Method and system for integrated energy remote control of accommodation facilities for global carbon neutrality

Abstract

An integrated energy remote control system for accommodation facilities for global carbon neutrality may include a memory and a processor for storing instructions. When executed by the processor, the instructions allow the system to receive a detection signal from a motion sensor installed in a guest room of the accommodation facility to determine the occupancy status, including the presence of a customer in the guest room and their movement path; collect comprehensive energy usage data, including electricity, water, and gas consumed in the guest room, in real time; derive the customer's energy usage pattern through time-series data analysis; compare the derived energy usage pattern with a preset reference pattern; if abnormal energy consumption outside the predicted range is detected, transmit a warning notification to a manager terminal or control the energy supply to be cut off; and if the determined occupancy status changes to an absence state, control the system to automatically cut off or adjust the supply of at least one of the electricity, water, and gas through an energy control device connected to the guest room.

Description

Technology Field

[0001] The present invention relates to the field of energy management utilizing artificial intelligence (AI) and Internet of Things (IoT) technologies.

[0002] More specifically, the invention relates to an artificial intelligence (AI)-based Internet of Things (IoT) system capable of monitoring the status of individual rooms within an accommodation facility in real time and integrally remotely controlling the use of water, electricity, and gas energy to solve the problem of energy waste occurring in accommodation facilities and reduce carbon emissions, and a method of operating the same. Background Technology

[0003] Recently, the practice of carbon neutrality to address climate change has emerged as an important task worldwide. In particular, the tourism service sector is known to account for approximately 5–8% of global greenhouse gas emissions, and among them, accommodation facilities are identified as a carbon source accounting for 10% of the entire tourism sector.

[0004] Energy consumption in accommodation facilities faces the problem of unnecessary energy waste because, although rooms are rented to guests, not all rooms are actually in constant use, and the use of electrical appliances such as heating, cooling, and lighting is not immediately controlled even after check-out. Furthermore, there are difficulties in implementing customized energy-saving systems due to costs and a lack of control.

[0005] Most existing energy management systems were limited to monitoring the integrated energy usage of the entire accommodation facility, and lacked intelligent control functions to identify real-time occupancy status at the individual room level and dynamically cut off or adjust the energy supply accordingly. The problem to be solved

[0006] The present invention aims to provide an IoT-based system capable of detecting the occupancy status of individual rooms within an accommodation facility in real time, automatically cutting off energy that is unnecessarily consumed during periods when the room is not in use, and remotely performing integrated energy control for water, electricity, and gas.

[0007] In addition, the present invention aims to automatically collect accommodation facility occupancy rates, check-in, and check-out information by linking with an external reservation service platform, optimize an energy saving algorithm based on this information, communicate the collected data through a cloud platform, and provide an interface that is easy for users to understand.

[0008] The present invention aims to continuously improve the energy efficiency of the entire system and provide a warning alarm to the user when abnormal energy consumption occurs by utilizing an artificial intelligence (AI) model to analyze and predict energy usage patterns. means of solving the problem

[0009] An integrated energy remote control system for accommodation facilities for global carbon neutrality may include a memory and a processor for storing instructions. When executed by the processor, the instructions allow the system to receive a detection signal from a motion sensor installed in a guest room of the accommodation facility to determine the occupancy status, including the presence of a customer in the guest room and their movement path; collect comprehensive energy usage data, including electricity, water, and gas consumed in the guest room, in real time; derive the customer's energy usage pattern through time-series data analysis; compare the derived energy usage pattern with a preset reference pattern; if abnormal energy consumption outside the predicted range is detected, transmit a warning notification to a manager terminal or control the energy supply to be cut off; and if the determined occupancy status changes to an absence state, control the system to automatically cut off or adjust the supply of at least one of the electricity, water, and gas through an energy control device connected to the guest room. Effects of the invention

[0010] The integrated energy remote control method and system for accommodation facilities for global carbon neutrality according to the present invention fundamentally prevents unnecessary energy waste occurring in accommodation facilities through real-time occupancy status monitoring at the individual room level and an automatic energy cutoff function, thereby enabling the reduction of carbon emissions.

[0011] In addition, the present invention can improve operational efficiency and convenience by collecting data in real time without the need for accommodation facility operators to manually input check-in and check-out information through automatic integration with an external reservation service platform, and by optimizing the energy saving system based on this data.

[0012] The present invention can assist in energy management by detecting abnormal energy consumption early and providing an alarm to the user through an energy usage pattern analysis and prediction function utilizing an AI model.

[0013] The present invention establishes a stable and scalable energy management infrastructure through the configuration of an IoT system integrating a motion sensor, a circuit board, a microcontroller, and a cloud platform, and can provide universality and compatibility applicable to various accommodation facility environments. Brief explanation of the drawing

[0014] FIG. 1 is a drawing for explaining a system utilizing an artificial intelligence model according to one embodiment. FIG. 2 is a diagram illustrating the learning of a neural network according to one embodiment. FIG. 3 is a diagram illustrating the configuration of an artificial intelligence model according to one embodiment. FIG. 4 is a block diagram showing the configuration of an integrated energy remote control system for accommodation facilities for global carbon neutrality according to one embodiment. FIG. 5 is a flowchart illustrating an integrated energy remote control method for accommodation facilities according to one embodiment. FIG. 6 is a flowchart illustrating a motion sensor data preprocessing method according to one embodiment. FIG. 7 is a flowchart illustrating an integrated control method based on reservation platform linkage according to one embodiment. FIG. 8 illustrates the process of analyzing the energy usage pattern of an artificial intelligence model according to one embodiment. Specific details for implementing the invention

[0015] Various embodiments are described in detail below with reference to the attached drawings. However, these embodiments may be modified in various ways, and therefore the scope of the patent application is not limited or restricted by these embodiments. All modifications and equivalent substitutes should be considered to be included within the scope of the rights. The specific structural or functional descriptions for each embodiment are merely for illustrative purposes and may be modified in various forms. Accordingly, the embodiments are not limited to a specific form, and the scope of this specification includes variations of the technical concept, equivalents, or substitutes.

[0016] The terms "first" or "second" may be used to describe various components, but are intended merely to distinguish each component. For example, the first component may be named the second component, and conversely, the second component may be named the first component. When a component is referred to as being "connected" to another component, this means that it may be directly connected or that there may be another component in between. The terms used are for descriptive purposes only and are not intended to be limiting. Singular expressions include the plural unless otherwise interpreted in the context. In this specification, expressions such as "comprising" or "having" indicate the presence of the features or elements described in the specification and should be understood as not excluding the presence of other features or elements.

[0017] The embodiments can be implemented in various forms, such as personal computers, laptops, tablets, smartphones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices. An artificial intelligence (AI) system is a computer system that implements human-level intelligence; unlike existing rule-based smart systems, it is a system in which the machine learns and makes decisions on its own. As the use of AI systems increases, recognition rates improve and user preferences are understood more accurately, leading to the gradual replacement of existing systems with deep learning-based AI.

