Algae bloom removal system capable of autonomous navigation and algae bloom removal method using same

Abstract

The present invention relates to an algae bloom removal system and an algae bloom removal method using same and, more specifically, to an eco-friendly and efficient algae bloom removal system and an algae bloom removal method using same, in which the occurrence of algal bloom can be effectively predicted, and algae bloom can be removed in an automated manner.

Description

Autonomous algae removal system and algae removal method using the same

[0001] The present invention relates to an algae removal system and a method for removing algae using the same, and more specifically, to an eco-friendly and efficient algae removal system and a method for removing algae using the same that can effectively predict the occurrence of algae and remove them in an automated manner.

[0002] In general, algal blooms are emerging as a serious environmental problem worldwide. Algal blooms occur rapidly, primarily due to rising summer water temperatures and anthropogenic pollutants—specifically, an increase in nutrients such as nitrogen and phosphorus—and have a devastating impact not only on aquatic ecosystems but also on human life. While various conventional technologies exist to address this problem, they still have significant limitations in providing a fundamental solution. The purpose of this invention is to overcome the shortcomings of these existing technologies and to provide a more effective and eco-friendly solution.

[0003] Most existing methods for removing algae are manual and have very low operational efficiency. Most conventional technologies suffer from the following problems.

[0004] Conventional algae removal operations are primarily carried out manually, mobilizing a large workforce. For instance, most methods involve physically scraping off algae floating on the water surface or removing them using capture devices. However, these methods have limitations in removing algae from large bodies of water, and particularly in the case of large lakes or rivers, they are time-consuming and incur excessive labor costs.

[0005] Furthermore, manual methods are inefficient because workers must individually inspect and remove algae blooms. Additionally, since algae spread rapidly over time, manpower alone is often insufficient to handle them in a timely manner. In particular, when algae blooms occur on a large scale, these manual methods are largely ineffective, often leading to a vicious cycle where new blooms reappear before the work is completed.

[0006] Conventional technologies have attempted to remove algal blooms using large-scale vessels or water purification devices. However, these devices are effective only in specific body of water and are inefficient for algal blooms occurring over wide areas. Large-scale equipment is costly and limited to specific regions, making it insufficient to solve the problem of algal blooms in vast bodies of water. Furthermore, there is a concern that the underwater ecosystem may be destroyed during the process of physically removing algae while moving mechanical devices.

[0007] Furthermore, a significant number of conventional technologies for removing algae use chemical treatments, which pose a serious problem as they can actually worsen water pollution.

[0008] One widely used method for removing algal blooms is the use of chemical coagulants. Chemical coagulants help to clump algae together into large masses, making them easier to remove from the water surface. However, these coagulants are mostly composed of chemicals and are highly likely to remain in the water. In particular, coagulants often remain in the water, affecting the ecosystem or degrading water quality.

[0009] While chemical coagulants may provide immediate effects, they have a significant long-term impact on aquatic ecosystems. If coagulants remain undegraded, they can be toxic to fish and other aquatic life, posing an indirect risk to humans who ingest them. Thus, although chemical treatments may be effective in the short term, they pose long-term problems of environmental pollution and ecosystem destruction.

[0010] Furthermore, conventional technology sometimes utilizes methods to decompose algae directly through chemical treatment. However, this method also entails many problems. While chemical spraying can rapidly decompose algae, toxic substances may be generated during the process. In particular, the toxic substances produced during the decomposition of algae can be harmful not only to aquatic life but also to humans. Moreover, if the sprayed chemicals remain in the water, water quality can severely deteriorate, potentially rendering long-term water quality improvement impossible.

[0011] In conclusion, conventional technologies for predicting and preventing algal blooms also have clear limitations. Existing systems rely on passive data collection and analysis, which leads to problems such as difficulty in accurate prediction and delayed response.

[0012] Most conventional technologies focus on removing algal blooms after they have already occurred. Systems capable of predicting and taking preventive measures before an bloom occurs are very limited. For example, while there have been attempts to predict the likelihood of algal blooms through simple statistical analysis based on weather and water quality data, their accuracy is very low. This is because conventional technologies remain at the level of merely analyzing past data.

[0013] Furthermore, existing prediction systems attempt to forecast using only limited data, making it difficult to accurately predict the occurrence of algal blooms. While water quality and meteorological data are closely related to algal bloom occurrences, there is a lack of technology capable of comprehensively analyzing them. Consequently, the reliability of predictions is low, leading to problems such as unnecessary work being performed due to inaccurate forecasts or a failure to respond in a timely manner when an algal bloom actually occurs.

[0014] Furthermore, conventional algae removal technologies cause many problems from an economic perspective. The costs associated with the equipment, chemicals, and manpower used for removal are very high, and these methods are not sustainable in the long term.

[0015] Furthermore, existing algae removal equipment has limitations when used in large-scale water bodies. Large-scale equipment has very high initial installation and maintenance costs, and requires frequent repairs.

[0016] Furthermore, the use of chemicals incurs ongoing purchasing and maintenance costs. While this method may provide short-term benefits, maintenance costs skyrocket in the long run. In particular, in areas prone to recurring algal blooms, these costs inevitably become a significant burden.

[0017] Therefore, conventional technologies suffer from low operational efficiency, making it difficult to effectively address algal bloom problems occurring in large bodies of water. Manual methods or simple mechanical removal techniques may be effective only in limited areas and are insufficient to respond when algal blooms spread rapidly. This ultimately leads to a situation where the algal bloom problem remains unresolved and neglected.

[0018] Finally, algae removal is a physically demanding task that can easily lead to worker fatigue. In particular, the risk of safety accidents during water-based operations is very high. The work poses significant risks whether removing algae manually or using large-scale equipment.

[0019] The present invention has been devised to improve upon the aforementioned problems, and the present invention aims to provide an algae removal system having the following objectives:

[0020] 1. Provides a system capable of effectively predicting and responding to the occurrence of algal blooms in advance.

[0021] 2. Implement an efficient system capable of removing algae in an automated manner.

[0022] 3. Presents an eco-friendly solution that can remove algae while minimizing water pollution.

[0023] 4. Develop an autonomous navigation system capable of effectively removing algae blooms even in large bodies of water.

[0024] 5. Provides a method to continuously improve system performance through real-time data collection and analysis.

[0025] 6. Utilize artificial intelligence technology to increase the accuracy and efficiency of algae removal operations.

[0026] 7. Establish a sustainable algal bloom management system that can contribute to water quality improvement and ecosystem protection in the long term.

[0027] To achieve the above objectives, one embodiment of the present invention relates to an autonomously navigable algae removal system, wherein

[0028] It includes an artificial intelligence (AI)-based algal bloom prediction module that predicts the occurrence of algal blooms in advance, wherein the prediction module is a machine learning model that predicts the probability of algal bloom occurrence by utilizing multiple machine learning models including Support Vector Machine (SVM), Random Forest, and Elman Recurrent Neural Network (ERNN);

[0029] A multi-sensor unit that collects and analyzes water quality data in real time in conjunction with the machine learning model described above, wherein the multi-sensor unit includes multiple sensors that measure chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth, and wherein the data collected from the sensors is transmitted to a central control server to form a real-time feedback loop, thereby continuously improving the prediction of algal bloom occurrence and removal efficiency;

[0030] An autonomous vessel capable of autonomous navigation to a location where algal bloom is predicted to occur, providing GPS-based route setting and collision avoidance functions, and including a decomposition device that decomposes algal blooms through a water quality improvement device mounted on the vessel and a coagulant sprayer for algal bloom removal;

[0031] It includes a central control server connected to the above-mentioned autonomous vessel and commanding the vessel's movement path and algae removal operations based on data collected in real time.

