Ecological pond water distribution method and regulation system based on internet of things

Abstract

The application relates to the technical field of water distribution regulation, and discloses an ecological pond water distribution method and a regulation system based on the Internet of Things. The method comprises the following steps: collecting water quality parameters at different depths of an ecological pond through vertical layered sensors to form a depth gradient data set; quantitatively analyzing the layered state of the water body by using a thermocline identification algorithm to obtain a thermocline position, a vertical gradient of dissolved oxygen and dynamic variables of a layered state of nutrient salts; comparing the dynamic variables with preset threshold values to generate a layered abnormal trigger signal and a target water distribution depth; cooperatively scheduling multi-depth water distribution equipment to generate water distribution execution instructions; executing directional water distribution treatment to obtain water quality improvement effect data and update a parameter optimization database. The application solves the problems that the layered state of the water body cannot be accurately identified, targeted water distribution control is lacked and the water distribution effect is difficult to quantitatively optimize in ecological pond water quality management.

Description

Technical Field

[0001] This application relates to the field of water distribution and regulation technology, and in particular to an ecological pond water distribution method and regulation system based on the Internet of Things. Background Technology

[0002] Existing ecological pond water quality management technologies mainly rely on regular manual testing and uniform water distribution for water quality regulation. This involves adding water conditioners or aeration equipment to the surface or at a fixed depth of the ecological pond to improve the overall water quality. Some advanced water quality management equipment can achieve automated water quality monitoring and basic water quality regulation functions, but these technologies still have significant limitations in water stratification identification, precise water distribution control, and effect evaluation.

[0003] The main shortcomings of existing technologies include: the inability to accurately identify and quantify the vertical stratification of water bodies, especially the dynamic changes in the thermocline and the vertical gradient distribution of dissolved oxygen; the lack of precise water distribution control capabilities for water quality anomalies at different depths, and the inability to adjust water distribution strategies according to the type of stratification anomaly; and the lack of real-time monitoring and quantitative evaluation mechanisms for water distribution effects, making it impossible to establish the correlation between water distribution parameters and improvement effects, resulting in blind and inefficient water distribution operations.

[0004] Because existing technologies cannot accurately identify and quantify the stratification of water bodies, it is impossible to formulate corresponding water distribution strategies based on different types of stratification anomalies. Furthermore, the lack of real-time effect monitoring and parameter optimization mechanisms during the water distribution process leads to progressive technical problems in ecological pond water quality management, such as insufficient water distribution accuracy, unpredictable effects, and unreasonable parameter configuration. These problems are interconnected and deepen at each level, seriously affecting the scientific nature and effectiveness of ecological pond water quality management. Summary of the Invention

[0005] This application provides an IoT-based method and control system for water distribution in ecological ponds, which addresses the problems of inaccurate identification of water stratification, lack of targeted water distribution control, and difficulty in quantifying and optimizing water distribution effects in ecological pond water quality management.

[0006] In a first aspect, this application provides an IoT-based method for water distribution in an ecological pond, which includes: collecting and processing water quality parameters at different depths of the ecological pond using vertically layered sensors to obtain a depth gradient dataset containing temperature, dissolved oxygen, pH value and turbidity.

[0007] Based on the depth gradient dataset, the water stratification state is quantitatively analyzed using a thermocline identification algorithm to obtain dynamic variables of thermocline location, dissolved oxygen vertical gradient, and nutrient stratification state. These variables are then compared with corresponding preset thresholds to obtain stratification anomaly trigger signals and corresponding target water distribution depths. Based on the stratification anomaly trigger signals, multi-depth water distribution equipment is coordinated and scheduled to obtain water distribution execution instructions containing water distribution depth, flow rate, and duration. These instructions are then used to perform directional water distribution at the target water distribution depth, yielding water quality improvement data and updating the water distribution parameter optimization database.

[0008] Secondly, this application provides an IoT-based ecological pond water distribution and control system, which includes:

[0009] The data acquisition module is used to collect and process water quality parameters at different depths of the ecological pond through vertical stratified sensors, and obtain a depth gradient dataset including temperature, dissolved oxygen, pH value and turbidity.

[0010] The quantification module is used to perform quantitative analysis and processing of the water stratification state based on the depth gradient dataset using a thermocline identification algorithm, and to obtain dynamic variables of thermocline location, dynamic variables of dissolved oxygen vertical gradient, and dynamic variables of nutrient stratification state.

[0011] The judgment module is used to compare and judge the dynamic variables of the thermocline position, the dynamic variables of the dissolved oxygen vertical gradient, and the dynamic variables of the nutrient stratification state with the corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth.

[0012] The scheduling module is used to perform coordinated scheduling processing on the multi-depth water distribution equipment according to the layered abnormality trigger signal, and obtain water distribution execution instructions including water distribution depth, flow rate and duration.

[0013] The water distribution module is used to perform directional water distribution treatment on the water body at the target water distribution depth through the water distribution execution command, obtain water quality improvement effect data, and update the water distribution parameter optimization database.

[0014] Thirdly, an IoT-based ecological pond water distribution device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor invokes the instructions in the memory to cause the IoT-based ecological pond water distribution device to execute the aforementioned IoT-based ecological pond water distribution method.

[0015] Fourthly, a computer-readable storage medium is provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the above-described IoT-based ecological pond water distribution method.

[0016] The technical solution provided in this application utilizes a vertical stratification sensor to collect and process water quality parameters at different depths in the ecological pond. This enables comprehensive perception of the vertical distribution of water bodies. Compared to traditional single-point or surface monitoring methods, it can obtain a depth gradient dataset containing temperature, dissolved oxygen, pH, and turbidity, providing a complete data foundation for subsequent accurate analysis. Simultaneously, the thermocline identification algorithm, as a core technical feature, accurately identifies the water stratification state and extracts dynamic variables of thermocline location, dissolved oxygen vertical gradient, and nutrient stratification state through quantitative analysis of the depth gradient dataset. This algorithm employs a method of comparing the average temperature gradient of three consecutive layers and identifying gradient abrupt change points, significantly improving the accuracy and stability of thermocline location and solving the technical problem of inaccurate thermocline identification in traditional methods. Furthermore, the technical feature of comparing dynamic variables with preset thresholds establishes a quantitative anomaly judgment standard, avoiding the subjectivity and inconsistency of manual judgment, and enabling timely and accurate identification of stratification anomalies and determination of target water distribution depth.

[0017] The multi-depth water distribution equipment collaborative scheduling technology, through refined control links such as equipment matching, depth positioning, flow allocation, and timing coordination, realizes differentiated water distribution strategies for different types of layered anomalies. Compared with the traditional unified water distribution method, it can formulate precise water distribution execution instructions including water distribution depth, flow rate, and duration based on the specific anomaly type and location, significantly improving the targeting and efficiency of water distribution operations. The technology feature of directional water distribution combined with real-time water quality monitoring not only achieves precise treatment of water bodies at target depths but also tracks the water distribution effect in real time. Through data processing methods such as difference calculation, improvement degree quantification, and water distribution efficiency calculation, a quantitative relationship between water distribution parameters and improvement effects is established. The application of machine learning models further enhances the system's self-optimization capability. Through parameter optimization and strategy update processing, the water distribution parameter optimization database is continuously improved, forming a closed-loop intelligent water quality management system. This fundamentally solves the technical problems of strong blindness in water distribution, unpredictable effects, and unreasonable parameter configuration in traditional ecological pond management, realizing precise, intelligent, and efficient water quality management of ecological ponds. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of one embodiment of the IoT-based ecological pond water distribution method in this application.