[0018] In this invention, Artificial Intelligence (AI) refers to a technology that implements human learning, reasoning, and perceptual abilities on a computer, and includes concepts such as machine learning and symbolic logic. Machine learning (ML) is an algorithmic technology that learns the characteristics of input data on its own. Through such machine learning algorithms, AI technology can analyze input data, learn from the results, and perform judgments or predictions based on them. Furthermore, technologies that mimic human cognitive and judgment functions using machine learning algorithms can also be included in the category of AI. For example, fields such as linguistic understanding, visual understanding, reasoning and prediction, knowledge representation, and motion control fall into this category. Machine learning refers to the process of training a neural network model through experience in processing data. Through this, computer software can improve its data processing capabilities on its own. A neural network model models the correlations between data and can be expressed by various parameters. The core of machine learning is to optimize the model's parameters by extracting and analyzing features from given data to derive relationships between the data and repeating this process. For example, a neural network model can learn the relationship between input and output, or learn relationships by deriving regularities solely from the input data.

[0019] Artificial intelligence learning models or neural network models are designed to implement the structure of the human brain on a computer and include multiple network nodes that mimic neurons. These nodes are interconnected to exchange signals and process data through layers of varying depths within the AI ​​learning model. Such models may include artificial neural networks and convolutional neural networks (CNNs). For example, AI learning models can be machine learned through methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms that may be used include decision trees, Bayesian networks, support vector machines, artificial neural networks, AdaBoost, perceptrons, genetic programming, and clustering.

[0021] FIG. 1 is a diagram illustrating a system utilizing an artificial intelligence model according to one embodiment.

[0022] As illustrated in FIG. 1, a system (100) utilizing an artificial intelligence model may include multiple user terminals (110-1, ���), a server (120), and a database (130). According to one embodiment, the database (130) is configured separately from the server (120), but it is not necessary to configure it this way, and the database (130) may be located inside the server (120). For example, the server (120) may include multiple artificial intelligences for performing machine learning algorithms. According to another embodiment, multiple user terminals (110-1, ���), a server (120), and a database (130) may be connected to communicate with each other through a network (N).

[0023] The network (N) can perform wireless or wired communication between multiple user terminals (110-1, ���), a server (120), and a database (130). For example, the network can support wireless communication in the form of LTE (long-term evolution), LTE-A (LTE Advanced), CDMA (code division multiple access), WCDMA (wideband CDMA), WiBro (Wireless BroadBand), WiFi (wireless fidelity), Bluetooth, NFC (near field communication), GPS (Global Positioning System), or GNSS (global navigation satellite system). Additionally, the network (N) can perform wired communication in the form of USB (universal serial bus), HDMI (high definition multimedia interface), RS-232 (recommended standard 232), or POTS (plain old telephone service).

[0024] The database (130) has the ability to store various data. The data stored in the database (130) is data acquired, processed, or used by at least one component of a multiple user terminal (110-1, ���) or server (120), and may include software (e.g., programs). The database (130) may include volatile and / or non-volatile memory.

[0026] FIG. 2 is a diagram illustrating the learning of a neural network according to one embodiment.

[0027] As shown in FIG. 2, the learning device can train a neural network (123) to process review responses received from multiple user terminals (110-1, ���) item by item. Additionally, the learning device can train a neural network (123) to extract user stay records based on user movement path information. According to one embodiment, the learning device may be a separate entity from the server (120), but this is not limited to this.

[0028] The neural network (123) includes an input layer (121) and an output layer (125), where training samples are input and training outputs are generated. The network is trained based on the difference between the training outputs and the labels. The labels are defined based on items related to review responses and may also be defined based on user dwell records according to movement path information. The neural network (123) is connected in groups of multiple nodes and is defined by weights and activation functions between the nodes.

[0029] The learning device can train a neural network (123) using the Gradient Descent (GD) technique or the Stochastic Gradient Descent (SGD) technique. The device can utilize a loss function designed by the outputs and labels of the neural network. The learning device can calculate the training error using a predefined loss function. The loss function can predefine the labels, outputs, and parameters as input variables, and these parameters can be set as weights within the neural network (123). For example, the loss function can be designed in the form of Mean Square Error (MSE) or entropy, and various techniques can be applied.

[0030] The learning device can find weights that affect the training error using the backpropagation technique. Here, weights represent the relationships between nodes within the neural network (123). The learning device can use the SGD technique by utilizing labels and outputs to optimize the weights found by the backpropagation technique. For example, the learning device can update the weights of a loss function defined based on labels, outputs, and weights using the SGD technique.

[0031] According to one embodiment, the learning device extracts a first object from a review response, obtains a first label corresponding to the object, and applies it to a first neural network to generate a first training output corresponding to the first object. Subsequently, the first neural network can be trained based on the first training output and the first label. The learning device extracts a second object from movement path information, obtains a second label which is a user dwell record corresponding to the object, and applies it to a second neural network to generate a second training output corresponding to the second object. Subsequently, the second neural network can be trained based on the second training output and the second label.

[0032] According to one embodiment, the learning device can generate a first training feature vector based on the constituent features, position features, and pattern features of a review response, and various methods may be used for feature extraction. Additionally, the learning device can generate a second training feature vector based on the constituent features, length features, and pattern features of movement path information, and various methods may be used to extract features.

[0033] According to one embodiment, the learning device can obtain a training output by applying a first training feature vector to a neural network (123). Subsequently, the learning device can train a review item extraction algorithm of the neural network (123) based on the training output and the first label.

[0035] FIG. 3 is a diagram illustrating the configuration of an artificial intelligence model according to one embodiment.

[0036] According to one embodiment, an artificial intelligence model may be composed of an input layer, a hidden layer, and an output layer.

[0037] The input layer is the layer associated with the values ​​input into the artificial intelligence model. In the hidden layer, feature maps can be generated by performing multiply-accumulate (MAC) operations and activation operations on the input values. The MAC operation is a process of multiplying the input values ​​by their corresponding weights and then summing these multiplied values. The activation operation is a process of inputting the results of the MAC operation into an activation function to output a final result, and various types of activation functions can be used. For example, activation functions may include, but are not limited to, the sigmoid function, tangent function, ReLU function, Leaky ReLU function, Max Out function, and ELU function.

[0038] A hidden layer may consist of at least one layer, for example, divided into a first hidden layer and a second hidden layer. In this case, the first hidden layer generates a feature map by performing MAC operations and activation operations based on the input values ​​of the input layer, and this feature map can be used as the input value of the second hidden layer. The second hidden layer can perform MAC operations and activation operations based on the feature map generated as a result of the first hidden layer.

[0039] The output layer may be a layer related to the result of an operation performed in the hidden layer.

[0040] In one embodiment, a learning model learns syllable (character) patterns that are frequently used together in a given corpus to automatically identify the boundaries between compound words and named entities, and integrates object information from a first UI source with object information rendered in a browser to generate a learning object information file. Using this learning object information file, data for training a deep learning network is generated, and based on data received from various domains, the data is standardized into an integrated format according to one or more standardization methods suitable for each domain. Subsequently, data of a specific domain is trained and inferred, information necessary for standardization in that domain is determined, and post-processing can be performed on the data received from various domains.

[0041] The first UI source includes an XML file, and the training object information file includes an input JSON file for feature learning and an output JSON file used as label data during training. This output JSON file is a file containing HTML DOM Tree information implemented in compliance with web standards. Various domains include at least one of a RAN (radio access network), a transport, or a core, and the post-processing process may include a correlation function.

[0043] FIG. 4 is a block diagram showing the configuration of an integrated energy remote control system for accommodation facilities for global carbon neutrality according to one embodiment.

