A remote intelligent control system and method for abalone breeding environment

Abstract

The application discloses a kind of abalone breeding environment remote intelligent regulation and control system and method, specifically related to aquatic environmental regulation and control technical field;It is the whole and the water quality of the surface of attached base of breeding pond, illumination and local flow parameter are synchronously collected, and attached micro area observation set is formed;Based on the registration of benthic diatom film day-night metabolism, attached micro area state sequence is generated, and the amount of mismatch of attached micro area is calculated and coupled fluctuation category is identified;According to the fluctuation category, the remote execution sequence is generated and executed by calling linkage rule, and the remote identification and closed-loop regulation and control of abalone breeding attached micro area environment are realized.The application is used to realize the remote identification, linkage regulation and closed-loop correction of abalone breeding attached micro area environment.

Description

Technical Field

[0001] This invention relates to the field of aquatic environment control technology, and more specifically, to a remote intelligent control system and method for abalone seedling environment. Background Technology

[0002] Abalone larvae are highly sensitive to water quality during the seedling stage, especially during the attachment metamorphosis stage. Larvae typically attach to benthic diatomaceous membranes on the surface of the substrate within the seedling pond to complete their metamorphosis. Existing methods for controlling the abalone seedling environment generally use remote monitoring to obtain basic parameters of the overall water quality in the seedling pond and implement water quality control based on these parameters. However, the monitoring locations are usually inside the water body of the seedling pond or near the inlet / outlet, obtaining only macroscopic average data for the entire water body, without addressing the specific attachment area of ​​the larvae and its interface environmental parameters for targeted monitoring and control.

[0003] Existing methods for controlling the abalone breeding environment lack continuous monitoring and accurate identification of the local microenvironment at the abalone larvae attachment interface. This results in the overall water quality parameters of the breeding pond meeting the standards, but the actual condition of the local microenvironment where the larvae attach may not be stable, ultimately affecting the metamorphosis attachment effect of abalone larvae and the stability of early cultivation.

[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a remote intelligent control system and method for abalone seedling environment to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for remote intelligent control of abalone seedling cultivation environment includes the following steps: S1: Collect overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and form a micro-area observation set according to the location of the substrate; S2: Based on the diurnal metabolic transformation of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-regions is registered to generate the state sequence of the attachment micro-regions; S3: Perform difference analysis on the overall water quality parameters of the seedling pond and the state sequence of the attached micro-region to obtain the mismatch of the attached micro-region, and identify the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch of the attached micro-region. S4: According to the coupling fluctuation category, the preset linkage rules are invoked to combine oxygenation, water exchange, shading and circulation paths to generate candidate control sequences; S5: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, candidate control sequences are screened and time period rearranged to generate remote execution sequences; S6: Send a remote execution sequence to the remote execution device and collect the surface water quality parameters of the substrate after execution and write them back to the preset linkage rules.

[0007] In a preferred embodiment, S1 specifically refers to: Water quality parameter collection points for the entire seedling pond are set up along the water body distribution of the seedling pond, and water quality parameter collection points for the surface of the substrate are set up along the surface of the substrate. Within the same sampling period, the overall water quality parameters of the seedling pond, the surface water quality parameters of the substrate, the light parameters, and the local flow parameters were collected simultaneously. The synchronous acquisition results are time-aligned and bound to the location of the substrate, and then merged according to the location of the substrate to form an attached micro-area observation set.

[0008] In a preferred embodiment, S2 specifically refers to: Arrange the continuous acquisition results corresponding to the same attachment substrate location in the attachment micro-area observation set according to the sampling time to form a location observation trajectory; Based on the changes in light parameters, the observation period was divided into light and dark observation periods, and the diurnal metabolic transition interval of benthic diatom film was determined by combining the changes in surface water quality parameters of the substrate. The location observation trajectory was segmented and registered according to the diurnal metabolic transition interval of the benthic diatom membrane to maintain the corresponding relationship of the attachment substrate and generate the state sequence of the attachment micro-region.

[0009] In a preferred embodiment, S3 specifically refers to: The overall water quality parameters of the seedling pond were mapped to the location of the attachment substrate at the same moment in the state sequence of the attachment micro-region according to the sampling time, forming a parameter comparison group; Based on the parameter comparison group, the deviation direction, deviation magnitude and continuous change relationship of the surface water quality parameters of the attached substrate and the overall water quality parameters of the seedling pond in terms of pH, dissolved oxygen and ammonia toxicity were extracted to generate the mismatch amount of the attached micro-region. Based on the combined variation characteristics of the mismatch amount of attached micro-regions during the light observation period, dark observation period, and diurnal metabolic transition period of benthic diatom membranes, we can identify the coupled fluctuation categories of pH, dissolved oxygen, and ammonia toxicity.

[0010] In a preferred embodiment, S4 specifically refers to: Based on the matching of pH, dissolved oxygen and ammonia toxicity coupling fluctuation categories, the rule items in the preset linkage rules corresponding to the mismatch amount of the attached micro-area are used to determine the action combination relationship corresponding to oxygenation, water exchange, shading and circulation paths; Based on the action combination relationship, distinguish the preceding, subsequent and synchronous relationships between oxygenation, water exchange, shading and circulation paths, and form a control arrangement relationship corresponding to the category; Based on the control arrangement relationship, oxygenation, water exchange, shading and circulation paths are combined to generate candidate control sequences.

[0011] In a preferred embodiment, S5 specifically refers to: Based on the state sequence of the attached micro-region, the continuous stable interval corresponding to the location of the attached substrate is extracted, and the abnormal attachment time window is determined by combining the coupled fluctuation categories of pH, dissolved oxygen and ammonia toxicity. Based on the temporal correspondence between the abnormal attachment time window and the action combination relationship in the candidate regulation sequence, candidate regulation content that conflicts with the abnormal attachment time window is screened out. The selected candidate control sequences are rearranged according to the start and end times of the abnormal attachment window to generate remote execution sequences.

[0012] In a preferred embodiment, S6 specifically refers to: The remote execution sequence is sent to the corresponding remote execution device according to the action combination relationship, and the execution time corresponding to the action combination relationship is recorded. After the remote execution device completes the remote execution sequence, it collects water quality parameters on the surface of the attached substrate according to the execution time, forming a post-execution parameter correspondence. Based on the parameter correspondence after execution, the surface water quality parameters of the attached substrate are written back to the rule items in the preset linkage rules that correspond to the action combination relationship.

[0013] On the other hand, the present invention provides a remote intelligent control system for abalone seedling cultivation environment, comprising: Micro-area acquisition module: Collects overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and forms an attachment micro-area observation set according to the location of the substrate; State registration module: Based on the diurnal metabolic transition of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-region is registered to generate the state sequence of the attachment micro-region; Mismatch Analysis Module: Performs difference analysis between the overall water quality parameters of the seedling pond and the state sequence of the attached micro-regions to obtain the mismatch amount of the attached micro-regions, and identifies the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch amount of the attached micro-regions. Linkage Combination Module: Based on the coupling fluctuation category, it calls preset linkage rules to combine oxygenation, water exchange, shading, and circulation paths to generate candidate control sequences; The time-series filtering module: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, it filters and rearranges the time periods of candidate control sequences to generate remote execution sequences; Execution write-back module: Sends remote execution sequence to remote execution device and collects water quality parameters of the attached substrate surface after execution to write back the preset linkage rules.

