System and method for monitoring and managing an aquaculture environment
An AI-driven aquaculture system optimizes feeding schedules and quantities using real-time data capture and machine learning, addressing inefficiencies in manual methods and improving profitability and growth rates.
Patent Information
- Application Number
- PCT/MY2026/050049
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-07
- Filing Date
- 2026-04-28
- Publication Date
- 2026-11-12
AI Technical Summary
Existing aquaculture practices face inefficiencies due to manual monitoring and estimation methods, leading to inconsistent feeding strategies, increased mortality, and reduced profitability, as well as challenges in accurately measuring Feed Conversion Ratio (FCR) and determining optimal harvest times.
An AI-based system that integrates real-time data capture, pre-processing, cloud-based analytics, and automated feed delivery to optimize feeding schedules and quantities, using cameras and sensors to monitor environmental and stock parameters, and machine learning to adjust feeding dynamically.
Enables precise, real-time feed optimization, reducing waste, improving growth rates, and enhancing profitability by minimizing feed wastage and ensuring optimal harvest times.
Smart Images

Figure MY2026050049_12112026_PF_FP_ABST
Abstract
Description
[0001] SYSTEM AND METHOD FOR MONITORING AND MANAGING AN AQUACULTURE ENVIRONMENT FIELD OF INVENTION
[0002] The present invention relates to a system and method for monitoring and managing an aquaculture environment. In particular, the present invention relates to the system and method for monitoring and managing an aquaculture environment that utilizes an Al-based analytics module for determining an optimized feeding schedule and the optimized quantity of feed required for aquaculture stock.
[0003] BACKGROUND ART
[0004] Aquaculture has become a vital industry for meeting global food demands, with fish farming playing a significant role in sustainable food production. Existing technologies in aquaculture primarily rely on manual monitoring and basic tools to manage pond conditions, fish feeding, and harvest timing. Farmers often use traditional methods such as visual inspections, manual record-keeping, and generalized growth assumptions to make critical decisions. While these methods have been employed for decades, they are increasingly inadequate in addressing the complexities of modern aquaculture operations. Tools for monitoring environmental factors like pH and temperature, estimating fish weight, and calculating feed efficiency are either limited or require significant manual effort, leaving room for inefficiencies and errors.
[0005] Current aquaculture practices involve periodic manual checks of pond conditions, such as water quality and fish behaviour, which are time-consuming and often impractical for large or remote ponds. Farmers typically estimate fish weight and feeding amounts based on experience or outdated methods, leading to inconsistent feeding strategies. Feed Conversion Ratio, FCR, a critical metric for evaluating feed efficiency, is often calculated manually using fragmented data, which is prone to inaccuracies. Additionally, harvest timing is determined using generalized growth assumptions or manual observations, which fail to account for realtime variations in fish growth and environmental conditions. While some advanced tools exist, they are often inaccessible or impractical for small to medium-scale farmers, leaving them reliant on traditional, inefficient methods.
[0006] The existing technology in aquaculture faces several significant challenges. Firstly, the lack of real-time pond visibility hinders farmers' ability to monitor critical environmental factors like pH and temperature continuously, leading to delayed detection of issues such as changes in fish behaviour or hunger. This results in increased fish mortality, missed opportunities for timelyinterventions, and higher operational costs. Secondly, the reliance on estimations or outdated methods for determining fish weight and feeding amounts leads to overfeeding or underfeeding, causing wasted feed, poor water quality, stunted fish growth, and increased environmental impact. Thirdly, the difficulty in accurately measuring Feed Conversion Ratio due to manual record-keeping and fragmented data prevents farmers from optimizing feed utilization, increasing costs and reducing growth rates. Lastly, the uncertainty in harvest timing, driven by generalized growth assumptions and manual observations, leads to missed market opportunities, prolonged grow-out periods, and overcrowding in ponds, ultimately impacting scalability and profitability.
[0007] Numerous systems and methods for monitoring aquaculture environment have been developed.
[0008] One example of such system is disclosed in EP Patent Publication No. EP 4304348 A1 (hereinafter referred to as EP 348 A1 Publication) entitled “Al based feeding system and method for land-based fish farms” having a filing date of March 07, 2022, Applicant: Reeldata Inc. The EP 348 A1 Publication discloses an Al-based system for feed monitoring in land-based fish farms utilizes a feed camera mounted on an effluent pipe to capture video footage of objects, such as uneaten feed pellets, traversing its field of view. The system incorporates a specialized computer running a pellet-tracking algorithm, which includes modules for proposing regions of interest, ROI, classifying ROIs, tracking ROIs, and analysing trajectories. The EP 348 A1 Publication further discloses that an object enhancer is positioned opposite the feed camera on the effluent pipe, providing a contrasting background that ensures objects in the captured images have clear, definable perimeters.
[0009] Another example of such system is disclosed in KR Patent No. KR 102528286 B1 (hereinafter referred to as KR 286 B1 Patent) entitled “Automatic feeding system and method for fish farms based on Al” having a filing date of August 30, 2022, Applicant: Jeong Mi-jeong. The KR 286 B1 Patent discloses an Al-based automatic feeding system and method for fish farms, configured to objectively analyse operational data and determine the optimal feed supply amount. The KR 286 B1 Patent further discloses that the system comprises three main components which are data input part for collecting relevant farm data, an Al analysis part for processing and interpreting this data to calculate the ideal feeding requirements, and a supply control part for executing the feed distribution based on the Al's recommendations.
[0010] A further example of such system is disclosed in CN Patent Publication No. CN 117478835 A (hereinafter referred to as CN 835 A Publication) entitled “Intelligent fish monitoring system and control method”, having a filing date of October 30, 2023, Applicant: China Institute ofWater Resources and Hydropower Research. The CN 835 A Publication discloses an intelligent fish monitoring system and control method, comprising a fish image acquisition system, a monitoring end control system, and a remote monitoring system. The image acquisition system includes a lamp panel at the bottom of a river fishway and a waterproof camera above it, while the monitoring end control system integrates components like a power supply UPS, NVR, Al industrial computer, and ultrasonic imaging equipment. The remote monitoring system consists of an upper computer and a display at the monitoring center, enabling real-time oversight and analysis. The CN 835 A Publication further discloses that the system is deployed on fishways to continuously monitor video images and extract fish passage data, overcoming challenges like varying natural light conditions that affect image contrast and target identification.
[0011] As outlined above, various systems and methods for monitoring aquaculture environment have been developed. However, none of the existing systems, disclose such a configuration that at least overcomes the aforementioned drawbacks.
[0012] In light of the foregoing discussion, there exists a need to provide an improved system and method for monitoring aquaculture environment, that overcome at least the above-mentioned drawbacks.SUMMARY OF INVENTION
[0013] The present invention relates to a system and method for monitoring and managing an aquaculture environment. In particular, the present invention relates to the system and method for monitoring and managing an aquaculture environment that utilizes an Al-based analytics module for determining an optimized feeding schedule and the optimized quantity of feed required for aquaculture stock.
