IoT AND AI-BASED SMART WATER QUALITY MANAGEMENT SYSTEM
The smart water quality environment control system addresses inefficiencies in conventional systems by dividing zones and using AI to optimize sterilization and resource allocation, ensuring efficient and rapid responses to contamination in public water features.
Patent Information
- Application Number
- KR1020260001893
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-01-06
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2045-10-28
AI Technical Summary
Conventional water quality management systems for public water features fail to differentiate between zones of high and low contamination risk, leading to inefficient resource use and delayed responses to contamination, and lack real-time zone-specific control capabilities.
A smart water quality environment control system utilizing IoT and AI that divides the spraying space into grid zones, collects data on user information, water quality, and environmental conditions, and adjusts sterilization intensity and resource allocation based on AI analysis for each zone, using ozone, chlorine, and UV sterilization methods.
Enables customized, efficient, and rapid responses to contamination, reducing resource waste and ensuring hygienic safety by optimizing sterilization based on real-time data analysis and zone-specific needs.
Smart Images

Figure 112026001574683-PAT00011_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a smart water quality environment control system utilizing IoT and artificial intelligence. Background Technology
[0002] Water features have established themselves as an important element providing citizens with rest and entertainment in various public spaces, such as parks, plazas, and waterfront areas. Facilities like ground fountains, water play areas, and mirror ponds become crowded spaces depending on seasonal factors or usage purposes; given the high volume of use by children and families, hygienic safety management is essential. Due to their nature as open spaces, these facilities are directly exposed to the influx of external pollutants, contact by multiple users, and rapid changes in weather conditions; therefore, a system for managing water quality safety continuously and precisely is absolutely necessary.
[0003] Conventional sterilization methods for hydroponic facilities generally involve treating circulating water in a batch using equipment such as AOP (ozone), CL (chemicals), and UV. However, because this single treatment structure applies the same sterilization intensity to the entire facility, it fails to distinguish between areas with a high probability of contamination and those without, leading to problems such as unnecessary use of sterilized water and waste of resources. Furthermore, there were limitations in operational stability, such as the difficulty of immediate response in the event of equipment failure and delays in responding to initial contamination during the restart process.
[0004] Furthermore, the dwell patterns, density, and age demographics of facility users can vary significantly by zone, and various variables exist due to the frequent coexistence of adults, infants, and pets. Nevertheless, conventional technology lacked a system capable of detecting and reflecting these zone-specific differences in real time, resulting in reduced efficiency and precision in the management and response to water pollution.
[0005] Meanwhile, with the recent proliferation of smart city technologies, there has been an increasing number of attempts to collect and analyze environmental data and user behavior in real time using IoT sensors and artificial intelligence. This trend demonstrates the potential to simultaneously achieve precise hygiene management and resource optimization in the operation of water features.
[0006] Therefore, in order to enhance operational efficiency while maintaining the public value and safety of water features, there is a call for the development of a system capable of managing water quality in a sustainable manner by precisely reflecting various environmental factors and usage patterns. The problem to be solved
[0007] The objective of the present invention is to provide a smart water quality environment control system utilizing IoT and artificial intelligence that enhances hygiene safety and operational efficiency by dividing and managing the spraying space of a water feature facility by zone, multidimensionally detecting on-site status data, and performing customized sterilization control and resource allocation for each zone based on the results of integrated analysis using artificial intelligence, which was devised to solve the aforementioned conventional problems. means of solving the problem
[0008] The above objective is achieved, according to the present invention, by a sensing unit that divides the spraying space of a water feature into a plurality of grid zones and collects user information, water quality information, environmental information, and equipment status information for each grid zone to generate sensing information; an analysis unit that calculates a sterilization priority and a sterilized water supply flow rate based on the sensing information; and a control unit that sterilizes circulating water based on the sensing information to generate the sterilized water and supplies the sterilized water to the grid zones according to the sterilization priority and the sterilized water supply flow rate. This is achieved by a smart water quality environment control system utilizing IoT and artificial intelligence, which includes an output unit that visually displays the sterilization status by grid zone, the sterilization water supply flow rate, user information, and warning notifications; a sensing unit that calculates the water quality pollution level based on the personnel density of the grid zone collected through a personnel detection sensor and water quality indicators collected from a water quality sensor; an analysis unit that calculates the sterilization priority and the sterilization water supply flow rate according to the water quality pollution level; and a control unit that includes a first sterilization device that generates the sterilization water using ozone treatment (AOP), a second sterilization device that generates the sterilization water using chlorine sterilization (CL), and a third sterilization device that generates the sterilization water using ultraviolet (UV).
[0009] In addition, the analysis unit is characterized by simultaneously activating the first sterilization device and the second sterilization device when the personnel density is above a preset threshold, and activating the third sterilization device alone when the grid area is unmanned.
[0010] Additionally, the analysis unit includes a switching module that predicts the possibility of failure of the first sterilization device based on the detection information, and if the first sterilization device is determined to be faulty, activates the second sterilization device to generate the sterilized water.
[0011] In addition, the analysis unit comprises: an overheating prediction module that predicts the possibility of overheating of the control unit based on the detection information and performs cooling response logic according to the possibility of overheating; a weather module that predicts the possibility of rainfall based on the detection information and determines whether the control unit operates; a response module that adjusts the sterilization intensity of the control unit to be temporarily strengthened upon restarting the control unit; a digital twin module that applies the detection information to a digital twin model to simulate water quality changes and sterilization effects and determines whether the control unit operates; a sensitivity adjustment module that calculates an environmental stress index, which is an index of the environmental conditions of the grid area, based on the detection information, and adjusts the sterilization priority and the sterilized water supply flow rate; an infant detection module that analyzes the user's body dimensions, body contour characteristics, gait characteristics, and motion rhythm based on the detection information to determine whether the user is an infant or a child; an animal detection module that analyzes quadrupedal walking characteristics and the body shape of small organisms based on the detection information to determine whether a pet has entered; a warning module that learns the manager's response pattern to the warning notification and adjusts the priority and frequency of the warning notification; and based on the detection information, the It includes an optimization module that optimizes the operation of the control unit.
[0012] In addition, the infant detection module is characterized by transmitting to the output unit whether to change the operation of the control unit when the ratio of infants and children is above a preset threshold. Effects of the invention
[0013] According to the present invention, by subdividing the spraying space into grid zones and combining and analyzing state data for each zone, it is possible to move away from uniform operation and enable customized control for each zone. This allows for rapid response to zones with concentrated contamination while stably maintaining overall water quality and ensuring hygienic safety.
[0014] In addition, chemical and power consumption can be reduced by selectively allocating resources to necessary areas and adjusting control intensity according to the situation, and operational continuity and reliability can be enhanced by limiting the scope of impact to zone units, even in the event of weather changes, equipment degradation, or a surge in users.
[0015] In addition, by monitoring the status and response of each zone through a consistent system and providing a basis for judgment to improve management efficiency, it has the effect of continuously providing a safe and pleasant environment for users and enhancing the operational stability, resource efficiency, and quality of public services of the facilities.
[0016] Meanwhile, the effects of the present invention are not limited to those mentioned above, and various effects may be included within the scope obvious to a person skilled in the art from the contents described below. Brief explanation of the drawing
[0017] FIG. 1 illustrates a smart water quality environment control system utilizing IoT and artificial intelligence according to an embodiment of the present invention. FIG. 2 illustrates the connections between the components of a smart water quality environment control system utilizing IoT and artificial intelligence according to an embodiment of the present invention, and FIG. 3 illustrates the arrangement of branch pipes and electronic valves by grid zone of a smart water quality environment control system utilizing IoT and artificial intelligence according to an embodiment of the present invention. Specific details for implementing the invention
[0018] Hereinafter, some embodiments of the present invention will be described in detail with reference to the exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings.
[0019] In addition, when describing embodiments of the present invention, if it is determined that a detailed description of related known configurations or functions would hinder understanding of the embodiments of the present invention, such detailed description is omitted.
[0020] In addition, terms such as first, second, A, B, (a), (b), etc., may be used when describing the components of the embodiments of the present invention. These terms are used merely to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the terms used.
[0021] Additionally, in the specification, the singular form includes the plural form unless specifically stated otherwise in the text. The terms “comprising” and / or “comprising” as used in the specification do not exclude the presence or addition of one or more other components in addition to the mentioned components.
[0022] Additionally, in describing embodiments of the present invention, each "part," "module," or "step" may be implemented through a processor and memory. The term "processor" should be broadly interpreted to include general-purpose processors, central processing units (CPUs), microprocessors, digital signal processors (DSPs), controllers, microcontrollers, state machines, and the like. In some contexts, the term "processor" may refer to an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), and the like. The term "processor" may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a DSP core, or any other combination of such configurations.
[0023] Furthermore, memory should be broadly interpreted to include any electronic component capable of storing electronic information. Memory may also refer to various types of processor-readable media such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), eraseable-programmable read-only memory (EPROM), electrically eraseable PROM (EEPROM), flash memory, magnetic or optical information storage devices, registers, etc. If a processor can read information from memory or write information to memory, memory is said to be in an electronic communication state with the processor, and each "part," "module," or "stage" may be implemented through a program or application based on the processor and memory in an electronic communication state.