[0032] The multi-sensor unit transmits data to a central server in real time via wireless communication, and the sensor data is linked with a data processing module that predicts the likelihood of algal bloom occurrence by reflecting meteorological and hydrological information. During the process of integrating and analyzing hydrological information, including meteorological information, water level, and flow velocity information, the data processing module constructs a prediction model for each data point through AI-based multidimensional regression analysis.

[0033] The autonomous navigation function of the above-mentioned vessel is controlled by an artificial intelligence algorithm, which optimizes the route by considering the frequency of algal blooms, water quality status, weather conditions, hydrological information, and performance data from previous algal bloom removal operations, and sets the optimal route by using criteria for prioritizing high-contamination areas and avoiding obstacles.

[0034] The above decomposition device utilizes specific microorganisms, including Bacillus subtilis and Pseudomonas fluorescens, which decompose algae clumps, and chemical agents including citric acid and hydrogen peroxide.

[0035] The present invention comprises the steps of: collecting a plurality of water quality data, utilizing multiple sensors to detect environmental factors including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth; inputting the collected water quality data into an artificial intelligence (AI) model in real time to predict the likelihood of algal bloom occurrence, wherein the AI ​​model makes predictions based on water quality and weather data learned through a machine learning algorithm; identifying areas with a high likelihood of algal bloom occurrence and setting a route to move an autonomous vessel to the corresponding location, wherein the autonomous vessel sets the route based on GPS and operates autonomously; and, upon arrival at the destination, performing the operation of decomposing the algal bloom by spraying a coagulant or microorganisms to remove the algal bloom or using physical means.

[0036] The aforementioned artificial intelligence model is based on a Multilayer Perceptron (MLP), which is advantageous for processing and analyzing various forms of water quality and environmental data, and analyzes the likelihood of algal bloom occurrence through a process of effectively learning and predicting the nonlinear interactions of complex environmental factors, including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth.

[0037] The path setting step of the autonomous vessel included in the above autonomous operation step uses artificial intelligence-based reinforcement learning to achieve collision avoidance and path optimization, and the reinforcement learning selects the optimal path by analyzing in real time obstacles, weather conditions, water quality changes, and other surrounding environmental factors that may occur on the vessel's operating path, and the reinforcement learning algorithm continuously adjusts the path to new environmental variables through iterative simulation and real-time data collection, and the vessel autonomously sets the path.

[0038] The step of redeploying the autonomous vessel or ordering additional work after the step of performing the above task monitors the work results in real time and dynamically adjusts the work plan in accordance with environmental changes or fluctuations in water quality conditions; the real-time monitoring collects and analyzes information including the vessel's location, work progress, and water quality change data, and based on this, resets the route of the autonomous vessel or executes additional work orders to maximize the efficiency of the algae removal operation.

[0039] According to one embodiment of the present invention, the present invention provides a method for autonomously removing algae by collecting and analyzing water quality data in real time to predict the likelihood of algae occurrence through an autonomous vessel and an artificial intelligence (AI)-based algae prediction and removal system.

[0040] In addition, the present invention improves the accuracy and efficiency of operations by utilizing a machine learning model (150) to analyze water quality data and detect the possibility of algal bloom occurrence in advance. Furthermore, through AI-based reinforcement learning, the autonomous vessel can set an optimal route to quickly reach the algal bloom area and effectively remove the algal bloom using a coagulant injection device (180) and a physical decomposition device (190).

[0041] The autonomous algae removal system of the present invention combines an environmentally friendly coagulant with a physical treatment method to enable efficient algae removal while minimizing water pollution. Through this, rapid water quality improvement is achieved, and long-term ecosystem protection and cost-saving effects can be expected.

[0042] FIGS. 1 and FIGS. 2 are a block diagram and a flowchart schematically showing the overall configuration of an algae removal system according to one embodiment of the present invention.

[0043] FIGS. 3 and 4 are a conceptual diagram and a flowchart illustrating the interaction between an autonomous vessel and related components of an algae removal system according to one embodiment of the present invention.

[0044] FIG. 5 is a flowchart illustrating the data collection and transmission process of an algae removal system according to one embodiment of the present invention.

[0045] FIGS. 6 and FIGS. 7 are flowcharts showing the step-by-step procedure for performing work of an algae removal system according to one embodiment of the present invention.

[0046] The present invention as described above will be explained in detail through the attached drawings and embodiments.

[0047] When a technical term used in this invention is a technical term that similarly expresses the concept of this invention, it should be understood as being replaced with a technical term that can be correctly understood by a person skilled in the art (e.g., ~ module, ~ server, ~ part).

[0048] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings, wherein identical or similar components and functions, regardless of the drawing symbols, are given the same reference number and function as modules, servers, parts, means, devices, etc. having specific functions.

[0049] Furthermore, in describing the present invention, detailed descriptions of related prior art are omitted if it is determined that such descriptions could obscure the essence of the invention. Additionally, it should be noted that the attached drawings are intended only to facilitate an understanding of the concept of the present invention and should not be interpreted as limiting the concept of the present invention.

[0050] In this case, each functional description divided by a distinguishing number describing each embodiment implies that it includes a function or module according to such description. Furthermore, these functions or modules are organically connected to the present invention via a network.

[0051] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. Identical or similar components are given the same reference number regardless of the drawing symbols, and redundant descriptions thereof will be omitted.

[0052] In this case, each functional description divided by a distinguishing number describing each embodiment implies that it includes a function or module according to such description. Furthermore, these functions or modules are organically connected to the present invention via a network.

[0053] The present invention relates to an autonomous navigation device and system for predicting the occurrence of algal blooms and autonomously removing them. In particular, the present invention includes an automated system that collects water quality data in real time to predict the possibility of algal bloom occurrence and provides the function of removing algal blooms through an autonomous navigation vessel. The main components of the present invention consist of a water quality data collection device (110), a multi-sensor unit (120), a wireless communication module (130), a central control server (140), a machine learning model (150), an autonomous navigation vessel (160), a GPS module (170), a coagulant injection device (180), and a physical decomposition device (190), thereby implementing an autonomous system that detects and removes algal blooms in real time.

[0054] The water quality data collection device (110) measures various water quality data in real time that are necessary to detect the possibility of algal blooms occurring in the water body.

[0055] The water quality data collection device described above provides data directly associated with the occurrence of algal blooms, including chlorophyll a and phycocyanin concentrations. Chlorophyll a reflects the biological activity of algae, and phycocyanin is a key indicator of cyanobacteria; these are important water quality data capable of detecting the potential for algal blooms. Additionally, data such as water temperature, dissolved oxygen (DO), and water depth are collected simultaneously, enabling real-time monitoring of environmental conditions affecting algal growth.

[0056] The above-described water quality data collection device has a structure in which various sensors for measuring various water quality elements are integrated, and is designed to monitor a wide area of ​​a water body. The device is combined with a multi-sensor unit (120) to collect various environmental data and has the characteristic of being able to transmit and process the data to a central control server in real time.

[0057] The multi-sensor unit (120) is a combination of several sensors included in a water quality data collection device, and each sensor measures various environmental factors such as chlorophyll a, phycocyanin, water temperature, dissolved oxygen, and water depth. The chlorophyll a sensor detects the amount of cyanobacteria to perform the function of detecting the possibility of algal bloom occurrence in advance. The phycocyanin sensor detects the activation of cyanobacteria to monitor algal bloom proliferation.

[0058] In addition, water temperature sensors measure water temperature to verify growth conditions for algae, while dissolved oxygen (DO) sensors monitor oxygen concentration in the water to provide crucial information for understanding the physiological state of the algae. This data enables rapid detection from the early stages of algal blooms, allowing for the planning of removal operations before the algae proliferate. Depth sensors provide depth information that affects water quality, playing a vital role in understanding the physical environment of algal blooms.