[0014] One aspect of the present invention provides a system (100) for monitoring and managing an aquaculture environment, comprising a monitoring module (102), configured to capture realtime data related to the aquaculture environment and aquaculture stock residing therein; a data pre-processing module (104) operably coupled to the monitoring module (102), configured to pre-process captured real-time data for further analysis; a cloud-based database arrangement (106) operably coupled to the data pre-processing module (104) via a data communication network (106a), the cloud-based database arrangement (106) is configured to store pre-processed real-time data received from the data pre-processing module (104) and facilitate real-time access thereof; an artificial intelligence, Al based analytics module (108) operably coupled to the cloud-based database arrangement (106), configured to analyse the pre-processed real-time data for optimizing feeding schedule and determining an optimized feed quantity; and a feed delivery module (110) operably coupled to the Al-based analytics module (108), configured to dynamically provide feed delivery to the aquaculture environment based on optimized feeding schedule and the feed quantity in real-time; and a user interface (112) operably coupled to the cloud-based database arrangement (106), the Al-based analytics module (108), and the feed delivery module (110), wherein the user interface (112) is configured to enable at least one user to monitor the aquaculture environment.
[0015] Another aspect of the present invention provides that the user interface is configured to receive the pre-processed real-time data from the cloud-based database arrangement and enable remote live-stream of the aquaculture environment including the pre-processed real-time data, the optimized feeding schedule, and the optimized feed quantity determined by the Al-based analytics module (108); and enable at least one user to remotely control the feed delivery module (110) based on user inputs.
[0016] Another aspect of the present invention provides that the monitoring module (102) further comprises at least one first camera (102a), configured to capture images and videos of movement and behaviour related to the aquaculture stock, wherein the at least one first camera is encased in a water resistant casing; at least one second camera (102b), configured to detect underwater aquaculture stock density and track the movement associated thereof; and anaquaculture environment monitoring module (102c) comprising a plurality of sensors configured to measure a plurality of parameters related to the aquaculture environment, wherein the plurality of parameters comprises at least one of pH, temperature, and turbidity -ammonia, dissolved oxygen, salinity.
[0017] Another aspect of the present invention provides that the at least one first camera (102a) is an Artificial Intelligence, Al-based camera.
[0018] Another aspect of the present invention provides that the data pre-processing module (104) pre-processes data , wherein the data pre-processing module (104) further comprises a data validation unit (104a), configured to determine whether the captured real-time data meets predefined accuracy and completeness criteria by detecting and flagging invalid data points; a data imputation unit (104b) operably coupled to the data validation unit (104a), configured to determine any missing data using statistical techniques or predictive modelling based on historical data; a duplication removal unit (104c) operably coupled to the data imputation unit (104b), configured to identify and eliminate redundant or duplicated data received from the data imputation unit (104b); a data profiling unit (104d) operably coupled to the duplication removal unit (104c), configured to analyse dataset for patterns, inconsistencies, or anomalies; and a data re-validation unit (104e), operably coupled to the data profiling unit (104d) and the data validation unit (104a), configured to repeat pre-processing steps upon determining that data received from the data profiling unit (104d) is not aligned with predefined standards, wherein the data re-validation unit (104e) initiates a feedback loop to reprocess the data starting from the data validation unit (104a) until compliance with the predefined standard is achieved.
[0019] Another aspect of the present invention provides that the Al based analytics module (108) further comprises an aquaculture parameter estimation module (108a) operably coupled to the cloud-based database arrangement (106), configured to analyse the pre-processed real-time data using an Al and machine learning model to estimate a type and corresponding number of aquaculture stock residing in the aquaculture environment; and monitor a change in parameters related to the aquaculture environment and at least one of aquaculture stock residing therein; a feed conversion ratio, FCR computation module (108b), operably coupled to the aquaculture parameter estimation module (108a), configured to compute an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock; store computed FCR value in the cloud-based database arrangement (106) for historical tracking and analysis; and dynamically adjust feeding parameters based on the FCR value; a feed optimization module (108c) operably coupled to the FCR computation module (108b) and the aquaculture parameter estimation module (108a), configured to determine the optimized feeding scheduleand the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module (108a) and the computed FCR value; and a predictive analytics module (108d), operably coupled to the feed optimization module (108), configured to forecast aquaculture stock growth and optimal harvest times based on the computed FCR value, historical data, and real-time inputs.
[0020] Another aspect of the present invention provides that the change in parameters comprises a change in at least one of size, weight, hunger, behaviour, growth, or fish counting of the at least one aquaculture stock; or a change in at least one of pH, temperature, or turbidity of the aquaculture environment.
[0021] Another aspect of the present invention provides that the feed delivery module (110) further comprises an automated feed control module (110a), configured to dispense feed in real time based on feeding parameters received from the feed optimization module (108c); a user override module (110b), operably coupled to the automated feed control module (110a), configured to enable manual control of feed distribution via the user interface (108); and a realtime feed adjustment module (110c), configured to adjust feeder settings dynamically based on real-time feeding recommendations from the Al-based analytics module (108), the real-time feed adjustment module is configured to increase feed delivery upon determining that the computed FCR value exceeds a predefined threshold, indicating suboptimal feed efficiency; decrease feed delivery upon determining that the computed FCR value is within an optimal range, ensuring reduced feed wastage; and continuously adjust feed quantity based on realtime FCR variations and aquaculture stock growth determined by the Al-based analytics module (108).
[0022] Another aspect of the present invention provides a method (200) for monitoring and managing an aquaculture environment, comprising steps of capturing real-time data related to aquaculture environment and aquaculture stock residing therein (202); pre-processing captured real-time data for further analysis (204); storing pre-processed real-time data and facilitating real-time access thereof; analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity (206); and enabling feed delivery to the aquaculture environment based on optimized feeding schedule and the feed quantity in real-time (208); and enabling at least one user to monitoring the aquaculture environment using a user interface (210).
[0023] Another aspect of the present invention provides that pre-processing captured real-time data for further analysis (204), further comprises method steps of (300) that comprise validating the captured real-time data by detecting and flagging invalid data points based on predefinedaccuracy and completeness criteria (302); imputing missing data using statistical techniques or predictive modelling based on historical data (304); removing redundant or duplicated data from the captured real-time dataset (306); profiling dataset to identify patterns, inconsistencies, or anomalies (308); and re-validating the processed data in a feedback loop until compliance with predefined data quality standards is achieved (310).