[0024] Furthermore, in the present invention, Artificial Intelligence (AI) refers to a technology that imitates human learning ability, reasoning ability, and perceptual ability, and implements them on a computer, and may include concepts such as machine learning and symbolic logic. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input information on its own. AI technology can analyze input information as a machine learning algorithm, learn from the results of that analysis, and make judgments or predictions based on the results of that learning. Additionally, technologies that mimic the functions of the human brain, such as cognition and judgment, by utilizing machine learning algorithms can also be understood as falling within the category of AI. For example, technological fields such as linguistic understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control may be included.
[0025] Artificial intelligence learning models or neural network models can be designed to implement the structure of the human brain on a computer and may include multiple network nodes that simulate neurons of a human neural network and have weights. These multiple network nodes can have interconnected relationships by simulating the synaptic activity of neurons, where neurons exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes can be located in layers of different depths and exchange information according to convolutional connections. An artificial intelligence learning model may be, for example, an Artificial Neural Network (ANN) or a Convolutional Neural Network (CNN). An artificial intelligence learning model can be machine learned according to methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms that can be used to perform machine learning include Decision Trees, Bayesian Networks, Support Vector Machines, Perceptrons, and Clustering.
[0027] Now, with reference to the attached drawings, a smart water quality environment control system (1000) utilizing IoT and artificial intelligence according to one embodiment of the present invention will be described in detail.
[0028] As illustrated in FIG. 1, a smart water quality environment control system (1000) utilizing IoT and artificial intelligence according to one embodiment of the present invention includes a sensing unit (1100), an analysis unit (1200), a control unit (1300) and / or an output unit (1400).
[0029] First, the detection unit (1100) divides the spraying space of the water feature into a plurality of grid zones (S) and collects user information, water quality information, environmental information and / or equipment status information for each grid zone (S) to generate detection information, and is electrically connected to the analysis unit (1200), control unit (1300) and / or output unit (1400).
[0030] The detection unit (1100) can set grid zones (S) by dividing the entire spraying space into a square grid-shaped unit area, for example, a 0.6m × 0.6m area or a 1.0m × 1.0m unit area depending on the facility structure.
[0031] The unit of such grid zone (S) can be flexibly set considering the user’s dwell range and the spray radius of the spray nozzle, and a relatively large unit grid zone (S) can be applied in small water features, and a smaller unit grid zone (S) can be applied in large water features to perform high-resolution detection.
[0032] Here, the grid area (S) is not limited to the area where the actual sensor is physically installed, but can be composed of a logical sensing unit defined according to the spatial interpolation of data collected from adjacent sensors or the result of an artificial intelligence-based estimation.
[0033] Therefore, the grid area (S) may not correspond one-to-one with the arrangement structure of the sensors and can function as a data unit for analyzing the spatial distribution of water quality and environmental data.
[0034] The boundaries of the grid area (S) can be set in an orthogonal coordinate system based on the floor area, and each grid area (S) can be assigned a unique identification number (Grid ID).
[0035] The detection unit (1100) can independently map and manage user information, water quality information, environmental information, and equipment status information collected in each grid zone (S) based on the identification number.
[0036] Therefore, even within the same water feature, it is possible to distinguish between areas with high population density and areas with relatively low population density, and to accurately identify situations where water pollution levels rise intensively only in specific areas.
[0037] In addition, the grid zones (S) are not simply divided in a planar manner, but can also be divided by reflecting three-dimensional elements such as water depth, spray pressure, and bottom slope as needed. For example, in the case of a floor fountain, even if the grid zones (S) are on the same plane, they can be divided into separate zones and managed if the spray pressure or flow rate conditions are different. Through this, the detection unit (1100) can perform precise area-specific detection that considers not only spatial location but also fluid dynamic characteristics.
[0038] Additionally, as shown in FIG. 3, a branch pipe (B) and an electronic valve (V) for supplying sterilized water may be installed embedded in each grid zone (S). The branch pipe (B) branched from the main pipe (M) may be connected to correspond to each grid zone (S), and an electronic valve (V) may be individually provided at the end of the branch pipe (B).
[0039] These electronic valves (V) open and close upon receiving a signal from the control unit (1300) based on the analysis results of the analysis unit (1200) described later, and can selectively spray sterilized water only in the corresponding grid area (S). Conversely, electronic valves (V) in grid areas (S) where sterilization is determined not to be necessary remain closed to block the supply of unnecessary sterilized water.
[0040] Through this branching structure for grid zones (S), unlike the conventional method of uniformly supplying sterilized water to the entire water feature, sterilized water can be intensively sprayed only in specific areas with a high possibility of contamination, thereby preventing resource waste and maximizing the efficiency of the sterilization effect.
[0041] In addition, the detection unit (1100) can generate detection information by collecting user information, water quality information, environmental information and / or equipment status information for each grid area (S).
[0042] At this time, each piece of information can be defined as a data item belonging to a specific category and can be combined with each other to be structured in a form that can be utilized by the analysis unit (1200) and the control unit (1300).
[0043] User information refers to basic data indicating the presence, location, movement, and / or stay patterns of a user located within a grid area (S).
[0044] User information may consist of values directly measured by the sensor or calculated values using the time difference or spatial difference of the measured values.
[0045] The detection unit (1100) can track the location coordinates of the user on a frame-by-frame basis through a LiDAR sensor, a radar sensor, a thermal imaging camera and / or an image recognition sensor, and each coordinate can be converted into a grid coordinate system and mapped to an identification number of the corresponding grid area (S).
[0046] Additionally, the detection unit (1100) can maintain a tracking ID to continuously identify the same object through sensor fusion and perform time synchronization and interpolation to correct for differences in sampling cycles per sensor.
[0047] Specifically, user information may include user dwell time, user movement patterns, population density, body dimensions, body contour characteristics, gait characteristics, movement rhythm, quadrupedal gait characteristics, and / or surface temperature distribution.
[0048] Each data item of user information may be collected by one or more of a LiDAR sensor, a radar sensor, a thermal imaging camera and / or an image recognition sensor, and may be stored as a record including an identification number and a timestamp of a grid area (S) to be utilized in a subsequent processing step.
[0049] The user movement pattern refers to an indicator that combines trajectory characteristics and speed changes that occur as the same object passes through a specific grid area (S) or moves within the grid area (S). The detection unit (1100) collects continuous location coordinates and direction change points through a person detection sensor and can calculate path length, average speed, stopping section ratio, direction change frequency and / or revisit frequency. The user movement pattern is a basic indicator for quantifying the degree of contact and disturbance of the grid area (S), and can be reflected as a weighting factor that increases the likelihood of contamination contribution to the grid area (S) as the revisit frequency and stopping section ratio increase.
[0050] Personnel density refers to the number of people staying simultaneously relative to a unit area of a grid zone (S). The detection unit (1100) can calculate personnel density by counting the number of objects at each point in time through a personnel detection sensor, and can apply a time moving average and outlier removal to reduce errors caused by sudden surges and drops in the number of users. Personnel density is a basic indicator that quantifies the degree of congestion persistence for each grid zone (S), and can be reflected as a weighting factor that increases the possibility of contamination contribution in the corresponding grid zone (S) as the value is higher and the duration is longer.
[0051] Body dimensions refer to numerical data of height and width calculated from the contour of an object within a grid area (S). The detection unit (1100) can collect the silhouette and contour point cloud of the object through a shape detection sensor and calculate the body dimensions by calculating the maximum height relative to the floor and the horizontal projection contour width. The measured values can be mapped to the identification number and timestamp of the grid area (S) after smoothing and outlier exclusion and stored as time-series data. The body dimensions can be provided as input values for determining whether the object is an infant to the analysis unit (1200) described later.
[0052] Body shape contour characteristics refer to shape characteristic values that quantify the silhouette and three-dimensional distribution of an object within a grid area (S). The sensing unit (1100) can collect the silhouette and contour point cloud of an object through a shape sensing sensor and calculate the body shape contour characteristics by calculating the perimeter, aspect ratio, volume approximation, curvature distribution, and / or centerline asymmetry of the silhouette. Each calculated characteristic value can be stored connected to an identification number and timestamp of the grid area (S).
[0053] Walking characteristics refer to temporal and spatial features expressed as the walking cycle, stride length approximation, speed variability, and / or path curvature of an object existing within a grid area (S). The detection unit (1100) can collect the continuous trajectory of the object through a motion detection sensor, estimate the walking cycle and stride length, and calculate the variability of speed and acceleration. Walking characteristics can be stored as a basic indicator for quantifying the degree of walking stability in a congested area.
[0054] The motion rhythm refers to an indicator that combines the temporal pattern of repeated stopping, acceleration, and deceleration of an object existing within a grid zone (S), as well as the variability of speed and direction. The detection unit (1100) can record sudden stopping events and changes in direction of the object through a motion detection sensor and calculate the motion rhythm by analyzing the speed spectrum on the time axis. The motion rhythm can be stored as a basic indicator for distinguishing between stay-oriented activities and pass-oriented activities, and can be used as a basis for assigning a higher risk weight to grid zones (S) with a high stopping ratio of the motion rhythm through the analysis unit (1200).