[0020] Figure 2 This is a schematic diagram of one embodiment of the IoT-based ecological pond water distribution and control system in this application.

[0021] Figure 3 This is a schematic block diagram of the structure of the IoT-based ecological pond water distribution device in an embodiment of the present invention. Detailed Implementation

[0022] This application provides an IoT-based method and control system for water distribution in an ecological pond. The terms "first," "second," "third," "fourth," etc. (if applicable) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0023] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the IoT-based ecological pond water distribution method in this application includes:

[0024] Step S101: Collect and process water quality parameters at different depths of the ecological pond using a vertical stratified sensor to obtain a depth gradient dataset containing temperature, dissolved oxygen, pH value and turbidity.

[0025] Step S102: Based on the depth gradient dataset, the water stratification state is quantitatively analyzed using the thermocline identification algorithm to obtain dynamic variables of thermocline location, dissolved oxygen vertical gradient, and nutrient stratification state.

[0026] Step S103: Compare and judge the dynamic variables of thermocline position, dissolved oxygen vertical gradient, and nutrient stratification state with the corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth.

[0027] Step S104: Perform coordinated scheduling processing on the multi-depth water distribution equipment according to the layered abnormality trigger signal to obtain a water distribution execution command containing water distribution depth, flow rate and duration;

[0028] Step S105: Perform directional water distribution treatment on the water body at the target water distribution depth by executing the water distribution command, obtain water quality improvement effect data, and update the water distribution parameter optimization database.

[0029] It is understood that the executing entity of this application can be an IoT-based ecological pond water distribution and control system, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiments use a server as an example for illustration.

[0030] Specifically, the IoT-based ecological pond water distribution method collects and processes water quality parameters at different depths in the ecological pond using vertically layered sensors. First, sensor nodes, including temperature, dissolved oxygen, pH, and turbidity sensors, are deployed at 0.5-meter intervals within the ecological pond. The raw data collected by each sensor node is time-calibrated using a clock synchronization protocol to ensure consistency of data at different depths in the time dimension. Then, depth identification and data association processing are performed based on the synchronized water quality data, establishing a mapping relationship between the water quality parameters of each depth layer and their corresponding depth coordinates to form a water quality parameter matrix arranged by depth. Next, the water quality parameter matrix undergoes data quality inspection and outlier removal. Valid water quality data is selected through the established data quality standards and then classified and integrated according to four dimensions: temperature, dissolved oxygen, pH, and turbidity, forming a depth gradient dataset.

[0031] In the process of quantitatively analyzing the water stratification state using a thermocline identification algorithm based on a depth gradient dataset, the thermocline identification algorithm is a water stratification identification method based on temperature gradient change characteristics. This algorithm first arranges the temperature data in the depth gradient dataset according to a depth sequence from shallow to deep, forming a temperature-depth distribution sequence. Then, it performs layer-by-layer temperature difference calculation and gradient change rate analysis on the temperature-depth distribution sequence. By calculating the temperature difference between adjacent depth layers, it determines the temperature gradient distribution data between each depth layer. Next, based on the temperature gradient distribution data, it performs a comparison of the mean temperature gradient of three consecutive layers and identifies gradient abrupt change points. The process involves grouping temperature gradient distribution data into three consecutive layers according to depth order, calculating the arithmetic mean of the three gradient values ​​in each group to obtain the three-layer gradient mean at each depth. By calculating the difference and analyzing the rate of change of the three-layer gradient mean at adjacent depths, the depth range of gradient abrupt change is identified. Within the depth range of temperature gradient abrupt change, the dynamic variables of the thermocline position are determined through precise positioning and depth interpolation. At the same time, the dissolved oxygen data in the depth gradient dataset is processed by interlayer difference accumulation calculation, and the pH and turbidity data are processed by distribution variance statistics to obtain the dynamic variables of dissolved oxygen vertical gradient and nutrient stratification state.

[0032] When comparing and judging the dynamic variables of thermocline location, dissolved oxygen vertical gradient, and nutrient stratification status with corresponding preset thresholds, the preset thresholds are critical values ​​determined based on ecological pond water quality management standards and historical data statistics. The comparison and judgment process first compares the dynamic variable of thermocline location with the thermocline location threshold. When the dynamic variable of thermocline location exceeds the preset range, a thermocline anomaly judgment result and a thermocline trigger flag are generated. At the same time, the dynamic variable of dissolved oxygen vertical gradient is compared with the dissolved oxygen gradient threshold. When the dissolved oxygen vertical gradient exceeds the normal range, an anoxic stratification judgment result and an anoxic trigger flag are generated. The dynamic variable of nutrient stratification status is compared with the nutrient stratification threshold. When the nutrient stratification status is abnormal, a nutrient anomaly judgment result and a nutrient trigger flag are generated. Then, based on the thermocline trigger flag, the anoxic trigger flag, and the nutrient trigger flag, logical combination and priority sorting are performed. The stratification anomaly trigger signal is determined according to the severity and urgency of different anomaly types. According to the trigger type of the stratification anomaly trigger signal, the corresponding depth layer location is queried and the depth coordinate is extracted to determine the target water distribution depth.

[0033] In the process of collaborative scheduling of multi-depth water distribution equipment based on the stratified anomaly trigger signal, the multi-depth water distribution equipment includes a variable depth water distributor and a flow regulating valve group. The collaborative scheduling process first performs equipment matching and water distribution mode selection on the trigger type in the stratified anomaly trigger signal. According to different anomaly types, the corresponding water distribution operation mode is selected and the equipment call list is determined. Based on the target water distribution depth, the variable depth water distributor is depth-positioned and position-adjusted. Through a precise depth control mechanism, the water distributor is adjusted to the specified depth position to obtain precise depth coordinates and equipment positioning confirmation signals. According to the water distribution operation mode, the flow regulating valve group is opened and the flow is distributed. The valve opening is determined by calculating the water distribution flow required for each depth layer, and the water distribution flow parameters and valve control commands for each depth layer are obtained. The equipment positioning confirmation signal and valve control commands are time-coordinated and synchronously started to ensure that multiple devices work collaboratively in the correct time sequence. Based on the multi-device collaborative operation schedule, the water distribution duration is calculated and the stop condition is set to form a water distribution execution command that includes water distribution depth, flow rate and duration.