[0024] Yet another aspect of the present invention provides that analysing the pre-processed realtime time for optimizing feeding schedule and determining an optimized feed quantity (206), further comprises method steps of (400), the method steps (400) comprising analysing the pre-processed real-time data using an Al and machine learning model to estimate a type and corresponding number of aquaculture stock residing in the aquaculture environment (402); and monitoring a change in parameters related to the aquaculture environment and at least one of aquaculture stock residing therein (404); computing an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock (406); storing computed FCR value in the cloudbased database arrangement (106) for historical tracking and analysis (408); dynamically adjusting feeding parameters based on the FCR value (410); determining the optimized feeding schedule and the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module and the computed FCR value (412); and forecasting aquaculture stock growth and optimal harvest times based on the computed FCR value, historical data, and real-time inputs (414).
[0025] The present invention consists of features and a combination of parts hereinafter fully described and illustrated in the accompanying drawings, it being understood that various changes in the details may be made without departing from the scope of the invention or sacrificing any of the advantages of the present invention.BRIEF DESCRIPTION OF ACCOMPANYING DRAWINGS
[0026] To further clarify various aspects of some embodiments of the present invention, a more particular description of the invention will be rendered by references to specific embodiments thereof, which are illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail through the accompanying drawings in which:
[0027] Figure 1 illustrates a block diagram of a system (100) for monitoring and managing an aquaculture environment, in accordance with an embodiment of the present invention;
[0028] Figure 1a illustrates a block diagram of a monitoring module (102), in accordance with an embodiment of the present invention;
[0029] Figure 1b illustrates a block diagram of a data pre-processing module (104), in accordance with an embodiment of the present invention;
[0030] Figure 1c illustrates a block diagram of an Al based analytics module (108), in accordance with an embodiment of the present invention;
[0031] Figure 1d illustrates a block diagram of a feed delivery module (110), in accordance with an embodiment of the present invention;
[0032] Figure 2 is a flowchart illustrated a method (200) for monitoring and managing aquaculture environment, in accordance with an embodiment of the present invention;
[0033] Figure 3 is a flowchart illustrated a method (300) for pre-processing captured real-time data for further analysis, in accordance with an embodiment of the present invention; and
[0034] Figure 4 is a flowchart illustrated a method (400) for analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity, in accordance with an embodiment of the present invention.DETAILED DESCRIPTION OF THE DRAWINGS
[0035] The present invention relates to a system and method for monitoring and managing an aquaculture environment. In particular, the present invention relates to the system and method for monitoring and managing an aquaculture environment that utilizes an Al-based analytics module for determining an optimized feeding schedule and the optimized quantity of feed required for aquaculture stock.
[0036] Hereinafter, this specification will describe the present invention according to the preferred embodiments. It is to be understood that limiting the description to the preferred embodiments of the invention is merely to facilitate discussion of the present invention and it is envisioned without departing from the scope of the appended claims.
[0037] The present invention discloses an aquaculture monitoring system that integrates Al, machine learning, loT-enabled monitoring devices, and cloud-based analytics to optimize / feed distribution and improve stock management. The system comprises a monitoring module, a data pre-processing module, a cloud-based database arrangement, an Al-based analytics module (108), a feed delivery module (110), and a user interface. The monitoring module is configured to capture real-time data related to the aquaculture environment and stock, including images, videos, and environmental parameters such as pH, temperature, and turbidity. This data is then pre-processed locally by the data pre-processing module to eliminate redundancy, validate data integrity, and ensure efficient cloud transmission for further analysis.
[0038] The Al-based analytics module processes the pre-processed data using Al and machine learning algorithms to detect changes in fish size, weight, hunger levels, and behavioural patterns. Additionally, the analytics module calculates the Feed Conversion Ratio, FCR, which is the ratio of feed input to weight gain, allowing for real-time adjustments to the feeding strategy. Based on these insights, the feed delivery module dynamically adjusts feed quantity and distribution in real time, ensuring that the aquaculture stock receives the optimal amount of nutrition while minimizing feed wastage. The system further includes a predictive analytics module, which forecasts growth trends, optimal harvest schedules, and long-term feeding strategies based on historical data and real-time environmental conditions.
[0039] The user interface provides farmers with remote access to real-time monitoring, feeding recommendations, and system overrides, and allows users to view live-streamed data, review computed FCR values, and adjust feeding parameters manually if necessary. Additionally, the cloud-based database arrangement stores historical feeding records, growth trends, and waterquality metrics, enabling farmers to make data-driven decisions for improved aquaculture management. The system may be implemented in various aquaculture environments, including fish farms, shrimp ponds, and recirculating aquaculture systems, offering a scalable and adaptive solution for sustainable fish farming.
[0040] Throughout the present invention, the term “database arrangement” as used herein relates to an organized body of digital information regardless of the manner in which the data or the organized body thereof is represented. Optionally, the database may be hardware, software, firmware and / or any combination thereof. For example, the organized body of related data may be in the form of a table, a map, a grid, a packet, a datagram, a file, a document, a list or in any other form. The database includes any data storage software and systems. Optionally, the database may be operable to support relational operations, regardless of whether it enforces strict adherence to the relational model, as understood by those of ordinary skill in the art.
[0041] The database arrangement comprising a columnar database configured for storing a plurality of documents. Herein, the columnar database is used to store terabytes of raw data which can support functionality of index lookup and persistent storage. Herein, the document is a description of an electronic copy and may be in a form of a web page, internet document, multimedia file and so forth. Additionally, document may be a set of information related to the biomedical entity, object and so forth. Herein, the raw data may be crawled and cleaned through a module which interacts with the database and is stored in the columnar database.
[0042] Throughout the present invention, the term "module” refers to a component or unit of the system that performs a specific function, in accordance with an embodiment of the present invention. Each module may be implemented in hardware, software, or a combination of both, depending on the requirements of the system. In one embodiment, a module is realized as a set of electronic circuits configured to execute predefined operations. Alternatively, the module may be implemented as software code executed by a processor or microcontroller. The modules are operably coupled to one another through communication links, such as electrical connections, data buses, or wireless communication channels, enabling seamless interaction among them.
[0043] A specific module typically comprises a combination of subcomponents designed to perform its assigned functionality. For example, a module may include one or more processors, memory components, data processing circuits, and input / output interfaces. The processor or microcontroller executes the instructions necessary to perform the desired function of the module. The memory components, which may include volatile or non-volatile storage, are used to store data or program instructions. The input / output interfaces facilitate communication withother modules or external devices, ensuring that the module can interact with the larger system. Additionally, in an embodiment, the modules may include specialized sensors or actuators when required by the specific function. For instance, a module responsible for data collection may incorporate optical sensors, imaging sensors, or other appropriate hardware for gathering information. Similarly, modules responsible for output generation may include display modules, printers, or communication transmitters.
[0044] Each module is configured to operate independently to perform its designated task but is also integrated within the system to contribute to the overall functionality. The design and configuration of the modules may vary based on the intended application, ensuring that they align with the operational requirements of the system as a whole.