[0055] Quadrupedal locomotion characteristics refer to the unique movement characteristics of an animal, which are manifested by a low contour height, a height-to-length ratio, and ground contact periodicity. The detection unit (1100) can record the animal's quadrupedal locomotion characteristics by extracting the low contour and length and height ratios through a shape detection sensor, and acquiring a short stride and high-frequency gait cycle through a motion detection sensor. The quadrupedal locomotion characteristics can be used as raw data to determine the possibility of a pet entering a grid area (S).
[0056] The surface temperature distribution refers to the thermal signal intensity and spatial distribution of an object. The detection unit (1100) can record the surface temperature distribution by acquiring the temperature distribution of the object's center and extremities, the temperature difference relative to the background, and temporal stability through a thermal detection sensor. The surface temperature distribution can be used as a cross-validation indicator to reinforce other signals in environments where visibility is limited, such as at night or during rain.
[0057] Meanwhile, the detection unit (1100) can calculate the water pollution level based on the density of people in the grid area (S) calculated through the people detection sensor and the water quality indicator of the grid area (S) collected from the water quality sensor.
[0058] Here, the water quality indicator may include not only the measured values from water quality sensors directly installed in each grid zone (S), but also values estimated through sensor data from adjacent zones and an artificial intelligence-based spatial interpolation algorithm.
[0059] People density refers to an indicator representing the degree of concentration of people per unit area within a grid zone (S). This is a quantified value of how many users are simultaneously present in a specific zone, implying that the higher the number of people or the longer such conditions persist, the greater the potential for the influx of contaminants. Therefore, people density (D) can be utilized as a key indicator reflecting the potential for contamination caused by user activities.
[0060] In addition, water quality indicators refer to metrics used to quantitatively evaluate the contamination status of water based on items such as pH, turbidity (NTU), residual chlorine, oxidation-reduction potential (ORP), dissolved oxygen (DO), and / or water temperature, which are measured by water quality sensors installed on-site. For each item, threshold values are set according to statutory standards or internal management standards, and a risk score is calculated based on the degree to which the measured value deviates from the standard range (deviation). In this case, the water quality indicator can be expressed as a range between 0 and 100 by aggregating the deviation scores of individual items, with a higher value indicating a state of greater contamination risk.
[0061] The detection unit (1100) counts the number of people per unit area within the grid zone (S) and can reflect the persistence of density by applying the average number of people over a certain time interval. If the number of people in the grid zone (S) is maintained above a preset threshold for a certain period of time or longer, the detection unit (1100) determines that the zone is a zone with a high probability of causing pollution and can assign a high weight to the number of people when calculating the water quality pollution level.
[0062] Additionally, the detection unit (1100) can calculate water quality indicators from the measurement values for each item obtained through the water quality sensor. For example, in the case of turbidity, 4 NTU or less is evaluated as 0 points, and in the range of 4 to 6 NTU, it is linearly converted from 0 to 100 points according to the deviation, and 6 NTU or more can be calculated as 100 points. In the case of pH, the range of 5.8 to 8.6 is considered as 0 points, and deviations outside this range are linearly converted from 0 to 100 points up to ±0.5, and if it deviates by ±0.5 or more, it can be evaluated as 100 points.
[0063] The detection unit (1100) can calculate the water pollution level (C) by weighting the population density and water quality indicators according to their respective pollution contributions. The formula for calculating the water pollution level can be expressed as Equation 1 below.
[0064]
[0065] Here, C represents the water quality pollution level of the grid zone (S) and can be expressed in a range from 0 to 100. D represents the population density and is a value calculated based on the concentration of people per unit area within the grid zone (S). represents the maximum value of the population density and can be set according to the facility's design capacity or operational data. Q represents a water quality indicator and is a risk score calculated based on items such as pH, turbidity, residual chlorine, oxidation-reduction potential (ORP), dissolved oxygen (DO), and / or water temperature measured by a water quality sensor. represents a weight to reflect the pollution contribution of population density (D), and represents a weight to reflect the pollution contribution of the water quality indicator (Q).
[0066] In mathematical formula 1 represents the normalization standard and can be set as a weighted sum obtained by multiplying the maximum values of each indicator (D, Q) by their respective weights and adding them together. This refers to a normalization process designed to enable comparison within the same scale (range 0–100) even if population density (D) and water quality indicators (Q) have different units and ranges. In other words, corresponding to the denominator in Equation 1 The two input indicators each have their maximum value ( By normalizing the entire expression based on the weighted sum when it reaches , 100, the resulting water quality pollution level can be consistently maintained within the range of 0 to 100.
[0067] These normalization criteria serve to balance the relative influence between personnel density and water quality indicators in actual operating environments, and can determine which indicator has a greater impact on the result based on the ratio of each term.
[0068] For example, in an environment where the population density (D) has a relatively small numerical range Because it is small, the influence of the water quality indicator (Q) ) becomes relatively larger. Conversely, in cases where personnel density acts as a major contamination factor due to the characteristics of the facility The value of can be set high so that the contribution of D becomes relatively larger. Accordingly, the detection unit (1100) can perform a dynamic correction function according to actual environmental conditions.
[0069] Meanwhile, the constant in Equation 1 expressed as 100 represents the maximum risk score of the water quality indicator (Q), signifying a state where deviations from water quality items, such as very high turbidity and pH outside the standard, completely exceed the management standard. By setting the upper limit of the water quality indicator (Q) to 100, all items of the water quality indicator (Q) can be expressed on the same percentage scale, thereby allowing the influence of the water quality indicator (Q) to be consistently reflected in the normalization standard calculation formula.
[0070] The detection unit (1100) can manage the calculated water quality pollution score by normalizing it to a range between 0 and 100, and can automatically adjust the sterilization intensity by transmitting a control signal to the control unit of the sterilization device of the water facility according to the score. For example, if the water quality pollution score is 70 or higher and 100 or lower, it can be classified as a high-risk range and the output of the sterilization device can be increased by 30 to 40%, and if it is 40 or higher and less than 70, it can be classified as a medium-risk range and the output can be increased by 10 to 20%. If the water quality pollution score is less than 40, it can be classified as a low-risk range and the sterilization device can be maintained in a normal operation mode.
[0071] In addition, the detection unit (1100) can transmit the calculated water quality pollution level to the analysis unit (1200) to calculate the sterilization priority and the sterilization water supply flow rate described later, and accordingly, through precise pollution prediction, it is possible to perform efficient water quality management that reduces unnecessary sterilization and minimizes resource waste.
[0072] Next, water quality information refers to basic data that reflects the water quality status within a grid area (S) in real time. This water quality information is associated with the identification number (Grid ID) of the grid area (S), and can be managed in a time-series format by assigning a timestamp.
[0073] Water quality information can consist of values directly measured by sensors or calculated values using the time and spatial differences of the measured values.
[0074] The detection unit (1100) can collect water quality information through a pH sensor, a turbidity sensor, a residual chlorine sensor, a total organic carbon (TOC) sensor, a microorganism detection sensor and / or a temperature sensor.
[0075] Additionally, the detection unit (1100) can maintain a tracking ID to continuously identify changes in water quality in the same area through sensor fusion, and can perform time synchronization and interpolation to correct for differences in sampling cycles per sensor.
[0076] Specifically, water quality information may include pH values, turbidity levels, residual chlorine concentration, total organic carbon (TOC) concentration, microbial concentration, water temperature, and / or dissolved oxygen (DO) levels.
[0077] Each data item of water quality information can be collected by one or more of a pH sensor, a turbidity sensor, a residual chlorine sensor, a total organic carbon (TOC) sensor, a microorganism detection sensor, and / or a temperature sensor, and is stored as a record including an identification number and a timestamp of a grid zone (S) to be utilized in a subsequent processing step.
[0078] The pH value is an indicator representing the degree of acidity or alkalinity of the water quality within a grid zone (S). The detection unit (1100) measures the hydrogen ion concentration of a water quality sample in real time through a pH sensor and can calculate the average value, the range of variation, and / or the frequency of outlier occurrences. The pH value is a basic indicator for quantifying the stability of the water quality, and can be reflected as a weighting factor that increases the likelihood of pollution contribution in the corresponding grid zone (S) as the value deviates significantly from neutral (7.0).
[0079] The turbidity level refers to an indicator representing the turbidity and suspended solids concentration of the water quality within a grid zone (S). The detection unit (1100) can calculate the turbidity level by measuring optical transmittance through a turbidity sensor and converting it into Nephelometric Turbidity Units (NTU). The measured value can be mapped to the identification number and timestamp of the grid zone (S) after smoothing and outlier exclusion, and stored as time-series data. The turbidity level is a basic indicator for quantifying the degree of contamination spread caused by user activity, and the higher the value, the more it can be provided as an input value for calculating the sterilization priority in the analysis unit (1200).
