A big data-based breeding environment monitoring method and system

Abstract

The present application relates to the technical field of big data analysis, and more particularly to a breeding environment monitoring method and system based on big data, which comprises the following steps: extracting and collecting air temperature, snow thickness and soil data of alfalfa breeding environment in the overwintering period to obtain overwintering freeze-thaw stress data, and predicting the predicted green rate of the current period; determining the overwintering precision irrigation scheme based on the predicted green rate at the beginning of overwintering and the soil data; obtaining remote sensing images after the end of the alfalfa green period to obtain the actual green rate, and determining the accurate state of monitoring according to the deviation of the predicted green rate before the green period; when the monitoring is determined to be inaccurate, determining the abnormal reason based on the soil temperature attenuation degree and issuing corresponding processing instructions; when it is determined that the monitoring is still inaccurate after executing the processing instructions, determining that the abnormal reason is insufficient spatial resolution of the remote sensing image; and when the monitoring is determined to be accurate, continuously monitoring the alfalfa breeding environment. The present application realizes automatic and precise monitoring and active regulation of the alfalfa overwintering breeding environment.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, and in particular to a method and system for monitoring aquaculture environments based on big data. Background Technology

[0002] Most of my country's main alfalfa producing areas are located in the cold, high-altitude regions of the north, where winters are long and harsh, making winter frost damage a persistent problem for alfalfa. As a perennial forage crop, alfalfa's root collar is highly susceptible to irreversible damage during winter due to repeated freeze-thaw cycles in the soil. This damage can range from a decrease in regrowth rate and reduced yield and quality the following year to large-scale mortality. The annual losses in alfalfa production due to winter frost damage are substantial and have become one of the key bottlenecks restricting the development of the domestic high-quality alfalfa industry.

[0003] Meanwhile, traditional alfalfa winter management relies heavily on growers' experience and judgment, lacking precise perception of environmental factors and the ability to quantitatively assess frost damage risks. With the continuous expansion of large-scale alfalfa planting areas, manual experience alone is insufficient to meet the demands of refined management. The industry urgently needs scientific and intelligent methods for monitoring the winter environment and controlling risks. Improving the precision of alfalfa winter management, reducing frost damage losses, and ensuring a stable supply of high-quality alfalfa have become pressing needs for promoting the high-quality development of the alfalfa industry.

[0004] Chinese Patent Publication No. CN122366844A discloses a method and system for monitoring and assessing marine aquaculture environment using IoT big data. The main steps include: collecting multi-dimensional environmental parameters such as dissolved oxygen, nitrogen and phosphorus nutrients, and pH value through IoT sensing devices; performing parallel reception and status verification through a high-concurrency marine environmental data processing platform; using a multi-dimensional time-series data adaptive clustering algorithm to perform time-series segmentation and similarity clustering of the parameters; quantitatively assessing water quality status based on the clustering results using a water eutrophication assessment model; and then dynamically predicting stress intensity and duration using an environmental stress response prediction model, combined with parameters adapted to the growth cycle of aquaculture organisms. The platform integrates the assessment and prediction results to generate a structured report including parameter change trajectories, level identifiers, and predicted trend curves, achieving precise monitoring and assessment of the marine aquaculture environment throughout the entire process, providing reliable technical support for aquaculture environment regulation.

[0005] Therefore, although the proposed solution can assess and predict the degree of environmental stress and generate monitoring reports, it has the following problems: The scheme only outputs stress prediction results and cannot automatically generate and implement corresponding environmental control and intervention measures based on the prediction results. It fails to form a closed-loop control of prediction, intervention and effect verification, and therefore cannot actively avoid or reduce the actual damage of environmental stress to aquaculture objects. Summary of the Invention

[0006] To address this, the present invention provides a method and system for monitoring aquaculture environments based on big data, which overcomes the problem in existing technologies that cannot automatically generate and execute environmental regulation and intervention measures and complete closed-loop control based on stress prediction results.

[0007] To achieve the above objectives, the present invention provides a method for monitoring aquaculture environments based on big data, comprising: Real-time data collection of air temperature, snow depth, and soil data at several depths of soil layers during the wintering period of alfalfa farming environment. Soil data includes soil temperature and soil moisture content. Feature extraction is performed on the air temperature, snow thickness and soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The predicted greening rate for the current period is predicted based on the freeze-thaw stress data. Based on the predicted greening rate at the beginning of winter and the soil moisture content, a precise winter irrigation plan is generated and implemented, including irrigation time and irrigation amount. To obtain the actual greening rate, remote sensing images of alfalfa were acquired after the greening period ended. The accuracy of monitoring the alfalfa farming environment is determined based on the deviation between the predicted greening rate before the greening period and the actual greening rate. If the monitoring of the alfalfa farming environment is found to be inaccurate, the cause of the anomaly is determined based on the soil temperature attenuation and corresponding processing instructions are issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. After executing the corresponding processing instructions, if the monitoring of the alfalfa farming environment is still inaccurate, the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image. Continuous monitoring of the alfalfa farming environment should be conducted once the accuracy of monitoring for the alfalfa farming environment is confirmed.

[0008] Furthermore, the process of determining the accurate monitoring status of the alfalfa farming environment based on the prediction deviation rate of the greening rate includes: The predicted greening rate before the greening period is obtained based on all freeze-thaw stress data during the overwintering period and is denoted as the expected greening rate. The predicted greening rate deviation rate is calculated based on the expected greening rate and the actual greening rate. The accuracy of monitoring the alfalfa farming environment is determined based on the predicted deviation rate of the greening rate.