[0045] Optionally, the “module” may include, but is not limited to, a microprocessor, a microcontroller, a complex instruction set computing, CISC microprocessor, a reduced instruction set, RISC microprocessor, a very long instruction word microprocessor, or any other type of processing circuit. Furthermore, the module may refer to one or more individual microprocessor, a part of a microprocessor, processing devices and various elements associated with a processing device that may be shared by other processing devices. Optionally, the “module” is configured to establish communication and networking between the columnar database, the ontological database and the search database.
[0046] Throughout the present invention, the term “user interface” relates to an arrangement that allows for interaction between the user and the automated aquaculture environment monitoring system. The user interface allows for obtaining inputs from the user and providing user-friendly, systematic, easily comprehensible, and customisable representations of information to the user. As a result, the user interface facilitates the user in better organizing, viewing, analysis, and processing of information related to various fields. In another embodiment, the user interface described herein can be easily implemented by way of the hardware system of the aquaculture monitoring system.
[0047] Throughout the present invention, the term “data communication network” refers to individual networks or a collection thereof interconnected with each other and functioning as a single large network. Optionally, such a data communication network is implemented by way of a wired communication network, wireless communication network, or a combination thereof. It will be appreciated that a physical connection is established for implementing the wired communication network, whereas the wireless communication network is implemented using electromagnetic waves. Examples of such data communication networks include, but are not limited to, Local Area Networks, LANs, Wide Area Networks, WANs, Metropolitan AreaNetworks, MANs, Wireless LANs, Wireless WANS, Wireless MANS, the Internet, second-generation telecommunication networks, third generation telecommunication networks, fourthgeneration telecommunication networks, fifth-generation telecommunication networks and Worldwide Interoperability for Microwave Access networks.
[0048] Optionally, the data communication network is implemented as a cellular network. It will be appreciated that the cellular network refers to a radio communication network, wherein the cellular network is distributed over land through cells. Specifically, each cell includes a fixed location transceiver, for example, a base station.
[0049] Reference is made to Figure 1, where there is illustrated a system (100) for monitoring and managing an aquaculture environment in accordance with an embodiment of the present invention.
[0050] The system (100) is configured to capture real-time data related to the aquaculture environment and aquaculture stock, process the captured data for further analysis, store and retrieve the processed data via a cloud-based infrastructure, apply Al-based analytics for optimizing feeding schedules, and dynamically adjust the delivery of feed in real time. The system (100) further enables remote monitoring and control through a user interface (112), thereby providing an integrated and automated solution for aquaculture management.
[0051] The system (100) comprises a monitoring module (102) configured to capture real-time data associated with the aquaculture environment and aquaculture stock residing therein. In an instance, the monitoring module (102) include a combination of image-capturing devices, environmental sensors, and motion-tracking technologies to gather comprehensive data regarding stock behaviour, physical attributes, and water conditions. The captured data is transmitted to the data pre-processing module (104) for further analysis.
[0052] The system (100) further comprises a data pre-processing module (104) operably coupled to the monitoring module (102), is configured to pre-process the captured real-time data for further analysis. Advantageously, the pre-processing module (104) ensures data integrity, accuracy, and efficiency by refining raw data, removing anomalies, and preparing the dataset for subsequent Al-based analytics. The pre-processed data is thereafter transmitted to the cloud-based database arrangement (106).
[0053] In an embodiment, the pre-processing (104) module conducts pre-processing locally with the monitoring module (102) to reduce latency, and for effective transmission of pre-processed data for further analysis.In another embodiment, the pre-processing module (104) is configured to conduct preprocessing analysis remotely in a cloud-based environment. For example, the pre-processor module (104) is configured to conduct pre-processing analysis on a cloud-base database arrangement (106).
[0054] The system (100) also comprises a cloud-based database arrangement (106) that is operably coupled to the data pre-processing module (104) via a data communication network (106a) and is configured to store the pre-processed real-time data received from the data preprocessing module (104). The cloud-based database arrangement (106) provides high-availability storage and facilitates real-time access to the processed data for analysis and decision-making.
[0055] Further, the Al-based analytics module (108) is operably coupled to the cloud-based database arrangement (106) and is configured to analyse the pre-processed real-time data to optimize the feeding schedule and determine an optimized feed quantity. The Al-based analytics module (108) utilizes machine learning models and data-driven algorithms to assess aquaculture stock parameters, detect behavioural patterns, and generate real-time feeding recommendations.
[0056] Notably, the Al-based analytics module (108) is implemented in a cloud-based environment.
[0057] Furthermore, the feed delivery module (110) is operably coupled to the Al-based analytics module (108) and is configured to provide dynamic feed delivery within the aquaculture environment based on the optimized feeding schedule and feed quantity in real-time.
[0058] In an embodiment, the feed delivery module (110) may include automated feeders, dispensing mechanisms, and real-time feedback controls to ensure that feeding operations are continuously adjusted based on Al-based optimization results.
[0059] In another embodiment, the feed delivery module (110) is manually operated by at least one user to dispense feed delivery within the aquaculture environment based on the optimized feeding schedule and the optimized feed quantity in real-time.
[0060] Moreover, the user interface (112) is operably coupled to the cloud-based database arrangement (106), the Al-based analytics module (108), and the feed delivery module (110), and is configured to provide remote access and control functionalities. The user interface (112) enables at least one user to receive the pre-processed real-time data from the cloud-based database arrangement (106) and enable a remote live-stream of the aquaculture environment, including the pre-processed real-time data, the optimized feeding schedule, and the optimizedfeed quantity determined by the Al-based analytics module (108). Also, the user interface (112) is configured to enable remote control of the feed delivery module (110) based on user inputs, allowing manual adjustments to feeding operations when necessary.
[0061] In describing further on Figure 1, reference is made to Figure 1a, where there is illustrated a block diagram of the monitoring module (102) in accordance with an embodiment of the present invention. The monitoring module (102) is configured to capture real-time data associated with the aquaculture environment and the aquaculture stock residing therein. The monitoring module (102) comprises at least one first camera (102a), at least one second camera (102b), and an aquaculture environment monitoring module (102c), each of which is configured to monitor specific parameters related to aquaculture stock behaviour, density, and environmental conditions.
[0062] According to one embodiment, the at least one first camera (102a) is configured to capture images and videos of movement and behaviour related to the aquaculture stock. The at least one first camera (102a) is encased in a water-resistant casing, thereby ensuring protection against water exposure, humidity, and temperature fluctuations commonly encountered in aquaculture environments.
[0063] In one instance, the first camera (102a) is an Al-based camera equipped with object recognition and tracking capabilities. The first camera (102a) may use computer vision techniques to detect fish movement, feeding activity, and abnormal behaviours such as erratic swimming patterns or reduced motion, which may indicate stress, disease, or unfavourable environmental conditions.
[0064] In another instance, the first camera (102a) is an infrared camera, configured to capture images in low-light or nighttime conditions. This may be particularly beneficial in aquaculture environments where natural or artificial lighting is limited, such as deep-sea aquaculture cages or enclosed water tanks.
