and B is the data on the change of aquatic animal behavioral entropy. These are weighting coefficients for the entropy changes in aquatic animal behavior. All values ​​were obtained through historical data calculation and analysis, and all are greater than 0.

[0046] In a preferred embodiment, the process of automatically issuing an early warning signal when an abnormal change in the rice-fish growth environment is predicted is as follows:

[0047] The comprehensive ecological disturbance index is compared with the preset ecological disturbance threshold. If the comprehensive ecological disturbance index is greater than the preset ecological disturbance threshold, it indicates that the current environmental changes have exceeded the normal frequency of changes in the rice-fish growth environment. Intervention measures should be taken immediately, and an early warning signal will be automatically issued.

[0048] If the comprehensive ecological disturbance index is less than the preset ecological disturbance threshold, it indicates that the rice-fish growth environment is in a relatively stable state, and the monitoring of changes in the rice-fish growth environment can be maintained.

[0049] In a preferred embodiment, the process of dynamically adjusting the monitoring frequency is as follows:

[0050] When the comprehensive ecological disturbance index exceeds the preset ecological disturbance threshold, the comprehensive ecological disturbance index and the preset ecological disturbance threshold are obtained based on the real-time feedback of prediction results and early warning signals.

[0051] The monitoring frequency adjustment coefficient is generated based on the comprehensive ecological disturbance index and the preset ecological disturbance threshold.

[0052] The monitoring frequency is dynamically adjusted based on the monitoring frequency adjustment coefficient and the initial monitoring frequency. The specific calculation formula is as follows:

[0053]

[0054] In the formula, This is the adjusted monitoring frequency; G is the comprehensive ecological disturbance index. It is a preset ecological disturbance threshold. It is the ecological disturbance regulation coefficient. It is the initial monitoring frequency weighting coefficient. This is the initial monitoring frequency. Among them, , These are obtained through historical monitoring frequency adjustment information.

[0055] The technical effects and advantages of the wireless multi-sensor data monitoring terminal system for smart rice-fish farming according to the present invention are as follows:

[0056] 1. This invention uses a data setting module to set the initial monitoring frequency based on historical rice-fish farming experience and the growth environment, avoiding the randomness of monitoring frequency. The environmental monitoring module collects real-time and comprehensive data on rice-fish growth, soil data, and aquatic animal status, ensuring that the acquired data matches actual needs and provides a reliable foundation for subsequent analysis. For example, during critical rice growth periods, the system can specifically increase the collection frequency of soil nutrient and moisture data to accurately capture the impact of soil environmental changes on rice growth.

[0057] The data processing module analyzes feature data to quantify the degree of change in the rice-fish growing environment, making environmental changes clearly identifiable. The early warning module uses machine learning models to predict changes in the growing environment, enabling early identification of potential risks. For example, by analyzing characteristic data such as water quality and soil, it predicts potential problems such as eutrophication and soil salinization, and issues timely warnings. Compared to traditional manual judgment or simple threshold judgment, the accuracy and foresight of the predictions are significantly improved. The dynamic adjustment module dynamically adjusts the monitoring frequency of the monitoring equipment based on prediction results and early warning signals. When environmental changes are predicted or anomalies have already occurred, the monitoring frequency is increased to acquire data more intensively, track environmental change trends in real time, and take timely countermeasures. When the environment is stable, the monitoring frequency is reduced to save system resources and extend equipment lifespan. For example, before extreme weather events, the system automatically increases the monitoring frequency of meteorological and water quality data, providing sufficient data support for disaster prevention and mitigation.