[0059] The wireless communication module (130) performs the role of transmitting data collected from multiple sensor units to a central control server in real time. The wireless communication module uses wireless communication technology such as LTE, 5G, or LoRa to enable stable communication even over a wide area of ​​water. Through this, collected water quality data can be transmitted without delay, allowing the central control server to analyze the situation in real time and perform necessary tasks immediately.

[0060] The wireless communication module is designed to enable the vessel to continuously transmit data while moving autonomously, thereby allowing for the effective removal of algae blooms even in large bodies of water. The wireless communication module also transmits the vessel's location data and, in combination with a GPS module, provides data that can control autonomous navigation.

[0061] The central control server (140) performs the core analysis and control role of the system. The central control server processes data collected through a water quality data collection device and a wireless communication module in real time and predicts the possibility of algal bloom occurrence. In addition, the central control server learns past data using a machine learning model (150) and analyzes the pattern of algal bloom occurrence based on currently collected data to optimize the navigation route and removal operation of the autonomous vessel.

[0062] The central control server includes not only data collection and analysis but also the capability to remotely control autonomous vessels. Through this, it rapidly detects areas where algal blooms have occurred and commands autonomous vessels to move to those areas and perform appropriate removal operations. By performing these control and analysis functions, the central control server enables the prediction of potential algal bloom occurrences in advance and allows for an autonomous response.

[0063] The machine learning model (150) is integrated into a central control server and is responsible for analyzing real-time collected water quality data and predicting the likelihood of algal bloom occurrence. The machine learning model learns from a large amount of past water quality data, weather data, and records of algal bloom occurrences to predict current and future algal bloom occurrences.

[0064] The machine learning models used in the present invention may include deep neural networks, multilayer perceptrons (MLPs), and recurrent neural networks (RNNs). These models learn the major causes of algal blooms based on data accumulated in the past and are used to predict the likelihood of algal bloom occurrence by considering variables such as water temperature, dissolved oxygen (DO), and chlorophyll a.

[0065] The prediction results of machine learning models are reflected in the navigation routes and operational plans of autonomous vessels. For example, if it is determined that there is a high probability of an algal bloom in a specific area, autonomous vessels can be deployed to preemptively detect and remove the bloom in that region. Through this prediction-based response, measures can be taken at the early stages of an algal bloom to prevent large-scale outbreaks.

[0066] The autonomous vessel (160) is the main operating device of the present invention and performs the task of removing algae autonomously according to the command of the central control server. The autonomous vessel includes a GPS module (170) and can move autonomously along a designated path, and moves by calculating the optimal path in real time based on data provided by the central control server. The path setting is based on prediction data from a machine learning model, and the vessel detects an area with a high probability of algae occurrence and moves to that area.

[0067] In this case, the present invention uses a multi-sensor network to measure water quality indicators such as chlorophyll-a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth in real time. This data is input into an algal bloom prediction model, reflects changes in water quality in real time, and is transmitted to a central control server.

[0068] The water quality data collected in this way is analyzed using AI-based machine learning models such as SVM, Random Forest, and Elman Recurrent Neural Network. These models predict the likelihood of algal bloom occurrences by performing multidimensional regression analysis that includes meteorological data and hydrological information. Models trained on diverse data analyze algal bloom patterns to enable more accurate predictions.

[0069] The results of this prediction model are quantified as the risk of algal bloom occurrence for each location and visualized on a map to identify high-risk areas. This visualization supports spatial decision-making and enables proactive preventive measures for high-risk areas.

[0070] This quantification of risk is a process of quantitatively evaluating the likelihood of algal bloom occurrence at each location. A prediction model combines environmental factors at each location (e.g., chlorophyll-a concentration, phycocyanin concentration, water temperature, etc.) to represent the likelihood of an algal bloom at that location as a probability value. This probability value is expressed as a value between 0 and 1, and a higher value indicates a higher likelihood of an algal bloom occurring in that area.

[0071] Meanwhile, autonomous ships are equipped with a collision avoidance system that can detect and avoid obstacles while in operation. The collision avoidance system ensures the safety of autonomous operation and enables the ship to operate autonomously for a long period of time.

[0072] The GPS module (170) provides location data so that the autonomous vessel can accurately follow a designated path. The GPS module checks the coordinates where the vessel is located in real time and communicates with a central control server based on this to adjust or change the path.

[0073] The GPS module also helps the vessel accurately designate the area where it will perform algae removal operations. For example, if a machine learning model determines that there is a high probability of an algae bloom occurring in a specific area, the coordinates of that area are transmitted to the vessel via the GPS module, and the vessel automatically moves along those coordinates.

[0074] The coagulant spraying device (180) is a device mounted on an autonomous vessel and is used to physically perform the task of removing algae. The coagulant spraying device is designed to coagulate clumps of algae to remove them quickly, and the coagulant is composed of environmentally friendly materials that do not adversely affect water quality. When the coagulant is sprayed, the algae clump together into clumps, thereby allowing for efficient removal.

[0075] The coagulant is used as a physically harmless substance and plays a role in maximizing the effectiveness of algae removal while preventing water pollution. The coagulant spraying device can automatically adjust the spray volume according to the autonomous vessel's speed and operating environment, and if necessary, check the operating status after spraying and spray additionally.

[0076] The physical decomposition device (190) is a device that physically decomposes clumps of algae remaining after the injection of a coagulant or large clumps that are difficult to remove. The device is installed on an autonomous vessel and is designed so that the decomposed algae can be disposed of in a manner harmless to the environment. The physical decomposition device provides the ability to rapidly remove algae by destroying their cellular structure and has the advantage of being able to remove algae by a physical method without using a coagulant.

[0077] The physical decomposition device can adjust its decomposition power according to water depth and temperature, enabling it to operate effectively in various water quality environments. For example, if algae blooms occur at deeper depths, the device enhances its decomposition power to decompose green algae even in deep water.

[0078] After the water quality monitoring and feedback operations following the algae removal are completed, the multi-sensor unit (120) collects water quality data again and transmits it to the central control server, and analyzes the results of the operation. Based on the collected data, the central control server determines whether the algae have been completely removed and can order additional removal operations as needed. This feedback system ensures the effectiveness of the operation and saves resources by preventing unnecessary additional work.

[0079] Even after the work is completed, water quality conditions are continuously monitored, and water quality data is recorded in real time. This enables long-term analysis of algal bloom patterns in specific areas and provides important data for preventing future occurrences.

[0080] Accordingly, the present invention provides a system capable of predicting the occurrence of algal blooms in real time and effectively removing them through an autonomous vessel. Water quality data is collected in real time through a water quality data collection device (110) and a multi-sensor unit (120), and the possibility of algal bloom occurrence is predicted using a machine learning model (150). The autonomous vessel (160) sets a route in conjunction with a GPS module (170) and physically removes the algal bloom through a coagulant injection device (180) and a physical decomposition device (190).

[0081] The autonomous system of the present invention provides an environmentally friendly solution that minimizes water pollution and can perform efficient algae removal operations over a long period of time.

[0082] As illustrated in FIG. 3, the invention relates to a system for transmitting data collected through a multi-sensor unit (120) in real time to a central control server (140) via a wireless communication module (130). The communication module transmits the collected data immediately, enabling fast and efficient analysis and processing.

[0083] The multi-sensor unit (120) collects data such as chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth.

[0084] The above multi-sensor unit transmits real-time data from each sensor to a wireless communication module (130), which forms the basis for the detection and removal of algae blooms on an autonomous vessel.