[0065] In yet another embodiment, the first camera (102a) is a stereoscopic camera configured to capture three-dimensional, 3D images of aquaculture stock. The stereoscopic camera may facilitate depth perception analysis, allowing for precise measurement of fish size, body structure, and density estimation within a given water volume.
[0066] The monitoring module (102) further comprises at least second camera (102b), configured to detect underwater aquaculture stock density and track movement associated therewith.In an embodiment, the at least second camera (102b) is selected from at least one of, but not limited to, a sonar device or an underwater camera. In another embodiment, the at least one second camera (102b) is configured in an arrangement of one or more underwater camera, based on the requirements.
[0067] In another embodiment, the at least second camera (102b) is selected from at least one of an underwater camera.
[0068] The monitoring module (102) further comprises an aquaculture environment monitoring module (102c), which includes a plurality of sensors configured to measure a plurality of parameters related to the aquaculture environment. The plurality of parameters includes at least one of pH, temperature, and turbidity, each of which is critical for maintaining optimal water quality and aquaculture stock health.
[0069] In one instance, the pH sensor measures the acidity or alkalinity of the water to ensure it remains within a biologically acceptable range for the aquaculture stock. The pH sensor may be integrated with an automated alert system, which notifies farmers when pH levels exceed or drop below the optimal threshold.
[0070] In another instance, a temperature sensor continuously monitors the water temperature to prevent thermal stress in fish, which could lead to slowed growth, increased mortality rates, or disrupted feeding behaviour. The temperature sensor may be integrated with a climate control system, which regulates water heating or cooling mechanisms.
[0071] In yet another instance, a turbidity sensor measures clarity of the water, which is an important indicator of sediment levels, organic matter, and microbial activity. The turbidity sensor may be used to detect water pollution, excessive algae growth, or the presence of uneaten feed that may degrade water quality. In another instance, additional sensor may be provided selected from at least one of, but not limited to, ammonia sensor, dissolved oxygen sensor, and so forth.
[0072] In some embodiments, the monitoring module (102) operates in coordination with other system components to enhance data accuracy and provide a comprehensive assessment of the aquaculture environment.
[0073] For example, in one embodiment, the first camera (102a) and the second camera (102b) operate simultaneously, where the first camera (102a) captures visual data of fish movement, and the second camera (102b) provides depth-based tracking information. This combination allows for precise mapping of fish distribution and behaviour analysis.In another embodiment, the pH sensor and turbidity sensor operate in conjunction, where a detected increase in water turbidity triggers an automatic pH measurement to determine if the change is due to organic waste accumulation.
[0074] While the embodiments described above illustrate various implementations of the monitoring module (102), it should be understood that the invention is not limited to these configurations.
[0075] In one alternative embodiment, the monitoring module (102) is implemented, where individual components can be selectively enabled or disabled based on the specific aquaculture setup. For example, a small-scale indoor aquaculture facility may utilize only first camera-based monitoring (102a) and temperature sensing (102c), while an offshore aquaculture system may rely on second camera-based tracking (102b) and turbidity monitoring (102c).
[0076] In another alternative implementation, the monitoring module (102) is configured as a fully wireless system, where all sensor data is transmitted via loT-enabled communication protocols, eliminating the need for wired connections and allowing for greater flexibility in sensor deployment.
[0077] Reference is made to Figure 1b, where there is illustrated a block diagram of the data preprocessing module (104) in accordance with an embodiment of the present invention. The data pre-processing module (104) is configured to pre-process real-time data locally before transmission to the cloud-based database arrangement (106). The pre-processing module (104) plays a crucial role in reducing data redundancy, ensuring accuracy, and optimizing data integrity, thereby enabling efficient transmission and storage of aquaculture-related data for further Al-based analysis.
[0078] The data pre-processing module (104) comprises a plurality of processing units, each responsible for different aspects of data refinement, including validation, imputation, redundancy removal, profiling, and re-validation. These units operate sequentially and iteratively, ensuring that the data meets predefined accuracy and completeness criteria before being stored or analysed.
[0079] The data pre-processing module (104) comprises a data validation unit (104a) that is configured to determine whether the captured real-time data meets predefined accuracy and completeness criteria by detecting and flagging invalid data points.
[0080] In one embodiment, the data validation unit (104a) applies range validation checks, ensuring that captured sensor values, e.g., pH, temperature, turbidity, fall within expected biologicallimits. If a data point exceeds a predefined threshold, the validation unit (104a) flags it as an anomaly and prevents it from propagating to subsequent processing stages.
[0081] In another instance, the data validation unit (104a) performs outlier detection using statistical techniques, such as mean deviation analysis or Z-score computation, to identify values that significantly differ from expected trends. This ensures that erroneous sensor readings or corrupted image data do not influence downstream analytics.
[0082] In yet another embodiment, the data validation unit (104a) executes format verification to ensure that captured image / video files, numerical sensor values, and metadata conform to predefined encoding standards, reducing potential issues during cloud storage or Al-based processing.
[0083] The data pre-processing module (104) further comprises a data imputation unit (104b) that is operably coupled to the data validation unit (104a) and is configured to handle missing or incomplete data using statistical techniques or predictive modelling based on historical data.
[0084] In one embodiment, the data imputation unit (104b) applies mean or median imputation, where missing sensor values are replaced with the average of previously recorded values, ensuring minimal data loss.
[0085] In an instance, the data imputation unit (104b) utilizes machine learning-based predictive modelling, for example, linear regression, k-nearest neighbors, KNN, or random forest algorithms, to estimate missing values based on historical trends and correlated sensor readings. This technique is particularly useful in scenarios where temporary sensor failures or transmission losses result in partial data gaps.
[0086] In yet another embodiment, the data imputation unit (104b) applies interpolation techniques such as polynomial regression or spline interpolation to reconstruct missing time-series data from environmental sensors, ensuring continuous trend analysis without abrupt discontinuities.
[0087] The data pre-processing module (104) further comprises a duplication removal unit (104c) that is operably coupled to the data imputation unit (104b) and is configured to identify and eliminate redundant or duplicated data received from the data imputation unit (104b).
[0088] In one embodiment, the duplication removal unit (104c) performs exact match detection, where identical data records are flagged and merged to prevent unnecessary storage redundancy.
[0089] In another instance, the duplication removal unit (104c) applies similarity-based comparison algorithms, such as hashing techniques, i.e., MD5, SHA-256, or cosine similarity, to detectnear-duplicate entries in image / video datasets. This approach is beneficial when multiple Al cameras (102a) capture overlapping images of the same aquaculture stock, ensuring that duplicate frames are filtered out before cloud transmission.
[0090] In yet another embodiment, the duplication removal unit (104c) integrates timestamp-based filtering, ensuring that only the latest data instance is retained when multiple data entries are recorded within a short time interval.