[0058] 2. This invention, through precise monitoring and intelligent prediction, enables aquaculture operators to promptly identify and resolve environmental problems, reducing issues such as poor rice-fish growth and disease outbreaks caused by unsuitable environments, thereby improving rice-fish yield and quality. For example, it provides early warnings of water quality deterioration, preventing fish deaths due to oxygen deficiency and ensuring fishery profits; it also allows for timely adjustments to soil management measures, promoting healthy rice growth and increasing rice yield. Automated data collection, analysis, prediction, and adjustment functions reduce the workload of manual inspections and data processing, lowering labor costs. Dynamically adjusting monitoring frequency avoids unnecessary resource waste; for instance, reducing equipment operating frequency during stable environmental conditions reduces power consumption and equipment wear, while also lowering data storage and transmission costs. The system integrates advanced technologies across multiple modules, achieving intelligent management throughout the entire process from data collection to decision response. It provides scientific and efficient management tools for rice-fish farming, contributing to the transformation and upgrading of traditional rice-fish farming towards intelligence and digitalization, enhancing the competitiveness and sustainable development capabilities of the entire industry. The data setting module allows for flexible setting of initial monitoring frequencies and parameters based on historical rice-fish farming experience and growth environments in different regions and with different farming models, making the system applicable to diverse rice-fish farming scenarios. Whether it's a large-scale professional breeding base or a small-scale farmer breeding operation, this system can achieve precise and intelligent management. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the structure of a wireless multi-sensor data monitoring terminal system for smart rice-fish farming according to the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0061] Example 1, Figure 1 This invention presents a wireless multi-sensor data monitoring terminal system for smart rice-fish farming.

[0062] The data setting module is used to set the initial monitoring frequency based on historical rice-fish farming experience and the rice-fish growth environment;

[0063] Data on growth indicators of rice and fish at different growth stages were collected, such as plant height, number of tillers, and yield of rice, and body length, weight, and survival rate of naturally born animals. At the same time, management operations during the breeding process were recorded, including feeding time and amount, water change frequency, and drug use, to obtain data on the growth of rice and fish.

[0064] Organize and preprocess historical meteorological and soil data;

[0065] The fluctuation range and trend of various environmental parameters under different categories were analyzed based on the preprocessed historical meteorological data and historical soil data to obtain the fluctuation characteristics of environmental parameters.

[0066] An initial monitoring frequency is set based on the fluctuation characteristics of environmental parameters and the growth of rice and fish.

[0067] It should be noted that setting a reasonable monitoring frequency helps to detect abnormal changes in environmental parameters in a timely manner, enabling aquaculture personnel to take corresponding measures as soon as possible, such as adjusting water quality, increasing dissolved oxygen, and adjusting feeding strategies, to maintain a suitable environment for rice-fish growth, reduce problems such as poor growth, disease occurrence, and even death caused by unsuitable environment, and improve the yield and quality of rice-fish farming.

[0068] The environmental monitoring module monitors the growth of rice and fish, soil data, and the status of aquatic animals in real time, and promptly captures changes in the growth environment of rice and fish.

[0069] Sensors are deployed at key locations based on the topography, water flow direction, and planting and aquaculture distribution of the rice-fish farming area.

[0070] For example, water quality sensors are installed at the inlets and outlets of rice paddies and in the center of fishponds; soil sensors are buried at different depths in rice planting areas; weather stations are erected above aquaculture areas; and insect monitoring lights and cameras are set up in the fields.

[0071] Wireless communication technology is used to connect scattered sensors into a network, establish a data transmission channel, and realize the real-time transmission of sensor data to the data processing center, ensuring the timeliness and integrity of data acquisition.

[0072] Based on the monitoring frequency predetermined by the data setting module, each sensor collects corresponding environmental parameters at regular intervals. To ensure the correlation and analytical value of the data, multiple environmental parameters are collected simultaneously at the same time point.

[0073] The raw data (such as analog signals) collected by the sensors is converted into a digital format that the system can recognize and uniformly encoded for easy storage and transmission;

[0074] The system performs preliminary verification of the collected data using preset rules and algorithms to identify abnormal data. For example, if the data collected by the water temperature sensor exceeds the normal fluctuation range (such as the water temperature being displayed as 50℃ in winter), it is determined to be an abnormal value, and the system automatically marks it and attempts to re-collect the data or issues a sensor fault warning.

[0075] Using the established wireless communication network, the processed environmental data is transmitted to the data processing center or cloud server in real time. For important abnormal data, a priority transmission strategy is adopted to ensure that it is delivered as soon as possible so that timely countermeasures can be taken.

[0076] When the network is unstable or interrupted, the sensor device will temporarily store the collected data in a local cache. Once the network is restored, it will automatically resume transmission from the point of interruption to ensure data integrity and avoid data loss due to network problems.