[0014] The technical effects and advantages of the remote intelligent control system and method for abalone seedling cultivation environment of the present invention are as follows: This invention constructs a complete regulatory chain around the stable control of the micro-environment of abalone larvae attachment. By collecting overall water quality parameters of the nursery pond, surface water quality parameters of the attachment substrate, light parameters, and local flow parameters to form an observation set of attachment micro-areas, it can correlate the overall environment of the nursery pond with the actual attachment area of ​​abalone larvae. By generating a sequence of attachment micro-area states and analyzing the mismatch of attachment micro-areas, it can identify the differences between the overall parameters of the nursery pond and the states of attachment micro-areas. By identifying the coupled fluctuation categories of pH, dissolved oxygen, and ammonia toxicity and generating candidate regulatory sequences, it can make the regulation process of oxygenation, water exchange, shading, and circulation paths more targeted. By combining the abnormal attachment time window for screening and time period rearrangement, it can match the remote execution sequence with the attachment process of abalone larvae. By writing back the preset linkage rules after execution, it can form a continuously corrected closed-loop regulation process, improving the accuracy and adaptability of remote intelligent regulation of the abalone nursery environment. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a remote intelligent control method for abalone seedling cultivation environment according to the present invention; Figure 2 This is a schematic diagram of the structure of a remote intelligent control system for abalone seedling cultivation environment according to the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example 1

[0017] Figure 1 This invention discloses a method for remote intelligent control of abalone seedling cultivation environment, which includes the following steps: S1: Collect overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and form a micro-area observation set according to the location of the substrate; S2: Based on the diurnal metabolic transformation of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-regions is registered to generate the state sequence of the attachment micro-regions; S3: Perform difference analysis on the overall water quality parameters of the seedling pond and the state sequence of the attached micro-region to obtain the mismatch of the attached micro-region, and identify the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch of the attached micro-region. S4: According to the coupling fluctuation category, the preset linkage rules are invoked to combine oxygenation, water exchange, shading and circulation paths to generate candidate control sequences; S5: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, candidate control sequences are screened and time period rearranged to generate remote execution sequences; S6: Send a remote execution sequence to the remote execution device and collect the surface water quality parameters of the substrate after execution and write them back to the preset linkage rules.

[0018] S1: Collect overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period. Form a micro-area observation set based on the location of the substrate, including: The layout of water quality parameter collection points in the seedling pond follows the principle of both uniform spatial coverage and stratified representativeness. Horizontally, the seedling pond is divided into several equal-area grid areas based on its bottom area. One water quality parameter collection point is placed at the center of each grid area. For example, for a rectangular seedling pond with an area of ​​20 square meters, four columns can be evenly arranged along the long axis and three rows along the short axis, for a total of 12 collection points. Vertically, one layer of water quality parameter collection points is placed at 0.1 meters below the water surface, in the middle layer of the water, and 0.05 meters above the bottom of the pond. This captures the gradient changes in water quality parameters at different depths. The depth of the middle layer is determined based on the actual water depth of the seedling pond, using the arithmetic mean of the surface depth and the bottom depth of the pond. The overall water quality parameters collected at the seedling pond overall water quality parameter collection points include, but are not limited to, pH, dissolved oxygen concentration, and ammonia nitrogen concentration. The range and accuracy of the sensors mounted on the seedling pond overall water quality parameter collection points must meet the following basic requirements: the pH sensor resolution is not less than 0.01 pH units, the dissolved oxygen sensor resolution is not less than 0.01 mg / L, and the ammonia nitrogen sensor range covers 0 to 10 mg / L with a resolution not less than 0.01 mg / L.

[0019] The placement of water quality parameter collection points on the substrate surface utilizes the substrate itself as a carrier, tailored to the microenvironmental characteristics of the substrate surface. The substrate refers to the corrugated or flat substrate within the nursery pond used for abalone larvae to attach and grow. These substrates are arranged vertically suspended or obliquely stacked within the nursery pond. When distributing water quality parameter collection points along the substrate surface, one collection point is placed at each of the upper, middle, and lower edges of each substrate. These collection points are installed flush against the substrate surface, with a distance of no more than 5 mm between them and the surface. This ensures that the collected water quality parameters accurately reflect the local microenvironmental water quality conditions of the benthic diatomaceous membrane and the abalone larvae. The water quality parameters collected at these collection points correspond in type to the overall water quality parameters of the nursery pond, including substrate surface pH, dissolved oxygen concentration, and ammonia nitrogen concentration. Each water quality parameter collection point on the surface of the substrate simultaneously records the substrate location identifier. The substrate location identifier is composed of three segments: the seedling pond number, the substrate number, and the area number where the water quality parameter collection point on the surface of the substrate is located. For example, the number format is a string of "pond number-substrate number-area number", which uniquely identifies the spatial location of each water quality parameter collection point on the surface of the substrate within the seedling pond.

[0020] Lighting parameter collection points are evenly distributed above the seedling pond to collect light intensity and direction within the pond. The density of these collection points is designed to cover the entire horizontal projection area of ​​the substrate in the seedling pond, ensuring that there is at least one light parameter collection point directly above each substrate within its horizontal projection area. Local flow parameter collection points are placed in the water flow channels between the substrates to collect the flow velocity and direction of the water around them. One local flow parameter collection point is placed in the middle of the channel between every two adjacent substrates, with the probe of each local flow parameter collection point facing the main flow direction.

[0021] Within the same sampling period, the overall water quality parameters of the seedling pond, the surface water quality parameters of the substrate, the light parameters, and the local flow parameters are all collected synchronously using hardware triggering. The timing deviation of the collection start time for all collection points within the same sampling period does not exceed 10 milliseconds to ensure that the overall water quality parameters of the seedling pond, the surface water quality parameters of the substrate, the light parameters, and the local flow parameters correspond in the time dimension. The time interval of the sampling period is determined based on the metabolic response rate of the benthic diatom film. For example, it can be set to 5 minutes, meaning that synchronous collection of all collection points is performed once every 5 minutes. Each synchronous collection generates one set of raw collection results, with the timestamp of the collection trigger time serving as the unique time marker. The timestamp format is accurate to the second.

[0022] When aligning the synchronous acquisition results by time, the timestamp of the same acquisition trigger time is used as the benchmark. Data values ​​with the same timestamp among the overall water quality parameters of the seedling pond, the surface water quality parameters of the substrate, the light parameters, and the local flow parameters are merged into the same time node record, forming the original acquisition record aligned by time node. When the data value of the acquisition point at the same timestamp is missing due to sensor failure or communication anomaly, the missing value is filled by the arithmetic mean of the valid data values ​​acquired by the acquisition point in the three sampling periods before and after it. If there are fewer than one valid data value in each of the three sampling periods before and after it, the time node record is marked as invalid and will not participate in the generation of the substrate micro-area observation set.

[0023] When binding the original data collection records that have been aligned with the completion time to the location of the substrate, the original data collection values ​​of each substrate surface water quality parameter collection point are associated with the substrate location identifier carried by the substrate surface water quality parameter collection point, so that each substrate surface water quality parameter collection value is accompanied by clear substrate location information. The data collection values ​​of the illumination parameter collection points are allocated to the substrates directly below the illumination parameter collection point based on the correspondence between the horizontal coordinates of the illumination parameter collection point and the horizontal projection range of the substrate. When the horizontal projection range of one illumination parameter collection point covers multiple substrates at the same time, the illumination parameter collection data values ​​are simultaneously allocated to all covered substrates without weighting or splitting. The collected data values ​​of the local flow parameter collection points are distributed to all adjacent attachment bases on both sides of the local flow parameter collection point according to the correspondence between the channel position of the local flow parameter collection point and the number of the adjacent attachment base. Each attachment base obtains the local flow parameter collection data value corresponding to the channel on both sides of the attachment base. When there is a difference in the local flow parameter collection data value corresponding to the channel on both sides of the attachment base, all values ​​are retained and recorded separately under the attachment base position mark, without taking the average value, so as to retain the spatial non-uniformity information of the local flow parameters.

[0024] When merging attachment micro-area observation sets according to the location of the attachment substrate, the attachment substrate location identifier is used as the primary key. The surface water quality parameters, corresponding light parameters, and corresponding local flow parameters of the attachment substrate under the same attachment substrate location identifier are combined with the overall water quality parameters of the seedling pond at the same time stamp to form one attachment micro-area observation record. One attachment micro-area observation record contains 12 data fields, including timestamp, attachment substrate location identifier, pH of the attachment substrate surface, dissolved oxygen concentration of the attachment substrate surface, ammonia nitrogen concentration of the attachment substrate surface, light intensity at the corresponding location, light direction at the corresponding location, flow velocity at the corresponding location, flow direction at the corresponding location, overall pH of the seedling pond, overall dissolved oxygen concentration of the seedling pond, and overall ammonia nitrogen concentration of the seedling pond. All attachment micro-area observation records corresponding to the attachment base location identifiers within the same sampling period are summarized to form an attachment micro-area observation set. The attachment micro-area observation set is stored in a two-dimensional structure, with rows corresponding to different attachment base location identifiers and columns corresponding to the above 12 data fields. After each sampling period is completed, synchronous acquisition, time alignment, attachment base location binding and merging are completed, and a new attachment micro-area observation record is added to the attachment micro-area observation set.