[0085] The chlorophyll a sensor detects the proliferation of algal blooms, while the phycocyanin sensor determines the occurrence of algal blooms by measuring the density of cyanobacteria. In addition, the water temperature sensor and DO sensor comprehensively monitor the water quality environment and detect conditions suitable for algal growth.

[0086] The above wireless communication module (130) transmits data to a central control server using high-speed wireless communication technology such as LTE, 5G, or LoRa. Through this, data collected by the multi-sensor unit is analyzed in real time without delay. This function enables a rapid response when an algal bloom is predicted or detected, thereby contributing to minimizing environmental problems.

[0087] The central control server (140) is a device that performs the central control function of the autonomous navigation system. It receives data transmitted via a wireless communication module in real time and analyzes it to predict the possibility of an algal bloom. The data analyzed based on a machine learning model is immediately reflected in the route setting and work commands of the autonomous navigation vessel.

[0088] The central control server, receiving data in real-time via a wireless communication module, can efficiently carry out algae removal operations on autonomous vessels through rapid analysis and response. Data transmission stability is critical in this process, and the system is designed to enable stable communication even over long distances. This allows for efficient data collection and analysis across vast waters.

[0089] The wireless communication module uses high-performance antennas and multi-band communication technology to ensure the reliability and speed of data transmission.

[0090] Multi-band antennas can accommodate multiple frequencies, extending the communication range and reducing signal interference. In particular, LoRa technology offers long communication distances while consuming little power, enabling autonomous vessels to operate for extended periods. This allows for continuous data transmission even during algae removal operations, while the central control server continuously analyzes real-time data.

[0091] Meanwhile, the technology for optimizing the path in real time through AI-based reinforcement learning for an autonomous vessel (160) is described. The AI-based path optimization system of the present invention enables the autonomous vessel to avoid obstacles and arrive quickly at an area where algal blooms have occurred.

[0092] One of the most critical factors when autonomous vessels perform algae removal operations is route planning. Since algae blooms can occur in various locations, autonomous vessels must follow an optimal path to perform removal quickly. AI-based reinforcement learning enables vessels to learn from real-time data and select the optimal route within a given environment.

[0093] The aforementioned AI system determines the vessel's route by analyzing water quality and environmental data. The reinforcement learning model continuously learns based on feedback from the environment, enabling the vessel to respond quickly to obstacles or weather changes encountered during operation.

[0094] Reinforcement learning is an algorithm in which AI learns optimal behavior while interacting with the environment. The autonomous vessel of the present invention avoids obstacles based on data collected in real time and selects a route that allows it to quickly reach the algal bloom zone. A reward is provided when the vessel arrives quickly at the algal bloom zone, thereby facilitating learning.

[0095] The GPS module (170) plays an important role in this learning process. The GPS module provides coordinates between the ship's current location and the target area, and the AI ​​model can calculate the optimal path based on this. For example, if the ship arrives at the green algae area without hitting any obstacles, the reinforcement learning algorithm learns that path and incorporates it into subsequent navigation.

[0096] Autonomous ships are equipped with an obstacle avoidance system. This system detects the ship's surrounding environment in real time using sensors such as LiDAR, cameras, and ultrasonic sensors. When the ship approaches an obstacle, a reinforcement learning algorithm learns the situation and sets a new path to prevent collisions.

[0097] For example, if a vessel detects an obstacle while moving to perform algae removal work, the AI ​​immediately calculates a new route to avoid the obstacle and allows it to arrive at the target area quickly. This capability enables the vessel to move autonomously and operate safely, while increasing the efficiency of algae removal operations.

[0098] Reinforcement learning algorithms improve the efficiency of navigation route planning as autonomous vessels perform repetitive tasks. By learning from data accumulated in various situations, AI models become able to perform tasks such as obstacle avoidance or algal bloom detection increasingly faster and more accurately. As a result, vessels can reduce fuel consumption, shorten working times, and perform algal bloom removal operations more efficiently.

[0099] As illustrated in FIGS. 1 to 7, this is a detailed description of the coagulant spraying device (180) and physical decomposition device (190) used by an autonomous vessel to remove algae. The present invention enables effective removal of algae by minimizing water pollution through an environmentally friendly coagulant and a physical decomposition method.

[0100] The coagulant spraying device (180) used in the present invention is a device that sprays an environmentally friendly coagulant to remove algae. The coagulant physically coagulates the algae, making the removal process easier. The coagulant minimizes chemical components and is composed of natural mineral-based materials, so it does not have an adverse effect on water quality.

[0101] The coagulant is manufactured with highly biodegradable ingredients that decompose rapidly in water, leaving no residue. This resolves the problem where conventional coagulants could cause water pollution and ensures that the coagulant does not affect the aquatic ecosystem even after decomposition. In particular, the coagulant is designed to be harmless to fish and aquatic life, playing an important role in environmental protection.

[0102] The coagulant spraying device (180) is linked to the movement path of the autonomous vessel and automatically sprays a coagulant in areas where algae are detected. The device can adjust the amount of coagulant sprayed according to water quality data, and if the density of algae is high in a specific area, it sprays additional coagulant to remove algae more quickly and effectively.

[0103] The spraying device checks the status of algae removal even after operation and can receive and execute additional coagulant spraying orders if necessary. This allows for the achievement of maximum removal effectiveness while minimizing the use of coagulants.

[0104] The physical decomposition device (190) is a device that physically decomposes clumps of algae that are difficult to remove with only a coagulant. The device is mounted on an autonomous vessel and decomposes clumps of algae by applying strong pressure underwater.

[0105] The physical decomposition device breaks down clumps of algae into small particles using rotating blades or water currents. The device crushes the algae through the rotation of the blades, and through this process, the algae naturally decompose without further proliferation.

[0106] Physical decomposition devices remove algae solely through physical force, without using chemical treatment. This prevents water pollution and does not have a negative impact on the aquatic ecosystem after operation.

[0107] After the work is completed, the multi-sensor unit (120) collects water quality data again to verify whether the work was successfully completed. Based on the data, the central control server determines whether additional work is required and, if necessary, orders additional removal work.

[0108] Below, a detailed explanation of the method for predicting the occurrence of algal blooms and autonomously removing them is provided.

[0109] The present invention, in particular, includes an automated system that collects water quality data in real time to predict the likelihood of algal bloom occurrence and provides the function to remove algal blooms through autonomous vessels.

[0110] The main components of the present invention consist of a water quality data collection device (110), a multi-sensor unit (120), a wireless communication module (130), a central control server (140), a machine learning model (150), an autonomous vessel (160), a GPS module (170), a coagulant injection device (180), and a physical decomposition device (190), thereby implementing an autonomous system that detects and removes algal blooms in real time.

[0111] The present invention relates to a method for predicting the occurrence of algal blooms and automatically removing them. This method consists of a series of procedures comprising collecting and analyzing water quality data in real time, and an autonomous vessel moving to an area where algal blooms are likely to occur to perform removal operations.

[0112] 1. Water quality data collection and real-time analysis phase

[0113] The first step is to collect various water quality data using a water quality data collection device (110). The multi-sensor unit (120) measures data such as chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth in real time. Through this data, the possibility of an algal bloom can be detected in advance.

[0114] Chlorophyll a sensors are used to measure the concentration of algae (particularly cyanobacteria). Since chlorophyll a concentration is closely related to the photosynthetic activity of cyanobacteria, this allows for the identification of early signs of algal blooms. Phycocyanin sensors measure the presence and density of cyanobacteria. Phycocyanin is a substance found within the cells of cyanobacteria, and its concentration increases when these bacteria, the primary cause of algal blooms, become active.