[0077] It should be noted that by deploying high-density, multi-dimensional sensors and collecting real-time data, we can achieve comprehensive and seamless monitoring of the rice-fish growing environment, promptly capturing subtle changes such as water quality deterioration, soil drought, meteorological disasters, and outbreaks of pests and diseases, and providing accurate information for aquaculture management.

[0078] The data processing module extracts feature data related to changes in the rice-fish growth environment and abnormalities in the aquaculture species from the environmental monitoring module, and analyzes the extracted feature data to quantify the degree of change in the rice-fish growth environment.

[0079] Soil microbial communities are an important indicator of ecosystem health. Microbial diversity reflects soil health and its adaptability to environmental changes, while behavioral entropy is an indicator of the complexity and diversity of aquatic animal behavior. When aquatic animals are subjected to environmental stress or abnormalities, their behavioral patterns may become monotonous or irregular, leading to changes in entropy values. Therefore, based on the environmental monitoring module, characteristic data related to changes in the rice-fish farming environment and abnormalities in the cultured organisms were extracted. These characteristic data include soil microbial diversity change index and aquatic animal behavioral entropy value change data.

[0080] The process for obtaining the soil microbial diversity change index is as follows:

[0081] Sampling points were set up at key locations such as the intersection of ditches and the center of fish ponds in the rice-fish farming area, and buoy devices equipped with microbial sensors were deployed. The sensors use fluorescent labeling and flow cytometry technology to identify and count different microbial groups.

[0082] The microbial sensor collects soil samples at a set frequency and detects the fluorescent labeling signals of different microorganisms in the samples through the built-in fluorescence detection system. It records the number of pulse signals for each microorganism, and the number of these pulse signals corresponds to the number of individual microorganisms.

[0083] The sensor transmits the raw pulse signal data it collects to the data processing center via the LoRa wireless communication module and stores it in the time-series database. The data record includes the sampling time, sampling point number, and the number of pulse signals corresponding to each microorganism.

[0084] The data processing center processes the received pulse signal data and uses the sensor calibration parameters to convert the pulse signal into the actual number of microbial individuals. For data anomalies caused by signal interference, the 3σ principle is used to identify and remove them, that is, when the data deviates from the mean by more than 3 times the standard deviation, it is considered an anomaly.

[0085] The detected microorganisms are classified according to their morphological characteristics and gene sequences to obtain the total number of detected microbial species. Combined with the number of pulse signals corresponding to each microorganism, i.e., the number of individual microorganisms, the relative abundance of each microorganism is calculated using the following formula:

[0086]

[0087] In the formula, It represents the relative abundance of each type of microorganism. The number of pulse signals corresponding to the i-th microorganism is the number of individual microorganisms, and S is the total number of detected microorganism species.

[0088] The relative abundance of each microorganism in the initial state is obtained. The relative abundance factor is obtained by dividing the current relative abundance of each microorganism by the relative abundance of each microorganism in the initial state. The soil microbial diversity change index is calculated based on the relative abundance factor. The specific calculation formula is as follows:

[0089]

[0090] In the formula, H is the soil microbial diversity change index, and S is the total number of detected microbial species. It represents the relative abundance of each type of microorganism. It represents the relative abundance of each microorganism in its initial state.

[0091] The process of obtaining data on changes in the entropy of aquatic animal behavior is as follows:

[0092] High-definition cameras should be strategically deployed above fishponds and rice paddies to ensure coverage of the entire aquatic animal activity area.

[0093] The camera continuously collects video data of aquatic animal activities according to the set parameters;

[0094] The locally stored video data is transferred to the data processing center via the network and stored in a large-capacity hard drive array. To save storage space, the video data is compressed using H.265 encoding.

[0095] Video processing software is used to extract one frame at regular intervals from the captured video to obtain an image sequence. The extracted images are then converted to grayscale to reduce the amount of data and highlight the animal outlines. Gaussian filtering is then used to remove noise from the images.

[0096] The preprocessed image is processed using deep learning object detection algorithms (such as the YOLO model) to identify aquatic animals in the image and determine their outlines and position coordinates in the image.