[0044] A system (400) according to one embodiment may include a processor (420) and a memory (430), and some of the illustrated components may be omitted or substituted. A system (400) according to one embodiment may be a server or a terminal. According to one embodiment, the processor (420) is a component capable of performing operations or data processing regarding the control and / or communication of each component of the system (400), and may be composed of one or more processors. The memory (430) may store information related to the method described above or store a program in which the method described above is implemented. The memory (430) may be volatile memory or non-volatile memory. The memory (430) may store various file data, and the stored file data may be updated according to the operation of the processor (120).

[0045] According to one embodiment, the processor (420) can execute a program and control the device (400). The code of the program executed by the processor (120) can be stored in memory (430). Operations of the processor (420) can be performed by loading instructions stored in memory (430). The system (400) can be connected to an external device (e.g., a personal computer or a network) through an input / output device (not shown in the drawing) and exchange data.

[0046] A system (400) according to one embodiment includes a processor (420) and a memory (430). A system (400) according to one embodiment may be the server or terminal described above. The processor (420) may include at least one device described through FIGS. 1 to 3 or may perform at least one method described through FIGS. 1 to 3. The memory (430) may store information related to the method described above or store a program in which the method described above is implemented. The memory (430) may be volatile memory or non-volatile memory.

[0048] FIG. 5 is a flowchart illustrating an integrated energy remote control method for accommodation facilities according to one embodiment.

[0049] In operation 510, the system (400) can determine the occupancy status of a room by receiving a detection signal from a motion sensor under the control of a processor (e.g., processor (420) of FIG. 4). The system (400) can detect the movement of a customer in real time through a Passive Infrared (PIR) sensor or an ultrasonic sensor installed in the room, and can convert the detected signal into digital data and transmit it to a processor (e.g., processor (420) of FIG. 4). The system (400) can determine not only the presence of a user but also their movement path within the room by analyzing the intensity and frequency of the detection signal. The system (400) can perform periodic sensor data collection, for example, at intervals of one second, and can store the collected data in a memory (e.g., memory (430) of FIG. 4). The system (400) can determine the absence status if there is no detection signal from the sensor for a certain period of time or longer, and determine the occupancy status if a continuous detection signal occurs. The measurement cycle and judgment standard time are merely examples and are not limited thereto; they may vary depending on the settings or the accommodation's operational policies.

[0050] In operation 520, the system (400) can collect data on electricity, water, and gas usage in real time and derive energy usage patterns. The system (400) can store the derived data in a time-series database along with timestamps. The system (400) can collect energy usage at intervals of, for example, 30 seconds or 1 minute, and can analyze energy consumption patterns by time of day and day of the week based on the collected data. In the case of electricity usage, the system (400) can measure the individual consumption of heating and cooling devices, lighting, and home appliances separately, and in the case of water usage, it can monitor cold water and hot water separately. In the case of gas usage, the system (400) can determine the amount of gas consumed for heating and hot water supply in conjunction with the boiler operating time. The system (400) can derive pattern information including average usage, maximum usage, and usage frequency by statistically analyzing the accumulated time-series data. The collection cycle and measurement unit are merely examples and are not limited thereto, and may vary depending on the characteristics of the energy source or the specifications of the measurement equipment.

[0051] In operation 530, the system (400) can detect abnormal consumption by comparing the derived energy usage pattern with a reference pattern and send a notification or cut off the energy supply. The system (400) can store a pre-set normal range energy usage pattern as a reference pattern in memory, and can determine abnormal consumption if the real-time measured usage exceeds the allowable error range of the reference pattern. The system (400) can set the threshold for abnormal consumption as, for example, when usage exceeds 50% or more of the average usage, and this threshold can be adjusted by the user. When abnormal consumption is detected, the system (400) can send a warning message to an administrator terminal via SMS, push notification, or email. The warning message may include the room number, the type of energy source where abnormal consumption occurred, the excess usage, and the time of detection. Depending on the settings, the system (400) can execute a control command to automatically cut off the supply of the energy source simultaneously with sending the notification. The threshold and notification method are merely examples and are not limited thereto, and may vary depending on the operating policy or safety standards of the accommodation facility.

[0052] In operation 540, the system (400) can automatically cut off or control the supply of at least one of electricity, water, or gas through an energy control device when transitioning to an absent state. The system (400) can determine that the customer has gone out and transition to an absent state if it confirms that there is no detection signal from a motion sensor for a certain period of time or longer, for example, if no movement is detected for 30 minutes, it can determine that the customer is absent. The system (400) can execute an energy cut-off command immediately after determining the absent state or after a set waiting time has elapsed. In the case of power control, the system (400) can switch the heating and cooling device to standby mode or adjust the set temperature, and can automatically turn off the lights. In the case of water control, the system (400) can prevent water leakage accidents by shutting off the main valve, and in the case of gas control, it can stop the operation of the boiler and lock the gas valve.

[0053] According to one embodiment, the system (400) can transmit a wireless control signal to each energy control device via an IoT protocol and can verify normal operation by receiving feedback on whether a control command has been executed. The system (400) can store the control history as a log to accumulate data that can be used to analyze the energy saving effect in the future.

[0054] According to one embodiment, the system (400) receives a detection signal from a motion sensor installed in a guest room of an accommodation facility and can determine in real time the occupancy status, including whether a customer is present in the guest room and the movement path.

[0055] Additionally, the system (400) collects comprehensive energy usage data, including electricity, water, and gas consumed in the guest room, in real time and derives the customer's energy usage pattern through time-series data analysis. By comparing the derived energy usage pattern with a preset standard pattern, if abnormal energy consumption outside the predicted range is detected, the system can send a warning notification to the manager terminal or control the energy supply to be cut off. When the determined occupancy status changes to an absent state, the system (400) can control the supply of at least one of electricity, water, and gas to be automatically cut off or regulated through an energy control device connected to the guest room.

[0057] According to one embodiment, the system (400) can perform a data preprocessing process to remove noise, false positives, and false negatives included in raw data collected from a motion sensor and to correct missing values. The system (400) can control the system to minimize detection errors by calculating the amount of change in minute movement within the room based on valid data obtained through this data preprocessing process, and by confirming the occupancy status as a state where a user is present only when the calculated amount of change exceeds a preset detection threshold.

[0058] According to one embodiment, the system (400) may perform a data preprocessing process to remove noise, false positives, and false negatives included in raw data collected from a motion sensor and to correct missing values. Raw data may refer to unprocessed signal values ​​received directly from the sensor and may include unnecessary signals caused by electrical interference or environmental factors. Noise may refer to random signal fluctuations unrelated to actual movement and may be caused by voltage fluctuations of the power supply or interference from surrounding electronic devices. A false positive may refer to an error where the sensor detects movement even though there is actually no person present. A false negative may refer to an error where the sensor fails to detect a person even though a person is present. Missing values ​​may refer to a point in time when data was not recorded due to a communication error or a temporary sensor failure.

[0059] According to one embodiment, the system (400) can remove noise by applying a moving average filter to calculate the average of consecutive measurements, which can have the effect of mitigating instantaneous fluctuations in the signal. The system (400) can verify the signal duration to remove false positives and can determine only detection signals that last for a certain period of time or longer as valid movement. To correct false negatives, the system (400) can maintain the occupancy status even if no movement is detected for a certain period of time after being determined to be occupied. To correct missing values, the system (400) can fill in the missing values ​​by using the previous valid value or by linearly interpolating the values ​​at the preceding and subsequent points in time.