[0080] Residual chlorine concentration refers to an indicator representing the remaining amount of disinfectant within a grid zone (S). The detection unit (1100) measures the chlorine concentration using an electrochemical or optical method through a residual chlorine sensor and can calculate the residual chlorine concentration in ppm (Parts Per Million) units. The calculated value can be stored linked to the identification number and timestamp of the grid zone (S). Residual chlorine concentration is a basic indicator for evaluating the sustainability of the existing disinfection effect, and a lower value can be used as grounds for requiring additional disinfection in the corresponding grid zone (S).
[0081] Total organic carbon (TOC) concentration refers to an indicator representing the total amount of organic pollutants within a grid zone (S). The detection unit (1100) can measure the TOC concentration through oxidation and detection processes using a total organic carbon (TOC) sensor and calculate the total organic carbon (TOC) concentration by converting it into mg / L units. The total organic carbon (TOC) concentration can be stored as a basic indicator for quantifying the concentration of pollutants entering the user, and can be used as a basis for assigning a higher risk weight to grid zones (S) with high TOC concentrations through the analysis unit (1200).
[0082] Microbial concentration refers to an indicator representing the density of bacteria, viruses, and other microorganisms within a grid zone (S). The detection unit (1100) can calculate a microbial activity indicator by performing ATP (Adenosine Triphosphate) detection or optical fluorescence analysis through a microorganism detection sensor. Microbial concentration is a basic indicator for quantifying the degree of biological risk of water pollution, and a higher value may indicate a situation requiring immediate sterilization measures.
[0083] Water temperature refers to an indicator representing the temperature of the water quality within a grid zone (S). The detection unit (1100) measures real-time temperature values through a temperature sensor and can calculate the average value, range of variation, and / or frequency of outlier occurrences. Water temperature is a basic indicator that affects microbial growth and chemical reaction rates, and if the value is high, it can be reflected as a weight that increases the likelihood of pollution contribution in the corresponding grid zone (S).
[0084] The dissolved oxygen (DO) level refers to an indicator representing the amount of oxygen dissolved in the water within the grid zone (S). The detection unit (1100) can measure the DO level using an electrochemical or optical method through a dissolved oxygen sensor and calculate the dissolved oxygen (DO) level by converting it into mg / L units. The dissolved oxygen (DO) level is a basic indicator for evaluating the oxygen saturation and ecological stability of the water quality, and the lower the value, the more it can be provided as an input value for calculating the sterilization priority in the analysis unit (1200).
[0085] Next, environmental information refers to basic data that reflects environmental conditions within a grid area (S) in real time. This environmental information is associated with the identification number (Grid ID) of the grid area (S), and can be managed in a time-series format by assigning a timestamp.
[0086] Environmental information may consist of values directly measured by sensors or calculated values using the time and spatial differences of the measured values.
[0087] The sensing unit (1100) can track environmental conditions on a frame-by-frame basis through a temperature sensor, humidity sensor, illuminance sensor, wind sensor, rainfall sensor, air quality sensor, ultraviolet (UV) sensor and / or solar sensor, and each measurement value can be converted into a grid coordinate system and mapped to an identification number of the corresponding grid area (S).
[0088] Additionally, the detection unit (1100) can maintain a tracking ID to continuously identify environmental changes in the same area through sensor fusion, and can perform time synchronization and interpolation to correct for differences in sampling cycles per sensor.
[0089] Specifically, environmental information may include temperature, humidity, illuminance, wind speed, rainfall, air pollution levels (PM2.5, PM10), ultraviolet (UV) index, and / or solar heat.
[0090] Each data item of environmental information may be collected by one or more of a temperature sensor, humidity sensor, illuminance sensor, wind sensor, rainfall sensor, air quality sensor, ultraviolet (UV) sensor and / or solar sensor, and may be stored as a record including an identification number and timestamp of a grid zone (S) to be utilized in a subsequent processing step.
[0091] Temperature refers to an indicator representing the temperature of the atmosphere or floor surface within a grid zone (S). The sensing unit (1100) measures real-time temperature values through a temperature sensor and can calculate the average value, range of variation, and / or frequency of outlier occurrences. Temperature is a basic indicator that affects the microbial growth rate and chemical reactions of water quality, and can be reflected as a weighting factor where a higher value increases the likelihood of pollution contribution in the corresponding grid zone (S), and can be used to evaluate the possibility of overheating of the control unit in the overheating prediction module.
[0092] Humidity refers to an indicator representing the relative humidity of the atmosphere within a grid zone (S). The sensing unit (1100) can measure relative humidity through a humidity sensor and calculate the humidity level by converting it into a % unit. The measured value can be mapped to the identification number and timestamp of the grid zone (S) after smoothing and outlier exclusion, and stored as time-series data. Humidity is a basic indicator that indirectly reflects the microbial survival environment and water pollution levels, and can be provided to the analysis unit (1200) as an input value for calculating the environmental stress index and adjusting the sterilization priority.
[0093] Illumination refers to an indicator representing the intensity of light within a grid area (S). The detection unit (1100) can measure illuminance in units of lux through an illuminance sensor and calculate the average value and variability. Illuminance is a basic indicator for evaluating the indirect correlation between user activity patterns and water pollution; for example, low illuminance may reflect the difficulty of detecting pollution during nighttime operation and may be used as a basis for additional sterilization measures to compensate for this by the analysis unit (1200).
[0094] Wind speed refers to an indicator representing the strength and direction of the wind within a grid area (S). The detection unit (1100) measures the wind speed in m / s units through a wind sensor and can calculate the frequency of direction changes and the average wind speed. Wind speed is a basic indicator for quantifying water quality disturbance and the degree of diffusion of pollutants; the higher the value, the greater the possibility of increased water turbidity, and can be used as an input value for calculating the environmental stress index and adjusting the sterilization water supply flow rate in the analysis unit (1200).
[0095] Rainfall refers to an indicator representing the amount of precipitation within a grid area (S). The detection unit (1100) measures rainfall in mm units through a rainfall sensor and can calculate the amount of rainfall per hour and the duration. Rainfall is a basic indicator that directly affects the inflow of external pollutants and changes in water quality; the higher the value, the more it can be used as a basis for the weather module to predict the possibility of rainfall and determine whether the control unit operates.
[0096] Air pollution levels (PM2.5, PM10) refer to an indicator representing the concentration of fine dust within a grid area (S). The detection unit (1100) measures PM2.5 and PM10 concentrations in μg / m³ units through an air quality sensor and can calculate air pollution levels by applying a time moving average and outlier removal. Air pollution levels are a basic indicator for quantifying the potential for water pollution caused by external environmental factors, and a higher value can be reflected as a weighting factor that increases the potential for pollution contribution when calculating the environmental stress index of the corresponding grid area (S).
[0097] The ultraviolet (UV) index refers to an indicator representing the intensity of ultraviolet rays within a grid zone (S). The detection unit (1100) can measure the UV index through a UV sensor and calculate the variability over time. The UV index is a basic indicator for evaluating the indirect correlation between natural sterilization effects and water pollution, and a low UV index can be utilized by the analysis unit (1200) as grounds to increase the need for artificial sterilization measures.
[0098] Solar heat quantity refers to an indicator representing the amount of solar radiation energy within a grid area (S). The sensing unit (1100) measures solar heat quantity in units of W / m² through a solar sensor and can calculate cumulative heat quantity, peak value and / or variability over time. Solar heat quantity is a basic indicator for predicting the possibility of overheating of the control unit, and the higher the value, the more it can be used as a basis for performing cooling response logic in the overheating prediction module or giving weight when calculating the environmental stress index.
[0099] Next, equipment status information refers to basic data that reflects the operating status of equipment within a grid area (S) in real time. This equipment status information is associated with the identification number (Grid ID) of the grid area (S), and can be managed in a time-series format by assigning a timestamp.
[0100] Equipment status information may consist of values directly measured by sensors or calculated values using the time and spatial differences of the measured values.
[0101] The detection unit (1100) can track the equipment status in frames through a temperature sensor, a pressure sensor, a vibration sensor, a current sensor, an operation time counter and / or a fault detection sensor, and each measurement value can be converted into a grid coordinate system and mapped to an identification number of the corresponding grid area (S).
[0102] Additionally, the detection unit (1100) can maintain a tracking ID to continuously identify changes in equipment in the same area through sensor fusion, and perform time synchronization and interpolation to correct for differences in sampling cycles per sensor.
[0103] Specifically, equipment status information may include equipment temperature, pressure level, vibration intensity, current consumption, accumulated operating time, fault code, and / or valve open / closed status.
[0104] Each data item of equipment status information can be collected by one or more of a temperature sensor, pressure sensor, vibration sensor, current sensor, operating time counter and / or fault detection sensor, and stored as a record including an identification number and timestamp of a grid zone (S) to be utilized in a subsequent processing step.
[0105] Equipment temperature refers to an indicator representing the internal temperature of a sterilization device or pump, etc., within a grid zone (S). The detection unit (1100) measures real-time temperature values through a temperature sensor and can calculate the average value, the range of variation, and / or the frequency of occurrence of outliers. Equipment temperature is a basic indicator for quantifying the risk of overheating and operational stability, and can be reflected as a weighting factor where the higher the value, the greater the probability of equipment failure in the corresponding grid zone (S), and can be used to evaluate the possibility of overheating of the control unit in an overheating prediction module.