[0097] The activity area of ​​aquatic animals is divided into m equally sized grids on the image. The number of times each individual animal appears in different grids within a certain time period is counted. The frequency of animal appearance in each grid at the current time is calculated. Combined with the frequency of animal appearance in each grid at the initial state, the change in aquatic animal behavioral entropy is calculated. The specific calculation formula is as follows:

[0098]

[0099] In the formula, B represents the data on the change of aquatic animal behavioral entropy. It represents the frequency of animal occurrences within each grid. It represents the frequency of animal occurrences within each grid in the initial state. It is the frequency index in the initial state.

[0100] It is important to note that monitoring the diversity index of the microbial community can reveal changes in soil quality, nutrient cycling, and microbial ecological balance, which is crucial for paddy field productivity and ecological environmental protection. Monitoring the behavioral entropy values ​​of aquatic animals allows for the timely detection of their health status, especially abnormal behaviors such as escape, aggregation, and loss of appetite, which may reflect changes in environmental factors such as water quality, temperature, and oxygen levels. Extracting data on changes in soil microbial diversity indices and aquatic animal behavioral entropy values ​​provides multi-dimensional, early-stage environmental and biological monitoring information for smart rice-fish farming systems.

[0101] The early warning module inputs the analyzed feature data into a pre-trained machine learning model, which then predicts changes in the rice-fish growth environment. When the machine learning model predicts that the changes in the rice-fish growth environment are abnormal, it automatically issues an early warning signal to remind operators to pay attention to potential risks in a timely manner.

[0102] Long Short-Term Memory (LSTM) network was chosen as the machine learning model. The model was trained using historical feature datasets. During training, time series feature data was used as input and feature data for a future period was used as output. The model parameters were adjusted through backpropagation algorithm to minimize the prediction error of the model.

[0103] The feature dataset analyzed by the data processing module is organized in time series format and input into the trained LSTM model. Based on the input feature data, the model outputs a comprehensive ecological disturbance index through internal neuron calculations and state propagation, and makes predictions about changes in the growth environment.

[0104] The specific formula for calculating the comprehensive ecological disturbance index in the LSTM model is as follows:

[0105]

[0106] In the formula, G is the comprehensive ecological disturbance index, and H is the soil microbial diversity change index. B is the weighting coefficient of the soil microbial diversity change index, and B is the data on the change of aquatic animal behavioral entropy. These are weighting coefficients for the entropy changes in aquatic animal behavior. All values ​​were obtained through historical data calculation and analysis, and all are greater than 0.

[0107] The comprehensive ecological disturbance index is compared with the preset ecological disturbance threshold. If the comprehensive ecological disturbance index is greater than the preset ecological disturbance threshold, it indicates that the current environmental changes have exceeded the normal frequency of changes in the rice-fish growth environment. Intervention measures should be taken immediately, and an early warning signal will be automatically issued to remind operators to pay attention to potential risks in a timely manner.

[0108] If the comprehensive ecological disturbance index is less than the preset ecological disturbance threshold, it indicates that the rice-fish growth environment is in a relatively stable state, and the monitoring of changes in the rice-fish growth environment can be maintained.

[0109] It should be noted that by comparing and analyzing the comprehensive ecological disturbance index and threshold in real time, the system can accurately grasp the environmental status of rice and fish growth. When environmental anomalies occur, timely warnings and interventions are provided to create stable and suitable environmental conditions for rice and fish growth, reducing pests and diseases caused by environmental fluctuations, improving the survival rate and quality of rice and fish, and ultimately increasing yield and economic benefits. For example, when abnormal soil pH may affect rice root growth, timely intervention can ensure that rice absorbs nutrients normally, achieving increased production and income.

[0110] The dynamic adjustment module dynamically adjusts the monitoring frequency based on real-time feedback of prediction results and early warning signals, combined with the initial monitoring frequency, in order to cope with changes in the rice-fish growth environment.

[0111] When the comprehensive ecological disturbance index is greater than the preset ecological disturbance threshold, the comprehensive ecological disturbance index and the preset ecological disturbance threshold are obtained based on the real-time feedback prediction results and early warning signals. The monitoring frequency adjustment coefficient is generated based on the comprehensive ecological disturbance index and the preset ecological disturbance threshold.