[0025] S2: Based on the diurnal metabolic transition of benthic diatom membranes and the location of the attachment substrate, the observation set of attachment micro-regions is registered to generate a sequence of attachment micro-region states, including: When arranging continuous acquisition results corresponding to the same attachment base location in the attachment micro-area observation set according to sampling time to form the location observation trajectory, the attachment base location identifier is used as the retrieval primary key to extract all attachment micro-area observation records with the same attachment base location identifier from the attachment micro-area observation set. All extracted attachment micro-area observation records are then sorted in ascending order according to the value of the timestamp field in the attachment micro-area observation record, forming a one-dimensional ordered record sequence indexed by timestamp. This one-dimensional ordered record sequence is named the location observation trajectory. The time span of the location observation trajectory is determined by the interval between the earliest and latest timestamps under the same attachment substrate location identifier in the attachment micro-area observation set. The timestamp difference between two adjacent attachment micro-area observation records in the location observation trajectory should be consistent with the time interval of the sampling period. For example, when the time interval of the sampling period is set to 5 minutes, the timestamp difference between two adjacent attachment micro-area observation records in the location observation trajectory should be 5 minutes. If the timestamp difference between two adjacent attachment micro-area observation records exceeds 1.5 times the time interval of the sampling period, it is determined that there is a sampling discontinuity between the two attachment micro-area observation records. The location with the sampling discontinuity is marked in the location observation trajectory by inserting a discontinuity flag bit in the timestamp field. Each attachment substrate location identifier in the seedling pond corresponds to an independent location observation trajectory. The location observation trajectories corresponding to all attachment substrate location identifiers together constitute the location observation trajectory set. The time span of each location observation trajectory in the location observation trajectory set is consistent and covers the complete time range of the attachment micro-area observation set.

[0026] When dividing the observation period into bright and dark observation periods based on changes in illumination parameters, the illumination intensity field value of the corresponding location is extracted from any one location observation trajectory in the location observation trajectory set, forming an illumination intensity time series with the timestamp as the horizontal axis and the corresponding location illumination intensity value as the vertical axis. Continuous trend analysis is performed on the illumination intensity time series. The moment when the corresponding location illumination intensity value exceeds a preset illumination start threshold is taken as the start time of the bright period, and the moment when the corresponding location illumination intensity value decreases to below a preset illumination end threshold is taken as the end time of the bright period. The preset illumination start threshold and preset illumination end threshold are set as follows: the arithmetic mean of the daily maximum and daily minimum illumination intensity values ​​of the corresponding location in the illumination intensity time series collected over seven consecutive natural days is used as the baseline illumination value. The value corresponding to 10% of the baseline illumination value is set as the preset illumination start threshold and preset illumination end threshold. For example, when the baseline illumination value is 5000 lux, both the preset illumination start threshold and preset illumination end threshold are set to 500 lux. The time period from the start time of the bright period to the end time of the bright period is defined as the bright period observation segment, and the time period from the end time of the bright period to the start time of the bright period of the next natural day is defined as the dark period observation segment. The bright period observation segment and the dark period observation segment appear alternately, together covering the complete time span of the position observation trajectory.

[0027] When determining the diurnal metabolic transition range of benthic diatom films by combining changes in surface water quality parameters, the transitional changes in dissolved oxygen concentration and pH on the substrate surface between the light and dark observation periods are used as the basis for identifying the diurnal metabolic transition range of benthic diatom films. During the light period, benthic diatom films perform photosynthesis, and the dissolved oxygen concentration and pH on the substrate surface show an increasing trend. During the dark period, benthic diatom films perform respiration, and the dissolved oxygen concentration and pH on the substrate surface show a decreasing trend. In the transitional stage from the light to the dark period or from the dark to the light period, the trends of changes in dissolved oxygen concentration and pH on the substrate surface reverse. The time period during which the reversal occurs is the diurnal metabolic transition range of the benthic diatom film. The method for determining the diurnal metabolic transition interval of benthic diatomaceous membranes is as follows: Dissolved oxygen concentration values ​​on the substrate surface are collected for six consecutive sampling periods before and after the end of each light observation period. The absolute value of the difference between the average dissolved oxygen concentration on the substrate surface for the six sampling periods before the end of the light observation period and the average dissolved oxygen concentration on the substrate surface for the six sampling periods after the start of the dark observation period is calculated. A difference exceeding 0.3 mg / L is used as a condition for determining that metabolic transition has occurred. The first time point when the trend of dissolved oxygen concentration on the substrate surface changes from increasing to decreasing is defined as the start time of the light-to-dark metabolic transition. The third time point when the trend of dissolved oxygen concentration on the substrate surface continues to decrease for three consecutive sampling periods after the trend reverses is defined as the end time of the light-to-dark metabolic transition. The time period between the start time and the end time of the light-to-dark metabolic transition constitutes the diurnal metabolic transition interval of the light-to-dark benthic diatomaceous membrane. Using the same method, dissolved oxygen concentration values ​​on the substrate surface are collected for six consecutive sampling periods before and after the end of each dark observation period to determine the diurnal metabolic transition interval of the dark-to-light benthic diatomaceous membrane. When there are differences in the start time of the diurnal metabolic transition interval of benthic diatom film determined by the position observation trajectory corresponding to different attachment substrate location markers in the same seedling pond, the earliest value of the start time of the diurnal metabolic transition interval of benthic diatom film corresponding to all attachment substrate location markers is taken as the unified start time of the diurnal metabolic transition interval of benthic diatom film, and the latest value of the end time of the diurnal metabolic transition interval of benthic diatom film corresponding to all attachment substrate location markers is taken as the unified end time of the diurnal metabolic transition interval of benthic diatom film, so as to ensure that all position observation trajectories in the same seedling pond use the same boundary of the diurnal metabolic transition interval of benthic diatom film for segmented registration.

[0028] When segmenting and registering location observation trajectories according to the diurnal metabolic transition interval of benthic diatom membranes, all attached micro-area observation records in each location observation trajectory are divided into four time period labels based on their timestamps: attached micro-area observation records belonging to the light period observation segment but not within the diurnal metabolic transition interval of benthic diatom membranes are marked as light period stable segment labels; attached micro-area observation records belonging to the dark period observation segment but not within the diurnal metabolic transition interval of benthic diatom membranes are marked as dark period stable segment labels; attached micro-area observation records belonging to the diurnal metabolic transition interval of benthic diatom membranes from light to dark are marked as light to dark transition segment labels; and attached micro-area observation records belonging to the diurnal metabolic transition interval of benthic diatom membranes from dark to light are marked as dark to light transition segment labels. After completing the time period labeling, the continuity of attached micro-area observation records with the same time period label within the same location observation trajectory is verified. The verification standard is: the timestamp difference between attached micro-area observation records with the same time period label should not exceed twice the sampling period time interval. If the timestamp difference between attached micro-area observation records with the same time period label exceeds twice the sampling period time interval, it is determined that there is a registration breakpoint between the two attached micro-area observation records. A breakpoint mark is inserted at the registration breakpoint position of the location observation trajectory, and the attached micro-area observation record sequences on both sides of the registration breakpoint are treated as independent registration segments and processed separately without merging across the registration breakpoint.

[0029] To maintain the correspondence between attachment base locations, when generating the attachment micro-region state sequence, the attachment base location identifier is used as a constraint. Continuous attachment micro-region observation record sequences with the same time period label and no registration breakpoints are arranged sequentially according to the time period label category and the order of appearance of the time period labels on the time axis, forming the attachment micro-region state sequence corresponding to the attachment base location identifier. The storage structure of the attachment micro-region state sequence adds a time period label field to the 12 data fields of the location observation trajectory. The time period label field takes one of four values: light-period stable segment label, dark-period stable segment label, light-to-dark transition segment label, and dark-to-light transition segment label. The time period label field and the timestamp field together form a dual index for the attachment micro-region state sequence, uniquely locating the time and metabolic stage of any attachment micro-region observation record in the sequence. All attachment micro-region state sequences corresponding to attachment base location identifiers within the seedling pond are aggregated and stored. The attachment micro-region observation records in any two attachment micro-region state sequences with the same timestamp maintain a one-to-one correspondence with the attachment base location identifier.