[0106] The pressure level refers to an indicator representing the water pressure of pipes or nozzles within a grid zone (S). The sensing unit (1100) can measure pressure in units of bar or psi through a pressure sensor and calculate the average value and variability. The measured values can be mapped to the identification number and timestamp of the grid zone (S) after smoothing and outlier exclusion and stored as time-series data. The pressure level is a basic indicator reflecting the stability of the sterilization water supply, and abnormal values can be provided to the analysis unit (1200) as input values for fault prediction and sterilization priority adjustment.
[0107] Vibration intensity refers to an indicator representing the degree of vibration of a pump or valve within a grid zone (S). The detection unit (1100) can measure vibration intensity in g-force units through a vibration sensor and calculate frequency distribution and peak values. Vibration intensity is a basic indicator for evaluating mechanical wear and early signs of failure; a higher value can be reflected as a weighting factor that increases the likelihood of equipment failure in the corresponding grid zone (S), and can be used by the analysis unit (1200) as a basis for maintenance measures to supplement this.
[0108] Current consumption refers to an indicator representing the power usage of electrical equipment within a grid zone (S). The detection unit (1100) measures the current in units of A (amperes) through a current sensor and can calculate the consumption and variability per hour. Current consumption is a basic indicator for quantifying operational efficiency and failure risk, and if the value is abnormally high or low, it can be used as an input value for a failure prediction module in the analysis unit (1200). The measured value can be stored as time-series data by mapping it to the identification number and timestamp of the grid zone (S) after smoothing and outlier exclusion.
[0109] The accumulated operating time refers to an indicator representing the total operating time of equipment within a grid zone (S). The detection unit (1100) measures the accumulated time in hourly units through an operating time counter and can calculate daily / weekly averages and reset events. The accumulated operating time is a basic indicator for evaluating equipment lifespan and maintenance cycles; the higher the value, the more likely it is to be used as a basis for predicting the probability of failure through the analysis unit (1200) and adjusting the sterilization intensity. The calculated value can be stored linked to the identification number and timestamp of the grid zone (S).
[0110] A fault code refers to an indicator representing the error status of equipment within a grid zone (S). The detection unit (1100) can collect standard fault codes (e.g., E01, E02, etc.) through a fault detection sensor and calculate the frequency of occurrence and duration. The fault code serves as a basic indicator for immediate fault diagnosis and can be used as a basis for the analysis unit (1200) to activate a switching module as specific codes occur. The fault code is stored as a record containing the identification number and timestamp of the grid zone (S) and can be utilized in subsequent processing steps.
[0111] The valve open / closed status refers to an indicator representing whether the electronic valve (V) within the grid zone (S) is open or closed. The detection unit (1100) measures binary values (open / closed) and the number of switching operations through a status sensor, and can calculate the cumulative number of open / closed operations and abnormal switching operations. The valve open / closed status is a basic indicator reflecting the control stability of the sterilized water supply, and abnormal status can be provided as an input value to the optimization module in the analysis unit (1200). The measured value can be mapped to the identification number and timestamp of the grid zone (S) and stored as time-series data.
[0112] Finally, the detection unit (1100) can generate detection information by organizing collected user information, water quality information, environmental information and / or equipment status information into a single dataset through sensor fusion and time synchronization, and can provide structured data that can be used for analysis through the analysis unit (1200) described later.
[0113] Next, the analysis unit (1200) calculates the sterilization priority and the sterilization water supply flow rate based on the detection information received from the detection unit (1100), and is electrically connected to the detection unit (1100), the control unit (1300), and / or the output unit (1400).
[0114] The analysis unit (1200) can sort grid zones (S) according to the water quality pollution level received from the detection unit (1100) and calculate the sterilization priority by applying an AI-based priority calculation algorithm. This algorithm utilizes machine learning models such as random forests or neural networks to learn complex correlations between user information, water quality information, environmental information, and / or equipment status information included in the detection information, and can quantify the water quality pollution level for each grid zone (S).
[0115] In addition, by using the determined sterilization priority and water quality contamination level as inputs, an optimized sterilization water supply flow rate for each grid zone (S) can be calculated through an AI-based regression model such as XGBoost or deep learning. In this process, the analysis unit (1200) can compare past data with real-time data to increase the accuracy of flow rate distribution and minimize resource waste.
[0116] The sterilization priority is a ranking indicator representing the urgency of sterilization treatment in a grid zone (S). The analysis unit (1200) can assign a high rank to a grid zone (S) with a high level of contamination based on the magnitude of water contamination received from the detection unit (1100). For example, the priority can be calculated by dynamically adjusting weights such as population density, turbidity, and microbial concentration, and the reliability of the priority can be increased by identifying abnormal data patterns through an outlier detection algorithm.
[0117] The sterilization water supply flow rate refers to the distribution amount of sterilization water to be supplied to grid zones (S), and can be calculated using an AI-based reinforcement learning algorithm with sterilization priority and water quality contamination level as input values. The sterilization water supply flow rate can be designed to supply a larger flow rate to grid zones (S) with higher priority and higher contamination levels, and the flow rate adjustment can be dynamically optimized by reflecting real-time feedback.
[0118] The analysis unit (1200) transmits the calculated sterilization priority and sterilization water supply flow rate to the control unit (1300) to enable sterilization to be performed for each grid zone (S), and can provide visualized results to the manager through the output unit (1400).
[0119] Meanwhile, the analysis unit (1200) may include a plurality of modules to analyze detection information received from the detection unit (1100) to calculate the sterilization priority and sterilization water supply flow rate for each grid zone (S), and to perform additional judgments for operation decisions and safety protection.
[0120] In the present invention, a module is a conceptual configuration in which software functions within an analysis unit (1200) are logically partitioned according to purpose and processing stage. It is executed in a single AI-based processor and memory space and can operate sequentially or in parallel as needed. In actual implementation, some modules may be integrated or divided into detailed functions for execution, and data exchange between modules may be performed through a predefined interface. Even if the number, names, and boundaries of modules are changed according to performance requirements and operating policies, they do not deviate from the technical concept of the present invention.
[0121] Specifically, the analysis unit (1200) may include an overheating prediction module (1201), a weather module (1202), a response module (1203), a digital twin module (1204), a sensitivity adjustment module (1205), an infant detection module (1206), an animal detection module (1207), a warning module (1208) and / or an optimization module (1209).
[0122] The overheating prediction module (1201) can predict the possibility of overheating of the control unit (1300) described later based on detection information and perform cooling response logic according to the possibility of overheating.
[0123] The overheating prediction module (1201) can analyze solar heat collected from the sensing unit (1100), ambient temperature, and / or internal temperature data of the control unit (1300) using an artificial intelligence-based prediction model such as an LSTM or a support vector machine, and calculate the possibility of overheating based on time series data. At this time, if the calculated value exceeds a preset threshold, a cooling response logic consisting of driving a cooling fan, switching to a sterilization output Low mode, and pausing the equipment can be automatically executed.
[0124] The overheating prediction module (1201) learns past overheating data to increase prediction accuracy and can detect abnormal temperature rise in advance through outlier detection. The analysis unit (1200) can prevent equipment damage and enhance the reliability of continuous operation through the overheating prediction module (1201).
[0125] The weather module (1202) can predict the possibility of rainfall based on detection information and determine whether to operate the control unit (1300).
[0126] The weather module (1202) can analyze rainfall, wind speed, and / or humidity data of environmental information using a classification model such as a decision tree or deep learning, and probabilistically calculate the probability of rainfall. For example, if the probability of rainfall is 80% or higher or if wind speed fluctuations are rapid, the weather module (1202) can decide to temporarily suspend the operation of the control unit (1300), and recommend maintaining normal operation when rainfall is below a threshold and wind speed is stable.
[0127] The weather module (1202) learns the time series pattern of weather data to increase prediction accuracy and can reduce energy and chemical usage by preventing water tank dilution and over-injection of chemicals caused by rainwater inflow.
[0128] The corresponding module (1203) can be adjusted to temporarily increase the sterilization intensity of the control unit (1300) when the control unit (1300) is restarted.
[0129] The response module (1203) can evaluate the initial water quality state after restarting the control unit (1300) based on detection information and execute logic to strengthen the sterilization intensity in situations where there is a high possibility of contamination.
[0130] The response module (1203) can calculate the initial possibility of contamination after the restart of the control unit (1300) by analyzing detection information through an artificial intelligence-based decision model, such as reinforcement learning. If the possibility of contamination exceeds a preset threshold or if a restart event is confirmed, the response module (1203) can execute sterilization enhancement logic, such as increasing the chemical injection speed or increasing the output of ozone and ultraviolet lamps. Subsequently, when the water quality information reaches a stabilization standard, it switches to a normal operation mode.
[0131] The response module (1203) can dynamically adjust the optimal reinforcement time by learning past restart data and water quality response patterns, thereby ensuring both hygienic safety and operational efficiency.