[0009] Furthermore, in cases where monitoring of the alfalfa farming environment is determined to be inaccurate, the process of identifying the cause of the anomaly based on soil temperature attenuation includes: Soil temperature in the alfalfa root collar layer and its corresponding collection time were obtained to construct a time-root collar layer soil temperature curve, and air temperature and its corresponding collection time were obtained to construct a time-air temperature curve. In the continuous time of the time-air temperature curve, the turning point where the air temperature changes from a continuous decrease to a continuous increase is identified as a trough, and the turning point where the air temperature changes from a continuous increase to a continuous decrease is identified as a peak. The temperature change during this time period is recorded as a fluctuation event. The highest and lowest values ​​of soil temperature within the same time period were obtained from the time-root collar soil temperature curve. Calculate the peak-valley difference of air temperature based on air temperature peaks and valleys; Calculate the peak-valley difference of soil temperature based on the highest and lowest soil temperature values; The soil temperature attenuation rate is calculated based on the peak-to-valley difference of air temperature and the peak-to-valley difference of soil temperature during all fluctuation events in the overwintering period. Based on the soil temperature attenuation, the cause of the anomaly is determined and a corresponding processing instruction is issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount.

[0010] Furthermore, the process of adjusting the soil temperature data filtering window based on soil temperature fluctuation includes: The time-root collar soil temperature curve was processed by moving average to obtain a trend curve; The residual curve is determined based on the temperature values ​​at each moment on the time-root collar soil temperature curve and the trend values ​​at the corresponding moments on the trend curve. Soil temperature fluctuation is determined based on the residual curve. The soil temperature data filtering window is increased based on the soil temperature fluctuation, and the increase in the soil temperature data filtering window is positively correlated with the soil temperature fluctuation.

[0011] Furthermore, the process of further correcting the increase in the soil temperature data filtering window based on the deep baseline fluctuation ratio includes: The soil temperature of the deep reference layer of alfalfa and its corresponding collection time were obtained to construct a time-deep reference layer soil temperature curve; The time-deep reference layer soil temperature curve was processed by moving average to obtain the trend curve of the deep reference layer; The residual curve of the deep reference layer is determined based on the temperature values ​​at each moment on the time-deep reference layer soil temperature curve and the trend values ​​at the corresponding moments on the trend curve of the deep reference layer. The soil temperature fluctuation of the deep reference layer is determined based on the residual curve of the deep reference layer. The deep reference fluctuation ratio is determined based on the soil temperature fluctuation of the root collar layer and the soil temperature fluctuation of the deep reference layer. The increase in the soil temperature data filtering window is based on the deep reference fluctuation ratio, and the increase in the soil temperature data filtering window is positively correlated with the deep reference fluctuation ratio.

[0012] Furthermore, after adjusting the soil temperature data filtering window, if the monitoring of the alfalfa farming environment is still inaccurate based on the greening rate prediction deviation rate, a processing instruction to adjust the irrigation amount is issued.

[0013] Furthermore, the process of adjusting irrigation amount based on irrigation deterioration ratio includes: The soil temperature of the root collar layer before and after irrigation and the corresponding time were obtained to construct the time-root collar layer soil temperature curve before irrigation and the time-root collar layer soil temperature curve after irrigation. The cumulative duration before irrigation and the cumulative duration after irrigation are determined based on the cumulative duration before and after irrigation when the soil temperature in the root collar layer is lower than the average temperature in the curve. The irrigation deterioration ratio is calculated based on the cumulative duration after irrigation and the cumulative duration before irrigation; The irrigation amount is reduced based on the irrigation deterioration ratio, and the reduction in irrigation amount is negatively correlated with the irrigation deterioration ratio.

[0014] Furthermore, the process of further adjusting the reduction in irrigation amount based on the warming rate ratio includes: Identify the segments of continuous temperature rise before irrigation in the time-pre-irrigation root collar soil temperature curve. The temperature recovery rate of each pre-irrigation rising segment is determined based on the temperature rise amplitude and duration of each rising segment, and the average pre-irrigation temperature recovery rate is calculated based on the temperature recovery rate of each pre-irrigation rising segment. Identify the segments of continuous temperature rise after irrigation in the time-root collar soil temperature curve. The temperature recovery rate of each post-irrigation rising segment is determined based on the temperature rise amplitude and duration of each rising segment, and the average temperature recovery rate after irrigation is calculated based on the temperature recovery rate of each post-irrigation rising segment. The ratio of the recovery rate is calculated based on the average recovery rate after irrigation and the average recovery rate before irrigation. The reduction in irrigation amount is corrected based on the warming rate ratio, and the reduction in irrigation amount is negatively correlated with the warming rate ratio.

[0015] Furthermore, after adjusting the irrigation amount, if the monitoring of the alfalfa farming environment is still inaccurate based on the predicted deviation rate of the greening rate, the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image, and a processing instruction to adjust the pixel aggregation window of the remote sensing image is issued.

[0016] Furthermore, the big data-based aquaculture environment monitoring system of the method is characterized by comprising: The data acquisition module is used to collect real-time data on air temperature, snow thickness, and soil data at several depths in the alfalfa growing environment during the wintering period. The soil data includes soil temperature and soil moisture content. The feature extraction module is used to extract features from the air temperature, the snow thickness and the soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The greening rate prediction module is used to predict the greening rate for the current period based on the freeze-thaw stress data. The irrigation scheme decision module is used to generate and execute a precise winter irrigation scheme based on the predicted greening rate at the beginning of winter, including irrigation time and irrigation amount. The actual greening acquisition module is used to acquire remote sensing images of alfalfa after the greening period ends in order to obtain the actual greening rate. An assessment module is used to determine the accurate status of monitoring of the alfalfa farming environment based on the deviation between the predicted greening rate before the greening period and the actual greening rate. The anomaly analysis module is used to determine the cause of the anomaly based on the obtained soil temperature attenuation when the monitoring is inaccurate, and to issue corresponding processing instructions, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. The execution module is used to execute the processing instructions issued by the exception analysis module.