[0112] The monitoring frequency is dynamically adjusted based on the monitoring frequency adjustment coefficient and the initial monitoring frequency. The specific calculation formula is as follows:

[0113]

[0114] In the formula, This is the adjusted monitoring frequency; G is the comprehensive ecological disturbance index. It is a preset ecological disturbance threshold. It is the ecological disturbance regulation coefficient. It is the initial monitoring frequency weighting coefficient. This is the initial monitoring frequency. Among them, , These are obtained through historical monitoring frequency adjustment information.

[0115] It should be noted that when the Ecological Disturbance Index (EDCI) exceeds the threshold, the system adjusts the coefficient to increase the monitoring frequency, ensuring that rapid changes in environmental parameters are captured.

[0116] When environmental parameters fluctuate only slightly, the adjustment coefficient reduces the monitoring frequency, minimizing data redundancy and equipment energy consumption. Increasing the adjustment coefficient addresses sudden changes (such as abrupt changes in water quality caused by heavy rain), ensuring critical data is not lost. Traditional fixed-frequency monitoring may miss abnormal peaks due to excessively long sampling intervals; dynamic adjustment can reduce this risk by over 70%. Reducing the monitoring frequency during stable periods can extend sensor sleep time by 30%-50%, significantly reducing battery replacement frequency and wireless transmission traffic. High-frequency data provides denser time-series data, enabling machine learning models to capture subtle features of parameter changes, thereby improving the accuracy of ecological disturbance predictions.