[0115] Water temperature sensors play a crucial role in understanding the growth environment of algae. Since cyanobacteria can multiply rapidly within a certain temperature range, monitoring changes in water temperature is advantageous for predicting the occurrence of algal blooms. DO sensors measure the amount of dissolved oxygen in the water. As a decrease in DO creates an environment suitable for algal growth, DO data serves as an important factor in predicting the likelihood of algal blooms.

[0116] The water quality data collection device (110) transmits data collected from the multi-sensor unit (120) to the central control server in real time. This is done via a wireless communication module (130), and the data is analyzed at the central server immediately after collection. This real-time data collection enables the prediction of algal bloom occurrence in advance and allows for preemptive response.

[0117] 2. AI-based algal bloom prediction stage

[0118] The collected water quality data is transmitted to a central control server (140) and analyzed through a machine learning model (150). In the present invention, the possibility of algal bloom occurrence is predicted using a machine learning model that has learned past algal bloom occurrence patterns and weather data.

[0119] Machine learning models can predict current and future conditions by learning from historical water quality data, meteorological information (temperature, humidity, wind, etc.), and records of algal bloom occurrences. These models utilize Multilayer Perceptrons (MLPs) or Recurrent Neural Networks (RNNs) to analyze patterns in water quality data and predict areas with a high probability of algal bloom occurrence.

[0120] Based on the analyzed data, the central control server identifies areas where there is a high probability of algal blooms occurring. This allows the autonomous vessel (160) to set a route to move to the area. The machine learning model continuously learns and improves the accuracy of the prediction whenever new data is input. As a result, the vessel can arrive at the area preemptively before algal blooms occur and perform algae removal operations.

[0121] The present invention features a feedback system that continuously monitors water quality conditions even after the actual algae removal operation is completed following prediction. Data collected after the operation is fed back into a machine learning model to improve the model's prediction accuracy and reflect it in future operations. This enables the vessel to perform operations more efficiently over time.

[0122] 3. Route setting and movement phases of autonomous vessels

[0123] When an area with a high probability of algal bloom is predicted, the autonomous vessel (160) moves to that area. At this time, the vessel moves along an accurate path using a GPS module (170) and monitors the situation in real time through communication with a central control server (140).

[0124] The GPS module (170) tracks the real-time location of the vessel and sets a route based on data transmitted from the central control server. The vessel moves to an area with a high probability of algal bloom and can modify the route in real time depending on obstacles or weather conditions along the route. The GPS module is a key element for accurately controlling the movement of the autonomous vessel, which enables the vessel to arrive at its destination quickly and safely.

[0125] Autonomous vessels are equipped with built-in obstacle avoidance systems that can detect and avoid obstacles or hazards along their path. They sense the surrounding environment using sensors such as LiDAR, ultrasonic sensors, and cameras, and change their route or establish a new one if necessary. This is accomplished automatically by AI-based path optimization algorithms.

[0126] In this invention, route optimization is designed to enable a vessel to reach a target area using minimal energy. The autonomous vessel modifies its route in real-time based on GPS data and uses an optimized algorithm to move to areas affected by algal blooms as quickly and safely as possible. This route optimization system shortens work time, reduces energy consumption, and maximizes operational efficiency.

[0127] 4. Steps to Perform Algae Removal Operation

[0128] When the autonomous vessel arrives at an area where an algal bloom has occurred, it performs an algal bloom removal operation using a coagulant spraying device (180) and a physical decomposition device (190).

[0129] The coagulant spraying device (180) is mounted on the autonomous vessel and automatically sprays the coagulant when it arrives at an area where an algal bloom has occurred. The coagulant physically coagulates the algal bloom into larger clumps, allowing them to be easily removed from the water. The coagulant used in the present invention is composed of environmentally friendly materials, so that the algal bloom can be effectively removed while minimizing water pollution.

[0130] The coagulant used in this invention is based on natural minerals, has a rapid decomposition rate, and does not have an adverse effect on the aquatic ecosystem. This is an important characteristic that differentiates it from conventional chemical coagulants. The coagulant completely decomposes within a certain period after application and leaves no residue, thereby preventing additional water pollution.

[0131] The physical decomposition device (190) is a device that physically decomposes large clumps of algae that are difficult to remove with coagulants alone. The device is mounted on an autonomous vessel and decomposes the clumps of algae into small particles to facilitate the natural decomposition process. The physical decomposition device uses a rotary blade or a strong water stream to decompose unaggregated clumps of algae. This device breaks down the clumps of algae into small pieces so that they can naturally decompose in the water.

[0132] Since the physical decomposition device removes algae using only physical force without the use of chemicals, it does not cause water pollution. Furthermore, it does not have a negative impact on the aquatic ecosystem after the decomposition process, and water quality is restored immediately afterward.

[0133] 5. Verification and feedback phase after work completion

[0134] When the algae removal operation is completed, the multi-sensor unit (120) collects data again to check whether the algae has been completely removed. The central control server (140) analyzes the results of the operation based on the collected data and automatically issues commands to the autonomous vessel if additional work is required.

[0135] Water quality monitoring continues for a certain period even after the work is completed. The multi-sensor unit (120) measures the water quality after the work and analyzes whether the coagulant or algae or fine particles remaining after physical decomposition have an adverse effect on the water quality. This feedback system increases the effectiveness of the work and prevents unnecessary additional work.

[0136] The central control server feeds the data collected after operations back into machine learning models to enhance the efficiency of future operations. This enables the system to perform increasingly accurate and efficient algae removal over time. Furthermore, if additional tasks are ordered as needed, the autonomous vessel can immediately return to the relevant area to carry out the work.

[0137] The present invention presents a method for predicting the occurrence of algal blooms by collecting and analyzing water quality data in real time and automatically removing algal blooms using an autonomous vessel. Components such as a water quality data collection device (110), a multi-sensor unit (120), a central control server (140), a machine learning model (150), an autonomous vessel (160), a GPS module (170), a coagulant injection device (180), and a physical decomposition device (190) operate organically to automate the algal bloom removal process, thereby minimizing water pollution while maximizing work efficiency.

[0138] As an example, the present invention relates to a method of using a machine learning model (150) to predict the possibility of an algal bloom.

[0139] In particular, it is a method that uses machine learning algorithms such as Multilayer Perceptrons (MLPs) and Recurrent Neural Networks (RNNs) to analyze past and current water quality data and predict the likelihood of future algal blooms.

[0140] A multilayer neural network (MLP) is a model composed of multiple layers of artificial neurons, consisting of an input layer, a hidden layer, and an output layer. Various water quality data (chlorophyll a, phycocyanin, water temperature, DO, etc.) collected from a water quality data collection device (110) is transmitted to the input layer, and each neuron extracts features of the water quality data and transmits them to the hidden layer.

[0141] MLPs learn from historical data through supervised learning. Historical water quality and weather data are stored on a central control server, which is recorded along with the occurrence of algal blooms. Based on this data, the MLP learns the impact of each water quality element on the occurrence of algal blooms. Through this process, it can analyze current data input in real time and predict the likelihood of algal bloom occurrence.

[0142] The MLP predicts areas with a high probability of algal bloom occurrence based on analyzed data. The prediction results are reflected in the route setting of the autonomous vessel, and the vessel can move to the corresponding area through the GPS module (170). The predicted location is gradually improved according to accuracy, thereby preventing the possibility of algal bloom occurrence in advance.

[0143] Recurrent Neural Networks (RNNs) are algorithms specialized for processing time-series data and are suitable for analyzing changes in water quality data over time. In this invention, changes in water temperature, dissolved oxygen (DO), and chlorophyll a concentration are analyzed using an RNN, and based on this, patterns of algal bloom occurrence are predicted.