[0017] Compared with existing technologies, the beneficial effects of this invention are as follows: the method of this invention takes the freeze-thaw stress on the root collar layer of alfalfa during the overwintering period as the core monitoring object. It collects real-time data on air temperature, snow thickness, and soil at several depths during the overwintering period, extracts freeze-thaw stress data such as the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and corresponding time, and the number of days of snow cover, and predicts the greening rate based on this data, forming a closed-loop monitoring system from environmental perception, stress feature extraction, greening prediction, irrigation intervention to effect verification. Furthermore, through the coordinated operation of various functional modules, the system achieves automated, precise monitoring and proactive control of the alfalfa overwintering environment, solving the deficiency of existing technologies that only output prediction results without implementing closed-loop control. This provides a complete data-driven decision-making tool for alfalfa overwintering management, improves the accuracy of alfalfa overwintering environment monitoring and the timeliness of abnormal handling, thereby increasing the alfalfa greening rate and reducing overwintering frost damage losses.

[0018] Furthermore, this invention determines the accuracy of monitoring by using the prediction deviation rate of the greening rate, objectively reflecting the degree of consistency between the predicted results and the actual greening situation, and promptly identifying whether the monitoring process is inaccurate. After determining that the monitoring is inaccurate, the cause of the anomaly is further distinguished by the soil temperature attenuation rate, quickly locating the source between data acquisition distortion and irrigation plan deviation, avoiding blind investigation of the source of the anomaly, improving the pertinence of anomaly diagnosis and the efficiency of handling, and ensuring the reliable operation of the monitoring system in complex field environments, thereby further improving the accuracy of alfalfa overwintering environment monitoring and the timeliness of anomaly handling.

[0019] Furthermore, when the cause of the anomaly is determined to be distortion of the collected soil data, the present invention adaptively adjusts the data filtering window based on the noise intensity of the soil temperature data itself, and uses the stability characteristics of deep soil to correct the adjustment range. While effectively filtering out abnormal fluctuations, it avoids excessive smoothing, thereby improving the authenticity and usability of the soil temperature data, and thus further improving the accuracy of alfalfa overwintering environment monitoring and the timeliness of anomaly handling.

[0020] Furthermore, when the cause of the anomaly is determined to be a deviation in the irrigation plan, the present invention automatically adjusts the amount of irrigation based on the changing trend of the low-temperature stress in the root collar layer after irrigation, and makes fine corrections to the adjustment amount in combination with the soil warming capacity. This allows for timely reduction of irrigation amount when irrigation measures have a adverse effect, avoiding secondary damage caused by over-irrigation or improper timing, improving the accuracy and safety of irrigation decisions, and playing a positive role in ensuring the safe overwintering of alfalfa. This further improves the accuracy of alfalfa overwintering environment monitoring and the timeliness of anomaly handling.

[0021] Furthermore, if the monitoring remains inaccurate after executing the processing instructions, it indicates that the prediction results are not the problem, but rather an anomaly occurred when collecting actual greening data. Therefore, the cause of the anomaly was determined to be insufficient spatial resolution of the remote sensing image. Based on the dispersion of the local spectrum of the remote sensing image itself, the pixel aggregation window is adaptively adjusted to effectively reduce local noise interference caused by the mixed pixel effect, improve the data quality of the remote sensing image, ensure the accuracy of the actual greening rate calculation, and enhance the system's adaptability and fault tolerance to anomalies in the remote sensing data source. This further improves the accuracy of monitoring the alfalfa overwintering environment and the timeliness of anomaly handling. Attached Figure Description

[0022] Figure 1 This is a structural block diagram of the big data-based aquaculture environment monitoring system described in an embodiment of the present invention; Figure 2 This is a flowchart of the big data-based aquaculture environment monitoring method described in an embodiment of the present invention; Figure 3This is a flowchart illustrating the method described in this embodiment of the invention for determining the accurate status of monitoring the alfalfa farming environment; Figure 4 This is a flowchart illustrating the method described in this embodiment of the invention for determining the reasons for inaccurate monitoring of the alfalfa farming environment. Detailed Implementation

[0023] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0024] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0026] Please see Figure 1 As shown, it is a structural block diagram of the aquaculture environment monitoring system based on big data according to an embodiment of the present invention, including: The data acquisition module is used to collect real-time data on air temperature, snow thickness, and soil data at several depths in the alfalfa growing environment during the wintering period. The soil data includes soil temperature and soil moisture content. The feature extraction module, which is connected to the acquisition module, is used to extract features from the air temperature, the snow thickness and the soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The greening rate prediction module, which is connected to the feature extraction module, is used to predict the greening rate for the current period based on the freeze-thaw stress data. The irrigation scheme decision module, which is connected to the greening rate prediction module, is used to generate and execute a precise winter irrigation scheme based on the predicted greening rate at the beginning of winter, including irrigation time and irrigation amount. The actual greening acquisition module is used to acquire remote sensing images of alfalfa after the greening period ends in order to obtain the actual greening rate. An evaluation module, which is connected to the greening rate prediction module and the actual greening acquisition module, is used to determine the accurate status of monitoring the alfalfa farming environment based on the deviation between the predicted greening rate before the greening period and the actual greening rate. An anomaly analysis module, which is connected to the evaluation module, is used to determine the cause of the anomaly based on the obtained soil temperature attenuation degree when the monitoring is determined to be inaccurate, and to issue corresponding processing instructions, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. An execution module, which is connected to the exception analysis module, is used to execute the processing instructions issued by the exception analysis module.