[0030] S3: Perform difference analysis on the overall water quality parameters of the seedling pond and the state sequence of the attached micro-regions to obtain the mismatch of the attached micro-regions, and identify the coupled fluctuation categories of pH, dissolved oxygen, and ammonia toxicity based on the mismatch of the attached micro-regions, including: When mapping the overall water quality parameters of the seedling pond to the location of the attached substrate at the same moment in the attached micro-area state sequence according to the sampling time, and forming a parameter comparison group, the attachment micro-area observation records in the attached micro-area state sequence are extracted one by one based on the dual index composed of the time period label field and the timestamp field in the attached micro-area state sequence. For each attachment micro-area observation record, the value of the timestamp field in the attachment micro-area observation record is taken. The attachment micro-area observation records corresponding to all attachment substrate location identifiers with the same timestamp field value are located in the attachment micro-area state sequence. The values ​​of the attached substrate surface pH, attached substrate surface dissolved oxygen concentration, and attached substrate surface ammonia nitrogen concentration fields in the attachment micro-area observation records corresponding to all attached substrate location identifiers under the same timestamp are matched one-to-one with the values ​​of the overall pH, overall dissolved oxygen concentration, and overall ammonia nitrogen concentration fields of the seedling pond in the attachment micro-area observation records, forming one parameter comparison group. The data structure of the parameter comparison group includes nine fields: timestamp, substrate location identifier, time period label, substrate surface pH, overall seedling pond pH, substrate surface dissolved oxygen concentration, overall seedling pond dissolved oxygen concentration, substrate surface ammonia nitrogen concentration, and overall seedling pond ammonia nitrogen concentration. The timestamp, substrate location identifier, and time period label fields are derived from the corresponding observation records of the substrate micro-regions in the substrate micro-region state sequence. The overall seedling pond pH, overall dissolved oxygen concentration, and overall ammonia nitrogen concentration fields are taken as the values ​​of the corresponding fields in the observation records of the substrate micro-regions with the same timestamp as the parameter comparison group. When multiple seedling pond overall water quality parameter collection points have values ​​for overall pH, overall dissolved oxygen concentration, or overall ammonia nitrogen concentration at the same timestamp, the arithmetic mean of the corresponding field values ​​from all seedling pond overall water quality parameter collection points at the same timestamp is taken as the reference value for the overall pH, overall dissolved oxygen concentration, and overall ammonia nitrogen concentration of the seedling pond in the parameter comparison group. The arithmetic mean retains the same decimal precision as the original collected data values. The parameter comparison groups corresponding to all timestamps in the attached micro-region state sequence are arranged in ascending order of timestamps to form a parameter comparison group sequence. Each parameter comparison group in the parameter comparison group sequence maintains a one-to-one correspondence with the attached micro-region observation record of the corresponding timestamp in the attached micro-region state sequence.

[0031] Based on the parameter comparison groups, the deviation direction, deviation magnitude, and continuous change relationship of the water quality parameters on the surface of the attached substrate and the overall water quality parameters of the seedling pond in terms of pH, dissolved oxygen, and ammonia toxicity were extracted. When generating the mismatch amount of the attached micro-region, for each parameter comparison group in the parameter comparison group sequence, the deviation direction, deviation magnitude, and continuous change relationship were extracted for the three parameter dimensions of pH, dissolved oxygen concentration, and ammonia nitrogen concentration. The method for extracting the deviation direction is as follows: the difference between the pH of the attached substrate surface and the pH of the overall seedling pond is calculated. When the difference is positive, the pH deviation direction is recorded as positive; when the difference is negative, the pH deviation direction is recorded as negative; when the difference is 0, the pH deviation direction is recorded as no deviation. The same method is used to calculate the difference between the dissolved oxygen concentration on the attached substrate surface and the dissolved oxygen concentration of the overall seedling pond to determine the dissolved oxygen concentration deviation direction. The difference between the ammonia nitrogen concentration on the attached substrate surface and the ammonia nitrogen concentration of the overall seedling pond is calculated to determine the ammonia nitrogen concentration deviation direction. A positive deviation in ammonia nitrogen concentration indicates that the ammonia nitrogen concentration on the attached substrate surface is higher than that in the overall seedling pond, representing that the ammonia toxicity characterization of the attached substrate surface area is higher than that of the overall seedling pond. The deviation range was extracted as follows: the absolute value of the difference between the pH of the substrate surface and the pH of the seedling pond in the parameter comparison group was taken as the pH deviation range, in pH units; the absolute value of the difference between the dissolved oxygen concentration of the substrate surface and the dissolved oxygen concentration of the seedling pond was taken as the dissolved oxygen concentration deviation range, in milligrams per liter; and the absolute value of the difference between the ammonia nitrogen concentration of the substrate surface and the ammonia nitrogen concentration of the seedling pond was taken as the ammonia nitrogen concentration deviation range, in milligrams per liter. The method for extracting continuous variation relationships is as follows: Five consecutive parameter comparison groups in the parameter comparison group sequence, identified by the same substrate location, are used as one calculation window. The pH deviation values ​​of the five consecutive parameter comparison groups within the calculation window are calculated in chronological order of sampling time. If three or more of the five differences are positive, the continuous variation direction of pH deviation is recorded as an expanding trend; if three or more of the five differences are negative, the continuous variation direction of pH deviation is recorded as a contracting trend; otherwise, the continuous variation direction of pH deviation is recorded as a fluctuating trend. The continuous variation directions of dissolved oxygen concentration deviation and ammonia nitrogen concentration deviation are determined using the same method. Nine extraction results corresponding to the same parameter comparison group, namely pH deviation direction, pH deviation magnitude, continuous change direction of pH deviation magnitude, dissolved oxygen concentration deviation direction, dissolved oxygen concentration deviation magnitude, continuous change direction of dissolved oxygen concentration deviation magnitude, ammonia nitrogen concentration deviation direction, ammonia nitrogen concentration deviation magnitude, and continuous change direction of ammonia nitrogen concentration deviation magnitude, are combined with the timestamp, substrate location identifier, and time period label fields of the parameter comparison group to form a single attachment micro-region mismatch record. All attachment micro-region mismatch records corresponding to the parameter comparison group sequence are arranged in ascending order by timestamp to form an attachment micro-region mismatch sequence. The attachment micro-region mismatch sequence and the parameter comparison group sequence maintain a one-to-one correspondence in timestamp and substrate location identifier.

[0032] Based on the combined variation characteristics of the mismatch amount of attached micro-regions during the light observation period, dark observation period, and diurnal metabolic transition interval of benthic diatom membranes, when identifying the coupled fluctuation categories of pH, dissolved oxygen, and ammonia toxicity, the time period label field in the mismatch amount sequence of attached micro-regions is used as the grouping basis. All mismatch amount records of attached micro-regions in the mismatch amount sequence are classified into four subsets according to the value of the time period label field: mismatch quantum set of stable light period, mismatch quantum set of stable dark period, mismatch quantum set of light-to-dark transition, and mismatch quantum set of dark-to-light transition. The same attachment substrate location is identified in each of the mismatch quantum sets of stable light period, stable dark period, light-to-dark transition, and dark-to-light transition, forming a continuous subset trajectory. For each subset trajectory, the dominant values ​​of pH deviation direction, dissolved oxygen concentration deviation direction, and ammonia nitrogen concentration deviation direction in the mismatch records of the attached micro-regions within the subset trajectory are counted. The method for determining the dominant value is as follows: count the occurrence frequency of the three deviation directions (positive deviation, negative deviation, and no deviation) within the subset trajectory, and take the deviation direction with the most occurrence frequency as the dominant deviation direction. When the occurrence frequency is the same, take the deviation direction value corresponding to the most recent sampling period as the dominant deviation direction.The dominant deviation directions of the mismatch quantum sets during the light-period stable phase, the dominant deviation directions of the mismatch quantum sets during the dark-period stable phase, the continuous change directions of pH deviation and dissolved oxygen concentration deviation in the mismatch quantum sets during the light-to-dark transition phase, and the continuous change direction of ammonia nitrogen concentration deviation in the mismatch quantum sets during the dark-to-light transition phase are combined to form a coupled fluctuation characteristic combination. The pH, dissolved oxygen, and ammonia toxicity coupled fluctuation categories are identified based on this coupled fluctuation characteristic combination, with the following identification rules: When the dominant value of both the pH and dissolved oxygen concentration deviation directions in the mismatch quantum sets during the light-period stable phase are positive, and simultaneously, the continuous change direction of both pH and dissolved oxygen concentration deviations in the mismatch quantum sets during the light-to-dark transition phase shows an increasing trend, it is identified as a photosynthetic excess type coupled fluctuation category, indicating a systemic increase in pH and dissolved oxygen concentration on the substrate surface due to a persistently high photosynthetic rate of the benthic diatom film; when the dominant value of both the dissolved oxygen concentration and ammonia nitrogen concentration deviation directions in the mismatch quantum sets during the dark-period stable phase are negative, and the dominant value of the ammonia nitrogen concentration deviation direction is positive, the same... When the deviation of ammonia nitrogen concentration in the mismatch quantum set during the dark-to-light transition period shows an increasing trend, it is identified as a coupled fluctuation type of oxygen consumption and ammonia accumulation during the dark period, characterizing the superimposed effect of respiration consumption by benthic diatoms during the dark period and ammonia nitrogen excretion by abalone larvae. When the pH deviation in both the mismatch quantum sets during the light-to-dark and dark-to-light transition periods exceeds 0.3 pH units and the dissolved oxygen concentration deviation exceeds 1.0 mg / L, and the continuous change direction of the three deviations in both the mismatch quantum sets during the light-to-dark and dark-to-light transition periods is a wave-like pattern... When the trend is dynamic, it is identified as a high-amplitude coupled fluctuation category throughout the entire time period, which indicates that there is a significant deviation between the microenvironment of the substrate surface and the overall water quality parameters of the seedling pond throughout the day and night. Coupled fluctuation feature combinations that do not meet any of the above identification rules are identified as low-amplitude stable coupled fluctuation categories, which indicates that the deviation between the microenvironment of the substrate surface and the overall water quality parameters of the seedling pond is within an acceptable range and does not show a continuous expanding trend. The identification results of the above four pH, dissolved oxygen and ammonia toxicity coupled fluctuation categories are recorded with the substrate location identifier and the current timestamp as indexes.