[0034] When targeted water distribution is performed on water bodies at a target depth using water distribution execution commands, targeted water distribution is a process of precisely delivering treated water to a specific depth layer. This process involves targeted water distribution operations and real-time water quality monitoring based on water distribution execution commands. Water quality parameter changes are monitored simultaneously during the water distribution process to obtain water quality change data. The difference between the water quality change data and the baseline water quality value before water distribution is calculated to quantify the degree of improvement. Improvement quantification indicators are calculated by comparing changes in temperature, dissolved oxygen, pH, and turbidity before and after water distribution. Water distribution efficiency is calculated and parameter correlation analysis is performed on these improvement quantification indicators to analyze the correlation between water distribution depth, flow rate, duration, and the degree of water quality improvement. The correlation data is input into a machine learning model for parameter optimization and strategy updates. The machine learning model optimizes the water distribution strategy by analyzing historical water distribution data and effect data to obtain water quality improvement effect data. Based on the water quality improvement effect data, data is written to the water distribution parameter optimization database and model parameters are updated to continuously improve the water distribution strategy and parameter configuration.

[0035] In one specific embodiment, the process of performing step S101 may specifically include the following steps:

[0036] Vertically layered sensors were deployed in the ecological pond at 0.5-meter intervals to obtain sensor nodes that include temperature sensors, dissolved oxygen sensors, pH sensors, and turbidity sensors.

[0037] The raw data collected by the sensor nodes is time-calibrated using a clock synchronization protocol to obtain synchronized water quality data.

[0038] Based on synchronous water quality data, depth identification and data association processing are performed to obtain a water quality parameter matrix arranged in depth layers;

[0039] Data quality checks and outlier removal were performed on the water quality parameter matrix to obtain valid water quality data.

[0040] The effective water quality data were classified and integrated according to four dimensions: temperature, dissolved oxygen, pH value, and turbidity, to obtain a deep gradient dataset.

[0041] Specifically, when vertically layered sensors are deployed in the ecological pond at 0.5-meter intervals, these sensors are multi-layered sensor arrays installed vertically along the water body. By deploying sensor nodes at different depths, comprehensive monitoring of the water body's stratification is achieved. Each sensor node integrates four water quality detection devices: a temperature sensor, a dissolved oxygen sensor, a pH sensor, and a turbidity sensor. The temperature sensor detects changes in water temperature, the dissolved oxygen sensor measures the oxygen content in the water, the pH sensor monitors the acidity or alkalinity of the water, and the turbidity sensor detects the degree of turbidity. The sensor nodes are arranged at fixed intervals of 0.5 meters, starting from the surface of the ecological pond and proceeding downwards, ensuring the coverage and continuity of water quality parameter collection at each depth.

[0042] The process of time-calibrating the raw data collected by sensor nodes using a clock synchronization protocol is a communication protocol that ensures the time consistency of all nodes in a distributed sensor network. This protocol eliminates time deviations between different sensor nodes by setting a unified time base in the sensor network. The raw data includes temperature, dissolved oxygen, pH, and turbidity values ​​collected by each sensor node at different times. The time calibration process establishes the correspondence between data and collection time by adding a precise timestamp to each data point. Synchronized water quality data refers to a set of water quality parameter data with a unified time base after time calibration. This data set ensures that the data collected by sensor nodes at different depths at the same time have time consistency.

[0043] When performing depth labeling and data association processing based on synchronous water quality data, depth labeling refers to marking the corresponding depth coordinate information for each water quality data point. Data association processing is the process of establishing a mapping relationship between water quality parameters and their spatial location and temporal information. This process first reads the sensor node location information in the synchronous water quality data, then assigns a corresponding depth label to each data point according to the installation depth of the sensor node, and then establishes the association relationship between the data point and the depth coordinate, forming a multi-dimensional data structure containing depth, time, temperature, dissolved oxygen, pH value and turbidity information. The water quality parameter matrix arranged by depth layer arranges the associated data in order from shallow to deep, forming a two-dimensional data matrix structure with depth as the row and water quality parameters as the column.

[0044] In the process of data quality inspection and outlier removal of the water quality parameter matrix, data quality inspection is the process of evaluating the accuracy, completeness, and rationality of the data through established quality standards. This process includes three aspects: data range inspection, data continuity inspection, and data consistency inspection. Data range inspection identifies obviously erroneous data by judging whether the water quality parameter values ​​are within a reasonable physical range. Data continuity inspection identifies abnormally jumping data points by analyzing the variation range of data at adjacent time points or adjacent depth layers. Data consistency inspection identifies logically contradictory data by comparing the correlation between different parameters at the same depth. Outlier removal uses statistical methods to calculate the standard deviation and mean of each parameter, and marks data points exceeding three times the standard deviation range as outliers and removes them from the dataset. Valid water quality data refers to the reliable set of data retained after quality inspection and outlier removal.

[0045] When classifying and integrating effective water quality data according to four dimensions—temperature, dissolved oxygen, pH, and turbidity—the classification and integration process refers to the reorganization and classification of mixed water quality parameter data according to parameter type. This process first extracts temperature data from the effective water quality data and arranges it in depth order to form a temperature depth distribution array. Then, it extracts dissolved oxygen data and arranges it in depth order to form a dissolved oxygen depth distribution array. Next, it extracts pH data and arranges it in depth order to form a pH depth distribution array. Finally, it extracts turbidity data and arranges it in depth order to form a turbidity depth distribution array. The depth gradient dataset integrates the four depth distribution arrays according to a unified depth coordinate system to form a comprehensive data structure containing depth variation information of all water quality parameters. This data structure uses depth as an index, with each depth position corresponding to a complete set of water quality parameter values.

[0046] In one specific embodiment, the process of performing step S102 may specifically include the following steps:

[0047] Arrange the temperature data in the depth gradient dataset according to the depth sequence from shallow to deep to obtain the temperature depth distribution sequence;

[0048] By performing layer-by-layer temperature difference calculation and gradient change rate analysis on the temperature depth distribution sequence, the temperature gradient distribution data between layers at each depth is obtained.

[0049] Based on the temperature gradient distribution data, the mean of three consecutive temperature gradients is compared and gradient abrupt change points are identified to obtain the temperature gradient abrupt change depth range.

[0050] The maximum gradient point within the temperature gradient abrupt change depth range is precisely located and depth interpolation is performed to obtain the dynamic variable of the thermocline position.

[0051] The dissolved oxygen data in the depth gradient dataset were processed by accumulating the interlayer difference and the pH turbidity data were processed by statistical analysis of the distribution variance to obtain the dynamic variables of the dissolved oxygen vertical gradient and the nutrient stratification state.

[0052] Specifically, when arranging the temperature data in the depth gradient dataset according to the depth sequence from shallow to deep, the temperature depth distribution sequence refers to the ordered data sequence formed by reorganizing the temperature values ​​of each depth layer in ascending order of depth coordinates. This sequence uses depth as the index and temperature as the value, forming a one-dimensional array structure that reflects the vertical distribution characteristics of water temperature. The arrangement process first reads all temperature data points in the depth gradient dataset, extracts the depth coordinates and temperature values ​​corresponding to each data point, and then reorders the temperature data in ascending order of depth coordinates to form a sequence of correspondence between depth and temperature.