[0027] In this embodiment of the invention, the alfalfa cultivation environment can be divided into four stages according to the alfalfa growth process and field management needs: the establishment period, the greening period, the multi-crop growth period, the overwintering preparation period, the overwintering period, and the harvesting period. The establishment period is from sowing to the establishment of the root system during the seedling stage. The greening period is the stage where the soil temperature rises in the following spring and new shoots sprout from the root collar. The multi-crop growth period is the above-ground growth stage from the greening period to each harvest. The overwintering preparation period is the stage from the last harvest to before the soil freezes, during which the plants transfer nutrients to the root collar. The overwintering period is the dormant stage from soil freezing to the greening period in the following spring. The harvesting period spans between each growth stage.

[0028] Specifically, although alfalfa is a perennial plant, environmental conditions and field management priorities vary significantly at different stages, with the overwintering period playing the most critical role in ensuring safe production and yield formation in the following year. During the overwintering period, the above-ground parts cease growth and enter dormancy, but the root collar, as the core organ for overwintering, continues to carry out physiological activities and directly undertakes the function of sprouting new shoots in the following year.

[0029] In this embodiment of the invention, the acquisition module collects soil data from the near-surface layer, the root collar layer, and the deep reference layer. The root collar layer is determined based on the depth of alfalfa root collar penetration into the soil. The near-surface layer extends from the surface downwards to the upper boundary of the root collar layer, and the deep reference layer extends from the lower boundary of the root collar layer downwards to the position where the daily variation of soil temperature tends to stabilize.

[0030] In this embodiment of the invention, freeze-thaw cycle data obtained by feature extraction from environmental data collected during the overwintering period demonstrates the stress level of the freeze-thaw cycle process on the alfalfa root collar during the overwintering period. The freeze-thaw cycle refers to the repeated freezing and thawing of the soil during the overwintering period due to temperature fluctuations around 0°C; stress refers to the environmental pressure exerted on the alfalfa root collar by this freeze-thaw cycle process.

[0031] Specifically, the number of freeze-thaw cycles in the root collar soil is determined by identifying a freeze-thaw cycle occurring on a given day when the highest soil temperature in the root collar is greater than 0°C and the lowest soil temperature is less than 0°C, based on the time-series data of root collar soil temperature. This results in the number of days during the entire overwintering period when freeze-thaw cycles are identified. The number of days with snow cover is the cumulative number of days during the overwintering period when the snow thickness is greater than zero.

[0032] In this embodiment of the invention, the predicted greening rate is obtained by predicting using a pre-trained greening prediction model. Specifically, the greening prediction model is trained based on historical winter freeze-thaw stress data and the actual greening rate of the corresponding plots obtained through remote sensing imagery or manual surveys after the greening period of the following year. In this embodiment of the invention, when the soil is not completely frozen in the early stage of overwintering and there are still conditions for irrigation water to infiltrate, it is determined whether irrigation is needed based on the predicted greening rate and the soil moisture content, and an overwintering precision irrigation plan is generated when it is determined that irrigation is needed.

[0033] Specifically, the repeated fluctuations in root collar soil temperature around 0℃ during winter lead to multiple freeze-thaw cycles. The mechanical stress generated by these freeze-thaw cycles and the prolonged period of sub-zero soil temperature are the main reasons for root collar cell damage and a decreased greening rate the following year. Higher soil moisture content results in greater soil heat capacity, leading to smaller fluctuations in root collar soil temperature under the same air temperature changes. This reduces the number of freeze-thaw cycles and the duration of low temperatures, thus mitigating freeze-thaw stress damage to the root collar. Therefore, when the predicted greening rate is low and the root collar soil moisture content is insufficient to provide adequate heat capacity buffering, irrigation is deemed necessary. Increasing the root collar soil moisture content enhances soil heat capacity, reducing the degree of freeze-thaw stress on the root collar during winter and thus improving the greening rate. After determining the need for irrigation, the irrigation amount is determined based on the difference between the root collar soil moisture content and field capacity, and the irrigation time is determined by considering the timing before the soil is completely frozen, generating and implementing a precise winter irrigation plan.

[0034] Specifically, field capacity refers to the maximum water content that soil can stably maintain through capillary action after sufficient irrigation or rainfall, draining gravity water, and allowing for a certain period of natural drainage. It reflects the soil's water retention capacity and the upper limit of effective water available for crop absorption and utilization. In this embodiment of the invention, field capacity can be determined through laboratory ring sampler method, in-situ field measurement method, or model estimation based on historical data. Alternatively, corresponding relationships can be pre-established for different soil types and soil depths; this invention does not impose specific limitations. When generating a precision irrigation plan for overwintering, the difference between the soil moisture content in the root collar layer and the field capacity is used as the amount of water to be supplemented, ensuring that the soil moisture content in the root collar layer after irrigation is close to the field capacity. This increases the soil's heat capacity, buffers the impact of winter temperature fluctuations on the root collar layer soil temperature, and reduces damage to the root collar due to freeze-thaw stress.