[0091] The data pre-processing module (104) also comprises a data profiling unit (104d) that is operably coupled to the duplication removal unit (104c) and is configured to analyze datasets for patterns, inconsistencies, or anomalies.
[0092] In one embodiment, the data profiling unit (104d) performs distribution analysis, generating statistical summaries of aquaculture data to ensure uniformity across different sensors and data sources.
[0093] In another embodiment, the data profiling unit (104d) executes correlation analysis, identifying relationships between environmental factors, for example, pH levels vs. fish activity patterns, to optimize Al-based analytics models.
[0094] Additionally, a data re-validation unit (104e) is provided that operably coupled to the data profiling unit (104d) and the data validation unit (104a) and is configured to repeat preprocessing steps upon determining that the data received from the data profiling unit (104d) is not aligned with predefined standards. The data re-validation unit (104e) initiates a feedback loop, ensuring that erroneous or incomplete data is reprocessed until compliance with predefined accuracy standards is achieved.
[0095] In one embodiment, the data re-validation unit (104e) triggers a reprocessing cycle, where flagged anomalies are sent back to the data validation unit (104a) for re-evaluation. In another instance, the data re-validation unit (104e) dynamically adjusts imputation techniques if previous methods fail to reconstruct missing data adequately.
[0096] In yet another embodiment, the data re-validation unit (104e) applies threshold-based logic, where data flagged multiple times for inconsistencies may be temporarily quarantined or excluded from Al-based processing to prevent unreliable predictions.
[0097] In an embodiment, the data pre-processing module (104) is implemented as a decentralized architecture, where edge devices which are embedded micro-processors in aquaculture tanks, handle pre-processing tasks locally before cloud transmission, reducing latency and bandwidth usage.In another alternative implementation, the data profiling unit (104d) integrates an Al-driven selflearning mechanism, which continuously refines validation and profiling techniques based on historical performance trends.
[0098] Reference is made to Figure 1c, where there is illustrated a block diagram of the Al-based analytics module (108) in accordance with an embodiment of the present invention. The Al-based analytics module (108) is configured to analyse pre-processed real-time data to optimize feeding schedules, determine optimized feed quantities, and predict aquaculture stock growth trends. The Al-based analytics module (108) comprises multiple functional sub-modules, each responsible for different aspects of Al-driven aquaculture management, including parameter estimation, FCR computation, feed optimization, and predictive analytics.
[0099] The Al-based analytics module (108) comprises an aquaculture parameter estimation module (108a) that is operably coupled to the cloud-based database arrangement (106) and is configured to analyse pre-processed real-time data using Al and machine learning models to, for estimating a type and corresponding number of aquaculture stock residing in the aquaculture environment, and monitor changes in parameters related to the aquaculture environment and aquaculture stock.
[0100] In an embodiment, the change in parameters comprises a change in at least one of size, weight, hunger, behaviour, or growth of the at least one aquaculture stock; or a change in at least one of pH, temperature, or turbidity of the aquaculture environment.
[0101] In one embodiment, the aquaculture parameter estimation module (108a) utilizes computer vision techniques to analyse image and video data from the monitoring module (102), identifying fish species, size, and weight distribution based on pattern recognition models and convolutional neural networks, CNNs.
[0102] In another instance, the aquaculture parameter estimation module (108a) applies second camera-based analysis to estimate aquaculture stock density and biomass distribution within the aquaculture environment. By integrating Al-driven second camera interpretation algorithms, the system can track stock movement, identify potential overcrowding, and detect anomalies such as escape events.
[0103] In yet another embodiment, the aquaculture parameter estimation module (108a) incorporates behavioural analysis, where Al algorithms detect deviations in movement patterns, such as reduced swimming activity or erratic behaviour, which may indicate stress, disease, or environmental imbalances.The Al based analytics module (108) further comprises an FCR computation module (108b) operably coupled to the aquaculture parameter estimation module (108a) and is configured to compute an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock, store computed FCR values in the cloud-based database arrangement (106) for historical tracking and analysis, and dynamically adjust feeding parameters based on the computed FCR value.
[0104] In one embodiment, the FCR computation module (108b) calculates feed conversion efficiency by correlating feeding data from the feed delivery module (110) with real-time weight estimates from the aquaculture parameter estimation module (108a).
[0105] The Al based analytics module (108) further comprises a feed optimization module (108c) operably coupled to the FCR computation module (108b) and the aquaculture parameter estimation module (108a) and is configured to determine the optimized feeding schedule and the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module (108a), and the computed FCR value from the FCR computation module (108b).
[0106] In one embodiment, the feed optimization module (108c) applies reinforcement learning algorithms that continuously adjust feeding schedules based on real-time FCR variations and behavioural cues. In another instance, the feed optimization module (108c) incorporates realtime environmental parameters, where pH, temperature, and turbidity data influence feeding decisions, ensuring that feeding occurs under optimal water conditions. In yet another embodiment, the feed optimization module (108c) includes adaptive feeding controls, where the system dynamically increases or decreases feed distribution based on observed growth rates and stock metabolism.
[0107] The Al based analytics module (108) also comprises a predictive analytics module (108d) operably coupled to the feed optimization module (108c) and is configured to forecast aquaculture stock growth and optimal harvest times based on the computed FCR value from the FCR computation module (108b), historical feeding data and real-time environmental conditions., and machine learning-based growth modelling techniques.
[0108] In one embodiment, the predictive analytics module (108d) employs Al-driven time-series forecasting models, such as LSTM, Long Short-Term Memory neural networks or Auto Regressive Integrated Moving Average, ARIMA models, to predict future aquaculture stock weight and biomass distribution.In another instance, the predictive analytics module (108d) utilizes climate data integration, where seasonal temperature fluctuations and water chemistry changes are factored into harvest timing recommendations. In yet another embodiment, the predictive analytics module (108d) optimizes long-term stocking and culling strategies, ensuring that aquaculture operations maintain sustainable growth rates while minimizing resource expenditure.
[0109] Reference is made to Figure 1 d, where there is illustrated a block diagram of the feed delivery module (110) in accordance with an embodiment of the present invention. The feed delivery module (110) is configured to dispense feed in real time, enable manual feeding adjustments, and dynamically modify feed quantity based on real-time FCR values and aquaculture stock growth patterns.
[0110] It is worth noting that the feed delivery module (110) may either be manually controlled or automatically controlled by the at least one user.
[0111] The feed delivery module (110) comprises an automated feed control module (110a) that is configured to dispense feed in real time based on feeding parameters received from the feed optimization module (108c). The automated feed control module (110a) ensures precise and efficient feed distribution to optimize aquaculture stock growth while minimizing feed wastage.
[0112] In one embodiment, the automated feed control module (110a) comprises mechanical or pneumatic feeding mechanisms, such as rotary feeders, belt feeders, or demand feeders, which release controlled amounts of feed into the aquaculture environment in response to feeding commands generated by the Al-based analytics module (108).