[0033] S4: Based on the coupled fluctuation category, invoke preset linkage rules to combine aeration, water exchange, shading, and circulation paths to generate candidate control sequences, including: The preset linkage rules are stored in tabular form. Each rule item in the preset linkage rules contains four fields: coupling fluctuation category, mismatch condition, action type, and action parameter. The coupling fluctuation category field takes one of four types: photosynthetic excess coupling fluctuation, dark period oxygen consumption and ammonia accumulation coupling fluctuation, all-time high amplitude coupling fluctuation, and low amplitude stable coupling fluctuation. The mismatch condition field records the numerical range constraints of pH deviation, dissolved oxygen concentration deviation, and ammonia nitrogen concentration deviation in the mismatch record of the attached micro-area that must be met to trigger the rule item. The action type field takes one of four actions: oxygenation, water exchange, shading, and circulation path. The action parameter field records three parameters: duration, intensity level, and range of action corresponding to the action type field. The initial content of the preset linkage rules is determined based on the water quality parameters suitable for the growth of abalone larvae. The suitable pH range is 7.8 to 8.3, the suitable dissolved oxygen concentration range is 6.0 to 9.0 mg / L, and the suitable ammonia nitrogen concentration range is 0 to 0.5 mg / L. Deviations exceeding the suitable range will trigger the corresponding rule item, while deviations within the suitable range will not trigger the rule item.

[0034] When matching the rule items corresponding to the mismatch amount of the attached micro-region in the preset linkage rules, the identification results of the pH, dissolved oxygen, and ammonia toxicity coupling fluctuation categories are used as search conditions. All rule items whose values ​​in the coupling fluctuation category field are consistent with the identification results of the pH, dissolved oxygen, and ammonia toxicity coupling fluctuation categories are selected from the preset linkage rules to form a candidate rule item set. Within the candidate rule item set, based on the pH deviation, dissolved oxygen concentration deviation, and ammonia nitrogen concentration deviation recorded in the attached micro-region mismatch amount sequence corresponding to the current timestamp, the numerical range constraints recorded in the mismatch amount condition field of each rule item in the candidate rule item set are verified one by one to see if they are satisfied. Rule items whose numerical range constraints recorded in the mismatch amount condition field are all satisfied are retained, while rule items whose numerical range constraints recorded in any one of the mismatch amount condition fields are not satisfied are removed from the candidate rule item set, forming a matching rule item set. The action type and action parameter fields of each rule item in the matching rule item set are extracted. The extracted values ​​of the action type field corresponding to oxygenation, water exchange, shading, and circulation paths are combined with the values ​​of duration, intensity level, and range of action parameter fields to form action combination relationships. The action combination relationships are stored in list form, where each element corresponds to one action type and its corresponding duration, intensity level, and range. The number of elements in the action combination relationship list is equal to the number of rule items in the matching rule item set. When the matching rule item set is empty, meaning that the combination of the current coupled wave category and the mismatch of the attached micro-area has not triggered any rule item in the preset linkage rules, the action combination relationship list is set to an empty list, and the matching is re-executed after the attached micro-area mismatch sequence of the next sampling period is updated.

[0035] Based on the action combination relationships, the preceding, succeeding, and synchronizing relationships among oxygenation, water exchange, shading, and circulation paths are distinguished. When forming the corresponding category-based control arrangement relationships, the execution sequence of the four action types (oxygenation, water exchange, shading, and circulation) in the action combination relationship list is analyzed according to the water quality response mechanism of abalone seedling environmental control. A preceding relationship means that the execution of one action type must be completed before the execution of another action type. The basis for determining the preceding relationship is that after the execution of the preceding action type, the water quality parameters related to the preceding action type in the seedling pond must reach a stable state before the subsequent action type can effectively exert its regulatory effect. The stabilization waiting time of the preceding relationship is determined based on the duration value in the action parameter field, and the waiting time is equal to the duration of the preceding action type. A succeeding relationship means that the execution of one action type must be initiated only after the execution of another action type. The succeeding relationship and the preceding relationship are inversely related; one pair of preceding relationships corresponds to one pair of succeeding relationships. Synchronization refers to the complete overlap and simultaneous initiation of two action types. The basis for determining synchronization is that the water quality control effects of the two action types are independent and do not interfere with each other; simultaneous initiation will not reduce the control effect of either action type. Specific rules for determining precedence, succession, and synchronization relationships are as follows: When the action combination relationship list includes both water exchange and aeration actions, a precedence relationship is established between them. The water exchange action is the preceding action type, and the aeration action is the following action type. This is because the fresh water introduced during water exchange needs to first mix and stabilize with the existing water in the nursery pond before the aeration action can accurately affect the mixed water. When the action combination relationship list includes both shading and circulation path actions, a synchronization relationship is established between them. The shading action reduces the photosynthetic rate of the benthic diatom film. The effects of the aeration and circulation path action types on uniform water flow are independent of each other, and the two action types can be started simultaneously without mutual interference. When the action combination relationship list contains both the aeration action type and the circulation path action type, a synchronous relationship is established between them. When the action combination relationship list contains both the water exchange action type and the shading action type, a priority relationship is established between them. The shading action type is the first action type, and the water exchange action type is the subsequent action type. This is because after shading reduces the photosynthetic oxygen production rate of the benthic diatom film, the impact of water disturbance caused by the water exchange operation on the dissolved oxygen concentration can be effectively controlled.Summarize the preceding, succeeding, and synchronizing relationships established between all action types in the action combination relationship list according to the above-defined rules to form the control and arrangement relationships corresponding to the categories. The control and arrangement relationships are stored in the form of a directed graph. Each node in the directed graph corresponds to one action type in the action combination relationship list. Each directed edge in the directed graph corresponds to the preceding or succeeding relationship between two nodes. Undirected edges in the directed graph correspond to the synchronizing relationship between two nodes. The number of nodes in the directed graph of the control and arrangement relationships is equal to the number of elements in the action combination relationship list.

[0036] When generating candidate control sequences by combining oxygenation, water exchange, shading, and circulation paths according to the control arrangement relationships, a directed graph of the control arrangement relationships is used as input. Topological sorting is performed on the directed graph. The method for topological sorting is as follows: nodes with an in-degree of 0 in the directed graph (i.e., action type nodes without any preceding action type nodes) are identified as the action types to be initiated in the first batch of the execution sequence. The maximum duration of the action parameter field of all action types initiated in the first batch is taken as the duration of the first batch. After the duration of the first batch ends, nodes in the directed graph whose predecessor nodes all belong to the first batch are identified as the action types to be initiated in the second batch, and the duration of the second batch is determined in the same way. This process is repeated until all nodes in the directed graph belong to their corresponding batches, completing the sorting of all batches. For nodes connected by undirected edges within the same batch, all nodes connected by undirected edges are marked as synchronously initiated, and the synchronous initiation flag is recorded in the candidate control sequence. The action type, duration, intensity level, and scope of action parameters corresponding to each batch, as well as the synchronization start flag within the batch, are arranged in ascending order by batch number to form a candidate control sequence. The data structure of the candidate control sequence includes six fields: batch number, action type within the batch, duration within the batch, intensity level within the batch, scope of action within the batch, and synchronization start flag. The value of the batch number field in the candidate control sequence starts from 1 and increments by integer. The maximum value of the batch number field is equal to the total number of batches obtained by topological sorting. The candidate control sequence is stored using timestamps and attachment base location identifiers as indexes.

