DIGITAL TWIN-BASED BUILDING MANAGEMENT SYSTEM

Abstract

This invention relates to a system (1) that collects temperature, humidity, carbon dioxide levels, lighting levels, occupancy status and energy consumption data in smart buildings via 5G (Fifth Generation Mobile Communication Network) to determine the climate control and lighting needs in different parts of the building with a digital twin and to adjust the settings in the building according to energy cost, carbon emissions and thermal comfort of the users.

Description

1 TARIFF DIGITAL TWIN-BASED BUILDING MANAGEMENT SYSTEM Technical Area This invention allows for monitoring temperature, humidity, carbon dioxide levels, and lighting levels in smart buildings. occupancy status and energy consumption data for 5G (Fifth Generation Mobile) Communication Network – Fifth Generation Mobile Communication Network) by collecting the air conditioning and lighting needs of different parts of the building Using a digital twin, we can determine the relevant building settings, energy costs, and carbon 10. a system that allows regulation according to emissions and the thermal comfort of users It is related to. Previous Technique Today, building management involves temperature, humidity, carbon dioxide, occupancy, and energy. Consumption data is collected from sensors for heating, ventilation and air conditioning. Control of (Heating, Ventilation and Air Conditioning – HVAC) systems managing energy consumption and indoor environmental conditions The arrangement is ensured. Building Information Modeling (Building Information 20) Building equipment and physical structures are analyzed using BIM (Building Information Modeling) based digital twins. The fields are modeled in a digital environment, and the predicted mean vote (Predicted Mean Vote) is calculated. – PMV) and estimated dissatisfaction rate (Predicted Percentage of Dissatisfied – User thermal comfort is evaluated using PPD (Product Pressure Percentage) values. However, In current systems, sensor data is continuously processed using a digital twin model for 25 seconds. inability to update energy consumption, carbon emissions and PMV / PPD values Inability to optimize, low-latency and uninterrupted data from numerous sensors. Failure to provide transmission and connection interruptions will affect the uninterrupted operation of the target system. Various technical shortcomings emerged, resulting in the inability to continue the work. It is coming out. 30 2 Therefore, considering the studies and shortcomings in the current technique... When these factors are considered, smart buildings can monitor temperature, humidity, and carbon dioxide levels. lighting level, energy consumption and regional occupancy data, and weather information. Heating and cooling temperatures can be determined by analyzing forecasts and electricity tariff information. settings, operating levels of ventilation fans and air dampers, 5 lighting levels and motorized blinds, if any, in the building energy cost, carbon emissions and user comfort of sunshade locations It appears that a system is needed that will allow for its regulation accordingly. Chinese patent CN120524722A, which is included in the prior art, 10 The document simulates the energy consumption of buildings using digital twin technology. The text refers to a system that improves energy efficiency. The invention uses a physical model of the building and neural data from multi-source sensors. It combines them into a common digital twin model using symbolic methods. Thanks to machine learning algorithms, the model is constantly being updated, state 15 Variables, model parameters, and structural information are automatically generated. It is being corrected. Dynamic hybrid self-evolving network structure, real-time sensor. The data should be reinforced by taking into account environmental conditions and equipment aging. It generates energy efficiency strategies through learning. Updated digital twin. The model predicts future electricity, heating and cooling loads, hybrid 20 integer linear programming and event-based model predictive control It determines the most appropriate control actions using algorithms. Weather conditions, energy External factors such as prices, production plans, and equipment failures also contribute to optimization. is included in the process. Brief Description of the Invention The purpose of this invention is to measure temperature and relative humidity in different parts of a building. carbon dioxide levels, lighting levels, air speed, and occupancy data, along with electricity. 30 from meters and heating, cooling, ventilation and lighting systems the consumption and work data received, their importance and the need for delay 3 according to 5QI (5G Quality of Service Identifier) transmitting by classifying them into categories via 5G; building walls, windows, indoor air and Heat transfer between reinforced concrete structural masses is measured using RC (Resistance-Capacity – The goal is to implement a system that calculates resistance (capacitance) using a thermal model. Another purpose of this invention is to analyze past occupancy, energy consumption, days, hours, and weather. status data LSTM (Long Short-Term Memory) by operating with its network, it provides heating, cooling and the expected number of people in each region. Estimating ventilation requirements; heat resistance and heat in the RC thermal model