[0113] In another instance, the automated feed control module (110a) utilizes water-based feed dispensers, which mix dry feed with water before dispersal, ensuring that feed is easily accessible to aquatic stock while preventing floating feed loss due to wind or currents. In yet another embodiment, the automated feed control module (110a) integrates with sensor-based feedback loops, where infrared, second camera, or optical sensors detect uneaten feed particles, allowing the system to pause feeding cycles if excess feed is detected in the water.
[0114] The feed delivery module (110) further comprises a user override module (110b) operably coupled to the automated feed control module (110a) and is configured to enable manual control of feed distribution via the user interface (108). The user override module (110b) provides an interactive mechanism for aquaculture operators to manually modify feeding schedules, adjust feed amounts, or pause feeding operations when necessary.In a specific embodiment, the user override module (110b) enables manual feed activation, where farmers can override Al-driven feeding recommendations and dispense feed manually using a mobile or web-based control panel.
[0115] In another embodiment, the user override module (110b) enables predefined feeding adjustments, where users can set fixed feeding times and quantities, overriding Al-based automation in situations where manual intervention is required due to special feeding conditions, such as medication or dietary changes. In yet another embodiment, the user override module (110b) features a safety cut-off mechanism, ensuring that feed delivery can be manually halted in emergency situations, such as water quality deterioration, mechanical malfunctions, or disease outbreaks in aquaculture stock.
[0116] The feed delivery module (110) also comprises a real-time feed adjustment module (110c) that is configured to dynamically adjust feeder settings based on real-time feeding recommendations from the Al-based analytics module (108). The real-time feed adjustment module (110c) continuously adapts feeding parameters in response to computed FCR values, aquaculture stock growth trends, and observed feeding behaviour.
[0117] In an embodiment, the real-time feed adjustment module (110c) is configured to increase feed delivery upon determining that the computed FCR value exceeds a predefined threshold, indicating suboptimal feed efficiency, decrease feed delivery upon determining that the computed FCR value is within an optimal range, ensuring reduced feed wastage, continuously adjust feed quantity based on real-time FCR variations and aquaculture stock growth, ensuring that feeding strategies remain optimized over time, and increasing Feed Delivery for Suboptimal FCR
[0118] In one embodiment, the real-time feed adjustment module (110c) increases feed quantity when the FCR computation module (108b) detects a feed conversion ratio less than a predefined threshold, where aquaculture stock growth is lower than expected relative to the provided feed quantity.
[0119] In another embodiment, the real-time feed adjustment module (110c) applies Al-driven adaptive feeding, where the system gradually increases feed amounts in small increments, allowing the Al-based analytics module (108) to assess the impact of increased feeding on growth rates while preventing excessive feed waste.
[0120] In one embodiment, the real-time feed adjustment module (110c) decreases feed quantity when the computed FCR value falls within an optimal range, ensuring that feed is not overprovided, which could lead to unnecessary waste accumulation and poor water quality.In another instance, the real-time feed adjustment module (110c) integrates with water quality sensors, where detected increases in turbidity or ammonia levels trigger an automatic feed reduction, preventing further degradation of water conditions. In another instance, the realtime feed adjustment module (110c) incorporates behavioral analysis, where Al-based pattern recognition tracks aquaculture stock feeding activity, modifying feeding rates dynamically based on observed consumption trends. In yet another instance, the real-time feed adjustment module (110c) synchronizes with climate-based feeding adjustments, where seasonal temperature fluctuations are accounted for in feeding decisions, ensuring feeding rates align with fish metabolic rates during different environmental conditions.
[0121] The present disclosure also relates to the method as described above. Various embodiments and variants disclosed above apply mutatis mutandis to the method.
[0122] Reference is made to Figure 2 where there is illustrated a method (200) for monitoring and managing aquaculture environment, in accordance with an embodiment of the present invention. The method (200) comprises capturing real-time data related to aquaculture environment and aquaculture stock residing therein, at step (202), pre-processing captured real-time data for further analysis, at step (204), storing pre-processed real-time data and facilitating real-time access thereof, and analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity, at step (206). Further, the method (200) comprises dynamically adjusting feed delivery to the aquaculture environment based on optimized feeding schedule and the feed quantity in real-time, at step (208), and receiving the pre-processed real-time data from the cloud-based database arrangement at a user interface and enabling remote live-stream of the aquaculture environment including the pre-processed real-time data, the optimized feeding schedule, and the optimized feed quantity, at step (210). Subsequently, the method (200) comprises enabling at least one user to remotely control feed delivery based on user inputs, at step (212).
[0123] In describing further on Figure 2, reference is made to Figure 4 where there is illustrated a method (300) for pre-processing captured real-time data for further analysis, in accordance with an embodiment of the present invention. The method (300) validating the captured realtime data by detecting and flagging invalid data points based on predefined accuracy and completeness criteria, at step (302), and imputing missing data using statistical techniques or predictive modelling based on historical data, at step (304). Thereafter, the method (300) comprises removing redundant or duplicated data from the captured real-time dataset, at step (306), followed by profiling dataset to identify patterns, inconsistencies, or anomalies, at step (308) and re-validating the processed data in a feedback loop until compliance with predefined data quality standards is achieved, at step (310).In describing further on Figure 2, reference is made to Figure 4 where there is illustrated a method (400) for analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity (206), in accordance with an embodiment of the present invention. The method (400) initiates at step (402) that comprises analysing the pre-processed real-time data using an Al and machine learning model to estimate a type and corresponding number of aquaculture stock residing in the aquaculture environment. Further, the method (400) comprises monitoring a change in parameters related to the aquaculture environment and at least one of aquaculture stock residing therein, at step (404), and computing an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock, at step (406). Thereafter, the method (400) comprises storing computed FCR value in the cloud-based database arrangement (106) for historical tracking and analysis, at step (408), followed by dynamically adjusting feeding parameters based on the FCR value, at step (410), and determining the optimized feeding schedule and the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module (108a) and the computed FCR value, at step (412). Subsequently, the method (400) comprises forecasting aquaculture stock growth and optimal harvest times based on the computed FCR value, historical data, and real-time inputs, at step (414).
[0124] Throughout this specification, unless the context requires otherwise, the word “comprise”, or variations such as "comprises” or “comprising”, will be understood to imply the inclusion of a stated step or element or integer or group of steps or elements or integers, but not the exclusion of any other step or element or integer or group of steps, elements or integers. Thus, in the context of this specification, the term “comprising” is used in an inclusive sense and thus should be understood as meaning “including principally, but not necessarily solely”.