[0053] When performing layer-by-layer temperature difference calculation and gradient rate of change analysis on the temperature depth distribution sequence, layer-by-layer temperature difference calculation refers to the process of calculating the temperature difference between adjacent depth layers. This calculation process reads the temperature values ​​of two adjacent depth positions in the temperature depth distribution sequence and calculates the difference between them to reflect the temperature change amplitude. Gradient rate of change analysis refers to the process of dividing the temperature difference by the corresponding depth interval to calculate the temperature gradient. The temperature gradient represents the rate of temperature change with depth, and the calculation formula is: temperature gradient equals temperature difference divided by depth interval. The temperature gradient distribution data between each depth layer is a data set containing the temperature gradient values ​​between all adjacent depth layers. This data set reflects the distribution law of the intensity of water temperature change in the vertical direction.

[0054] When performing three-layer continuous temperature gradient mean comparison and gradient abrupt change point identification based on temperature gradient distribution data, the three-layer continuous temperature gradient mean comparison refers to grouping the temperature gradient distribution data into three consecutive layers according to depth order, calculating the arithmetic mean of the three gradient values ​​in each group to obtain the three-layer gradient mean at each depth position, and then calculating the difference and rate of change of the three-layer gradient mean at adjacent depth positions to form a gradient mean change sequence. Gradient abrupt change point identification involves setting a gradient change threshold, comparing the gradient mean change sequence against the threshold, and marking abrupt change points. When the gradient mean change exceeds the preset threshold, it is marked as a gradient abrupt change point, and the corresponding depth coordinates are recorded. The starting depth and ending depth are determined based on the continuity analysis of the gradient abrupt change flag. The temperature gradient abrupt change depth range refers to the continuous depth interval containing the gradient abrupt change point.

[0055] When performing precise location and depth interpolation calculations for the maximum gradient point within the temperature gradient abrupt change depth range, precise location refers to finding the depth position with the largest gradient value within the temperature gradient abrupt change depth range. This process determines the maximum gradient point by traversing all temperature gradient values ​​within the abrupt change depth range and comparing the magnitudes of each gradient value. The depth interpolation calculation process uses a linear interpolation method to perform refined calculations near the maximum gradient point. By analyzing the temperature change trend before and after the maximum gradient point, a more accurate thermocline depth position is determined. The thermocline position dynamic variable is a numerical parameter representing the specific depth position of the thermocline in the water body. This parameter reflects the core positional characteristics of water temperature stratification.

[0056] When performing interlayer difference accumulation calculation on dissolved oxygen data in the depth gradient dataset, the interlayer difference accumulation calculation refers to the process of calculating the difference between the dissolved oxygen value of each depth layer and the dissolved oxygen value of adjacent depth layers, and then summing these differences. This calculation process first extracts the dissolved oxygen values ​​of all depth layers in the depth gradient dataset, calculates the dissolved oxygen difference between adjacent layers in depth order, and then accumulates and sums all the interlayer differences to obtain a cumulative index reflecting the intensity of vertical change of dissolved oxygen. The dynamic variable of dissolved oxygen vertical gradient is a quantitative parameter representing the distribution and change characteristics of dissolved oxygen in the vertical direction of the water body.

[0057] When performing distribution variance statistical processing on pH and turbidity data, distribution variance statistics refers to a statistical analysis method that calculates the dispersion of pH and turbidity data at different depth layers. This process first extracts pH and turbidity data from all depth layers in the depth gradient dataset, calculates the mean and variance of the pH and turbidity data, and then combines the pH and turbidity variances to form a comprehensive index reflecting the nutrient stratification state. The dynamic variable of nutrient stratification state is a quantitative parameter representing the vertical distribution state of nutrients in the water body. This parameter indirectly reflects the stratification characteristics of nutrients through the distribution variance of pH and turbidity.

[0058] In one specific embodiment, the process of performing the comparison of the mean of three consecutive temperature gradients and the identification of gradient abrupt change points based on the temperature gradient distribution data can specifically include the following steps:

[0059] The temperature gradient distribution data is grouped into three consecutive layers according to depth order to obtain combined temperature gradient data of three consecutive layers.

[0060] The arithmetic mean of the three gradient values ​​in each group of three consecutive temperature gradient combination data is calculated to obtain the three-layer gradient mean at each depth location.

[0061] The gradient mean values ​​of three adjacent depth locations are calculated by difference and analyzed by rate of change to obtain a gradient mean value change sequence.

[0062] The gradient mean change sequence is subjected to threshold comparison and abrupt point marking to obtain gradient abruptness flags and corresponding depth coordinates.

[0063] The starting and ending depths are determined by analyzing the continuity of gradient mutation flags to obtain the temperature gradient mutation depth range.

[0064] Specifically, when processing temperature gradient distribution data into three consecutive layers according to depth order, three consecutive layers grouping refers to the processing method of grouping the gradient values ​​in the temperature gradient distribution data into a data group of every three adjacent gradient values ​​according to depth order. This processing method uses a sliding window approach, starting from the first gradient value of the temperature gradient distribution data, selecting three consecutive gradient values ​​as a group, then moving down one position, and selecting the next group of three consecutive gradient values, repeating this process until all gradient data has been traversed. The combined temperature gradient data of three consecutive layers refers to the collection of multiple three-element gradient data groups obtained through grouping processing. Each data group contains temperature gradient values ​​from three adjacent depth layers. This grouping method can capture the variation characteristics of the temperature gradient within a local depth range. When calculating the arithmetic mean of the three gradient values ​​in each group of three consecutive temperature gradient combination data, the arithmetic mean calculation refers to the mathematical operation of adding the three gradient values ​​and dividing by three. This calculation process first reads the three temperature gradient values ​​in each data group, sums the three values, and then divides the sum by three to obtain the average value of the data group. The three-layer gradient mean at each depth position refers to the arithmetic mean of each consecutive three-layer data group. This mean reflects the average level of the temperature gradient within the local depth range. The average value calculation can reduce the impact of random fluctuations of individual gradient values ​​on the overall trend judgment.

[0065] When performing difference calculation and rate of change analysis on the three-layer gradient mean values ​​at adjacent depth locations, difference calculation refers to the process of calculating the numerical difference between the three-layer gradient mean values ​​at two adjacent depth locations. This calculation process involves reading the three-layer gradient mean value sequence arranged in depth order and calculating the difference between every two adjacent mean values. Rate of change analysis refers to the process of dividing the gradient mean difference by the corresponding depth interval to calculate the gradient mean change rate. The gradient mean change sequence is a data sequence containing the gradient mean change rate of all adjacent depth locations. This sequence reflects the dynamic change characteristics of the temperature gradient in the vertical direction.

[0066] When performing threshold comparison and abrupt change point marking on the gradient mean change sequence, threshold comparison refers to the process of comparing each rate of change value in the gradient mean change sequence with a preset rate of change threshold. The preset threshold is a critical value determined based on the normal temperature gradient change law of water bodies. When the rate of change value exceeds the preset threshold, it indicates that there is an abnormal gradient change at that depth location. Abrupt change point marking refers to the process of marking the depth location that exceeds the threshold and recording its depth coordinates. The gradient abrupt change flag is a Boolean flag indicating whether there is a gradient abrupt change, and the corresponding depth coordinates are the specific depth location of the abrupt change point in the water body. Through threshold comparison and marking, the depth location where the temperature gradient changes significantly can be accurately identified.