[0035] In this embodiment of the invention, the actual greening rate is determined by remote sensing images of the alfalfa-covered planting area obtained after the alfalfa greening period. The remote sensing images are image data obtained by imaging the ground surface using a multispectral sensor, with each pixel recording the reflectance information of the corresponding ground location in different spectral bands. Because healthy alfalfa plants have increased chlorophyll content in their leaves after greening, they have a strong absorption effect on the red light band and a strong reflective effect on the near-infrared band. Therefore, areas with different degrees of greening exhibit differentiated spectral characteristics in the remote sensing images. The vegetation index of each pixel within the alfalfa planting area is extracted from the remote sensing images based on these spectral differences to determine the growth status of the surface vegetation.

[0036] Specifically, the actual greening rate is the ratio of the number of pixels in the alfalfa planting area that are determined to have greened up to the total number of effective vegetation pixels before winter. In this process, the vegetation index of each pixel after the greening period is compared with the vegetation index benchmark of the same area before winter. When the vegetation index of a certain pixel rises to the level that indicates that the ground surface at that location has changed from a bare or withered state to a state with vegetation cover, it is determined that the location of that pixel has greened up, thereby obtaining the number of pixels that have greened up.

[0037] Please see Figure 2 The flowchart shown is a big data-based aquaculture environment monitoring method according to an embodiment of the present invention, including: Real-time data collection of air temperature, snow depth, and soil data at several depths of soil layers during the wintering period of alfalfa farming environment. Soil data includes soil temperature and soil moisture content. Feature extraction is performed on the air temperature, snow thickness and soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The predicted greening rate for the current period is predicted based on the freeze-thaw stress data. Based on the predicted greening rate at the beginning of winter and the soil moisture content, a precise winter irrigation plan is generated and implemented, including irrigation time and irrigation amount. To obtain the actual greening rate, remote sensing images of alfalfa were acquired after the greening period ended. The accuracy of monitoring the alfalfa farming environment is determined based on the deviation between the predicted greening rate before the greening period and the actual greening rate. If the monitoring of the alfalfa farming environment is found to be inaccurate, the cause of the anomaly is determined based on the soil temperature attenuation and corresponding processing instructions are issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. After executing the corresponding processing instructions, if the monitoring of the alfalfa farming environment is still inaccurate, the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image. Continuous monitoring of the alfalfa farming environment should be conducted once the accuracy of monitoring for the alfalfa farming environment is confirmed.

[0038] Please see Figure 3 As shown, it is a flowchart of the method described in this embodiment of the invention for determining the accurate status of monitoring the alfalfa farming environment, including: The predicted greening rate before the greening period is obtained based on all freeze-thaw stress data during the overwintering period and is denoted as the expected greening rate. The greening rate prediction deviation rate is calculated based on the difference between the expected greening rate and the actual greening rate and the ratio of the expected greening rate. The accuracy of monitoring the alfalfa farming environment is determined based on the predicted deviation rate of the greening rate.

[0039] In this embodiment of the invention, if the predicted deviation rate of the greening rate is less than the preset predicted deviation rate of the greening rate, it is determined that the monitoring of the alfalfa farming environment is accurate. If the predicted deviation rate of the greening rate is greater than or equal to the preset predicted deviation rate of the greening rate, it is determined that the monitoring of the alfalfa farming environment is inaccurate.

[0040] Specifically, the preset greening rate prediction deviation rate described in this embodiment of the invention is determined by the deviation between the predicted greening rate and the actual greening rate during several historical overwintering periods. If the preset greening rate prediction deviation rate is set too high, the criteria for judging the accuracy of monitoring will be too lenient, potentially misjudging situations with significant prediction deviations as accurate monitoring, failing to trigger timely diagnosis and handling of abnormal causes, leading to persistent problems and affecting the reliability of environmental monitoring during the overwintering period. Conversely, if the preset greening rate prediction deviation rate is set too low, the criteria for judging the accuracy of monitoring will be too strict, potentially misjudging normal prediction errors as inaccurate monitoring, causing the system to frequently enter the abnormal diagnosis and adjustment process, resulting in over-adjustment and adversely affecting alfalfa overwintering management.

[0041] Please see Figure 4 As shown, it is a flowchart of the method described in this embodiment of the invention for determining the reasons for inaccurate monitoring of the alfalfa farming environment, including: Soil temperature in the alfalfa root collar layer and its corresponding collection time were obtained to construct a time-root collar layer soil temperature curve, and air temperature and its corresponding collection time were obtained to construct a time-air temperature curve. In the continuous time of the time-air temperature curve, the turning point where the air temperature changes from a continuous decrease to a continuous increase is identified as a trough, and the turning point where the air temperature changes from a continuous increase to a continuous decrease is identified as a peak. The temperature change during this time period is recorded as a fluctuation event. The highest and lowest values ​​of soil temperature within the same time period were obtained from the time-root collar soil temperature curve. Calculate the peak-to-valley difference of air temperature based on the difference between the peaks and troughs of air temperature. Calculate the peak-valley difference of soil temperature based on the difference between the highest and lowest soil temperature values; Soil temperature attenuation is calculated based on the average ratio of the peak-to-valley difference of air temperature to the peak-to-valley difference of soil temperature during all fluctuation events in the overwintering period. Based on the soil temperature attenuation, the cause of the anomaly is determined, and corresponding processing instructions are issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount.

[0042] In this embodiment of the invention, if the soil temperature attenuation is less than the preset soil temperature attenuation, the fluctuation range of the root collar soil temperature is abnormally large relative to the fluctuation range of the air temperature, and the collected soil temperature data contains high-frequency fluctuation components that are not true temperature characteristics of the root collar. Therefore, the abnormality is determined to be due to the distortion of the collected soil data, and a processing instruction to adjust the filtering window of the collected soil data is issued. If the soil temperature attenuation is greater than or equal to the preset soil temperature attenuation, the attenuation of the root collar soil temperature to air temperature fluctuations is within the normal range, and the collected soil temperature data is reliable. However, the lower predicted greening rate before the greening period indicates a problem during irrigation. Therefore, the abnormality is determined to be due to a deviation in the irrigation plan, and an instruction to adjust the irrigation amount is issued.