[0132] The digital twin module (1204) applies detection information to the digital twin model to simulate water quality changes and sterilization effects, and accordingly determines whether the control unit (1300) operates.
[0133] The digital twin module (1204) can input water quality, environment, and equipment status data into a digital twin virtual model using an artificial intelligence-based simulation model such as a physical neural network, and can predict water quality changes and sterilization effects by combining various sterilization methods, chemical injection amounts, and operating times. For example, it can simulate the reduction rate of microbial concentration when the chemical injection amount is increased by 10% when turbidity is high in a specific grid area (S), or predict the stabilization time when the UV operating time is extended in a rainy environment.
[0134] The digital twin module (1204) learns the simulation results to derive optimal operating conditions and prevents unnecessary equipment operation. This reduces resource waste and increases the accuracy of water quality control.
[0135] The sensitivity control module (1205) can calculate an environmental stress index, which is an index of the environmental conditions of the grid zone (S), based on the detection information, and adjust the sterilization priority and the sterilization water supply flow rate.
[0136] The environmental stress index is an indicator that quantifies the potential for contamination and system load in a grid zone (S), and can be expressed as a numerical value from 0 to 100. For example, the environmental stress index may increase if there is high population density, high ambient temperature, strong solar heat, and a history of water pollution.
[0137] The sensitivity control module (1205) can calculate an environmental stress index by analyzing personnel density, ambient temperature, solar heat content and / or water pollution history through an artificial intelligence-based regression model such as LightGBM. If the environmental stress index exceeds a preset threshold, the sensitivity control module (1205) executes logic to increase the sterilization priority and increase the flow rate, and if the environmental stress index is low, it can prevent over-response by reducing the sterilization frequency.
[0138] The sensitivity adjustment module (1205) learns environmental change patterns and dynamically adjusts sensitivity, thereby improving operational stability and warning reliability.
[0139] The infant detection module (1206) can determine whether the user is an infant or a child by analyzing the user's body dimensions, body shape contour characteristics, walking characteristics, and movement rhythm based on the detection information.
[0140] Specifically, based on detection information collected through the LiDAR sensor, radar sensor, thermal imaging camera and / or image recognition sensor of the detection unit (1100), the characteristics of the user's body dimensions, body shape contour characteristics, gait characteristics, and motion rhythm can be analyzed to determine the characteristics of infants and children.
[0141] For example, a person may be classified as an infant or child if their maximum height from the floor in terms of body dimensions is less than 1.2 m, their movement speed is lower than the average speed of an adult (1.4 m / s), and their walking cycle is short and irregular. This data is mapped to the identification number (Grid ID) and timestamp of the grid zone (S) and stored in a time series format, and the infant detection module (1206) processes this in real time to calculate the number of infants and children in each grid zone (S) and the ratio relative to the total users.
[0142] In addition, the infant detection module (1206) can detect infants by automatically determining the age characteristics of the user through an artificial intelligence analysis algorithm based on the collected detection data.
[0143] The infant detection module (1206) can predict users who are likely to be infants or children by learning the correlation between physical indicators, such as body dimensions, walking speed, and / or walking rhythm, through an artificial intelligence analysis algorithm. The determination result is transmitted to the administrator dashboard in real time, allowing the administrator to verify whether the user is an actual infant through video data or on-site observation. Additionally, the administrator's verification result is fed back to the infant detection module (1206) and accumulated as training data for the artificial intelligence analysis algorithm, thereby enabling the infant detection module (1206) to continuously improve its own determination accuracy.
[0144] For example, if the infant detection module (1206) determines a specific user to be an infant but the administrator confirms that the user is an adult, that case is recorded as a false positive, and the algorithm can be automatically corrected to lower the detection sensitivity of that pattern during subsequent learning. Conversely, if the infant detection module (1206) fails to recognize a user confirmed by the administrator as an infant, it is learned as a false negative case, and the detection probability can be increased when a similar pattern recurs. Through this feedback learning structure, the system can reduce the false positive rate and secure highly reliable infant detection performance in the long term by learning on its own as operational data accumulates, while simultaneously performing administrator verification in the initial stages.
[0145] Additionally, the infant detection module (1206) can transmit to the output unit (1400) whether to change the operation of the control unit (1300) when the ratio of infants and children is above a preset threshold.
[0146] Specifically, the infant detection module (1206) can send a warning notification to the output unit (1400) suggesting a change in the operation of the control unit (1300) when the calculated infant and child ratio is above a preset threshold, and the output unit (1400) can display this in real time on an administrator dashboard to monitor the overall situation and, if necessary, convey to the administrator whether to switch to an infant-only sterilization mode.
[0147] The infant-only sterilization mode can be configured, for example, by reducing the drug injection concentration of the control unit (1300) by 10 to 20 percent or lowering the output intensity to minimize potential effects on the infant's sensitive skin and respiratory system. However, if the infant ratio is low or the mixing ratio with adults is balanced, the general sterilization mode can be maintained to prevent unnecessary output reduction.
[0148] Additionally, the infant detection module (1206) can continuously track the number and distribution of infants and children and provide visualized data on the administrator dashboard. For example, if the infant ratio in a specific grid zone (S) exceeds 50%, the zone can be highlighted and a warning notification can be generated to help the administrator immediately recognize the situation. The warning notification can be further enhanced when the infant ratio exceeds 70%, thereby prompting the administrator to select an immediate switch to an infant-only sterilization mode.
[0149] The analysis unit (1200) can accurately identify the presence of infants and children through the infant detection module (1206) and prevent safety issues that may arise when applying the same standards as for general adults. This ensures safe water quality management that takes into account the sensitivity of infants.
[0150] The animal detection module (1207) can determine whether a pet has entered by analyzing the quadrupedal walking characteristics and the body shape of a small creature based on the detection information.
[0151] The animal detection module (1207) analyzes the data collected from the detection unit (1100) and, when a low contour height, a height-to-length ratio, a short stride, and / or a high-frequency walking cycle are detected, recognizes and determines the characteristics of the animal.
[0152] For example, if the body dimensions of an object detected within the grid zone (S) are measured to be less than 0.5m above the floor, a quadrupedal walking pattern is detected, and the speed of movement is irregular and periodic, it can be classified as a pet. This data is mapped to the identification number (Grid ID) and timestamp of the grid zone (S) and stored in a time series format, and the animal detection module (1207) can process this in real time to determine whether a pet has entered.
[0153] The animal detection module (1207) can temporarily stop the operation of the sterilization device of the control unit (1300) or send a warning notification to the manager through the output unit (1400) when the entry of a pet is detected. For example, if quadrupedal locomotion characteristics are confirmed in a specific grid area (S), the animal detection module (1207) can execute logic to temporarily stop the operation of the control unit (1300) to prevent unintentional inflow of contaminants or potential harm to animals.
[0154] Additionally, the animal detection module (1207) can analyze the frequency and pattern of animal entry to predict the likelihood of an animal appearing in a specific grid zone (S), and based on this, adjust the frequency of warnings or the sensitivity of the sterilization device. For example, in a grid zone (S) where pets frequently enter during a specific time period, the priority of warning notifications can be increased or the temporary suspension conditions of the sterilization device can be strengthened.
[0155] The analysis unit (1200) can accurately identify the entry of pets or small animals through the animal detection module (1207) and prevent the risk of unintentional contamination or malfunction that may be overlooked in existing systems.
[0156] The warning module (1208) can learn the manager's response pattern to the warning notification and adjust the priority and frequency of the warning notification.
[0157] The warning module (1208) can store the administrator's response data in the cloud and dynamically adjust the intensity, frequency, and / or priority of warning notifications based on this. For example, if a warning notification is sent from the infant detection module (1206) because the infant ratio exceeds 70%, and the administrator shows a pattern of immediately switching to a sterilization mode dedicated to infants, the warning module (1208) increases the priority of the warning notification and increases the frequency of the notification. Conversely, if the pet entry warning notification from the animal detection module (1207) is frequently ignored, the warning module (1208) can adjust to reduce the frequency of the warning notification or store it only as cloud history.
[0158] The warning module (1208) learns the manager's reaction time, response method, and / or whether to ignore it through an AI-based behavioral analysis model such as a Bayesian network, and can dynamically adjust the warning intensity and frequency. For example, for a manager who reacts immediately to a warning about the infant ratio, the AI increases the priority, and for warnings about animal entry that are frequently ignored, it reduces the frequency. The warning module (1208) learns situational responsiveness by analyzing time-series data, and can increase effectiveness by preventing excessive warnings or omissions.
[0159] The optimization module (1209) can optimize the operation of the control unit (1300) based on detection information to minimize the use of chemicals and power consumption while maximizing the effect of water quality improvement.
[0160] The optimization module (1209) can comprehensively analyze the detection information collected from the detection unit (1100) and adjust the operation of the sterilization device of the control unit (1300) in real time.