[0067] When determining the starting and ending depths based on the continuity analysis of gradient mutation flags, the continuity analysis refers to the processing method that examines the continuous distribution characteristics of gradient mutation flags in the depth sequence. This analysis method searches for a sequence of flags that are continuously true by traversing all gradient mutation flags. The starting depth determination process refers to finding the depth coordinates corresponding to the first true flag in the continuous mutation flag sequence, and the ending depth determination process refers to finding the depth coordinates corresponding to the last true flag in the continuous mutation flag sequence. The temperature gradient mutation depth range refers to the continuous depth interval from the starting depth to the ending depth, which includes all depth locations where temperature gradient mutations occur.

[0068] For example, during the summer stratification of an ecological pond, temperature gradient distribution data showed that the temperature gradient values ​​at each depth layer were unevenly distributed. Three-layer grouping processing organized the gradient data into multiple three-element data groups using a sliding window method. Arithmetic mean calculation was performed on each group to obtain a smooth three-layer gradient mean sequence. Difference calculation revealed that the difference between adjacent gradient means significantly increased within a specific depth range. Rate of change analysis showed that the rate of change of the gradient mean in this depth range far exceeded that of other depths. Threshold comparison processing compared the rate of change with a preset threshold and found that the rate of change in this depth range exceeded the threshold. Abrupt point marking processing marked these depth locations as gradient abrupt change points. Continuity analysis revealed that these abrupt change points were continuously distributed in depth. The starting depth was determined as the first depth position of the abrupt change point sequence, and the ending depth was determined as the last depth position of the abrupt change point sequence, thus determining the temperature gradient abrupt change depth range. This range accurately reflects the distribution area of ​​the thermocline in the ecological pond.

[0069] In one specific embodiment, the process of executing step S103 may specifically include the following steps:

[0070] The dynamic variable of the thermocline position is compared with the thermocline position threshold to obtain the thermocline anomaly judgment result and the thermocline trigger flag.

[0071] The dynamic variable of dissolved oxygen vertical gradient is numerically compared with the dissolved oxygen gradient threshold to obtain the hypoxia stratification judgment result and hypoxia triggering flag.

[0072] The dynamic variables of nutrient stratification status are compared with the nutrient stratification threshold to obtain the nutrient anomaly judgment result and nutrient trigger flag.

[0073] Based on the thermocline trigger flag, hypoxia trigger flag and nutrient trigger flag, logical combination and priority sorting are performed to obtain the stratification anomaly trigger signal;

[0074] Based on the triggering type of the stratified anomaly triggering signal, the location of the corresponding depth layer is queried and the depth coordinates are extracted to obtain the target water distribution depth.

[0075] Specifically, when comparing the dynamic variable of the thermocline position with the thermocline position threshold, the thermocline position threshold is a critical value of the thermocline depth determined based on the normal water stratification state of the ecological pond. The numerical comparison process reads the value of the dynamic variable of the thermocline position and compares it with the preset thermocline position threshold. When the dynamic variable of the thermocline position exceeds the normal depth range, it indicates that the thermocline position is abnormal. The thermocline abnormality judgment result is the logical output of the comparison process, including three states: normal, shallow, and deep. The thermocline trigger flag is a Boolean flag indicating whether water distribution treatment is needed for thermocline abnormality. When the judgment result is abnormal, the trigger flag is set to true.

[0076] In the process of numerically comparing the dynamic variable of dissolved oxygen vertical gradient with the dissolved oxygen gradient threshold, the dissolved oxygen gradient threshold is a critical value of vertical gradient determined based on the normal oxygen distribution characteristics of the water body. This threshold reflects the normal variation range of dissolved oxygen in the vertical direction in the ecological pond. The numerical comparison process reads the value of the dynamic variable of dissolved oxygen vertical gradient and compares it with the value of dissolved oxygen gradient threshold. When the dissolved oxygen vertical gradient exceeds the normal range, it indicates that there is an oxygen deficiency stratification phenomenon in the water body. The oxygen deficiency stratification judgment result is the output state of the comparison process, including three stratification states: normal distribution, mild oxygen deficiency, and severe oxygen deficiency. The oxygen deficiency trigger flag is a marker indicating whether oxygenation and water distribution treatment is required. When the judgment result shows oxygen deficiency, the trigger flag is set to true.

[0077] When comparing the dynamic variable of nutrient stratification status with the nutrient stratification threshold, the nutrient stratification threshold is a critical value for stratification status determined based on the normal distribution law of nutrients in water bodies. This threshold is determined by combining historical data statistics and water quality management standards. The numerical comparison process reads the value of the dynamic variable of nutrient stratification status and compares it with the nutrient stratification threshold. When the nutrient stratification status exceeds the normal range, it indicates that the nutrient distribution in the water body is abnormal. The nutrient abnormality judgment results include three states: uniform distribution, slight stratification, and severe stratification. The nutrient trigger flag is a marker indicating whether nutrient regulation and water distribution treatment are required. When the judgment result is abnormal stratification, the trigger flag is set to true.

[0078] In the logical combination and priority ranking process based on thermocline trigger flags, anoxic trigger flags, and nutrient trigger flags, logical combination refers to the processing method of combining the three trigger flags according to preset logical rules. This processing method first reads the status values ​​of the three trigger flags, and then determines the priority weight of each flag according to the importance order of ecological pond water quality management. Priority ranking sorts the triggered anomaly types according to their severity and urgency. Anoxic stratification has the highest priority because it directly affects the survival of aquatic organisms, thermocline anomalies have a medium priority because they affect water mixing, and nutrient stratification has a low priority because its impact is relatively slow. The stratification anomaly trigger signal is the output result of logical combination and priority ranking processing. This signal contains anomaly type identifiers and corresponding priority information.

[0079] When querying and extracting depth coordinates for the corresponding depth layer based on the trigger type of the stratification anomaly trigger signal, the trigger type refers to the specific anomaly type identified in the stratification anomaly trigger signal, including three types: thermocline anomaly, anoxic stratification anomaly, and nutrient stratification anomaly. The query process reads the trigger type information and searches for the depth location of the corresponding anomaly type in the depth gradient dataset. The depth coordinate extraction process extracts the specific depth value from the query results. The target water distribution depth is the water distribution operation depth coordinate determined based on the anomaly type and the location of occurrence. This depth coordinate points to the specific water layer location where water quality adjustment is required.

[0080] In one specific embodiment, the process of executing step S104 may specifically include the following steps:

[0081] The trigger types in the hierarchical abnormal trigger signals are processed by equipment matching and water distribution mode selection to obtain the corresponding water distribution operation mode and equipment call list.

[0082] Based on the target water distribution depth, the variable depth water distributor is subjected to depth positioning and position adjustment processing to obtain accurate depth coordinates and equipment positioning confirmation signal.

[0083] Based on the water distribution operation mode, the opening degree of the flow regulating valve group is calculated and the flow distribution is processed to obtain the water distribution flow parameters and valve control commands for each depth layer.

[0084] The timing coordination and synchronous start-up processing of equipment positioning confirmation signals and valve control commands are performed to obtain a multi-equipment collaborative operation schedule.