[0060] According to one embodiment, the system (400) can calculate the amount of change in minute movement within the room based on preprocessed valid data. The amount of change may refer to the difference in sensor output values ​​between two consecutive measurement points, and this may be an indicator for quantifying the intensity and frequency of movement. The system (400) can calculate the amount of change by subtracting the previous sensor value from the current sensor value, and can evaluate positive and negative changes in an integrated manner by taking the absolute value. The system (400) can compare the calculated amount of change with a preset detection threshold. The detection threshold may refer to a minimum change criterion for recognizing a movement as valid, and may be set considering sensor sensitivity and the installation environment. The system (400) can confirm the occupancy status only when the amount of change exceeds the threshold, which may have the effect of preventing false detection caused by minute environmental changes or residual noise. The system (400) can increase the reliability of the judgment by verifying whether multiple exceedances occur within a certain period of time, rather than making an immediate judgment based on a single exceedance event. The detection threshold and number of verifications are merely examples and are not limited thereto; they may vary depending on the cabin environment or sensor characteristics.

[0062] FIG. 6 is a flowchart illustrating a motion sensor data preprocessing method according to one embodiment.

[0063] In operation 610, the system (400) can perform noise removal and missing value correction of raw data collected from a motion sensor under the control of a processor. The system (400) can apply a moving average filter or a Kalman filter to remove electrical noise that occurs during the process of converting an analog signal received from the sensor into a digital value. The system (400) can remove instantaneous noise by, for example, calculating the average of the last 5 or 10 measurements, and the filter window size can be adjusted according to the measurement environment. The system (400) can remove false positives caused by the shaking of curtains or external heat sources even though there is no actual person present, and can correct false negatives where there is no detection because there is no movement for a long time even though a person is present. The system (400) can analyze the duration and intensity of the sensor signal and determine short signals of less than 0.5 seconds as false positives and ignore them, and can prevent false negatives by considering a state of sleep if there is no movement for more than 10 minutes while in a state of occupancy. The system (400) can correct data using linear interpolation or a recent valid value retention method when missing values ​​occur due to communication errors or sensor failures. The filter method and threshold value are merely examples and are not limited thereto, and may vary depending on the type of sensor or installation environment.

[0064] In operation 620, the system (400) can calculate the amount of change in minute movement within the room based on preprocessed valid data and check whether the detection threshold is exceeded. The system (400) can quantify the amount of change in movement by calculating the difference between sensor values ​​at two consecutive points in time, for example, by subtracting the previous sensor output value from the current sensor output value. The system (400) can compare whether the absolute value of the calculated amount of change exceeds a preset detection threshold, and the threshold can be set considering the sensor sensitivity and the installation environment. For example, in the case of a PIR sensor, the system (400) can determine that a change in output voltage of 0.5V or more is valid movement, and in the case of an ultrasonic sensor, a change in frequency due to the reflected wave Doppler effect of 100Hz or more is set as valid detection. The system (400) can store the result of the change calculation in memory and use it as data for time-series analysis. The sensor type and threshold are merely examples and are not limited thereto, and may vary depending on the room size or customer activity patterns.

[0065] In operation 630, the system (400) can minimize detection errors by confirming the occupancy status as occupied when the amount of change exceeds a threshold. The system (400) does not immediately determine occupancy based on a single sensor value, but can check whether a detection signal exceeding the threshold occurs continuously for a certain period of time. For example, the system (400) can confirm the occupancy status only when the threshold is exceeded three or more times within the last minute, and can prevent accidental false detection through a multi-verification method. In the occupancy state, the system (400) can disable the energy saving mode and switch to a normal supply state, return the heating and cooling device to a set temperature, and change the lighting control to a customer operation mode. The system (400) can transmit the confirmed status information to a cloud server to support real-time management of the room occupancy status of the entire accommodation facility. The number of verifications and the time range are merely examples and are not limited thereto, and may vary depending on the frequency of false detection or requirements to minimize customer inconvenience.

[0066] According to one embodiment, the system (400) converts collected total energy usage data into time-series data arranged in chronological order and inputs the time-series data into a deep learning model including at least one of a Convolutional Neural Network (CNN) or a Long Short-Term Memory (LSTM) to predict expected energy usage by time period. Through this, the system (400) can determine that energy usage measured in real time exceeds a preset tolerance range compared to the expected energy usage predicted by the artificial intelligence model as abnormal energy consumption and control the transmission of an abnormal sign to an administrator terminal.

[0067] According to one embodiment, the system (400) may activate a block encryption module for data security during data transmission and reception between a motion sensor, an energy control device, and a management server. A block encryption module may refer to a software or hardware component that performs encryption operations in units of data blocks of a fixed size. The system (400) may apply encryption to prevent man-in-the-middle attacks or eavesdropping that may occur during the data transmission and reception process.

[0068] According to one embodiment, the system (400) can process data blocks in units of 64 bits. A 64-bit block may refer to a fixed data unit of 8 bytes in size and may be suitable for processing short messages such as sensor measurements or control commands. If the data to be transmitted is larger than 64 bits, the system (400) may divide it into multiple blocks and encrypt them sequentially. The system (400) may divide each data block into an upper 32-bit block and a lower 32-bit block. The upper 32-bit block may refer to 4 bytes corresponding to the first half of the 64-bit data, and the lower 32-bit block may refer to the remaining half. This division structure enables parallel processing, thereby improving encryption speed.

[0069] According to one embodiment, the system (400) may allow two divided blocks to share the same logic operation circuit during the process of performing an encryption round function. An encryption round function may refer to a basic transformation operation that is repeatedly performed in an encryption algorithm. Sharing the logic operation circuit may refer to a method in which the upper block and the lower block use the same physical circuit in a time-sharing manner during hardware implementation, which may have the effect of reducing the circuit area by half. The system (400) may perform encryption and decryption by applying the Hight algorithm. The Hight algorithm refers to a 64-bit block lightweight encryption standard developed in Korea, which may refer to an algorithm that performs 32 rounds of operations using a 128-bit key. By utilizing the simple operation structure characteristic of the Hight algorithm, the system (400) can implement real-time encryption even on low-spec devices such as motion sensors. The block size and the number of rounds are merely examples and are not limited thereto, and may vary depending on security requirements or performance constraints.

[0071] FIG. 7 is a flowchart illustrating an integrated control method based on reservation platform linkage according to one embodiment.

[0072] In operation 710, the system (400) can receive real-time reservation data and check-in and check-out schedules by linking with an external reservation platform under the control of a processor. The system (400) can be linked with a reservation agency service platform via a RESTful API and can implement bidirectional data communication. The system (400) can receive real-time notifications whenever at least one event of reservation confirmation, change, or cancellation occurs from the reservation platform, and the received data may include room number, guest information, check-in time, check-out time, and number of guests. The system (400) may use digital signatures or API key authentication to verify the integrity of the received data. The system (400) can store the received reservation data in an internal database and update the room-specific check-in and check-out schedule table. The system (400) can maintain the latest information by updating the schedule when a reservation change or cancellation occurs. The method of communication with the linking platform is merely an example and is not limited thereto, and may vary depending on the contractual relationship of the accommodation facility or system compatibility.