[0043] Specifically, the preset soil temperature attenuation degree described in this embodiment of the invention is determined by the historical attenuation of the soil temperature in the root collar layer relative to the air temperature. If the preset soil temperature attenuation degree is set too high, it is easy to be overly sensitive, misjudging normal attenuation fluctuations caused by differences in soil thermal conductivity as data distortion, and frequently triggering filter window adjustments; if the preset soil temperature attenuation degree is set too low, it is easy to be overly insensitive, unable to effectively identify data distortion caused by the mismatch between the collection depth and the actual location of the root collar layer, causing abnormal data to enter the subsequent greening rate prediction and irrigation decision-making stages, affecting the accuracy of monitoring and control.

[0044] In this embodiment of the invention, when the cause of the anomaly is determined to be distortion of the collected soil data, the process of adjusting the soil temperature data filtering window based on soil temperature fluctuation includes: The time-root collar soil temperature curve was processed by moving average to obtain a trend curve; The residual curve is obtained by comparing the temperature values ​​at each moment on the time-root collar soil temperature curve with the trend values ​​at the corresponding moments on the trend curve. The root mean square value of the residual curve is calculated to obtain the soil temperature fluctuation. The soil temperature data filtering window is increased based on the soil temperature fluctuation, and the increase in the soil temperature data filtering window is positively correlated with the soil temperature fluctuation.

[0045] Specifically, the moving average processing involves dividing the curve into several segments according to time sequence, calculating the arithmetic mean of the soil temperature values ​​for each segment, and using this average as the trend value corresponding to the center time of that segment. The trend values ​​of each segment are then connected to obtain the trend curve. After the moving average processing, short-term high-frequency random fluctuations in the original curve are suppressed, while the main trend of root collar soil temperature changes with cold waves and warming processes is preserved.

[0046] Specifically, the greater the soil temperature fluctuation, the stronger the short-term high-frequency noise component superimposed in the root collar soil temperature data, the more serious the data distortion, and the greater the need to increase the filtering window to smooth the data over a larger time span, so as to more fully suppress noise and restore the true trend of root collar soil temperature changes with cold waves and warming processes.

[0047] In this embodiment of the invention, when the cause of the anomaly is determined to be distortion of the collected soil data, the process of further correcting the increase in the soil temperature data filtering window based on the deep reference fluctuation ratio includes: The soil temperature of the deep reference layer of alfalfa and its corresponding collection time were obtained to construct a time-deep reference layer soil temperature curve; The time-deep reference layer soil temperature curve was processed by moving average to obtain the trend curve of the deep reference layer; The residual curve of the deep reference layer is obtained by comparing the temperature value at each moment on the time-deep reference layer soil temperature curve with the trend value at the corresponding moment on the trend curve of the deep reference layer. The root mean square value of the residual curve of the deep reference layer is calculated to obtain the soil temperature fluctuation of the deep reference layer; The deep reference fluctuation ratio is determined based on the ratio of the soil temperature fluctuation of the root collar layer to the soil temperature fluctuation of the deep reference layer. The increase in the soil temperature data filtering window is based on the deep reference fluctuation ratio, and the increase in the soil temperature data filtering window is positively correlated with the deep reference fluctuation ratio.

[0048] Specifically, the larger the deep reference fluctuation ratio, the more prominent the fluctuation of the soil temperature in the root collar layer is relative to the deep reference layer. The more significant the abnormal noise introduced by non-environmental factors such as improper sensor burial depth or poor contact in the root collar layer data, the higher the confidence level of data distortion. Therefore, it is necessary to increase the filtering window by a larger margin to strengthen the suppression of abnormal noise.

[0049] Specifically, in the embodiments of the present invention, after adjusting the soil temperature data filtering window, if the monitoring of the alfalfa farming environment is still inaccurate based on the greening rate prediction deviation rate, a processing instruction to adjust the irrigation amount is issued.

[0050] In this embodiment of the invention, when the cause of the anomaly is determined to be a deviation in the irrigation plan, the process of adjusting the irrigation amount based on the irrigation deterioration ratio includes: The soil temperature of the root collar layer before and after irrigation and the corresponding time were obtained to construct the time-root collar layer soil temperature curve before irrigation and the time-root collar layer soil temperature curve after irrigation. The cumulative duration before irrigation and the cumulative duration after irrigation are determined based on the cumulative duration before and after irrigation when the soil temperature in the root collar layer is lower than the average temperature in the curve. The irrigation deterioration ratio is calculated based on the ratio of the cumulative duration after irrigation to the cumulative duration before irrigation; The irrigation amount is reduced based on the irrigation deterioration ratio, and the reduction in irrigation amount is negatively correlated with the irrigation deterioration ratio.

[0051] Specifically, the greater the irrigation deterioration ratio, the more significant the increase in the duration of the root collar soil temperature being lower than the average temperature value of the period after irrigation compared to before irrigation. Irrigation measures not only fail to effectively alleviate the low temperature stress of the root collar, but also exacerbate the persistent low temperature state. The more severe the irrigation reaction, the more necessary it is to reduce the amount of irrigation to reduce the adverse effects such as soil oversaturation and freezing caused by excessive irrigation, and to avoid further freeze-thaw damage to the root collar.