[0085] The water distribution duration is calculated and the stop conditions are set based on the multi-device collaborative operation schedule to obtain the water distribution execution command.

[0086] Specifically, when processing the trigger type in the stratified abnormal trigger signal for equipment matching and water distribution mode selection, equipment matching refers to the process of selecting the corresponding water distribution equipment according to different abnormality types. This process first reads the trigger type information in the stratified abnormal trigger signal, and then determines the corresponding water distribution equipment type according to the preset equipment configuration table. The water distribution mode selection process determines the corresponding water distribution strategy by analyzing the severity and impact range of the abnormality type. The water distribution operation mode includes three working modes: single-point directional water distribution, multi-point collaborative water distribution, and continuous stratified water distribution. The equipment call list is the specific equipment usage plan determined according to the water distribution operation mode. This list includes the number of variable depth water distributors to be activated, the configuration of flow regulating valve groups, and the corresponding control parameters.

[0087] In the process of depth positioning and position adjustment of the variable depth water distributor based on the target water distribution depth, the variable depth water distributor is a water distribution device that can adjust its vertical position in the water body. The device achieves precise depth control through motor drive and depth sensor feedback. The depth positioning process reads the target water distribution depth value, converts it into equipment control commands and sends them to the variable depth water distributor. The position adjustment process controls the water distributor to move to the specified depth position in the water body. The device reaches the target position by monitoring the feedback from the device's depth sensor in real time. The precise depth coordinate is the actual depth value after the water distributor reaches the target position. The device positioning confirmation signal is a status mark indicating that the water distributor has reached the specified position and is ready to carry out water distribution operations.

[0088] When performing opening degree calculation and flow distribution processing on the flow regulating valve group according to the water distribution operation mode, the flow regulating valve group is a flow control system composed of multiple adjustable flow valves. This system can independently control the water distribution flow at different depth layers. The opening degree calculation processing calculates the opening degree of each valve according to the flow requirements of the water distribution operation mode. The calculation process takes into account the severity of the abnormality type, the water quality of the target depth layer, and the water distribution effect requirements. The flow distribution processing distributes the total water distribution flow to the valves at different depth layers according to a preset ratio. The water distribution flow parameters of each depth layer are the specific flow values ​​corresponding to each depth position. The valve control commands include the opening angle and flow set value of each valve.

[0089] During the process of coordinating and synchronizing the equipment positioning confirmation signal and valve control command, the timing coordination refers to the method of arranging different equipment to perform operations in a reasonable time sequence. This method first confirms that all variable depth water distributors have issued equipment positioning confirmation signals, and then sends valve control commands according to the preset start-up sequence. The synchronous start-up process ensures that the water distribution equipment at each depth layer starts working at the same time, avoiding uneven water distribution effect caused by differences in start-up time. The multi-equipment collaborative operation schedule is a detailed time arrangement table that includes the start-up time, operation duration and stop time of each equipment. This schedule coordinates the working rhythm of each equipment.

[0090] When calculating the duration of water distribution and setting stop conditions based on the multi-device collaborative operation schedule, the calculation of the water distribution duration is determined according to the severity of the anomaly type, the water distribution flow rate, and the expected improvement effect. The calculation process considers factors such as the rate of change of water quality parameters and water distribution efficiency. The stop condition setting process establishes the termination judgment criteria for the water distribution operation, including stop conditions such as the time reaching the preset value, the water quality parameters reaching the target range, and equipment failure. The water distribution execution command is a comprehensive set of control commands that integrates water distribution depth, flow rate, duration, and stop conditions. This set of commands guides the entire water distribution operation execution process.

[0091] In one specific embodiment, the process of executing step S105 may specifically include the following steps:

[0092] Based on the water distribution execution command, directional water distribution operation and real-time water quality monitoring and processing are carried out on the water body at the target water distribution depth to obtain water quality change data during the water distribution process.

[0093] The difference between the water quality change data during the water distribution process and the water quality baseline value before water distribution is calculated and the degree of improvement is quantified to obtain quantitative indicators of improvement in temperature, dissolved oxygen, pH value and turbidity.

[0094] The water distribution efficiency was calculated and the parameters were analyzed to obtain the correlation data between water distribution depth, flow rate, duration and water quality improvement.

[0095] The associated data is input into the machine learning model for parameter optimization and strategy update to obtain data on water quality improvement effects.

[0096] Based on the water quality improvement effect data, the water distribution parameter optimization database is processed by writing data and updating model parameters to obtain the updated water distribution parameter optimization database.

[0097] Specifically, when performing directional water distribution operations and real-time water quality monitoring and treatment on water bodies at a target water distribution depth based on water distribution execution commands, directional water distribution operation refers to the process of injecting treated water into a specific depth layer according to the depth, flow rate, and duration parameters in the water distribution execution command. This operation releases the treated water at the target depth position through a variable depth water distributor, while the flow regulating valve group controls the injection rate according to preset flow parameters. Real-time water quality monitoring and treatment continuously collects changes in water quality parameters at each depth layer during the water distribution process through vertical stratification sensors. The monitoring data includes real-time values ​​of four dimensions: temperature, dissolved oxygen, pH value, and turbidity. The water quality change data during the water distribution process is a time series data of water quality parameters continuously collected during the water distribution operation. This data reflects the dynamic impact of the water distribution operation on water quality.

[0098] In the process of calculating the difference between the water quality change data during the water distribution process and the water quality baseline value before water distribution, and quantifying the degree of improvement, the water quality baseline value before water distribution refers to the water quality parameter values ​​of each depth layer collected before the start of the water distribution operation, which serves as a comparison benchmark for evaluating the water distribution effect. The difference calculation process reads the parameter values ​​in the water quality change data during the water distribution process and performs subtraction operations with the corresponding water quality baseline value before water distribution to obtain the change value of each water quality parameter. The degree of improvement quantification process converts the change value of water quality parameters into standardized improvement degree indicators. This process considers the normal range and improvement direction of each parameter. The improvement quantification indicators of temperature, dissolved oxygen, pH value and turbidity are standardized values ​​representing the degree of improvement of each water quality parameter. These indicators are uniformly expressed in the form of improvement percentage.

[0099] When performing water distribution efficiency calculation and parameter correlation analysis on the quantitative indicators of improvement, the water distribution efficiency calculation refers to the calculation process of assessing the contribution of unit water distribution volume to water quality improvement. This calculation process divides the quantitative indicator of improvement by the product of water distribution flow rate and duration to obtain the efficiency value. The parameter correlation analysis process uses statistical methods to analyze the quantitative relationship between water distribution depth, flow rate, duration and the degree of water quality improvement. This analysis process determines the correlation strength by calculating the correlation coefficient between each parameter and the improvement effect. The correlation data between water distribution depth, flow rate, duration and the degree of water quality improvement is a data set containing the correlation between each water distribution parameter and the improvement effect. This data set reveals the influence pattern of different water distribution strategies on water quality improvement.