[0144] RNNs have a cyclic structure that stores information learned in the previous step and incorporates it into the prediction of the current step. This is highly useful for analyzing how water quality variables, such as water temperature and dissolved oxygen (DO), change over time. For example, if water temperature gradually rises over a certain period, it can be considered a precursor to an algal bloom. RNNs learn these temporal patterns to predict the likelihood of an algal bloom more accurately.

[0145] The RNN processes real-time data collected from the multi-sensor unit (120) to analyze changes in water quality over time. Through this, the likelihood of algal blooms occurring in the short and long term can be predicted, and the predicted results are reflected in the work path of the autonomous vessel. In addition, the RNN has strengths in analyzing long-term patterns, so it can comprehensively analyze not only temporary changes in water quality but also interactions with weather conditions.

[0146] The probability of algal bloom occurrence predicted through MLP and RNN models is directly reflected in the route optimization and work planning of the autonomous vessel. Based on the prediction results, the autonomous vessel moves to a danger area and performs algae removal work in real time. After the work is completed, the multi-sensor unit (120) collects water quality data again, and a feedback system is established to evaluate the accuracy of the prediction and improve the model.

[0147] Meanwhile, this invention relates to a method for setting a route so that an autonomous vessel learns an optimal path through AI-based reinforcement learning and arrives quickly at an area affected by algal blooms. In this invention, reinforcement learning is used to modify the vessel's route in real time and to optimize the route by considering obstacles or weather conditions.

[0148] Reinforcement learning is a method in which AI learns by interacting with the environment, and the autonomous vessel receives feedback on route setting and work efficiency while performing the algae removal task. The central control server (140) collects real-time data of the vessel and evaluates how efficient the route selected by the vessel is to award a reward.

[0149] The basic structure of reinforcement learning consists of State, Action, and Reward. The ship's current location and surrounding environmental information (water quality data, obstacle locations, etc.) are defined as the State, while the paths the ship can choose are represented as Actions. The AI ​​receives a reward for each Action, and over time, it comes to prefer paths with higher rewards.

[0150] When an autonomous vessel moves toward an area affected by algal blooms, the AI ​​selects the route that allows it to arrive in the shortest possible time among the possible routes. To this end, the route is set based on the vessel's location information in conjunction with a GPS module (170), and the AI ​​is evaluated for a reward for the selected route. If the vessel arrives faster by selecting a route that avoids obstacles or considers weather conditions, that route receives a higher reward, and the AI ​​learns this route selection.

[0151] Through reinforcement learning, autonomous vessels can analyze their surroundings in real time and modify their routes in real time. For example, if a vessel detects an obstacle on its path, the AI ​​immediately sets a new route to avoid it. Using various sensors such as LiDAR, ultrasonic sensors, and cameras, the vessel perceives its environment and performs real-time route optimization based on this information.

[0152] Route correction is performed automatically through AI algorithms. The AI ​​evaluates the efficiency of the current route and immediately establishes a new path if obstacles are detected or weather conditions change. At this stage, a route optimization model trained through reinforcement learning is applied to rapidly calculate the most efficient route available to the vessel.

[0153] The obstacle avoidance system consists of LiDAR, cameras, and ultrasonic sensors, and helps the vessel automatically avoid obstacles when detected along its path. AI analyzes data collected from each sensor in real time and establishes a new path to avoid the obstacles. During this process, the vessel continuously learns, and its obstacle avoidance and path optimization capabilities improve over time.

[0154] Optimized route planning through reinforcement learning maximizes the operational efficiency of autonomous vessels. Ships can travel along the fastest route using minimal fuel, enabling them to perform algae removal tasks more quickly and efficiently. Furthermore, reinforcement learning enhances the vessel's adaptability to obstacles and weather changes, allowing it to perform stable operations in diverse environments.

[0155] The present invention relates to a method for removing algae in an environmentally friendly manner using a coagulant injection device (180) and a physical decomposition device (190) mounted on an autonomous vessel. The present invention provides a technology capable of effectively removing algae while minimizing water pollution.

[0156] The coagulant spraying device (180) is a device that sprays a coagulant to physically coagulate the algae when the autonomous vessel arrives at an area where algae have occurred. The coagulant used in the present invention is composed of environmentally friendly materials and does not have an adverse effect on water quality.

[0157] The coagulant is composed of naturally derived substances, such as natural minerals, and is characterized by rapid decomposition and the absence of residue. This distinguishes it from conventional chemical coagulants and ensures that it does not cause water pollution even after the removal of algae. When the coagulant is sprayed into the water, the algae aggregate into large clumps, which can then be easily removed from the water.

[0158] The coagulant injection device (180) automatically adjusts the amount of coagulant injected based on water quality data collected from the multi-sensor unit (120). More coagulant is injected in areas with high algae density, and less is injected in areas with low density. This allows for the optimization of coagulant usage and minimizes environmental impact.

[0159] The physical decomposition device (190) is a device that physically decomposes large clumps of algae that are difficult to remove with coagulants alone. The device is mounted on an autonomous vessel and applies strong physical force to decompose the algae into small particles.

[0160] The physical decomposition device uses rotary blades or a powerful water stream to break down unaggregated clumps of algae. This device finely crushes the algae clumps, facilitating their natural decomposition in the water. The device physically destroys the cellular structure of the algae, enabling its rapid removal.

[0161] Since the physical decomposition device removes algae using only physical force without the use of chemicals, it does not cause water pollution. Furthermore, it does not have a negative impact on the aquatic ecosystem after the decomposition process, and water quality is restored immediately afterward.

[0162] The present invention provides a method for effectively removing algae while minimizing water pollution by combining a coagulant and a physical decomposition device. The coagulant is composed of environmentally friendly materials, and the physical decomposition device can physically remove algae without chemical treatment. This prevents water pollution and allows for rapid restoration of water quality after the operation.

[0163] [Example 1]

[0164] The present invention comprises an artificial intelligence (AI)-based algal bloom prediction module that predicts the occurrence of algal blooms in advance, wherein the prediction module includes a machine learning model that predicts the probability of algal bloom occurrence by utilizing a plurality of machine learning models including Support Vector Machine (SVM), Random Forest, and Elman Recurrent Neural Network (ERNN); a multi-sensor unit that collects and analyzes water quality data in real time in conjunction with the machine learning model, wherein the multi-sensor unit includes a plurality of sensors that measure chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth, and wherein the data collected from the sensors is transmitted to a central control server to form a real-time feedback loop, thereby continuously improving the efficiency of algal bloom occurrence prediction and removal; and an autonomous vessel capable of autonomous navigation to a location where algal bloom occurrence is predicted, providing GPS-based route setting and collision avoidance functions, and including a decomposition device that decomposes algal blooms through a water quality improvement device mounted on the vessel and a coagulant sprayer for algal bloom removal. A central control server connected to the autonomous vessel, which predicts the risk of algal bloom occurrence based on data collected in real time and autonomously commands the vessel's movement path and algal bloom removal operations; wherein the central control server includes an artificial intelligence (AI)-based algal bloom prediction module that predicts the occurrence of algal blooms in advance, and the prediction module utilizes a plurality of machine learning models including Support Vector Machine (SVM), Random Forest, and Elman Recurrent Neural Network (ERNN) to receive water quality data collected in real time and predict the probability of algal bloom occurrence;The system includes a multi-sensor unit that collects and analyzes water quality data in real time in conjunction with the machine learning model, wherein the multi-sensor unit includes multiple sensors that measure chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth, and the data collected from the sensors is transmitted to a central control server via a wireless communication module to form a real-time feedback loop that improves the prediction accuracy of the machine learning model; an autonomous vessel capable of autonomous navigation to a location where algal bloom occurrence is predicted, providing GPS-based route setting and collision avoidance functions, and removing algal blooms through a coagulant injection device and a physical decomposition device mounted on the vessel; and a central control server connected to the autonomous vessel, wherein the central control server inputs water quality data collected in real time into the machine learning model to predict the risk of algal bloom occurrence, sets the movement path of the autonomous vessel according to the prediction result, and commands the algal bloom removal operation.