[0037] S5: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, candidate control sequences are screened and time-period rearranged to generate remote execution sequences, including: When extracting continuous stable intervals corresponding to the location of the attachment base based on the state sequence of the attachment micro-region, the attachment base location identifier field in the state sequence of the attachment micro-region is used as the retrieval primary key. The time period label field of the state sequence of the attachment micro-region corresponding to each attachment base location identifier is scanned one by one. The attachment micro-region observation record sequence whose time period label field value is continuously the bright period stable segment label or the dark period stable segment label is identified as the candidate stable interval. The validity determination method for candidate stable intervals is as follows: count the number of observation records of continuously attached micro-areas within the candidate stable interval, multiply the number of observation records of continuously attached micro-areas within the candidate stable interval by the time interval of the sampling period to obtain the duration of the candidate stable interval. The condition for determining the validity of the candidate stable interval is that the duration of the candidate stable interval is not less than the preset continuous stable duration threshold. The preset continuous stable duration threshold is set according to the minimum time requirement for environmental stability of abalone larvae metamorphic attachment behavior. The completion of abalone larvae metamorphic attachment behavior requires the surface water quality parameters of the attachment substrate to remain stable within a continuous time period. It is usually required that the continuous stable duration of water quality parameters is not less than 3 sampling periods. For example, when the time interval of the sampling period is set to 5 minutes, the preset continuous stable duration threshold can be set to 15 minutes, corresponding to a number of observation records of continuously attached micro-areas within the candidate stable interval not less than 3. Candidate stable intervals that meet the duration requirement are marked as continuous stable intervals. The data structure of a continuous stable interval includes five fields: attachment base location identifier, continuous stable interval start timestamp, continuous stable interval end timestamp, continuous stable interval duration, and continuous stable interval time period label type. The continuous stable interval time period label type field takes the value of either the bright period stable segment label or the dark period stable segment label, corresponding to the consistent value of the time period label field of all attached micro-area observation records within the continuous stable interval. Candidate stable intervals that do not meet the duration requirement are marked as invalid stable intervals, and invalid stable intervals are not included in the subsequent determination of the abnormal attachment time window. All valid continuous stable intervals under the same attachment base location identifier are sorted in ascending order according to the value of the continuous stable interval start timestamp field, forming a continuous stable interval sequence corresponding to the attachment base location identifier. The continuous stable interval sequence and the attached micro-area state sequence maintain a one-to-one relationship through the attachment base location identifier field.

[0038] When determining the metamorphic attachment window by combining pH, dissolved oxygen, and ammonia toxicity coupled fluctuation categories, the attachment substrate location identifier field and the continuous stable interval start timestamp field of each continuous stable interval in the continuous stable interval sequence are used as indexes. The coupled fluctuation category corresponding to the start timestamp of the continuous stable interval is located in the pH, dissolved oxygen, and ammonia toxicity coupled fluctuation category identification results. The located coupled fluctuation category is then associated with the continuous stable interval to form a continuous stable interval labeled with the coupled fluctuation category. The method for determining the metamorphic attachment window is as follows: the metamorphic attachment behavior of abalone larvae is dually dependent on water quality stability and a specific water quality range. The metamorphic attachment window must simultaneously meet two conditions: the continuous stable interval is valid and the coupled fluctuation category is a low-amplitude stable coupled fluctuation category. When the coupled fluctuation category corresponding to the continuous stable interval is a low-amplitude stable coupled fluctuation category, the time period from the start timestamp of the continuous stable interval to the end timestamp of the continuous stable interval is determined as the metamorphic attachment window. The data structure of the metamorphic attachment window includes an attachment substrate location identifier field, a metamorphic attachment window start timestamp field, a metamorphic attachment window end timestamp field, and a metamorphic attachment time... The window duration field consists of four fields. The value of the abnormal attachment window duration field is equal to the difference between the abnormal attachment window end timestamp field and the abnormal attachment window start timestamp field. When the coupled fluctuation category corresponding to the continuous stable interval is a photosynthetic excess coupled fluctuation category, a dark period oxygen consumption ammonia accumulation coupled fluctuation category, or a full-time high amplitude coupled fluctuation category, the continuous stable interval does not meet the coupled fluctuation condition of the abnormal attachment window. Therefore, the continuous stable interval is not marked as an abnormal attachment window, and the time period corresponding to the continuous stable interval is regarded as a non-abnormal attachment window period. No abnormal attachment protective constraints are applied to the candidate control sequences within the non-abnormal attachment window period. All abnormal attachment windows that meet the conditions under the same attachment base location identifier are sorted in ascending order according to the value of the abnormal attachment window start timestamp field, forming an abnormal attachment window sequence corresponding to the attachment base location identifier. The abnormal attachment window sequence is stored with the attachment base location identifier field as the primary key.

[0039] Based on the temporal correspondence between the abnormal attachment time window and the action combination relationship in the candidate control sequence, when filtering out candidate control content that conflicts with the abnormal attachment time window, the batch number field in the candidate control sequence is used as the traversal order, and the execution time interval corresponding to each batch in the candidate control sequence is calculated batch by batch. The start time of the execution time interval is determined by the sum of the batch number field and the intra-batch duration field values ​​of the preceding batches. The start time of the execution time interval of the first batch is equal to the current timestamp corresponding to the candidate control sequence. The start time of the execution time interval of the second batch and subsequent batches is equal to the current timestamp plus the sum of the intra-batch duration field values ​​of all preceding batches. The end time of the execution time interval is equal to the start time of the execution time interval plus the intra-batch duration field value. The execution time interval of each batch is overlapped with the time interval between the start and end timestamps of all abnormal attachment windows under the same attachment base position in the abnormal attachment window sequence. The overlap detection method is as follows: if the start time of the batch's execution time interval is earlier than the end timestamp of the abnormal attachment window and the end time of the batch's execution time interval is later than the start timestamp of the abnormal attachment window, then the execution time interval of the batch is determined to have time overlap with the abnormal attachment window. All fields in the row of the batch number field in the candidate control sequence corresponding to the batch with time overlap are marked as conflict flags, and the batch with conflict flags is removed from the candidate control sequence and will not be included in the subsequent time period rearrangement process. If the execution time interval of the batch does not have time overlap with any abnormal attachment window, then all fields of the batch in the candidate control sequence are retained, and no conflict flag is applied. After removing batches with conflict flags, the batch number field of the remaining batches is renumbered consecutively starting from 1. The value of the batch number field in the candidate control sequence after renumbering is continuously increasing and there are no gaps. The candidate control sequence after removing conflicting batches and completing the batch numbering rearrangement is named the filtered candidate control sequence. The filtered candidate control sequence is stored with the timestamp and attachment base location identifier fields as indexes.