[0073] In operation 720, the system (400) can determine unauthorized occupancy and perform a blocking process by cross-verifying the received reservation data and the detection signal of the motion sensor. The system (400) can check whether there is a valid reservation at that time by comparing the current time with the reservation schedule for each room, and can determine unauthorized occupancy if movement is detected by the motion sensor during a non-reservation period when there is no reservation. The system (400) can determine illegal extended stay if energy consumption in the room continues or the motion sensor is activated after the scheduled checkout time. When unauthorized occupancy is detected, the system (400) can immediately send an emergency notification to the administrator terminal, and the notification may include the room number, time of detection, reservation status, and sensor data information. If the automatic blocking mode is activated, the system (400) can execute a control command to immediately cut off the power, water, and gas supply to the room simultaneously with sending a warning notification. The system (400) may, for example, provide a grace period of 30 minutes after the checkout time has elapsed, and if the user does not vacate the premises even after the grace period has ended, the energy supply may be cut off in stages. The system (400) may record the history of the cutoff in a log so that it can be used as evidence in the event of a dispute. The grace period and the method of cutoff are merely examples and are not limited thereto, and may vary depending on facility policies or legal requirements.

[0074] In operation 730, the system (400) can perform preset control to adjust the environment of the room prior to the scheduled check-in time. The system (400) can determine the scheduled check-in time from the reservation data and automatically start adjusting the room environment by taking into account the pre-set preparation time. The system (400) can, for example, operate the heating and cooling unit starting two hours before check-in to adjust the room temperature to a comfortable level. The system (400) can automatically turn on the lights to create a bright atmosphere upon check-in and, if necessary, preheat the hot water boiler to prepare for immediate use of hot water. If the guest has a past stay history, the system (400) can provide a customized environment by retrieving personalized settings, including preferred temperature and lighting brightness. The system (400) can analyze energy usage efficiency by storing the execution status and results of the preset control in a log. The temperature setting value and preparation time are merely examples and are not limited thereto, and may vary depending on the season or customer preferences.

[0075] In operation 740, the system (400) can detect whether a vehicle matched to a specific room has left by analyzing video footage of a parking lot or a common area. The system (400) can receive a real-time video stream from a CCTV camera and recognize the vehicle by applying computer vision technology, and can use an object detection algorithm (e.g., YOLO (You Only Look Once) or Faster R-CNN). The system (400) can automatically read the number of an entering or exiting vehicle and convert it into text using license plate recognition technology, and can match it with the vehicle number registered at check-in. For example, when a guest registers a vehicle number at check-in, the system (400) can link the vehicle and the room in the database and detect the exit of the registered vehicle in real time by analyzing the parking lot entry and exit video. The system (400) can record the time when the vehicle leaves the parking lot and use it as an indicator to determine the likelihood of the room being vacated. During the video analysis process, the system (400) can automatically mask video of people other than vehicles to protect personal information, and the video data can be encrypted and stored. The object detection algorithm and recognition accuracy are merely examples and are not limited to this; they may vary depending on the parking lot environment or camera resolution.

[0076] In operation 750, the system (400) can perform weight-based control logic to shorten the waiting time or cut off the energy supply by determining a definite state of going out when vehicle departure information and a non-detection signal from a motion sensor in the cabin occur simultaneously. The system (400) can calculate the degree of certainty of going out by comprehensively evaluating indicators such as vehicle departure and non-detection of motion, and can make a final decision by assigning weights to each indicator. For example, the system (400) can assign a weight of 60% to non-detection of motion and 40% to vehicle departure, and determine a definite state of going out if the weighted score is 80 points or higher. The weights and the weighted score are merely examples and may vary depending on the settings.

[0077] According to one embodiment, the system (400) may apply a shorter waiting time than a normal absence state when it is determined to be in a definite absence state; for example, it may be set to shut off after 30 minutes for a normal absence state and after 10 minutes for a definite absence state. After the shortened waiting time has elapsed, the system (400) may switch the heating and cooling device to a low-power mode and shut off the power to unnecessary lights and home appliances. The system (400) may prevent safety accidents by closing the main water valve and shutting off the gas supply. The system (400) may release the absence state and automatically restore the energy supply as soon as the vehicle re-enters the facility or the motion sensor is activated. The system (400) may store the execution history of the weight-based control logic in a database and use it as data to quantitatively analyze the energy saving effect. The weight ratio and waiting time are merely examples and are not limited thereto, and may vary depending on the operational strategy of the accommodation facility or consideration of customer satisfaction.

[0078] According to one embodiment, the system (400) can receive real-time reservation data and check-in / check-out schedules for a room by linking with an external reservation agency service platform through an application programming interface. The system (400) cross-verifies the received reservation data and the detection signal of a motion sensor to determine that movement detection during a non-reservation period or energy consumption after the check-out time constitutes unauthorized occupancy and performs a blocking process, while simultaneously performing preset control to adjust the environment of the room before the scheduled check-in time. The system (400) can detect whether a vehicle matched to a specific room has left the room through an intelligent video analysis module that analyzes video footage of the accommodation facility's parking lot or common areas, and determines a definite state of being out when the detected vehicle departure information and the undetected signal from the motion sensor inside the room occur simultaneously, thereby shortening a preset waiting time or blocking the energy supply.

[0079] According to one embodiment, the system (400) can integrally control heterogeneous energy source supply lines including water valves, gas valves, and power circuit breakers through an energy control device, and can operate a software compatibility interface that converts the system's control commands into a protocol corresponding to the specifications of each device so as to be compatible with third energy control devices having different hardware specifications. The system (400) can provide a visualized user interface for abnormal energy consumption situations through a dedicated application installed on a manager terminal, and can control the execution of remote shutdown based on approval input from the manager terminal.

[0080] According to one embodiment, the system (400) may provide abnormal energy consumption situations as a visualized user interface through a dedicated application installed on a manager terminal. The dedicated application may refer to a mobile app running on an Android or iOS operating system or a web browser-based management program. The visualized user interface may refer to a screen configuration that allows for intuitive understanding by expressing data using graphic elements such as graphs, charts, or heatmaps. The system (400) may apply color coding to highlight rooms where abnormal consumption has occurred in red and normal rooms in green. The system (400) may display energy usage as a real-time line graph and indicate the predicted range with shading so that the current value can be visually determined whether it has deviated from the normal range. The system (400) may list a list of abnormal events in chronological order on a notification panel and display detailed information when each item is touched.

[0081] According to one embodiment, the system (400) can control the execution of a block remotely based on an administrator's approval input. The system (400) may be configured to select between an automatic block mode and a manual approval mode, and in the manual approval mode, the block may not be executed without explicit confirmation from the administrator. The system (400) may require additional security verification upon approval input and may restrict the execution of the block command to only authorized administrators through fingerprint recognition or PIN code input.

[0082] According to one embodiment, the system (400) can transmit a remote blocking command to the energy control device of the corresponding room via wireless communication and receive the result of the command execution as feedback to verify whether the blocking was successfully completed. The system (400) can record the blocking execution history in a log to store information regarding the administrator ID, execution time, target room, and reason for blocking. The authentication method and log items are merely examples and are not limited thereto, and may vary depending on security policies or management requirements.