[0165] [Example 2]

[0166] The above-described multi-sensor unit transmits data to a central server in real time via wireless communication, and the algae removal system is characterized by being linked with a data processing module that predicts the possibility of algae bloom occurrence by reflecting meteorological and hydrological information based on the sensor data.

[0167] In the process of integrating and analyzing hydrological information including meteorological information, water level, and flow velocity information, the above data processing module derives interrelationships between data through AI-based multidimensional regression analysis and constructs an accurate prediction model by cross-validating the meteorological information and hydrological information.

[0168] The above multi-sensor unit transmits chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth data collected from the sensors to a central control server in real time via a wireless communication module;

[0169] The data processing module included in the central control server integrates and analyzes received sensor data, meteorological information (temperature, humidity, precipitation), and hydrological information (water level, flow velocity), performs AI-based multidimensional regression analysis to quantify the interrelationships between variables, and thereby constructs a prediction model that predicts the likelihood of algal bloom occurrence;

[0170] The above prediction model uses a cross-validation technique to verify the interrelationships between sensor data, meteorological information, and hydrological information, and improves the accuracy of the prediction;

[0171] Based on the above prediction results, the movement path of the autonomous vessel and the algae removal operation are optimized.

[0172] The above multi-sensor unit transmits chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth data collected from the sensors to a central control server in real time via a wireless communication module;

[0173] The data processing module included in the central control server integrates and analyzes received sensor data, meteorological information (temperature, humidity, precipitation), and hydrological information (water level, flow velocity), performs AI-based multidimensional regression analysis to quantify the correlation between each variable, and thereby constructs a model that predicts the likelihood of algal bloom occurrence;

[0174] The above data processing module statistically verifies the correlation between sensor data, meteorological information, and hydrological information using a cross-validation technique, calculates performance indicators such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R²) to evaluate the performance of the prediction model, and quantitatively evaluates the prediction accuracy of the model based on these indicators to optimize the model parameters;

[0175] Based on the results of the above-mentioned optimized prediction model, the movement path of the autonomous vessel is set, and the algae removal operation is performed efficiently.

[0176] [Example 3]

[0177] The autonomous navigation function of the above-mentioned vessel is controlled by an artificial intelligence algorithm, and the algorithm optimizes the route by considering the frequency of algal bloom occurrence, water quality status, weather conditions, hydrological information, and performance data of existing algal bloom removal operations, and in establishing the optimal route by using criteria for prioritizing high-contamination areas and avoiding obstacles,

[0178] The autonomous navigation function of the above-mentioned autonomous vessel is controlled by an artificial intelligence algorithm;

[0179] The above algorithm receives data on the frequency of algal bloom occurrence, water quality status, weather conditions, hydrological information, and the performance of previous algal bloom removal operations from a central control server, calculates the risk of algal bloom occurrence for each region, and;

[0180] The above risk calculation is performed by assigning weights to each variable, and utilizes the analysis of data from areas where algae removal was successful and areas where it was not in previous algae removal operations for the training of the algorithm;

[0181] Determining the order of ship movement by assigning priority to critically contaminated areas with a high calculated risk of algal bloom occurrence;

[0182] When setting the route, the location of obstacles is identified using sensors mounted on the vessel and map data, and a route capable of avoiding obstacles is calculated by applying minimum safety distance criteria;

[0183] Finally, the optimal travel route is established by integrating the priority of critical contamination areas and obstacle avoidance paths to minimize fuel consumption and travel time.

[0184] [Example 4]

[0185] The above decomposition device utilizes specific microorganisms, including Bacillus subtilis and Pseudomonas fluorescens, that decompose algal blooms, and chemical agents including citric acid and hydrogen peroxide,

[0186] The above decomposition device comprises a microbial spraying device that stores and sprays specific microorganisms, including Bacillus subtilis and Pseudomonas fluorescens, to decompose green algae clumps;

[0187] It includes a chemical spraying device that sprays a solution of citric acid and hydrogen peroxide mixed at a certain concentration onto an area where algal blooms occur to destroy the cell structure of the algal blooms and promote their decomposition;

[0188] The above chemical spraying device effectively decomposes green algae by uniformly spraying a solution mixed with citric acid at a concentration of 1-3% and hydrogen peroxide at a concentration of 2-5% into a body of water where green algae have occurred through a spray nozzle mounted on an autonomous vessel, thereby destroying the cell walls of the green algae through oxidation and acidification reactions.

[0189] [Example 5]

[0190] A step of collecting multiple water quality data, wherein the data utilizes multiple sensors to detect environmental factors including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth; a step of inputting the collected water quality data into an artificial intelligence (AI) model in real time to predict the likelihood of algal bloom occurrence, wherein the AI ​​model makes predictions based on water quality and weather data learned from machine learning algorithms; a step of identifying areas with a high probability of algal bloom occurrence and setting a route to move an autonomous vessel to the corresponding location, wherein the autonomous vessel sets the route based on GPS and operates autonomously; and a step of, upon arrival of the autonomous vessel at the destination, performing an operation to decompose the algal bloom by spraying coagulants or microorganisms to remove the algal bloom or using physical means; wherein, after the removal operation is completed, the processed water quality data is collected and analyzed again to determine whether the algal bloom has been completely removed, or, if the algal bloom removal operation is not completed, the autonomous vessel is redeployed or additional operations are ordered based on reference values ​​of algal bloom concentration, dissolved oxygen (DO) levels, and water turbidity.

[0191] Step 1, collecting multiple water quality data including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth in real time using multiple sensors;

[0192] Step 2, which involves inputting the collected water quality data into an artificial intelligence (AI) model to predict the likelihood of algal bloom occurrence, wherein the AI ​​model performs real-time predictions based on water quality and weather data learned from a machine learning algorithm;

[0193] Step 3, which involves identifying areas with a high probability of algal bloom occurrence and setting a route to move an autonomous vessel to the corresponding location, wherein the autonomous vessel sets the route based on GPS and operates autonomously, and uses an obstacle avoidance algorithm to ensure safe movement;

[0194] Step 4: When the autonomous vessel arrives at its destination, perform the task of decomposing the algae using a coagulant sprayer or microbial sprayer mounted on the vessel to remove the algae, or using a physical decomposition device;

[0195] Step 5, after the above removal operation is completed, collects water quality data processed through multiple sensors, measures chlorophyll a concentration, phycocyanin concentration, dissolved oxygen (DO) levels, and water turbidity, and determines whether the algal bloom has been completely removed by determining whether each water quality indicator is below a preset standard value;

[0196] Step 6, if, as a result of the above judgment, it is determined that one or more water quality indicators exceed a standard value and the algae removal work is not completed, identifies an area requiring additional removal work based on the exceeded water quality indicators, and redeploys an autonomous vessel to that area or orders additional work to perform the algae removal work again.

[0197] [Example 6]

[0198] The above artificial intelligence (AI) model is based on machine learning algorithms including a multilayer perceptron or a recurrent neural network to analyze past water quality and weather data and, through this, to predict the likelihood of current and future algal bloom occurrences.