[0052] In this embodiment of the invention, when the cause of the anomaly is determined to be a deviation in the irrigation plan, the process of further correcting the reduction in irrigation amount based on the warming rate ratio includes: Identify the segments of continuous temperature rise before irrigation in the time-pre-irrigation root collar soil temperature curve. The warming rate of each pre-irrigation rising segment is calculated based on the ratio of the temperature rise to the duration of that rising segment, and the average warming rate before irrigation is calculated based on the average warming rate of each pre-irrigation rising segment. Identify the segments of continuous temperature rise after irrigation in the time-root collar soil temperature curve. The warming rate of each post-irrigation rising segment is calculated based on the ratio of the temperature rise to the duration of that rising segment, and the average warming rate after irrigation is calculated based on the average warming rate of each post-irrigation rising segment. The recovery rate ratio is calculated based on the ratio of the average recovery rate after irrigation to the average recovery rate before irrigation. The reduction in irrigation amount is corrected based on the warming rate ratio, and the reduction in irrigation amount is negatively correlated with the warming rate ratio.

[0053] Specifically, the smaller the warming rate ratio, the slower the soil temperature rises in the root collar layer after irrigation compared to before irrigation. This indicates that the irrigation measures have an adverse effect on the soil heat recovery process. The stronger the irrigation reaction, the greater the need to reduce the amount of irrigation to further reduce the amount of irrigation and avoid further hindering soil warming and prolonging the low temperature exposure time of the root collar due to improper irrigation.

[0054] Specifically, in the embodiments of the present invention, after adjusting the irrigation amount, if the monitoring of the alfalfa farming environment is still inaccurate based on the predicted deviation rate of the greening rate, the method determines that the cause of the anomaly is insufficient spatial resolution of the remote sensing image and issues a processing instruction to adjust the pixel aggregation window of the remote sensing image.

[0055] In this embodiment of the invention, when the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image, the process of adjusting the pixel aggregation window of the remote sensing image based on pixel variability includes: The remote sensing image is evenly divided into several identical sub-regions according to its spatial location; The local variation value of each sub-region is calculated based on the standard deviation of the spectral value of each pixel in each sub-region under the target band. Pixel variability is calculated based on the mean of the local variation values ​​of each sub-region; The pixel aggregation window of the remote sensing image is increased based on the pixel variability, and the increase in the pixel aggregation window of the remote sensing image is positively correlated with the pixel variability.

[0056] Specifically, the greater the pixel variability, the higher the dispersion of pixel spectral values ​​in local areas of remote sensing images, the stronger the local spectral noise caused by the mixed pixel effect, the worse the image data quality, and the greater the need to increase the pixel aggregation window to average pixel spectral values ​​over a larger spatial range, thereby more effectively suppressing local noise and stabilizing regional spectral characteristics.

[0057] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for monitoring aquaculture environments based on big data, characterized in that, include: Real-time data collection of air temperature, snow thickness, and soil data at several depths of soil layers during the wintering period of alfalfa farming environment. Soil data includes soil temperature and soil moisture content. Feature extraction is performed on the air temperature, snow thickness and soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The predicted greening rate for the current period is predicted based on the freeze-thaw stress data. Based on the predicted greening rate at the beginning of winter and the soil moisture content, a precise winter irrigation plan is generated and implemented, including irrigation time and irrigation amount. To obtain the actual greening rate, remote sensing images of alfalfa were acquired after the greening period ended. The accuracy of monitoring the alfalfa farming environment is determined based on the deviation between the predicted greening rate before the greening period and the actual greening rate. If the monitoring of the alfalfa farming environment is found to be inaccurate, the cause of the anomaly is determined based on the soil temperature attenuation and corresponding processing instructions are issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. After executing the corresponding processing instructions, if the monitoring of the alfalfa farming environment is still inaccurate, the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image. Continuous monitoring of the alfalfa farming environment should be conducted once the accuracy of monitoring for the alfalfa farming environment is confirmed.

2. The method for monitoring aquaculture environments based on big data according to claim 1, characterized in that, The process of determining the accurate status of monitoring for alfalfa farming environments based on the prediction deviation rate of the greening rate includes: The predicted greening rate before the greening period is obtained based on all freeze-thaw stress data during the overwintering period and is denoted as the expected greening rate. The predicted greening rate deviation rate is calculated based on the expected greening rate and the actual greening rate. The accuracy of monitoring the alfalfa farming environment is determined based on the predicted deviation rate of the greening rate.

3. The method for monitoring aquaculture environments based on big data according to claim 2, characterized in that, When monitoring of the alfalfa farming environment is found to be inaccurate, the process of determining the cause of the anomaly based on soil temperature attenuation includes: Soil temperature in the alfalfa root collar layer and its corresponding collection time were obtained to construct a time-root collar layer soil temperature curve, and air temperature and its corresponding collection time were obtained to construct a time-air temperature curve. In the continuous time of the time-air temperature curve, the turning point where the air temperature changes from a continuous decrease to a continuous increase is identified as a trough, and the turning point where the air temperature changes from a continuous increase to a continuous decrease is identified as a peak. The temperature change during this time period is recorded as a fluctuation event. The highest and lowest values ​​of soil temperature within the same time period were obtained from the time-root collar soil temperature curve. Calculate the peak-valley difference of air temperature based on air temperature peaks and valleys; Calculate the peak-valley difference of soil temperature based on the highest and lowest soil temperature values; The soil temperature attenuation rate is calculated based on the peak-to-valley difference of air temperature and the peak-to-valley difference of soil temperature during all fluctuation events in the overwintering period. Based on the soil temperature attenuation, the cause of the anomaly is determined and a corresponding processing instruction is issued, including adjusting the filter window of the collected soil data and adjusting the irrigation amount.