capacity values ​​EKF (Extended Kalman Filter) (Filter) method updates according to actual temperature measurements and each building Temperature and energy consumption in the region; users' environment is hot or cold. the percentage of people expected to be dissatisfied with the level of feeling and environmental conditions The goal is to implement a system that identifies on a hybrid digital twin. Another objective of this invention is to determine regional temperature on a hybrid digital twin. Occupancy and energy consumption information, as well as outside air temperature, humidity, and solar radiation, electricity tariff and grid carbon emission values ​​MPC (Model Predictive) By analyzing each building zone using the Control – Model Predictive Control) method. Heating and cooling temperature settings, ventilation fan speed, air damper 20 the opening, the level of light, and motorized blinds if available. sunshade positioning according to energy cost, carbon emissions and ISO 7730 standard. The goal is to create a system that determines values ​​based on defined PMV and PPD parameters. Another purpose of this invention is to ensure that in the event of a disconnection, the connection remains active for 25 seconds before the disconnection occurs. specified heating and cooling temperature settings, fan speed, damper opening, lighting Indoor temperature and carbon dioxide levels, along with information on level and sunshade position. and building control based on occupancy status using MEC (Multi-access Edge Computing – Continuing via the Multi-Access Edge Computing server; connection renewed When installed, the work logs generated on the MEC server during the outage are 30. records kept at the center CRDT (Conflict-free Replicated Data Type – 4 Recording time, security decisions, and (Non-Conflict Replicated Data Type) method. Determines the current building status by combining it with authorized user interventions. It is about creating a system. Detailed Description of the Invention 5 The "Digital Twin-Based Building" project was carried out to achieve the purpose of this invention. The "Management System" is shown in the attached diagram; Figure 1 shows a schematic view of the system that is the subject of the invention. 10 The parts shown in the figure are individually numbered, and the corresponding numbers correspond to these numbers. given below: 1. System 15 2. Electronic device 3. Environmental Sensor 4. Optimization Module 5. Database 6. Edge Server 20 7. Cloud Server D. External Server In smart buildings, temperature, humidity, carbon dioxide levels, lighting levels, and occupancy rates are monitored. By collecting status and energy consumption data via 5G, the building's 25 different 25 Determining the air conditioning and lighting needs in the departments using a digital twin. and the settings in the building in question, energy cost, carbon emissions and users' thermal The system in question (1) enables adjustment according to its comfort; - Temperature, humidity, and carbon dioxide levels of sections in the form of buildings, blocks, floors, or rooms. quantity, lighting level, occupancy rate, energy consumption, thermal comfort, Display of alarm and operating setting information; authorized user approval, rejection and manual control decisions by at least one electronic device that provides an interface for entering operating settings device (2), - Temperature, relative humidity, and carbon dioxide levels in different parts of the building, 5 lighting level, air speed, occupancy rate, and door or window condition To measure at least one of the related data points; the measurement value is given by the sensor ID, measurement to generate it using information such as time, building area, unit of measurement and measurement quality. at least one configured environmental sensor (3), − By analyzing the data received from the environmental sensor (3) and the external server (D) 10 The building's temperature, energy needs, and thermal comfort status are digitally analyzed based on zones. On the twin, determine the energy settings for climate control and lighting. to determine the cost based on carbon emissions and PMV and PPD values. at least one optimization module configured (4), - User permissions and manual settings entered via electronic device (2), 15 Measurement data received from the environmental sensor (3) is optimized by the optimization module (4) the prediction and operating settings, building characteristics, model created by its parameters are configured to store connection interruption logs. a small database (5), − electronic device (2), environmental sensor (3), optimization module (4), database 20 (5) and to exchange data with the external server (D); measurement data to organize and prioritize, run the optimization module (4), Converting the created work settings into control commands, communication. at least one edge configured to maintain building control in the event of a power outage. server (6), 25 − Historical building use, environmental measurement and energy data stored in the database (5) LSTM used in the optimization module (4) using consumption data training the Long Short-Term Memory (LST) network, the trained transmit model information to the edge server (6) and energy consumption, carbon Creating periodic analyses and reports on emissions and thermal comfort 30 It includes at least one cloud server (7) configured for this purpose. 