Claims
CLAIMS1. A system (100) for monitoring and managing an aquaculture environment, comprising: a monitoring module (102), configured to capture real-time data related to the aquaculture environment and aquaculture stock residing therein;a data pre-processing module (104) operably coupled to the monitoring module (102), configured to pre-process captured real-time data for further analysis;a cloud-based database arrangement (106) operably coupled to the data pre-processing module (104) via a data communication network (106a), the cloud-based database arrangement (106) is configured to store pre-processed real-time data received from the data pre-processing module (104) and facilitate real-time access thereof;an artificial intelligence, Al based analytics module (108) operably coupled to the cloudbased database arrangement (106), configured to analyse the pre-processed real-time data for optimizing feeding schedule and determining an optimized feed quantity; and a feed delivery module (110) operably coupled to the Al-based analytics module (108), configured to dynamically provide feed delivery to the aquaculture environment based on optimized feeding schedule and the feed quantity in real-time; anda user interface (112) operably coupled to the cloud-based database arrangement (106), the Al-based analytics module (108), and the feed delivery module (110), wherein the user interface (112) is configured to enable at least one user to monitor the aquaculture environment.
2. The system (100) according to Claim 1, wherein the monitoring module (102) further comprises:at least one first camera (102a), configured to capture images and videos of movement and behaviour related to the aquaculture stock, wherein the at least one first camera is encased in a water resistant casing;at least one second camera (102b), configured to detect underwater aquaculture stock density and track the movement associated thereof; andan aquaculture environment monitoring module (102c) comprising a plurality of sensors configured to measure a plurality of parameters related to the aquaculture environment, wherein the plurality of parameters comprises at least one of pH, temperature, and turbidity.
3. The system (100) according to Claim 2, wherein the at least one first camera (102a) is an Artificial Intelligence, Al-based camera.
4. The system (100) according to Claim 1, wherein the data pre-processing module (104) pre-processes data, wherein the data pre-processing module (104) further comprises:a data validation unit (104a), configured to determine whether the captured real-time data meets predefined accuracy and completeness criteria by detecting and flagging invalid data points;a data imputation unit (104b) operably coupled to the data validation unit (104a), configured to determine any missing data using statistical techniques or predictive modelling based on historical data;a duplication removal unit (104c) operably coupled to the data imputation unit (104b), configured to identify and eliminate redundant or duplicated data received from the data imputation unit (104b);a data profiling unit (104d) operably coupled to the duplication removal unit (104c), configured to analyse dataset for patterns, inconsistencies, or anomalies; and a data re-validation unit (104e), operably coupled to the data profiling unit (104d) and the data validation unit (104a), configured to repeat pre-processing steps upon determining that data received from the data profiling unit (104d) is not aligned with predefined standards, wherein the data re-validation unit (104e) initiates a feedback loop to reprocess the data starting from the data validation unit (104a) until compliance with the predefined standard is achieved.
5. The system (100) according to Claim 1, wherein the Al based analytics module (108) further comprises:an aquaculture parameter estimation module (108a) operably coupled to the cloudbased database arrangement (106), configured to:analyse the pre-processed real-time data using an Al and machine learning model to estimate a type and corresponding number of aquaculture stock residing in the aquaculture environment; andmonitor a change in parameters related to the aquaculture environment and at least one of aquaculture stock residing therein;a feed conversion ratio, FCR computation module (108b), operably coupled to the aquaculture parameter estimation module (108a), configured to:compute an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock;store computed FCR value in the cloud-based database arrangement (106) for historical tracking and analysis; anddynamically adjust feeding parameters based on the FCR value;a feed optimization module (108c) operably coupled to the FCR computation module (108b) and the aquaculture parameter estimation module (108a), configured to determine the optimized feeding schedule and the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module (108a) and the computed FCR value; anda predictive analytics module (108d), operably coupled to the feed optimization module (108c), configured to forecast aquaculture stock growth and optimal harvest times based on the computed FCR value, historical data, and real-time inputs.
6. The system (100) according to Claim 1 , wherein the user interface (112) is further configured to:receive the pre-processed real-time data from the cloud-based database arrangement and enable remote live-stream of the aquaculture environment including the pre- processed real-time data, the optimized feeding schedule, and the optimized feed quantity determined by the Al-based analytics module (108); andenable at least one user to remotely control the feed delivery module (110) based on user inputs.
7. The system (100) according to Claim 5, wherein the feed delivery module (110) further comprises:an automated feed control module (110a), configured to dispense feed in real time based on feeding parameters received from the feed optimization module ( 108c);a user override module (110b), operably coupled to the automated feed control module (110a), configured to enable manual control of feed distribution via the user interface (108); anda real-time feed adjustment module (110c) operably coupled to the automated feed control module (110a), configured to adjust feeder settings dynamically based on realtime feeding recommendations from the Al-based analytics module (108), the real-time feed adjustment module is configured to:increase feed delivery upon determining that the computed FCR value exceeds a predefined threshold, indicating suboptimal feed efficiency;decrease feed delivery upon determining that the computed FCR value is within an optimal range, ensuring reduced feed wastage; andcontinuously adjust feed quantity based on real-time FCR variations and aquaculture stock growth determined by the Al-based analytics module (108).
8. A method (200) for monitoring and managing aquaculture environment, comprising steps of:capturing real-time data related to aquaculture environment and aquaculture stock residing therein (202);pre-processing captured real-time data for further analysis (204);storing pre-processed real-time data and facilitating real-time access thereof, and analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity (206);enabling feed delivery to the aquaculture environment based on optimized feeding schedule and the feed quantity in real-time (208); andenabling at least one user to monitoring the aquaculture environment using a user interface (210).
9. The method (200) according to Claim 8, wherein pre-processing captured real-time data for further analysis (204), further comprises method steps of (300):validating the captured real-time data by detecting and flagging invalid data points based on predefined accuracy and completeness criteria (302);imputing missing data using statistical techniques or predictive modelling based on historical data (304);removing redundant or duplicated data from the captured real-time dataset (306); profiling dataset to identify patterns, inconsistencies, or anomalies (308); and re-validating the processed data in a feedback loop until compliance with predefined data quality standards is achieved (310).
10. The method (200) according to Claim 8, wherein analysing the pre-processed real-time time for optimizing feeding schedule and determining an optimized feed quantity (206), further comprises method steps of (400), the method steps (400) comprising:analysing the pre-processed real-time data using an Al and machine learning model to estimate a type and corresponding number of aquaculture stock residing in the aquaculture environment (402);monitoring a change in parameters related to the aquaculture environment and at least one of aquaculture stock residing therein (404);computing an FCR value as a ratio of total provided feed to weight gained by the aquaculture stock (406);storing computed FCR value in the cloud-based database arrangement (106) for historical tracking and analysis (408);dynamically adjusting feeding parameters based on the FCR value (410);determining the optimized feeding schedule and the optimized feed quantity being fed to the aquaculture stock based on the analysis of the aquaculture parameter estimation module (108a) and the computed FCR value (412); andforecasting aquaculture stock growth and optimal harvest times based on the computed FCR value, historical data, and real-time inputs (414).