4. The method for monitoring aquaculture environments based on big data according to claim 3, characterized in that, The process of adjusting the soil temperature data filtering window based on soil temperature fluctuation includes: The time-root collar soil temperature curve was processed by moving average to obtain a trend curve; The residual curve is determined based on the temperature values ​​at each moment on the time-root collar soil temperature curve and the trend values ​​at the corresponding moments on the trend curve. Soil temperature fluctuation is determined based on the residual curve. The soil temperature data filtering window is increased based on the soil temperature fluctuation, and the increase in the soil temperature data filtering window is positively correlated with the soil temperature fluctuation.

5. The method for monitoring aquaculture environments based on big data according to claim 4, characterized in that, The process of further adjusting the increase in the soil temperature data filtering window based on the deep baseline fluctuation ratio includes: The soil temperature of the deep reference layer of alfalfa and its corresponding collection time were obtained to construct a time-deep reference layer soil temperature curve; The time-deep reference layer soil temperature curve was processed by moving average to obtain the trend curve of the deep reference layer; The residual curve of the deep reference layer is determined based on the temperature values ​​at each moment on the time-deep reference layer soil temperature curve and the trend values ​​at the corresponding moments on the trend curve of the deep reference layer. The soil temperature fluctuation of the deep reference layer is determined based on the residual curve of the deep reference layer. The deep reference fluctuation ratio is determined based on the soil temperature fluctuation of the root collar layer and the soil temperature fluctuation of the deep reference layer. The increase in the soil temperature data filtering window is based on the deep reference fluctuation ratio, and the increase in the soil temperature data filtering window is positively correlated with the deep reference fluctuation ratio.

6. The method for monitoring aquaculture environments based on big data according to claim 5, characterized in that, After adjusting the soil temperature data filtering window, if the monitoring of the alfalfa farming environment is still inaccurate based on the greening rate prediction deviation rate, an instruction to adjust the irrigation amount is issued.

7. The method for monitoring aquaculture environments based on big data according to claim 6, characterized in that, The process of adjusting irrigation amount based on irrigation deterioration ratio includes: The soil temperature of the root collar layer before and after irrigation and the corresponding time were obtained to construct the time-root collar layer soil temperature curve before irrigation and the time-root collar layer soil temperature curve after irrigation. The cumulative duration before irrigation and the cumulative duration after irrigation are determined based on the cumulative duration before and after irrigation when the soil temperature in the root collar layer is lower than the average temperature in the curve. The irrigation deterioration ratio is calculated based on the cumulative duration after irrigation and the cumulative duration before irrigation; The irrigation amount is reduced based on the irrigation deterioration ratio, and the reduction in irrigation amount is negatively correlated with the irrigation deterioration ratio.

8. The method for monitoring aquaculture environments based on big data according to claim 7, characterized in that, The process of further adjusting the reduction in irrigation amount based on the warming rate ratio includes: Identify the segments of continuous temperature rise before irrigation in the time-pre-irrigation root collar soil temperature curve. The temperature recovery rate of each pre-irrigation rising segment is determined based on the temperature rise amplitude and duration of each rising segment, and the average pre-irrigation temperature recovery rate is calculated based on the temperature recovery rate of each pre-irrigation rising segment. Identify the segments of continuous temperature rise after irrigation in the time-root collar soil temperature curve. The temperature recovery rate of each post-irrigation rising segment is determined based on the temperature rise amplitude and duration of each rising segment, and the average temperature recovery rate after irrigation is calculated based on the temperature recovery rate of each post-irrigation rising segment. The ratio of the recovery rate is calculated based on the average recovery rate after irrigation and the average recovery rate before irrigation. The reduction in irrigation amount is corrected based on the warming rate ratio, and the reduction in irrigation amount is negatively correlated with the warming rate ratio.

9. The method for monitoring aquaculture environments based on big data according to claim 8, characterized in that, If, after adjusting the irrigation amount, the monitoring of the alfalfa farming environment is still inaccurate based on the predicted deviation rate of the greening rate, the cause of the anomaly is determined to be insufficient spatial resolution of the remote sensing image, and a processing instruction to adjust the pixel aggregation window of the remote sensing image is issued.

10. A big data-based aquaculture environment monitoring system using the method described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to collect real-time data on air temperature, snow thickness, and soil data at several depths in the alfalfa growing environment during the wintering period. The soil data includes soil temperature and soil moisture content. The feature extraction module is used to extract features from the air temperature, the snow thickness and the soil data to obtain overwintering freeze-thaw stress data, including the number of freeze-thaw cycles in the root collar soil, the root collar soil temperature and the corresponding time, and the number of days of snow cover. The greening rate prediction module is used to predict the greening rate for the current period based on the freeze-thaw stress data. The irrigation scheme decision module is used to generate and execute a precise winter irrigation scheme based on the predicted greening rate at the beginning of winter, including irrigation time and irrigation amount. The actual greening acquisition module is used to acquire remote sensing images of alfalfa after the greening period ends in order to obtain the actual greening rate. An assessment module is used to determine the accurate status of monitoring of the alfalfa farming environment based on the deviation between the predicted greening rate before the greening period and the actual greening rate. The anomaly analysis module is used to determine the cause of the anomaly based on the obtained soil temperature attenuation when the monitoring is inaccurate, and to issue corresponding processing instructions, including adjusting the filter window of the collected soil data and adjusting the irrigation amount. The execution module is used to execute the processing instructions issued by the exception analysis module.

Images