[0040] After screening, the candidate control sequences are rearranged according to the start and end changes of the abnormal attachment time window. When generating remote execution sequences, the start and end timestamp fields of each abnormal attachment time window under the attachment base position identifier corresponding to the current timestamp in the abnormal attachment time window sequence are used as the boundary parameters for time period rearrangement. Time period offset processing is performed on the execution time interval of all batches in the candidate control sequences after screening. The time period offset handling method is as follows: If the start time of the execution time interval of a batch in the selected candidate control sequence falls before the start time stamp of the abnormal attachment time window of a certain abnormal attachment time window in the abnormal attachment time window sequence, and the end time of the batch's execution time interval falls before the start time stamp of the abnormal attachment time window, then the batch does not need time period offset and the execution time interval remains unchanged; if the start time of the execution time interval of a batch in the selected candidate control sequence falls after the end time stamp of the abnormal attachment time window, then the batch does not need time period offset and the execution time interval remains unchanged; if the start time of the execution time interval of a batch in the selected candidate control sequence falls between the start time stamp and the end time stamp of the abnormal attachment time window, then the start time of the batch's execution time interval is shifted to the end time stamp of the abnormal attachment time window, and the execution time intervals of all subsequent batches are shifted backward by the same amount of time, the shift time being equal to the difference between the end time stamp of the abnormal attachment time window and the start time of the original execution time interval of the batch. After completing the time period offset processing, all batches in the filtered candidate control sequences are sorted in ascending order according to the start time of the execution time interval after the time period offset processing. The batch number field of all batches is reassigned continuously starting from 1. The candidate control sequences that have completed the time period rearrangement are named remote execution sequences. The data structure of the remote execution sequence adds an execution start timestamp field to the six fields of the candidate control sequence: batch number field, action type field within the batch field, duration field within the batch field, intensity level field within the batch field, scope field within the batch field, and synchronization start flag field. The execution start timestamp field records the start time of the execution time interval of each batch after the time period rearrangement. The remote execution sequence is stored with the attachment base position identifier field and the execution start timestamp field as dual indexes.

[0041] S6: Send a remote execution sequence to the remote execution device, and collect the surface water quality parameters of the substrate after execution and write them back to the preset linkage rules, including: The remote execution sequence is sent to the corresponding remote execution device according to the action combination relationship. When recording the execution time corresponding to the action combination relationship, the dual index composed of the attachment base position identifier field and the execution start timestamp field in the remote execution sequence is used as the basis for sending. Sending instructions are generated batch by batch in the remote execution sequence according to the value of the batch number field from small to large. The data structure of each sending instruction includes seven fields: target device identifier field, action type instruction field, duration instruction field, intensity level instruction field, scope instruction field, synchronization start flag instruction field, and planned execution start timestamp field. The value of the target device identifier field is determined according to the action type instruction field. The oxygenation action type corresponds to the oxygenation remote execution device, the water change action type corresponds to the water change remote execution device, the shading action type corresponds to the shading remote execution device, and the circulation path action type corresponds to the circulation path remote execution device. The target device identifier field records the device address of the remote execution device corresponding to the action type instruction field in the seedling pond management network. The device address format is a string composed of the seedling pond number and the device type number, which uniquely identifies the communication entry point of each remote execution device in the seedling pond management network. The values ​​of the duration instruction field, intensity level instruction field, and scope instruction field are directly inherited from the batch duration field, batch intensity level field, and batch scope field of the corresponding batch in the remote execution sequence. The value of the planned execution start timestamp field is equal to the execution start timestamp field value of the corresponding batch in the remote execution sequence. Each sending command is unicasted through the seedling pool management network to the remote execution device corresponding to the target device identifier field. The sending method employs a reliable transmission method with acknowledgment. After receiving the sending command, the remote execution device returns a reception acknowledgment message to the control terminal. The reception acknowledgment message includes the remote execution device identifier field and a reception timestamp field, accurate to the second. The maximum waiting time for the control terminal to wait for the reception acknowledgment message after issuing the sending command is set as follows: three times the historical average round-trip time between the control terminal and various remote execution devices in the seedling pool management network. For example, when the historical average round-trip time is 2 seconds, the maximum waiting time can be set to 6 seconds. If no reception acknowledgment message is received within the maximum waiting time, the sending command is resent to the remote execution device corresponding to the target device identifier field. The maximum number of resentments is set to 3. If no reception acknowledgment message is received after 3 resentments, the remote execution device corresponding to the target device identifier field is marked as having a communication error, the communication error record is written to the error log, and the sending commands for the remaining batches in the remote execution sequence continue to be sent.

[0042] When recording the execution time corresponding to the action combination relationship, the planned execution end timestamp corresponding to each sent instruction is calculated based on the planned execution start timestamp field value and the duration instruction field value in the sent instruction. The planned execution end timestamp is equal to the planned execution start timestamp field value plus the duration instruction field value. The planned execution start timestamp field value and the planned execution end timestamp field value are used together as the execution time corresponding to the action combination relationship. The data structure of the execution time record includes six fields: attachment base location identifier field, batch number field, action type field, planned execution start timestamp field, planned execution end timestamp field, and synchronization start flag field. The execution time records corresponding to all batches in the remote execution sequence are sorted in ascending order according to the planned execution start timestamp field value, forming an execution time record sequence. The execution time record sequence is stored with the attachment base location identifier field and the batch number field as dual indexes. The number of rows in the execution time record sequence is equal to the maximum value of the batch number field in the remote execution sequence, that is, the total number of all batches.

[0043] After the remote execution device completes the remote execution sequence, it collects surface water quality parameters of the attached substrate according to the execution time. When forming the post-execution parameter correspondence, the planned execution termination timestamp field value of each execution time record in the execution time record sequence is used as the collection trigger time. At the time corresponding to the planned execution termination timestamp field value, a collection trigger command is sent to all surface water quality parameter collection points of the attached substrate corresponding to the attached substrate location identifier field in the execution time record. The method of sending the collection trigger command is the same as that of sending the command, and a reliable transmission method with acknowledgment is adopted. The maximum waiting time and the upper limit of the number of resends are set in the same way as those of sending the command. Upon receiving the collection trigger command, the water quality parameter collection points on the substrate surface immediately perform one simultaneous collection of pH, dissolved oxygen concentration, and ammonia nitrogen concentration on the substrate surface. The collection results, along with the substrate location identifier and actual collection timestamp carried by the collection points, are transmitted back to the control terminal. The actual collection timestamp is accurate to the second. The deviation between the actual collection timestamp and the planned execution termination timestamp does not exceed 0.5 times the sampling cycle time interval. Collection results exceeding the deviation range are marked as time deviation anomalies. Collection results with time deviation anomalies are not included in the generation of parameter correspondence after execution. The collected values ​​of pH, dissolved oxygen concentration, and ammonia nitrogen concentration on the substrate surface are correlated one-to-one with execution time records in the execution time record sequence that share the same substrate location identifier field and whose planned execution termination timestamp value deviates from the actual collection timestamp within the allowable range. The correlated execution time records and their corresponding collected values ​​are combined into a single post-execution parameter correspondence record. This record contains nine fields: substrate location identifier, batch number, action type, planned execution start timestamp, planned execution termination timestamp, actual collection timestamp, post-execution substrate surface pH, post-execution substrate surface dissolved oxygen concentration, and post-execution substrate surface ammonia nitrogen concentration. All post-execution parameter correspondence records for each batch are sorted in ascending order of batch number value, forming a post-execution parameter correspondence relationship. This relationship is stored using both the substrate location identifier field and the batch number field as dual indexes.

[0044] When writing back the surface water quality parameters of the attached substrate to the rule item corresponding to the action combination relationship in the preset linkage rules based on the parameter correspondence after execution, the attached substrate location identifier field and action type field in the parameter correspondence after execution are used as the retrieval basis. The rule item with the same value of the action type field in the parameter correspondence after execution and the value of the coupling fluctuation category field is consistent with the pH, dissolved oxygen and ammonia toxicity coupling fluctuation category identification results corresponding to the current execution is located in the preset linkage rules, and the target rule item for writing back is determined. The method for generating the write-back content is as follows: Calculate the deviation of the pH field value on the substrate surface after execution from the upper limit (8.3 pH units) and lower limit (7.8 pH units) of the suitable pH range for abalone seedling growth in the parameter correspondence relationship after execution. Use whether the absolute value of the deviation exceeds 0.1 pH units as the condition for triggering the write-back. When the absolute value of the deviation of the pH field value on the substrate surface after execution exceeds 0.1 pH units, update the lower limit of the pH deviation range constraint in the mismatch condition field of the write-back target rule to the absolute value of the difference between the pH field value on the substrate surface after execution and the overall pH reference value of the seedling pond, minus 0.05 pH units, to refine the trigger sensitivity of the rule. The same method is used... When the absolute value of the dissolved oxygen concentration field on the substrate surface deviates by more than 0.2 mg / L after execution, the lower limit of the range constraint for the dissolved oxygen concentration deviation in the mismatch condition field of the write-back target rule item will be updated to the absolute value of the difference between the dissolved oxygen concentration field value on the substrate surface and the reference value of the overall dissolved oxygen concentration in the seedling pond minus 0.1 mg / L; when the absolute value of the ammonia nitrogen concentration field on the substrate surface deviates by more than 0.05 mg / L after execution, the lower limit of the range constraint for the ammonia nitrogen concentration deviation in the mismatch condition field of the write-back target rule item will be updated to the absolute value of the difference between the ammonia nitrogen concentration field value on the substrate surface and the reference value of the overall ammonia nitrogen concentration in the seedling pond minus 0.02 mg / L. The updated mismatch condition field value is written into the corresponding position of the write-back target rule item in the preset linkage rule. The write-back operation is performed in the overwrite mode. After the write-back is completed, the write-back timestamp and write-back source batch number are recorded in the preset linkage rule. The write-back timestamp is accurate to the second. After the preset linkage rule generates the pH, dissolved oxygen and ammonia toxicity coupled fluctuation category identification results in the next sampling cycle, the updated rule item content is called when matching the rule item corresponding to the mismatch of the attached micro-region in the preset linkage rule. This realizes the continuous adaptive update of the preset linkage rule according to the feedback of the execution effect. Example 2

[0045] The difference between Embodiment 2 and Embodiment 1 is that this embodiment introduces a remote intelligent control system for abalone seedling cultivation environment.