[0084] FIG. 8 illustrates the process of analyzing the energy usage pattern of an artificial intelligence model according to one embodiment.

[0085] Referring to FIG. 8, the system (400) can perform an intelligent process that can detect and predict abnormal energy consumption through artificial intelligence (AI)-based energy usage pattern analysis.

[0086] In operation 810, the system (400) can perform a data collection step under the control of a processor to accumulate customer stay status, energy usage, and time-series data. As customer stay status data, the system (400) can collect information regarding check-in time, check-out time, duration of stay, number of guests, and customer type (e.g., business, tourism, family unit) by linking with a reservation system. As energy usage data, the system (400) can measure real-time consumption of electricity, water, and gas through smart meters. As time-series data, the system (400) can build a database sorted in chronological order by assigning accurate timestamps to all collected information. The system (400) can store all energy usage records in a time-series format, and each data point may include attributes such as date, time, room number, energy type, usage amount, and occupancy status.

[0087] In operation 820, the system (400) can perform a preprocessing process on the collected raw data. As a data cleaning operation, the system (400) can remove or correct abnormal values ​​caused by sensor errors or communication failures, and, for example, if power usage is recorded as a negative number or a physically impossible value such as 10,000 kWh is measured, it can determine this as an outlier and remove it. The system (400) can perform data normalization to convert variables with different units and ranges into values ​​between 0 and 1 or values ​​with a mean of 0 and a standard deviation of 1.

[0088] In operation 830, the system (400) can develop a time series prediction model as an AI modeling step. The system (400) can extract spatial features of energy usage patterns using a Convolutional Neural Networks (CNN) model and learn local patterns of time series data through a one-dimensional convolution layer. The system (400) can, for example, build a CNN model that takes past 24 hours of power usage data as input and predicts the usage for the next 1 hour, and the model structure can consist of an input layer, 23 convolution layers, a pooling layer, a fully connected layer, and an output layer. The system (400) can learn long-term dependencies of time series data using a Long Short-Term Memory (LSTM) model. LSTM is a type of Recurrent Neural Network (RNN) that can effectively model patterns and trends over time and can control the flow of information through an input gate, a forget gate, and an output gate. The system (400) may use a stacked LSTM structure with, for example, 24 stacked LSTM layers, and the number of hidden units in each layer may be set to 64, 128, 256, etc. The system (400) may build a hybrid model combining CNN and LSTM, and may implement a CNN-LSTM architecture that extracts local features with CNN and learns temporal dependencies with LSTM.

[0089] In operation 840, the system (400) can perform a verification and optimization process through a trained AI model. The system (400) can calculate various metrics to evaluate model performance and may use Mean Squared Error (MSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Coefficient of Determination. The system (400) may determine that the model shows excellent performance, for example, if the MAPE is 5% or less and the Coefficient of Determination is 0.9 or more in the test set. The system (400) can evaluate the generalization performance of the model by applying cross-validation techniques and can perform cross-validation using a time-series partitioning method that considers the characteristics of time-series data. The system (400) can gradually update the model whenever new data is collected, thereby enabling adaptation to seasonal changes or changes in customer behavior patterns.

[0090] In operation 850, the system (400) can perform energy usage prediction and analysis through an optimized AI model. The system (400) can estimate, for example, the amount of energy consumed after 1 hour, 6 hours, or 24 hours. The system (400) can calculate a deviation by comparing the predicted usage with the actual usage measured in real time, and can determine an abnormal pattern if the deviation exceeds the allowable error range. The system (400) can classify, for example, cases where the actual value exceeds the predicted value by more than 30% as overconsumption, and cases where it falls short by more than 30% as abnormal underconsumption. The system (400) can automatically analyze the cause of the abnormal pattern and can identify situations such as excessive operation of heating and cooling devices, water leakage, gas leakage, and standby power consumption of unused appliances. The system (400) can generate optimization recommendations by analyzing energy usage patterns by season, day of the week, and time of day, and, for example, can suggest dispersing peak hours since power usage is concentrated between 10 a.m. and 12 p.m. on weekdays. The system (400) can visualize the analysis results and display them on a dashboard.

[0091] In operation 860, the system (400) can generate and transmit a warning alarm when an abnormality occurs based on the analysis results. The system (400) can immediately generate an alarm message when abnormal energy consumption is detected, and the alarm message may include detailed information regarding the time of occurrence, the room number, the type of energy (e.g., electricity, water, gas), the deviation of the actual value from the predicted value, and the type of abnormal pattern (e.g., overconsumption, leakage, leak). The system (400) can generate a specific alarm (e.g., 'November 19, 2024, 15:00, Room 402, actual usage 2,400W detected as 100% over the predicted power usage of 1,200W. Suspected over-operation of the heating and cooling unit'). The system (400) can classify the severity level of the alarm into three stages, and the Warning level can be set to a deviation of 20–30% compared to the prediction, the Caution level to a deviation of 30–50%, and the Critical level to a deviation of 50% or more. The system (400) can apply a differentiated notification method according to each severity level.

[0092] According to one embodiment, the system (400) can continuously monitor and evaluate the prediction accuracy of an AI model. The system (400) can calculate model performance metrics by aggregating the error between the predicted value and the actual value accumulated over a certain period (e.g., one week, one month), and can automatically trigger model retraining when performance degradation is detected. The system (400) can determine, for example, when the MAPE increases from 5% to 10%, that it is time for retraining. The system (400) can apply an incremental learning method to quickly update the model using only recently collected data, thereby saving time and computing resources compared to full retraining.

[0094] According to one embodiment, the system (400) can perform an initial energy load pattern classification by calculating an average value of energy usage collected during a preset initial learning period from the time of check-in of a new customer, and comparing the calculated average value with a preset standard usage threshold value to classify the case where the average value exceeds the standard usage threshold value as a high load class and the case where it is less than the standard usage threshold value as a low load class. The system (400) can individually optimize the threshold by mitigating monitoring sensitivity for rooms classified as high load class by adding a positive (+) weight to the threshold that serves as a criterion for judging abnormal energy consumption and adjusting it upward, and by strengthening monitoring sensitivity for rooms classified as low load class by adding a negative (-) weight to the threshold and adjusting it downward.

[0095] According to one embodiment, when an approval signal indicating that the situation is normal is received from an administrator terminal after a warning notification is transmitted, the system (400) stores time-series data at the time the notification occurred as an exception-allowed pattern in a database, calculates the similarity between the real-time energy usage pattern that occurs thereafter and the stored exception-allowed pattern, and controls the system to withhold notification generation even if it exceeds a threshold if the calculated similarity is greater than or equal to a preset reference similarity.

[0096] According to one embodiment, the system (400) can calculate the average value of energy usage collected during a preset initial learning period from the time of a new customer's check-in. The initial learning period may refer to an initial observation period for collecting data to identify the customer's energy usage patterns, and may be set, for example, to the first 24 or 48 hours after check-in. During this period, the system (400) can measure the hourly usage of electricity, water, and gas and sum them to calculate the average value. The system (400) can compare the calculated average value with a preset standard usage threshold value. The standard usage threshold value may refer to the average energy consumption derived by statistically analyzing past accumulated guest data, and may be adjusted considering factors such as room size or seasonality. If the average value exceeds the standard usage threshold value, the system (400) may classify it as a high-load class. A high-load class may refer to a type of user that consumes more energy than average, which may apply to family guests or those staying in a room for an extended period. The system (400) can classify a low-load class if the average value is less than the reference value. A low-load class may refer to a type of user that consumes less energy than the average. The learning period and reference value are examples only and are not limited to this, and may vary depending on the characteristics of the accommodation facility or the degree of data accumulation.