[0161] Specifically, the optimization module (1209) can evaluate the amount of chemical input and the amount of equipment power consumed relative to the water quality improvement effect by utilizing an artificial intelligence-based reinforcement learning and / or regression model. In this process, standard efficiencies such as the reduction rate of microbial concentration per 1 mg / L of chemical and the turbidity improvement rate per 1 W of power are analyzed in real time, and if the efficiency is lower than the standard efficiencies, the AI can execute feedback loop control to dynamically adjust the chemical injection speed, the output cycle of the sterilization device, and the operating time.
[0162] For example, if a situation is detected in a specific grid zone (S) where the microbial concentration is low and the population density is low, the optimization module (1209) can determine that the amount of chemical injected by the control unit (1300) is below the standard efficiency through a reinforcement learning model that has learned past data and real-time detection information, and can minimize resource waste by reducing the injection speed by 10-20% or extending the cycle. Conversely, in a zone with high turbidity and a high environmental stress index, the water quality improvement effect can be maximized by executing a logic that temporarily increases the sterilization intensity by 20-30%. In this process, the optimization module (1209) can derive an optimal adjustment value by analyzing time-series data and develop a customized optimization strategy suitable for the characteristics of each grid zone (S) through long-term learning.
[0163] The analysis unit (1200) can maximize the efficiency of resource usage and prevent unnecessary waste of medicine and power by utilizing the learning and prediction capabilities of artificial intelligence through the optimization module (1209), thereby reducing maintenance costs and realizing eco-friendly operation.
[0164] Meanwhile, the analysis unit (1200) can simultaneously activate the first sterilization device (1301) and the second sterilization device (1302) of the control unit (1300) described later when the personnel density is above a preset threshold, and can activate the third sterilization device (1303) alone when the grid area (S) is unmanned.
[0165] Specifically, the analysis unit (1200) can dynamically determine the operation method of the sterilization device by analyzing the density of people in the grid area (S) based on user information collected from the detection unit (1100).
[0166] The analysis unit (1200) utilizes an artificial intelligence-based classification model to determine whether the population density exceeds a preset threshold. If the population density is high, the analysis unit (1200) considers the high possibility of contamination and transmits to the control unit (1300) to simultaneously activate the first sterilization device (1301) and the second sterilization device (1302), which will be described later. Through this, the high water quality contamination can be quickly resolved by combining the powerful sterilization effects of ozone and chlorine. On the other hand, if the grid area (S) is confirmed to be unmanned, the analysis unit (1200) may prioritize energy efficiency and instruct the third sterilization device (1303), which does not use chemicals, to operate independently. In this case, the analysis unit (1200) analyzes real-time user data to accurately determine the unmanned state and prevents unnecessary resource consumption.
[0167] Additionally, the analysis unit may further include a switching module (12010) that predicts the possibility of failure of the first sterilization device (1301) based on detection information, and if the first sterilization device (1301) is determined to be in failure, activates the second sterilization device (1302) to generate sterilized water.
[0168] The switching module (12010) can analyze the probability of failure of the first sterilization device (1301) in real time by utilizing an artificial intelligence-based prediction model. The switching module (12010) can probabilistically calculate the probability of failure through a time series analysis model using equipment status information collected from the detection unit (1100) as an input value. For example, if the current consumption of the first sterilization device (1301) increases abnormally or the internal temperature exceeds a preset threshold, the artificial intelligence can predict the probability of failure to be 80% or higher and instruct the control unit (1300) to switch to the second sterilization device (1302).
[0169] Through this, a dual sterilization structure utilizing combined AOP and CL equipment can be configured, and the Ministry of Environment's statutory standards can be met without any gaps in sterilization.
[0170] The switching module (12010) learns past failure data to improve prediction accuracy and detects abnormal equipment conditions in advance through an anomaly detection algorithm. In addition, it simulates failure scenarios in conjunction with the digital twin module (1204) and optimizes the chlorine injection concentration and operating cycle of the second sterilization device (1302) through the optimization module (1209) to maintain resource efficiency. Through this, the switching module (12010) can prevent interruptions in water quality management due to equipment failure and support stable and sustainable operation of the water feature facility.
[0171] Next, the control unit (1300) generates sterilized water by sterilizing circulating water based on detection information and supplies sterilized water to grid zones (S) according to the sterilization priority and the sterilized water supply flow rate, and is electrically connected to the detection unit (1100), analysis unit (1200) and / or output unit (1400).
[0172] Specifically, the control unit (1300) includes a first sterilization device (1301) that generates sterilized water using ozone treatment (AOP), a second sterilization device (1302) that generates sterilized water using chlorine sterilization (CL), and a third sterilization device (1303) that generates sterilized water using ultraviolet rays (UV).
[0173] The first sterilization device (1301) sterilizes circulating water through an Advanced Oxidation Process (AOP) and generates and injects ozone (O3) to powerfully decompose organic matter, microorganisms, and other contaminants. This device is effective in grid zones (S) where high turbidity and microorganism concentrations are detected and can provide rapid water quality improvement through strong oxidizing power. However, since power consumption is high and maintenance may be complex, the injection amount and operating time can be efficiently controlled by linking with an AI-based optimization module (1209).
[0174] The second sterilization device (1302) sterilizes circulating water using a chlorine (Cl) compound and can provide a continuous sterilization effect due to its high residual properties. It enables cost-effective and stable water quality management and can be automatically switched to an emergency response in the event of a failure of the first sterilization device (1301), or can be activated simultaneously with the first sterilization device (1301) in a grid zone (S) where the sterilization intensity needs to be increased.
[0175] The third sterilization device (1303) utilizes an ultraviolet (UV) lamp to inactivate microorganisms and enables sterilization without chemicals, making it suitable for grid zones (S) that are unmanned or have many infants or sensitive users. Although the third sterilization device (1303) is highly energy efficient, its effectiveness may decrease in environments with high turbidity, so it can predict optimal operating conditions by linking with the digital twin module (1204).
[0176] The control unit (1300) can precisely distribute sterilized water through a piping system branched out by grid zone (S), and can perform control suitable for the characteristics of each grid zone (S) by utilizing the AI-based analysis results of the analysis unit (1200).
[0177] In each grid zone (S), an independent electronic valve (V) and a pump may be installed, and the control unit (1300) can dynamically adjust them based on the sterilization priority and the sterilization water supply flow rate received from the analysis unit (1200). For example, if the analysis unit (1200) calculates that the water quality contamination level of a specific grid zone (S) is 80 or higher and the environmental stress index is 70 or higher, the control unit (1300) opens the electronic valve (V) of the zone to increase the ozone injection amount of the first sterilization device (1301) by 20% and extend the operating time of the ultraviolet lamp of the third sterilization device (1303) by 10 minutes. On the other hand, in grid zones (S) confirmed to have low population density or be unmanned, the pump output can be minimized so that only the third sterilization device (1303) operates independently to increase energy efficiency.
[0178] Additionally, the control unit (1300) can perform dynamic control through real-time feedback. When turbidity or microbial concentration increases rapidly in the detection unit (1100), the optimization module (1209) of the analysis unit (1200) immediately reflects this and requests the control unit (1300) to increase the flow rate. For example, if turbidity increases to 50 NTU or more in a specific grid area (S), the control unit (1300) can increase the pump speed of the area by 15% and adjust the ozone injection amount of the first sterilization device (1301) to 10 mg / L. Additionally, if the infant detection module (1206) detects an infant ratio of 70% or more, the control unit (1300) can reduce the drug concentration by 15% and use the ultraviolet output of the third sterilization device (1303) preferentially to minimize the impact on the infant's sensitive skin and respiratory system. When the entry of a pet is detected by the animal detection module (1207), the control unit (1300) temporarily stops all sterilization devices in the area and sends a warning to the manager through the output unit (1400) to prevent unintentional contamination or harm to the animal.
[0179] Accordingly, the control unit (1300) performs optimized sterilization water supply for each grid zone (S) in conjunction with the AI-based analysis of the analysis unit (1200), and can maintain hygiene standards by preferentially allocating resources to high-contamination zones.
[0180] Next, the output unit (1400) visually displays the sterilization status by grid zone (S), the sterilization water supply flow rate, user information and / or warning notifications, and is electrically connected to the detection unit (1100), the analysis unit (1200) and / or the control unit (1300).
[0181] The output unit (1400) can visualize data through a dashboard-type user interface (UI) and monitor the operating status of the water feature in real time.
[0182] The output unit (1400) can be implemented as a web-based or dedicated application and can provide compatibility so that an administrator can access it from various devices such as PCs, tablets, and smartphones. In addition, the output unit (1400) can intuitively display information separated by grid zones (S) and can support the administrator in making immediate situational awareness and decisions by utilizing color codes, graphs, map-based heatmaps and / or notification windows.
[0183] The output unit (1400) can visualize the real-time water quality status and the operating status of the sterilization device for each grid zone (S) through the sterilization status for each grid zone (S). For example, the output unit (1400) can display the water quality contamination level, turbidity, and microbial concentration for each grid zone (S) as a gauge or bar graph, and can visualize the operation status and output intensity of the first sterilization device (1301), the second sterilization device (1302), and / or the third sterilization device (1303) as icons or percentages. Additionally, if the water quality contamination level in a specific grid zone (S) rises to 80 or higher, that zone can be highlighted in red on a heatmap, and an administrator can click on it to view detailed data.