[0046] Figure 2A schematic diagram of a remote intelligent control system for abalone seedling cultivation environment is provided according to the present invention. The remote intelligent control system for abalone seedling cultivation environment includes: Micro-area acquisition module: Collects overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and forms an attachment micro-area observation set according to the location of the substrate; State registration module: Based on the diurnal metabolic transition of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-region is registered to generate the state sequence of the attachment micro-region; Mismatch Analysis Module: Performs difference analysis between the overall water quality parameters of the seedling pond and the state sequence of the attached micro-regions to obtain the mismatch amount of the attached micro-regions, and identifies the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch amount of the attached micro-regions. Linkage Combination Module: Based on the coupling fluctuation category, it calls preset linkage rules to combine oxygenation, water exchange, shading, and circulation paths to generate candidate control sequences; The time-series filtering module: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, it filters and rearranges the time periods of candidate control sequences to generate remote execution sequences; Execution write-back module: Sends remote execution sequence to remote execution device and collects water quality parameters of the attached substrate surface after execution to write back the preset linkage rules.

[0047] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0048] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0049] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0050] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0051] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0052] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0053] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 a computer device (which may 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 this application. 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.

[0054] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0055] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for remote intelligent control of abalone seedling cultivation environment, characterized in that, Includes the following steps: S1: Collect overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and form a micro-area observation set according to the location of the substrate; S2: Based on the diurnal metabolic transformation of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-regions is registered to generate the attachment micro-region state sequence; S3: Perform difference analysis on the overall water quality parameters of the seedling pond and the state sequence of the attached micro-region to obtain the mismatch of the attached micro-region, and identify the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch of the attached micro-region. S4: According to the coupling fluctuation category, the preset linkage rules are invoked to combine oxygenation, water exchange, shading and circulation paths to generate candidate control sequences; S5: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, candidate control sequences are screened and time period rearranged to generate remote execution sequences; S6: Send a remote execution sequence to the remote execution device and collect the surface water quality parameters of the substrate after execution and write them back to the preset linkage rules.

2. The method for remote intelligent control of abalone seedling cultivation environment according to claim 1, characterized in that, S1, specifically: Water quality parameter collection points for the entire seedling pond are set up along the water body distribution of the seedling pond, and water quality parameter collection points for the surface of the substrate are set up along the surface of the substrate. Within the same sampling period, the overall water quality parameters of the seedling pond, the surface water quality parameters of the substrate, the light parameters, and the local flow parameters were collected simultaneously. The synchronous acquisition results are time-aligned and bound to the location of the substrate, and then merged according to the location of the substrate to form an attached micro-area observation set.

3. The method for remote intelligent control of abalone seedling cultivation environment according to claim 2, characterized in that, S2, specifically: Arrange the continuous acquisition results corresponding to the same attachment substrate location in the attachment micro-area observation set according to the sampling time to form a location observation trajectory; Based on the changes in light parameters, the observation period was divided into light and dark observation periods, and the diurnal metabolic transition interval of benthic diatom film was determined by combining the changes in surface water quality parameters of the substrate. The location observation trajectory was segmented and registered according to the diurnal metabolic transition interval of the benthic diatom membrane to maintain the corresponding relationship of the attachment substrate and generate the state sequence of the attachment micro-region.

4. The method for remote intelligent control of abalone seedling cultivation environment according to claim 3, characterized in that, S3, specifically: The overall water quality parameters of the seedling pond were mapped to the location of the attachment substrate at the same moment in the state sequence of the attachment micro-region according to the sampling time, forming a parameter comparison group; Based on the parameter comparison group, the deviation direction, deviation magnitude and continuous change relationship of the surface water quality parameters of the attached substrate and the overall water quality parameters of the seedling pond in terms of pH, dissolved oxygen and ammonia toxicity were extracted to generate the mismatch amount of the attached micro-region. Based on the combined variation characteristics of the mismatch amount of attached micro-regions during the light observation period, dark observation period, and diurnal metabolic transition period of benthic diatom membranes, we can identify the coupled fluctuation categories of pH, dissolved oxygen, and ammonia toxicity.

5. The method for remote intelligent control of abalone seedling cultivation environment according to claim 4, characterized in that, S4, specifically: Based on the matching of pH, dissolved oxygen and ammonia toxicity coupling fluctuation categories, the rule items in the preset linkage rules corresponding to the mismatch amount of the attached micro-area are used to determine the action combination relationship corresponding to oxygenation, water exchange, shading and circulation paths; Based on the action combination relationship, distinguish the preceding, subsequent and synchronous relationships between oxygenation, water exchange, shading and circulation paths, and form a control arrangement relationship corresponding to the category; Candidate control sequences are generated by combining oxygenation, water exchange, shading, and circulation pathways according to the control arrangement relationship.

6. The method for remote intelligent control of abalone seedling cultivation environment according to claim 5, characterized in that, S5, specifically: Based on the state sequence of the attached micro-region, the continuous stable interval corresponding to the location of the attached substrate is extracted, and the abnormal attachment time window is determined by combining the coupled fluctuation categories of pH, dissolved oxygen and ammonia toxicity. Based on the temporal correspondence between the abnormal attachment time window and the action combination relationship in the candidate regulation sequence, candidate regulation content that conflicts with the abnormal attachment time window is screened out. The selected candidate control sequences are rearranged according to the start and end times of the abnormal attachment window to generate remote execution sequences.

7. The method for remote intelligent control of abalone seedling cultivation environment according to claim 6, characterized in that, S6, specifically: The remote execution sequence is sent to the corresponding remote execution device according to the action combination relationship, and the execution time corresponding to the action combination relationship is recorded. After the remote execution device completes the remote execution sequence, it collects water quality parameters on the surface of the attached substrate according to the execution time, forming a post-execution parameter correspondence. Based on the parameter correspondence after execution, the surface water quality parameters of the attached substrate are written back to the rule items in the preset linkage rules that correspond to the action combination relationship.

8. A remote intelligent control system for abalone seedling cultivation environment, used to implement the remote intelligent control method for abalone seedling cultivation environment as described in any one of claims 1-7, characterized in that, include: Micro-area acquisition module: Collects overall water quality parameters of the seedling pond, surface water quality parameters of the substrate, light parameters, and local flow parameters within the same sampling period, and forms an attachment micro-area observation set according to the location of the substrate; State registration module: Based on the diurnal metabolic transition of benthic diatom membranes and the location of the attachment substrate, the observation set of the attachment micro-region is registered to generate the state sequence of the attachment micro-region; Mismatch Analysis Module: Performs difference analysis between the overall water quality parameters of the seedling pond and the state sequence of the attached micro-regions to obtain the mismatch amount of the attached micro-regions, and identifies the coupling fluctuation categories of pH, dissolved oxygen and ammonia toxicity based on the mismatch amount of the attached micro-regions. Linkage Combination Module: Based on the coupling fluctuation category, it calls preset linkage rules to combine oxygenation, water exchange, shading, and circulation paths to generate candidate control sequences; The time-series filtering module: Based on the abnormal attachment time window corresponding to the attached micro-region state sequence, it filters and rearranges the time periods of candidate control sequences to generate remote execution sequences; Execution write-back module: Sends remote execution sequence to remote execution device and collects water quality parameters of the attached substrate surface after execution to write back the preset linkage rules.

Images