[0100] During the process of inputting correlated data into the machine learning model for parameter optimization and strategy update, the machine learning model is an artificial intelligence algorithm trained on historical water distribution data. This model can identify the complex nonlinear relationship between water distribution parameters and improvement effects. Parameter optimization is carried out by analyzing the parameter effect relationship in the correlated data to adjust the weight coefficients and threshold parameters inside the model. Strategy update is carried out by revising the selection logic and parameter setting rules of the water distribution strategy based on the new correlated data. The water quality improvement effect data is the comprehensive evaluation result output by the machine learning model, which includes the effect score and improvement suggestions of this water distribution operation.

[0101] When writing data and updating model parameters to the water distribution parameter optimization database based on water quality improvement effect data, the water distribution parameter optimization database is a data warehouse storing historical water distribution operation parameters and effect data. This database is classified and stored according to water distribution conditions, parameter configurations, and improvement effects. The data writing process stores the water quality improvement effect data into the corresponding tables in the database according to a predetermined format, while recording the time, conditions, and parameter information of the water distribution operation. The model parameter update process adjusts the parameter configuration of the machine learning model based on the newly added effect data, including key parameters such as learning rate, weight coefficients, and decision thresholds. The updated water distribution parameter optimization database contains new water distribution experience and optimization strategies, and this database supports the selection of strategies and parameter configurations for subsequent water distribution operations.

[0102] For example, during an anoxic stratified water distribution operation in an ecological pond, the water distribution command was set to inject oxygen-rich water into the bottom layer. The directional water distribution operation used a variable-depth distributor to inject treated water into the anoxic layer. Real-time water quality monitoring showed that the dissolved oxygen concentration in the bottom layer gradually increased during the water distribution process, while temperature and pH also changed accordingly. The water quality change data during the water distribution process recorded the time-varying curves of each parameter. Difference calculation compared the dissolved oxygen concentration after water distribution with the baseline value before water distribution, revealing a significant increase in dissolved oxygen concentration. The degree of improvement was quantified by converting the increase in dissolved oxygen concentration into a percentage improvement indicator. Water distribution efficiency calculation showed that a unit flow rate of oxygen-rich water significantly improved the anoxic conditions in the bottom layer. Parameter correlation analysis... The study revealed a positive correlation between water distribution flow rate and the degree of dissolved oxygen improvement. Correlation data showed that a larger water distribution flow rate corresponds to a more significant improvement effect. Based on this correlation data, the machine learning model adjusted the parameter setting strategy for anoxic stratified water distribution. Parameter optimization updated the coefficients in the flow calculation formula, and strategy update corrected the judgment criteria for the duration of anoxic stratified water distribution. The water quality improvement effect data comprehensively evaluated the success of this water distribution operation. Data writing stored the parameter configuration and improvement effect of this water distribution in the database. Model parameter update adjusted the decision weights of the machine learning model based on the new data. The updated water distribution parameter optimization database contains new experience in anoxic stratified water distribution.

[0103] The above describes the IoT-based ecological pond water distribution method in the embodiments of this application. The following describes the IoT-based ecological pond water distribution control system in the embodiments of this application. Please refer to [link / reference]. Figure 2 One embodiment of the IoT-based ecological pond water distribution and control system in this application includes:

[0104] The data acquisition module is used to collect and process water quality parameters at different depths of the ecological pond through vertical stratified sensors, and obtain a depth gradient dataset including temperature, dissolved oxygen, pH value and turbidity.

[0105] The quantification module is used to perform quantitative analysis and processing of the water stratification state based on the depth gradient dataset using a thermocline identification algorithm, and to obtain dynamic variables of thermocline location, dynamic variables of dissolved oxygen vertical gradient, and dynamic variables of nutrient stratification state.

[0106] The judgment module is used to compare and judge the dynamic variables of the thermocline position, the dynamic variables of the dissolved oxygen vertical gradient, and the dynamic variables of the nutrient stratification state with the corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth.

[0107] The scheduling module is used to perform coordinated scheduling processing on the multi-depth water distribution equipment according to the layered abnormality trigger signal, and obtain water distribution execution instructions including water distribution depth, flow rate and duration.

[0108] The water distribution module is used to perform directional water distribution treatment on the water body at the target water distribution depth through the water distribution execution command, obtain water quality improvement effect data, and update the water distribution parameter optimization database.

[0109] above Figure 2 The IoT-based ecological pond water distribution and control system in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The IoT-based ecological pond water distribution equipment in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0110] Reference Figure 3 This invention also provides an IoT-based ecological pond water distribution device, which can be a server, and its internal structure can be as follows: Figure 3As shown, the IoT-based ecological pond water distribution device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor, designed as a computer, provides computing and control capabilities. The memory of the IoT-based ecological pond water distribution device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the IoT-based ecological pond water distribution device stores the data corresponding to this embodiment. The network interface of the IoT-based ecological pond water distribution device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0111] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the IoT-based ecological pond water distribution device to which the present invention is applied.

[0112] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the Internet of Things-based ecological pond water distribution method.

[0113] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0114] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an IoT-based ecological pond water distribution device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for water distribution in an ecological pond based on the Internet of Things, characterized in that, The method includes: Water quality parameters at different depths in the ecological pond were collected and processed using vertically layered sensors to obtain a depth gradient dataset containing temperature, dissolved oxygen, pH value, and turbidity. Based on the depth gradient dataset, a thermocline identification algorithm is used to quantitatively analyze the water stratification state, obtaining dynamic variables for thermocline location, dissolved oxygen vertical gradient, and nutrient stratification state. This includes: arranging the temperature data in the depth gradient dataset according to a depth sequence from shallow to deep to obtain a temperature-depth distribution sequence; performing layer-by-layer temperature difference calculation and gradient change rate analysis on the temperature-depth distribution sequence to obtain temperature gradient distribution data between each depth layer; comparing the mean temperature gradients of three consecutive layers and identifying gradient abrupt change points based on the temperature gradient distribution data to obtain the temperature gradient abrupt change depth range; precisely locating and interpolating the maximum gradient point within the temperature gradient abrupt change depth range to obtain the dynamic variable for thermocline location; and performing interlayer difference accumulation calculation on dissolved oxygen data and distribution variance statistics on pH and turbidity data in the depth gradient dataset to obtain the dynamic variable for dissolved oxygen vertical gradient and the dynamic variable for nutrient stratification state. The dynamic variables of the thermocline position, the dynamic variables of the dissolved oxygen vertical gradient, and the dynamic variables of the nutrient stratification state are compared with the corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth. Based on the layered anomaly trigger signal, the multi-depth water distribution equipment is coordinated and scheduled to obtain a water distribution execution command that includes water distribution depth, flow rate and duration. The water distribution execution command is used to perform directional water distribution treatment on the water body at the target water distribution depth, thereby obtaining water quality improvement effect data and updating the water distribution parameter optimization database.