[0184] Additionally, the output unit (1400) can visualize the amount and flow of sterilized water supplied to each grid zone (S) in real time through the sterilized water supply flow rate. The output unit (1400) can represent the sterilized water supply flow rate as a line graph or a flow chart, and can display the open / closed status of the electronic valve (V) of each grid zone (S) and the pump output together. For example, if ozone sterilized water supplied from the first sterilization device (1301) is distributed to a specific grid zone (S) at 10 L / min and chlorine sterilized water supplied from the second sterilization device (1302) is distributed at 5 L / min, the output unit (1400) can display this as a pie chart and provide the flow rate change trend as a time series graph. Through this, the manager can evaluate the efficiency of resource distribution and issue an adjustment command immediately if an abnormal flow rate (e.g., exceeding the threshold by 20%) is detected.
[0185] Additionally, the output unit (1400) can visualize the population density, infant ratio, and / or animal entry for each grid zone (S) through user information. Based on data collected from the detection unit (1100), the infant detection module (1206), and the animal detection module (1207), the output unit (1400) can display the number of people in each grid zone (S) in real time, and generate a pop-up notification with a warning icon when the infant ratio is 70% or higher or when pet entry is detected. For example, if the population density in a specific grid zone (S) is confirmed to be 50 people / m² or higher, that zone is displayed in orange, and if the infant ratio is high, a suggestion to switch to “infant-only sterilization mode” is highlighted. When an animal enters, a “sterilization device stopped” status is displayed on the dashboard along with a red warning light so that the administrator can respond immediately.
[0186] Additionally, the output unit (1400) can visualize warning notifications according to priority by linking with the warning module (1208) of the analysis unit (1200). The dashboard can distinguish warning notifications by color according to priority and provide detailed information in the notification window. For example, if the overheating prediction module (1201) predicts that the probability of the first sterilization device (1301) overheating is 80% or higher, the dashboard can immediately display a red warning and suggest commands such as “operate cooling fan” or “reduce output.” If the warning module (1208) learns the manager’s reaction pattern and adjusts the warning frequency, the dashboard can minimize frequently ignored warnings and display urgent warnings fixed at the top.
[0187] The output unit (1400) can provide an interactive function that allows an administrator to input manual control commands through a dashboard. For example, the administrator can manually activate a sterilization device in a specific grid zone (S), adjust the flow rate, or check and dismiss warning notifications. These commands are immediately transmitted to the control unit (1300) to reflect the operation of the sterilization device. The dashboard includes an archive function that allows viewing past operation records based on time-series data, and supports operation analysis by visualizing water quality status, flow rate distribution, and warning history for a specific time period in graphs and tables.
[0188] Through this, the output unit (1400) intuitively conveys real-time operational status and potential risks to the manager of the water feature, and visualizes AI-based analysis results to enable fast and accurate decision-making. The output unit (1400) is designed to simultaneously satisfy hygiene safety, resource efficiency, and operational stability, and can maximize sustainable water quality management of the water feature and convenience for the manager.
[0189] As described above, according to the smart water quality environment control system (1000) utilizing IoT and artificial intelligence according to an embodiment of the present invention, by subdividing the spraying space into grid zones and combining and interpreting status data for each zone, it is possible to move away from uniform operation and enable customized control for each zone. This allows for rapid response to areas with concentrated pollution while stably maintaining overall water quality and ensuring hygienic safety.
[0190] In addition, chemical and power consumption can be reduced by selectively allocating resources to necessary areas and adjusting control intensity according to the situation, and operational continuity and reliability can be enhanced by limiting the scope of impact to zone units, even in the event of weather changes, equipment degradation, or a surge in users.
[0191] In addition, by monitoring the status and response of each zone through a consistent system and providing a basis for judgment to improve management efficiency, it has the effect of continuously providing a safe and pleasant environment for users and enhancing the operational stability, resource efficiency, and quality of public services of the facilities.
[0193] Although all components constituting the embodiments of the present invention have been described above as being combined or operating in combination, the present invention is not necessarily limited to such embodiments. That is, within the scope of the purpose of the present invention, all components may be selectively combined in one or more ways to operate.
[0194] Furthermore, terms such as "include," "compose," or "have" described above, unless specifically stated otherwise, mean that the relevant component may be inherent; therefore, they should be interpreted as allowing for the inclusion of additional components rather than excluding them. All terms, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains, unless otherwise defined. Commonly used terms, such as those defined in advance, should be interpreted in accordance with their meaning in the context of the relevant technology and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the present invention.
[0195] Furthermore, the above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential characteristics of the present invention.
[0196] Accordingly, the embodiments disclosed in this invention are intended to illustrate, not limit, the technical concept of the invention, and the scope of the technical concept of the invention is not limited by these embodiments. The scope of protection of this invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of this invention. Explanation of the symbols
[0197] 1000: Smart water quality environment control system utilizing IoT and artificial intelligence according to an embodiment of the present invention S: Grid area M: Main pipe B: Branch piping V: Electronic valve 1100: Detector 1200: Analysis Department 1201: Overheating Prediction Module 1202: Weather Module 1203: Response Module 1204: Digital Twin Module 1205: Sensitivity adjustment module 1206: Infant Detection Module 1207: Animal detection module 1208: Warning Module 1209: Optimization Module 1300: Control unit 1301: First sterilization device 1302: Second sterilization device 1303: Third sterilization device 1400: Output section
Claims
Claim 1 A sensing unit that divides the spraying space of a water feature into multiple grid zones and collects user information, water quality information, environmental information, and equipment status information for each grid zone to generate sensing information; A smart water quality environment control system utilizing IoT and artificial intelligence, comprising an analysis unit that calculates a sterilization priority based on the above-mentioned detection information, wherein the detection unit tracks the location coordinates of a user through a LiDAR sensor, a radar sensor, a thermal imaging camera, or an image recognition sensor, wherein the detection unit calculates the water quality pollution level based on the population density of the grid area collected through a population detection sensor and water quality indicators collected from a water quality sensor, and wherein the analysis unit calculates the sterilization priority according to the water quality pollution level, wherein the analysis unit calculates the sterilization water supply flow rate based on the above-mentioned detection information, and wherein the smart water quality environment control system utilizing IoT and artificial intelligence comprises a control unit that generates the sterilization water by sterilizing circulating water based on the above-mentioned detection information and supplies the sterilization water to the grid area according to the sterilization priority and the sterilization water supply flow rate, and an output unit that visually displays the sterilization status by grid area, the sterilization water supply flow rate, user information, and warning notifications, wherein the control unit includes a first sterilization device that generates the sterilization water using ozone treatment (AOP) and chlorine sterilization (CL) using the sterilization water A smart water quality environment control system utilizing IoT and artificial intelligence, comprising a second sterilization device that generates the sterilized water and a third sterilization device that generates the sterilized water using ultraviolet (UV) rays. Claim 2 delete Claim 3 A smart water quality environment control system utilizing IoT and artificial intelligence according to claim 1, wherein the analysis unit simultaneously activates the first sterilization device and the second sterilization device when the personnel density is above a preset threshold, and activates the third sterilization device alone when the grid area is unmanned. Claim 4 A smart water quality environment control system utilizing IoT and artificial intelligence according to claim 1, wherein the analysis unit predicts the possibility of failure of the first sterilization device based on the detection information, and if the first sterilization device is determined to be in failure, the system includes a switching module that activates the second sterilization device to generate the sterilized water. Claim 5 In claim 1, the analysis unit comprises: an overheating prediction module that predicts the possibility of overheating of the control unit based on the detection information and performs cooling response logic according to the possibility of overheating; a weather module that predicts the possibility of rainfall based on the detection information and determines whether the control unit operates; a response module that adjusts the sterilization intensity of the control unit to temporarily strengthen it upon restarting the control unit; a digital twin module that applies the detection information to a digital twin model to simulate water quality changes and sterilization effects and determines whether the control unit operates; a sensitivity adjustment module that calculates an environmental stress index, which is an index of the environmental conditions of the grid area, based on the detection information, and adjusts the sterilization priority and the sterilized water supply flow rate; an infant detection module that analyzes the user's body dimensions, body contour characteristics, gait characteristics, and motion rhythm based on the detection information to determine whether the user is an infant or a child; an animal detection module that analyzes quadrupedal walking characteristics and the body shape of small organisms based on the detection information to determine whether a pet has entered; a warning module that learns the manager's response pattern to the warning notification and adjusts the priority and frequency of the warning notification; and the detection A smart water quality environment control system utilizing IoT and artificial intelligence, comprising an optimization module that optimizes the operation of the control unit based on information.
Citation Information
Patent Citations
System of hydroponics culture with ECO-friendly
KR1020200128313A
Intelligent quarantine system and intelligent quarantine robot
KR1020230019356A
Method and system for monitoring water quality based on artificial intelligence
KR102504241B1
AI-based water quality prediction device and method using the same
KR1020240088343A
Water management system for fountain
KR102574102B1