[0117] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0118] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0119] Those skilled in the art will recognize that the modules and algorithm steps of the various examples 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 implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0121] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0122] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1.A wireless multi-sensing data monitoring terminal system for smart rice-fish, characterized in that, The module comprises a data setting module, an environment monitoring module, a data processing module, a warning module, and a dynamic adjustment module, and the modules are connected; The data setting module is configured to set an initial monitoring frequency according to historical rice-fish breeding experience and rice-fish growth environment; The environment monitoring module is configured to monitor rice-fish growth conditions, soil data, and aquatic animal states in real time, and capture changes in rice-fish growth environment in a timely manner; The data processing module is configured to extract feature data related to changes in rice-fish growth environment and abnormal breeding objects from the environment monitoring module, wherein the feature data comprises a soil microbial diversity change index and aquatic animal behavior entropy change data, and the soil microbial diversity change index is obtained by dividing the relative abundance of each microorganism in the current state by the relative abundance of each microorganism in the initial state; the soil microbial diversity change index is calculated according to the relative abundance factor, and the specific calculation formula is as follows: ， In the formula, H is a soil microbial diversity change index, S is the total number of detected microbial species, is the relative abundance of each microorganism at present, is the relative abundance of each microorganism at the initial state; The process of obtaining the aquatic animal behavior entropy change data is as follows: Video data of aquatic animal activities is obtained, and an image sequence is obtained by extracting one frame of image from the video data every certain time period; the extracted image is subjected to grayscale processing, and Gaussian filtering is used to remove noise in the image; a deep learning target detection algorithm is used to process the preprocessed image, identify the aquatic animals in the image, and determine the contour and position coordinates of the aquatic animals in the image; the activity area of the aquatic animals in the image is divided into m equal grids; the frequency of the appearance of each animal in each grid is calculated by counting the number of times each animal appears in different grids within a certain time period, and the aquatic animal behavior entropy change data is calculated by combining the frequency of the appearance of each animal in each grid in the initial state, and the specific calculation formula is as follows: ， In the formula, B is the change data of the entropy value of the behavior of aquatic animals, is the frequency of the appearance of animals in each grid at present, is the frequency of the appearance of animals in each grid at the initial state, is the frequency index at the initial state; The feature data is analyzed to quantify the degree of change in the rice-fish growth environment; The warning module is configured to input the analyzed feature data into a pre-trained machine learning model to predict changes in the growth environment, and the process is as follows: The long short-term memory network is selected as the machine learning model, and the historical feature data set is used to train the model; in the training process, the feature data in time series is used as the input, and the feature data in the future certain time period is used as the output; the model parameters are adjusted by the back propagation algorithm to minimize the prediction error of the model; The feature data set analyzed by the data processing module is arranged in time series format and input into the trained LSTM model; the model outputs an ecological disturbance comprehensive index by calculating and transferring the state of internal neurons according to the input feature data, to predict changes in the growth environment; When the predicted rice-fish growth environment changes abnormally, an automatic warning signal is sent out; The dynamic adjustment module is configured to dynamically adjust the monitoring frequency of the monitoring device according to the real-time feedback of the prediction result and the warning signal to cope with changes in the rice-fish growth environment, and the process of dynamically adjusting the monitoring frequency of the monitoring device is as follows: When the ecological disturbance comprehensive index is greater than the preset ecological disturbance threshold value, the ecological disturbance comprehensive index and the preset ecological disturbance threshold value are obtained according to the real-time feedback of the prediction result and the warning signal. According to the ecological disturbance comprehensive index and the preset ecological disturbance threshold, a monitoring frequency adjustment coefficient is generated; According to the monitoring frequency adjustment coefficient and the initial monitoring frequency, the monitoring frequency is dynamically adjusted, and the specific calculation formula is as follows: ， In the formula, is the adjusted monitoring frequency, G is the ecological disturbance comprehensive index, is the preset ecological disturbance threshold, is the ecological disturbance adjustment coefficient, is the initial monitoring frequency weight coefficient, is the initial monitoring frequency, wherein, , are respectively obtained by historical monitoring frequency adjustment information. 2.The wireless multi-sensing data monitoring terminal system for smart rice-fish farming according to claim 1, wherein, The initial monitoring frequency setting process is as follows: Collect the growth index data of rice-fish at different growth stages, and record the management operations during the breeding process to obtain the growth conditions of rice-fish; Organize and preprocess historical meteorological data and historical soil data; According to the environmental parameter fluctuation characteristics and the growth conditions of rice-fish, an initial monitoring frequency is set. The current relative abundance of each microorganism is obtained as follows: 3.The wireless multi-sensing data monitoring terminal system for smart rice-fish farming according to claim 2, wherein, According to the morphological characteristics and gene sequences of microorganisms, the detected microorganisms are classified by species, and the total number of species of the detected microorganisms is obtained; The relative abundance of each microorganism is calculated by combining the number of pulse signals corresponding to each microorganism, i.e. the number of individuals of the microorganism. The number of pulse signals corresponding to each microorganism, i.e. the number of individuals of the microorganism, is obtained as follows: 4.The wireless multi-sensing data monitoring terminal system for smart rice and fish farming according to claim 3, wherein, Set sampling points at key positions in the rice-fish breeding area and deploy buoy devices equipped with microorganism sensors. The microorganism sensors use fluorescence labeling and flow cytometry technology to identify and count different microorganism groups. The microorganism sensor collects soil samples at a set frequency, detects the fluorescence labeling signals of different microorganisms in the samples through the built-in fluorescence detection system, and records the number of pulse signals of each microorganism. These pulse signal numbers correspond to the number of individuals of the microorganism. The sensor transmits the collected raw pulse signal data to the data processing center through the LoRa wireless communication module and stores it in the time series database. The data processing center processes the received pulse signal data and converts the pulse signals into actual microorganism individual numbers using the calibration parameters of the sensor. The specific calculation formula of the ecological disturbance comprehensive index is as follows: 5.The wireless multi-sensing data monitoring terminal system for smart rice and fish farming according to claim 4, wherein, When the predicted rice-fish growth environment changes abnormally, the automatic warning signal process is as follows: ， In the formula, G is the comprehensive index of ecological disturbance, H is the index of soil microbial diversity change, is the weight coefficient of the index of soil microbial diversity change, B is the change data of the entropy value of aquatic animal behavior, is the weight coefficient of the change data of the entropy value of aquatic animal behavior, , are all obtained by historical data calculation and analysis, and are all greater than 0. 6.The wireless multi-sensing data monitoring terminal system for smart rice and fish farming according to claim 5, wherein, Compare the ecological disturbance comprehensive index with the preset ecological disturbance threshold. If the ecological disturbance comprehensive index is greater than the preset ecological disturbance threshold, it indicates that the current environmental change has exceeded the normal frequency of rice-fish growth environment change, and immediate intervention measures should be taken, and a warning signal is automatically sent out; If the ecological disturbance comprehensive index is less than the preset ecological disturbance threshold, it indicates that the rice-fish growth environment is in a relatively stable state, and the monitoring of the rice-fish growth environment change is maintained. ​

Images