2. The method for water distribution in an ecological pond based on the Internet of Things according to claim 1, characterized in that, The process involves collecting and processing water quality parameters at different depths in the ecological pond using vertically layered sensors to obtain a depth gradient dataset containing temperature, dissolved oxygen, pH, and turbidity. Vertically layered sensors were deployed in the ecological pond at 0.5-meter intervals to obtain sensor nodes that include temperature sensors, dissolved oxygen sensors, pH sensors, and turbidity sensors. The raw data collected by the sensor nodes is time-calibrated using a clock synchronization protocol to obtain synchronized water quality data. Based on the synchronous water quality data, depth identification and data association processing are performed to obtain a water quality parameter matrix arranged in depth layers; The water quality parameter matrix is ​​subjected to data quality inspection and outlier removal to obtain valid water quality data; The effective water quality data is classified and integrated according to four dimensions: temperature, dissolved oxygen, pH value, and turbidity to obtain the deep gradient dataset.

3. The IoT-based ecological pond water distribution method according to claim 2, characterized in that, The process of comparing the mean of three consecutive temperature gradients and identifying gradient abrupt change points based on the temperature gradient distribution data yields the temperature gradient abrupt change depth range, including: The temperature gradient distribution data is grouped into three consecutive layers according to depth order to obtain combined temperature gradient data of three consecutive layers. The arithmetic mean of the three gradient values ​​in each group of the three consecutive temperature gradient combination data is calculated to obtain the three-layer gradient mean at each depth position. The gradient mean values ​​of three adjacent depth locations are calculated by difference and analyzed by rate of change to obtain a gradient mean value change sequence. The gradient mean change sequence is subjected to threshold comparison and abrupt change point marking to obtain gradient abrupt change flag bits and corresponding depth coordinates; Based on the continuity analysis of the gradient mutation flag, the starting depth and ending depth are determined to obtain the temperature gradient mutation depth range.

4. The method for water distribution in an ecological pond based on the Internet of Things according to claim 1, characterized in that, The process of comparing and judging the dynamic variables of the thermocline position, the dissolved oxygen vertical gradient, and the nutrient stratification state with corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth includes: The dynamic variable of the thermocline position is compared with the thermocline position threshold to obtain the thermocline anomaly judgment result and the thermocline trigger flag. The dissolved oxygen vertical gradient dynamic variable is numerically compared with the dissolved oxygen gradient threshold to obtain the hypoxia stratification judgment result and hypoxia triggering flag. The dynamic variables of nutrient stratification status are compared with the nutrient stratification threshold to obtain the nutrient anomaly judgment result and nutrient trigger flag. The stratification anomaly trigger signal is obtained by logically combining and prioritizing the thermocline trigger flag, the hypoxia trigger flag, and the nutrient trigger flag. Based on the triggering type of the layered anomaly triggering signal, the corresponding depth layer position is queried and the depth coordinates are extracted to obtain the target water distribution depth.

5. The method for water distribution in an ecological pond based on the Internet of Things according to claim 1, characterized in that, The step of coordinating and scheduling the multi-depth water distribution equipment based on the layered anomaly trigger signal to obtain a water distribution execution command containing water distribution depth, flow rate, and duration includes: The trigger types in the layered abnormal trigger signals are processed by equipment matching and water distribution mode selection to obtain the corresponding water distribution operation mode and equipment call list. Based on the target water distribution depth, the variable depth water distributor is subjected to depth positioning and position adjustment processing to obtain accurate depth coordinates and equipment positioning confirmation signal. Based on the water distribution operation mode, the opening degree of the flow regulating valve group is calculated and the flow distribution is processed to obtain the water distribution flow parameters and valve control commands for each depth layer. The timing coordination and synchronous start-up processing of the equipment positioning confirmation signal and the valve control command are performed to obtain a multi-equipment collaborative operation schedule. The water distribution duration is calculated and the stop condition is set based on the multi-device collaborative operation schedule to obtain the water distribution execution command.

6. The method for water distribution in an ecological pond based on the Internet of Things according to claim 1, characterized in that, The process of performing directional water distribution treatment on the water body at the target water distribution depth through the water distribution execution command, obtaining water quality improvement effect data, and updating the water distribution parameter optimization database includes: Based on the water distribution execution command, directional water distribution operation and real-time water quality monitoring are performed on the water body at the target water distribution depth to obtain water quality change data during the water distribution process. The difference between the water quality change data during the water distribution process and the water quality baseline value before water distribution is calculated and the degree of improvement is quantified to obtain quantitative indicators of improvement in temperature, dissolved oxygen, pH value and turbidity. The water distribution efficiency was calculated and parameter correlation analysis was performed on the aforementioned improvement quantitative indicators to obtain correlation data between water distribution depth, flow rate, duration and water quality improvement degree; The associated data is input into a machine learning model for parameter optimization and strategy update to obtain the water quality improvement effect data. Based on the water quality improvement effect data, the water distribution parameter optimization database is processed by writing data and updating model parameters to obtain the updated water distribution parameter optimization database.

7. An Internet of Things-based ecological pond water distribution and control system, characterized in that, For implementing the IoT-based ecological pond water distribution method as described in any one of claims 1 to 6, the IoT-based ecological pond water distribution control system comprises: The data acquisition module is used to collect and process water quality parameters at different depths of the ecological pond through vertical stratified sensors, and obtain a depth gradient dataset including temperature, dissolved oxygen, pH value and turbidity. The quantification module is used to perform quantitative analysis of the water stratification state based on the depth gradient dataset using a thermocline identification algorithm, obtaining dynamic variables of thermocline location, dissolved oxygen vertical gradient, and nutrient stratification state. This includes: arranging the temperature data in the depth gradient dataset according to a depth sequence from shallow to deep to obtain a temperature-depth distribution sequence; performing layer-by-layer temperature difference calculation and gradient change rate analysis on the temperature-depth distribution sequence to obtain inter-layer temperature gradient distribution data; comparing the mean temperature gradients of three consecutive layers and identifying gradient abrupt change points based on the temperature gradient distribution data to obtain the temperature gradient abrupt change depth range; accurately locating and interpolating the maximum gradient point within the temperature gradient abrupt change depth range to obtain the dynamic variable of thermocline location; and performing inter-layer difference accumulation calculation on the dissolved oxygen data and distribution variance statistics on the pH and turbidity data in the depth gradient dataset to obtain the dynamic variable of dissolved oxygen vertical gradient and the dynamic variable of nutrient stratification state. The judgment module is used to compare and judge the dynamic variables of the thermocline position, the dynamic variables of the dissolved oxygen vertical gradient, and the dynamic variables of the nutrient stratification state with the corresponding preset thresholds to obtain the stratification anomaly trigger signal and the corresponding target water distribution depth. The scheduling module is used to perform coordinated scheduling processing on the multi-depth water distribution equipment according to the layered abnormality trigger signal, and obtain water distribution execution instructions including water distribution depth, flow rate and duration. The water distribution module is used to perform directional water distribution treatment on the water body at the target water distribution depth through the water distribution execution command, obtain water quality improvement effect data, and update the water distribution parameter optimization database.

8. An ecological pond water distribution device based on the Internet of Things, characterized in that, It includes a memory and a processor, the memory storing a computer program that can run on the processor, and the processor executing the computer program to implement the Internet of Things-based ecological pond water distribution method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is run by the processor, it causes the processor to execute the IoT-based ecological pond water distribution method as described in any one of claims 1 to 6.

Images