[0097] According to one embodiment, the system (400) can increase the abnormal consumption judgment threshold for rooms classified as high load class by adding a positive weight. Adding a positive weight may mean increasing the threshold value, for example, if the basic threshold is 1,500W, it may be increased to 1,800W by applying a 20% weight. The increase may have the effect of mitigating monitoring sensitivity, which can reduce the occurrence of unnecessary notifications for customers with high energy consumption.

[0098] According to one embodiment, the system (400) may lower the threshold for rooms classified as low load class by adding a negative weight to the threshold. Adding a negative weight may mean reducing the threshold value, for example, by reducing the threshold from the base threshold of 1,500W by 20% to 1,200W. Lowering the threshold may have the effect of enhancing monitoring sensitivity, which can more sensitively detect cases where a customer who normally uses less energy suddenly uses more energy. This individual optimization method may have the advantage of reducing false positives and false negatives compared to uniformly applying the same absolute value standard to all customers. The weight ratio is merely an example and is not limited to this, and may vary depending on the results of the deviation analysis by customer type.

[0099] According to one embodiment, if the system (400) receives an approval signal from an administrator terminal indicating that the situation is normal after a warning notification has been transmitted, the system (400) may store time-series data at the time the notification occurred as an exception-allowed pattern in the database. An approval signal may refer to feedback entered by an administrator after checking the notification content and determining that it is normal usage rather than abnormal. An exception-allowed pattern may refer to an energy usage pattern that exceeds a threshold but is confirmed to be actually problem-free, and this may be a normal high-consumption case occurring in a special situation. The system (400) may store information regarding electricity, water, and gas usage, external temperature, and time of day at that time as time-series data.

[0100] According to one embodiment, the system (400) can calculate the similarity between a subsequently occurring real-time energy usage pattern and a stored exception-allowed pattern. Similarity may refer to an indicator that quantifies how similar two patterns are and may be expressed as a value between 0 and 1. The system (400) may use methods such as cosine similarity or Euclidean distance to calculate similarity. The system (400) may calculate similarity by, for example, comparing the energy usage vectors of the current pattern and the stored pattern over time. The system (400) may withhold notification generation if the calculated similarity is greater than or equal to a preset threshold similarity. The threshold similarity may refer to a minimum similarity value that determines two patterns to be identical and may be set, for example, to 0.85 or 0.9. Withholding notification may mean not sending a notification even though the threshold has been exceeded. In this way, the system (100) can prevent unnecessary repeated notifications when a pattern previously identified as normal recurs.

[0102] According to one embodiment, the system (400) can diagnose the condition of a facility by defining the real-time usage of water, gas, and electricity as a first measurement value, a second measurement value, and a third measurement value, respectively. Specifically, under conditions where the user is absent, the system (400) can generate a facility inspection notification by determining that the water leak is caused by physical damage to the cold water pipe rather than artificial water usage, based on the fact that the first measurement value exceeds a first threshold value, which is a leak judgment criterion, and the second and third measurement values, which are linked to hot water heating or pump operation, are below a second threshold value, which is an operation criterion. Additionally, the system (400) can determine that the gas leak is caused by the absence of the first measurement value and the second measurement value exceeds a preset third threshold value, and can urgently shut off the gas valve. The system (400) can control the main power breaker to open by determining that there is a risk of electrical safety accidents caused by a short circuit in the wiring, based on the detection of a second pattern containing non-periodic high-frequency noise components rather than a first pattern in the form of a sinusoidal wave that appears during normal load operation, by converting the current waveform of the third measurement value into the frequency domain and analyzing it.

[0103] According to one embodiment, the system (400) can diagnose the condition of a facility by defining the real-time usage of water, gas, and electricity as a first measurement value, a second measurement value, and a third measurement value, respectively. The first measurement value may refer to water usage and may be measured in units of liters per minute through a flow meter. The second measurement value may refer to gas usage and may be measured in units of cubic meters per hour through a gas meter. The third measurement value may refer to electricity usage and may be measured in units of amperes or watts through a current sensor. The system (400) can detect an abnormal condition of the facility at an early stage by comprehensively analyzing these three measurements.

[0104] According to one embodiment, the system (400) can check whether a first measurement value exceeds a first threshold value, which is a leak judgment criterion, under conditions where the user is absent. An absent state may mean a state where the motion sensor does not detect movement for a certain period of time or longer, which may indicate that there is no customer in the room. The first threshold value may mean a standard value that exceeds the minute water flow that may occur in a normal standby state, and may be set, for example, to 2 liters per minute. The system (400) can check whether the second measurement value and the third measurement value are below a second threshold value, which is an operating standard. A measurement value linked to hot water heating or pump operation may mean gas and power consumption that increases together when the boiler or pressure pump is operating. The second threshold value may mean the consumption amount in a standby state where these equipment is not operating. The system (400) may determine that a combination where the first measurement value is high but the second and third measurement values ​​are low is due to physical damage to the cold water pipe rather than artificial water usage. The system (400) can determine these combinations of conditions as a water leak and generate a facility inspection notification. The facility inspection notification may be a message notifying the manager that pipe inspection and repair are required, and may include additional information for estimating the location of the leak. The threshold value is merely an example and is not limited thereto, and may vary depending on pipe age or water pressure conditions.

[0105] According to one embodiment, a low first measurement value may indicate that hot water or heating is not being used, and a high second measurement value in this state may indicate that gas is leaking without being consumed. Gas leakage without boiler operation may refer to a dangerous situation where gas is released into the atmosphere without combustion due to loose pipe connections or valve malfunctions. When the system (400) detects such conditions, it may urgently shut off the gas valve. Urgent shut-off may mean immediately closing the gas supply line, which can be completed within seconds by transmitting a shut-off signal to the electric valve. Simultaneously with the shut-off, the system (400) may send an emergency notification to the manager and activate the ventilation system of the room to lower the gas concentration. This may have the effect of preventing gas explosions or poisoning accidents.

[0106] According to one embodiment, the system (400) can analyze the current waveform of the third measurement value by converting it into the frequency domain. A current waveform may refer to a graph representing the change in current values ​​over time, and normal alternating current may have the form of a sine wave of a constant frequency. Frequency domain conversion may refer to the process of decomposing a signal in the time domain into frequency components and may be implemented using a Fast Fourier Transform technique. The system (400) may use a first pattern in the form of a sinusoidal wave that appears during normal load operation as a reference in the conversion result. The first pattern may refer to a clean spectrum in which only the fundamental frequency component is prominent and harmonic components are few. The system (400) may check whether a second pattern containing non-periodic high-frequency noise components is detected. The second pattern may refer to a spectrum containing many irregular high-frequency components, which may indicate signs of electrical abnormality. High-frequency noise may refer to abnormal frequency components in the range of several kHz to tens of kHz. The system (400) may determine the second pattern as a wiring short circuit or arc occurrence. A wiring short circuit can refer to a condition where two conductors come into direct contact due to damage to the insulation sheath of a wire, and an arc can refer to the phenomenon where electricity is discharged into the air between disconnected wires.