[0199] The above artificial intelligence (AI) model is based on machine learning algorithms including a multilayer perceptron (MLP) or a recurrent neural network (RNN);

[0200] Using water quality data (chlorophyll a concentration, phycocyanin concentration, water temperature, dissolved oxygen (DO) levels, water depth) and meteorological data (temperature, humidity, precipitation, solar radiation, wind speed) collected in the past as training data, the correlation between each variable and the occurrence of algal blooms is learned;

[0201] By inputting currently collected real-time water quality and weather data into a trained model, probabilistically predicting the probability of algal bloom occurrence based on combinations of each variable;

[0202] The predicted probability of algal bloom occurrence is visualized as a map quantifying the risk of algal bloom occurrence in specific regions, identifying areas where the risk exceeds a pre-set threshold, and designating those areas as priority targets for algal bloom removal by autonomous vessels.

[0203] [Example 7]

[0204] The path setting step of the above-mentioned autonomous vessel uses AI-based reinforcement learning for collision avoidance and path optimization to automatically set an optimal path considering obstacles and environmental conditions, and the path setting step of the above-mentioned autonomous vessel uses an AI-based reinforcement learning algorithm for collision avoidance and path optimization;

[0205] The above reinforcement learning algorithm receives real-time input of the ship's current location information, obstacle location and size data, weather conditions (wind speed, wave height, temperature), and hydrological information (water depth, current velocity);

[0206] The reward function is defined with the goal of minimizing travel time, minimizing fuel consumption, and maximizing safety (collision avoidance) and algae removal efficiency;

[0207] The algorithm performs learning through simulation and actual navigation data to learn a policy to select the action (determining the ship's speed and direction) that maximizes the reward function in a given environment;

[0208] When an obstacle is detected according to a learned policy, it calculates a detour route while maintaining a minimum safe distance, and generates an optimal movement path by adjusting the vessel's operating speed and direction in real time based on weather conditions and hydrological information;

[0209] Through this, autonomous vessels automatically set the optimal route to the algal bloom area by considering obstacles and environmental conditions.

[0210] [Example 8]

[0211] In the step of redeploying the above-mentioned autonomous vessel or ordering additional operations, the results of the operation are monitored in real time and dynamically adjusted in accordance with environmental changes to maintain optimal algae removal efficiency, wherein the reference value is dynamically adjusted in real time according to environmental changes to maintain optimal algae removal efficiency.

[0212] In the step of redeploying the above-mentioned autonomous vessel or ordering additional operations, to monitor the results of the operations in real time and dynamically adjust to environmental changes, collected water quality data (chlorophyll a concentration, phycocyanin concentration, dissolved oxygen (DO) levels, water turbidity), meteorological information (temperature, humidity, precipitation, solar radiation), and hydrological information (water level, flow velocity) are transmitted to a central control server;

[0213] The artificial intelligence algorithm of the central control server above calculates the risk of algal bloom occurrence due to environmental changes in real time based on collected data, and accordingly dynamically resets the standard values ​​of each water quality indicator;

[0214] Resetting threshold values ​​is performed through predefined functions or algorithms based on changes in environmental variables; for example, if the risk of algal blooms increases due to rising temperatures or increased solar radiation, the threshold values ​​for chlorophyll a and phycocyanin concentrations are lowered to detect algal blooms more sensitively;

[0215] It determines the necessity of algae removal operations by comparing reset reference values ​​with current water quality data, and maintains optimal algae removal efficiency by rapidly responding to the occurrence of algae caused by environmental changes through the redeployment of autonomous vessels to the relevant location or the issuance of additional operations if necessary.

Claims

1. In an autonomous algae removal system, It includes an artificial intelligence (AI)-based algal bloom prediction module that predicts the occurrence of algal blooms in advance, wherein the prediction module is a machine learning model that predicts the occurrence of algal blooms by utilizing multiple machine learning models including Support Vector Machine (SVM), Random Forest, and Elman Recurrent Neural Network (ERNN); A multi-sensor unit that collects and analyzes water quality data in real time in conjunction with the machine learning model described above, wherein the multi-sensor unit includes a plurality of sensors that measure chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth, and wherein the data collected from the sensors is transmitted to a central control server to form a real-time feedback loop; An autonomous vessel capable of autonomous navigation to a location where algal bloom is predicted to occur, providing GPS-based route setting and collision avoidance functions, and including a decomposition device that decomposes algal blooms through a water quality improvement device mounted on the vessel and a coagulant sprayer for algal bloom removal; An autonomous algae removal system characterized by including a central control server connected to the above-mentioned autonomous vessel and commanding the vessel's movement path and algae removal operations based on data collected in real time.

2. In Claim 1, The above-described multi-sensor unit transmits data to a central server in real time via wireless communication, and the sensor data is linked with a data processing module that predicts the possibility of algal bloom occurrence by reflecting weather and hydrological information, and the data processing module is characterized by constructing a prediction model through AI-based multidimensional regression analysis of each data during the process of integrating and analyzing hydrological information including weather information, water level, and flow velocity information, thereby enabling an autonomous algal bloom removal system.

3. In Claim 1, An autonomous algae removal system capable of autonomous navigation, wherein the autonomous navigation function of the above-mentioned vessel is controlled by an artificial intelligence algorithm, and the algorithm generates a route by considering the frequency of algae occurrence, water quality status, weather conditions, hydrological information, and performance data of existing algae removal operations, and sets the route using criteria for prioritizing high-risk contaminated areas and avoiding obstacles.

4. In Claim 1, The above decomposition device is an autonomous algae removal system characterized by utilizing specific microorganisms, including Bacillus subtilis and Pseudomonas fluorescens, which decompose algae clumps, and chemical agents including citric acid and hydrogen peroxide.

5. Regarding methods for removing algae, A step of collecting multiple water quality data, wherein the data utilizes multiple sensors to detect environmental factors including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth; A step of inputting the collected water quality data into an artificial intelligence (AI) model in real time to predict the possibility of algal bloom occurrence, and the AI ​​model making a prediction based on water quality and weather data learned from a machine learning algorithm; A step of identifying areas where the probability of algal bloom occurrence is higher than a certain value, setting a route to move an autonomous vessel to the said location, and the autonomous vessel setting the route based on GPS and operating autonomously; A method for removing algae characterized by including the step of, when an autonomous vessel arrives at a destination, performing a task to decompose the algae by spraying a coagulant or microorganism to remove the algae or using physical means.

6. In Claim 5, A method for removing algal blooms characterized by the above artificial intelligence model being based on a Multilayer Perceptron (MLP) that processes and analyzes various forms of water quality and environmental data, and analyzing the possibility of algal bloom occurrence through a process of effectively learning and predicting the nonlinear interactions of complex environmental factors including chlorophyll a, phycocyanin, water temperature, dissolved oxygen (DO), and water depth.

7. In Claim 5, A method for removing algae, characterized in that the path setting step of the autonomous vessel included in the above-mentioned autonomous operation step uses artificial intelligence-based reinforcement learning for collision avoidance and path setting, the reinforcement learning selects a path by analyzing in real time obstacles, weather conditions, water quality changes, and other surrounding environmental factors that may occur on the vessel's operating path, and the reinforcement learning algorithm continuously adjusts the path to new environmental variables through iterative simulation and real-time data collection, and the vessel autonomously sets the path.

8. In Claim 5, A method for removing algae, characterized in that, after the step of performing the above-mentioned work, the step of redeploying the autonomous vessel or ordering additional work involves monitoring the work results in real time and dynamically adjusting the work plan in accordance with environmental changes or fluctuations in water quality conditions, wherein the real-time monitoring involves collecting and analyzing information including the vessel's location, work progress, and water quality change data, and based on this, resetting the route of the autonomous vessel or ordering additional work.

Images