[0107] According to one embodiment, the system (400) may cut off the power supply to a room or an entire floor by opening the main power breaker upon determining a risk of an electrical safety accident. The system (400) may send an emergency notification immediately after the cutoff and may restrict power restoration until an electrical safety inspection is completed. The frequency range and pattern criteria are merely examples and are not limited thereto, and may vary depending on the characteristics of the electrical equipment or safety standards.

[0108] According to one embodiment, the system (400) can anonymize energy consumption data per room to compare and analyze the energy efficiency of the entire accommodation facility. The system (400) can identify rooms with excessive energy consumption relative to the average by comparing the energy consumption of rooms with the same structure and size. The system (400) can classify the equipment status of rooms with excessive consumption as inspection targets and suggest specific inspection items to the manager, such as cleaning the heating and cooling unit filters or inspecting the insulation. The system (400) can analyze energy consumption patterns by floor and determine that roof insulation reinforcement is necessary if the heating and cooling energy consumption of the top floor is higher than that of the middle floors. The system (400) can analyze the difference in energy consumption of rooms by orientation and recommend the installation of shading facilities if the summer cooling load of south-facing rooms is high. Through periodic comparative analysis, the system (400) can track rooms with improved and deteriorated energy efficiency and detect equipment aging or performance degradation at an early stage.

[0109] According to one embodiment, the system (400) can identify the correlation between energy saving and customer satisfaction by linking energy usage data with customer review data from a reservation platform. The system (400) can apply natural language processing techniques to extract energy-related keywords such as temperature, heating and cooling, lighting, and water pressure from customer reviews and classify positive or negative evaluations. The system (400) can determine that it is necessary to relax the intensity of the energy saving policy if the frequency of inconvenience-related keywords increases in reviews of rooms where the energy saving mode was actively operated.

[0110] According to one embodiment, the system (400) can analyze power usage patterns to automatically identify home appliances consuming standby power and cut off the power to the corresponding products when the user is absent. The system (400) can monitor hourly changes in power usage within the room to measure base power that continues to be consumed even after the user has left. If the base power exceeds the expected standby power level, the system (400) can determine that unnecessary home appliances are turned on or that products with excessive standby power are connected. The system (400) can measure power consumption for each individual outlet through a smart outlet and selectively cut off outlets that consume power while the user is absent. The system (400) can exclude products requiring a continuous power supply, such as refrigerators or air purifiers, from being cut off, and can selectively control only products capable of cutting off standby power, such as televisions or chargers. The system (400) can maintain user convenience by automatically restoring the outlets that were cut off after the user returns to the room.

[0111] According to one embodiment, the system (400) can quantitatively provide environmental contribution to accommodation facilities and customers by converting energy saving effects into carbon emission reductions. The system (400) can apply emission factors based on the power production structure of each country to convert electricity usage into carbon emissions. The system (400) can calculate the amount of carbon dioxide reduced in kilograms by multiplying the amount of energy saved by the emission factor. The system (400) can calculate the amount of carbon reduction contributed during the individual customer's stay and provide it in the form of a digital certificate at checkout. The system (400) can automatically generate a report in the form of accumulated carbon reduction performance data so that it can be used as eco-friendly certification system or ESG evaluation data.

[0113] Software may include computer programs, code, instructions, or a combination thereof, which may configure a processing unit in a desired manner or command it independently or collectively. Software and data may be interpreted by the processing unit or permanently or temporarily implemented on various machines, components, physical or virtual devices, computer storage media, or transmission signals to provide instructions and data.

[0114] Although the examples above have been illustrated with specific drawings, experts in the field may apply various technical modifications and variations based on them. For instance, appropriate results can be obtained even if the described techniques are performed differently from the presented order, or if the components of the described systems, structures, devices, or circuits are combined in different forms or replaced with other components or equivalents.

Claims

Claim 1 In an integrated energy remote control system for accommodation facilities for global carbon neutrality, a memory for storing instructions; The system includes a processor, and the instructions, when executed by the processor, receive a detection signal from a motion sensor installed in a guest room of an accommodation facility to determine the occupancy status, including the presence of a customer in the guest room and the movement path; collect comprehensive energy usage data, including electricity, water, and gas consumed in the guest room, in real time to derive the customer's energy usage pattern through time-series data analysis; compare the derived energy usage pattern with a preset reference pattern, and if abnormal energy consumption outside the predicted range is detected, transmit a warning notification to an administrator terminal or control the energy supply to be cut off; if the determined occupancy status switches to an absence state, control the system to automatically cut off or adjust the supply of at least one of the electricity, water, or gas through an energy control device connected to the guest room; define the water usage as a first measurement value, the gas usage as a second measurement value, and the electricity usage as a third measurement value; and when the determined occupancy status is an absence state, the first measurement value exceeds a first threshold value, which is a reference value exceeding the minute water flow that may occur in a normal standby state, and the A system that generates a facility inspection notification by determining that there is a water leak caused by physical damage to the piping when the second measurement value and the third measurement value are less than the second threshold value, which is the gas and power consumption in a standby state where the boiler or pressure pump is not operating. Claim 2 In claim 1, the above instructions, when executed by the processor, perform a data preprocessing process to remove noise, false positives, and false negatives included in the raw data collected from the motion sensor and correct missing values, calculate the amount of change of minute movement within the room based on the valid data obtained through the data preprocessing process, and control the system to minimize detection errors by confirming the occupancy state as an occupancy state where a user is present when the calculated amount of change exceeds a preset detection threshold. Claim 3 In claim 1, the above instructions are a system that, when executed by the processor, converts the collected total energy usage data into time-series data arranged in chronological order, inputs the time-series data into a deep learning model including at least one of a CNN (Convolutional Neural Networks) or an LSTM (Long Short-Term Memory) to predict the expected energy usage by time period, determines as abnormal energy consumption when the energy usage measured in real time exceeds a preset allowable error range compared to the expected energy usage predicted by the artificial intelligence model, and controls the system to send a warning notification to an administrator terminal. Claim 4 A system according to claim 1, wherein, when the instructions are executed by the processor, the system activates a block encryption module for data security during data transmission and reception between the motion sensor, the energy control device, and the management server, wherein the block encryption module divides the data block into an upper 32-bit block and a lower 32-bit block when processing a data block in 64-bit units, and performs encryption and decryption by applying a Hight algorithm designed such that the divided upper 32-bit block and the lower 32-bit block share the same logic operation circuit to perform operations during the process of performing an encryption round function. Claim 5 In claim 1, the above instructions are a system that, when executed by the processor, controls the system to receive real-time reservation data and check-in / check-out schedules for the room by linking with an external reservation agency service platform and an application programming interface, cross-verifies the received reservation data and the detection signal of the motion sensor to determine movement detection during the non-reservation period or energy consumption after the check-out time as unauthorized occupancy and performs a blocking process, performs preset control to adjust the environment of the room before the scheduled check-in time, detects whether a vehicle matched to a specific room has left the room through an intelligent video analysis module that analyzes video captured of the parking lot or common area of ​​the accommodation facility, and determines a definite state of being out when the detected vehicle departure information and the undetected signal of the motion sensor inside the room occur simultaneously, thereby controlling the system to perform weight-based control logic to shorten the preset waiting time or block the energy supply.

Images