6 The electronic device (2) in the system (1) which is the subject of the invention, any remote communication to communicate with the edge server (6) using the protocol and the established protocol A smartphone is configured to exchange data via communication channels. in the form of a tablet computer, desktop computer or portable computer It is a device. Electronic device (2), belonging to sections in the form of building, block, floor or room 5 temperature, relative humidity, carbon dioxide level, lighting level, occupancy rate, energy consumption, PMV and PPD values, alarm information, and connection status. Operating settings for climate control and lighting are determined according to user authorization. It is configured to provide an interface that allows it to be displayed. In the system that is the subject of the invention (1), the environmental sensor (3) located in the room, floor or block of the building by being positioned in usage areas such as these, temperature, relative humidity, Carbon dioxide levels, lighting levels, air speed, and whether a door or window is open or obtain at least one piece of data regarding the closed status and radar or Using the infrared method, the identities of the people in the relevant area are determined. 15 It is structured to perceive without determination and to create a regional population count. Environmental sensor (3) monitors the environmental conditions and occupancy of the building area in which it is located. by measuring the condition at specific time intervals and recording the value of each measurement. the unit, the time it was performed, its identifying information, and the measurement It is configured to generate measurement data that includes the building area where the measurement was taken. 20 The environmental sensor (3) transmits the measurement data it produces to 5G NPN (Non-Public Network – RedCap (Reduced Network) is used when connecting to a non-public network infrastructure. Capability – Reduced Capability) or NB-IoT (Narrowband Internet of Things – via Narrowband Internet of Things (Narrowband) connection; BLE (Bluetooth Low Energy) – In communication via Low Energy Bluetooth), Zigbee or Modbus, the word is 25 the subject is a gateway that transfers connections to the 5G NPN infrastructure. It is configured to be transmitted for processing within the system. The optimization module (4) in the system (1) which is the subject of the invention, is provided with Region-based temperature, relative humidity, carbon dioxide levels, lighting level, air 30 speed, occupancy, door or window status, energy consumption, and air conditioning operation. 7 Using data such as indoor air, building mass, walls, windows and exterior Heat transfer between environments is called RC (Resistance-Capacitance). to calculate using a thermal model and the heat resistance, heat capacity and Shading factor values ​​EKF (Extended Kalman Filter) To update according to actual measurement data using the Kalman Filter method 5 It is structured. The optimization module (4) uses trained LSTM (Long Short- Using the Term Memory (Long-Term Memory) network, past usage, days, Regional occupancy rates for future time intervals based on time and weather data. estimating internal heat gain and comparing these estimates with RC thermal model results. by combining them, determining the digital twin status of each area of ​​the building, air 10 temperature, mean radiant temperature, air velocity, relative humidity, clothing insulation, and user PMV (Predicted Mean Vote) based on activity level and PPD (Predicted Percentage Dissatisfied) Calculating percentage values ​​and digital twin status, weather forecast, electricity tariff and grid carbon emission factor MPC (Model Predictive 15 By evaluating the heating or cooling using the Control – Model Predictive Control) method. cooling temperature setting, air damper opening, fan speed, lighting level and to create a sunshade or curtain position if available It is being structured. The database (5) in the system (1) which is the subject of the invention is obtained from the environmental sensor (3). measured temperature, relative humidity, carbon dioxide level, lighting level, air speed, Measurement data regarding door or window conditions and the number of people in the area; electricity, natural gas and water consumption records; heating, cooling, ventilation and Data on lighting operations; past building usage and occupancy 25 information and outside air temperature, humidity and solar radiation received from external server (D) They record electricity tariff and grid carbon emission factor data along with forecasts. It is structured to keep it under control. Database (5), optimization module (4) regional occupancy, internal heat gain, energy demand, PMV and generated by PPD results, air conditioning, ventilation, lighting and availability 30 In this case, the operating settings related to the position of the sunshade or curtain; also BMS 8 (Building Management System) alarm and event logs, user approvals, manual interventions, situations that occur during connection interruptions records and reconciliation processes carried out after reconnection It is configured to store time and building location information. The edge server (6), electronic device (2), in the system (1) which is the subject of the invention, environmental sensor (3), optimization module (4), database (5), cloud server (7) and To communicate with the external server (D) and exchange data through this communication. It is configured to perform. The edge server (6) is located in the building or By positioning the environmental sensor (3) close to the building's location, the energy 10 Temperature, relative humidity, and carbon dioxide levels transmitted via meters and BMS, lighting level, air speed, occupancy rate, door or window status, energy consumption, Air conditioning operating status, alarm and event data are transmitted via 5G NPN (Non-Public). Receiving via Network (Non-Public Network); received from external server (D) outside air temperature, humidity and solar radiation forecasts, electricity tariff and 15 Obtaining grid carbon emission factor data, indoor and outdoor data. relevant building area, data source, measurement or data time, data value, unit of measurement and It is configured to organize according to quality information. Edge server (6), building characteristics, past usage data kept in the database (5) with instant data received data and model parameters and weather forecast within the target system, 20 Optimizing electricity tariff and grid carbon emission factor data. make available the module (4); run the optimization module (4) RC The thermal model was created using a trained LSTM network, EKF, and MPC methods. It is configured to instantly store the twin state in its memory. Edge heating or cooling 25 generated by the server (6), optimization module (4) temperature setting, air damper opening, fan speed, lighting level and current If applicable, the operating settings regarding the position of the sunshade and curtain for the target building control commands that include region, execution time, and set value to convert; the commands in question include digital signature, physical work limit and previously The timestamp and ID 30 prevent the previously used command from being executed again. to verify through the information and transmit the verified commands via the BMS 9 It is configured to detect when the 5G NPN connection is interrupted. The edge server (6) Switching to edge mode if determined; last valid digital twin By storing its status and operating settings in local memory, it maintains the RC thermal model and To maintain a simplified MPC process and monitor temperature, carbon dioxide level, and occupancy. and safe local control rules defined according to the latest valid operating conditions 5 It is configured to implement. The edge server (6) restarts the connection. if established, the situation created locally during the connection interruption. The records and the status records located at the center are CRDT (Conflict-free Replicated) Data Type – Non-Conflict Reproduced Data Type) within the scope of LWW-Register (Last- The Writer-Wins Register method (where the last writer's record is valid) takes 10 minutes. via the stamp, source ID, and Lamport logical clock information compare; manual interventions performed by an authorized user to reconcile conflicting records and transfer the reconciled situation to the database (5) and re-optimization module (4) based on the current digital twin status. It is being configured to run. 15 The invention includes the cloud server (7), electronic device (2), data in the system (1). to communicate with the base (5) and edge server (6) and the established communication It is configured to exchange data via the cloud. server (7), historical building usage and occupancy information stored in database (5) 20 with temperature, relative humidity, carbon dioxide levels, lighting level, energy consumption, data on air conditioning operational status and past outdoor conditions. to train the LSTM network used in the optimization module (4) is configured. Cloud server (7), historical occupancy profile, outside air temperature, Humidity and solar radiation, day type and time cycle data were input into the LSTM network. 25 to provide; region-based network occupancy rates for future time intervals, so that it can estimate internal heat gain and corrected external temperature and humidity values. The difference between the predicted results and the actual values ​​is called MSE (Mean Squared Equation). Error – Mean Squared Error) and the difference between occupancy classification and cross-reference. By evaluating the weights of the LSTM network using the entropy loss method, we determined them to be 30. It is configured to update. The cloud server (7) has been updated for the past three months. Trained LSTM network that is retrained weekly using data. to transmit the weights to the edge server (6) and the edge server (6) weights Using the optimization module (4), regional occupancy, internal heat gain and external to enable it to generate local predictions regarding environmental conditions It is configured. The cloud server (7) stores past energy data in the database (5). consumption, electricity tariff, grid carbon emission factor, PMV, PPD, work settings, alarm and event data for specific periods and building zones by evaluating energy consumption, energy cost, carbon emissions and thermal comfort. It is structured to generate periodic analyses and reports on the subject. Industrial Application of the Invention Thanks to the system (1) that is the subject of the invention, business centers, hospitals, hotels, shopping Temperature in centers, factories, university campuses and residential areas, humidity, carbon dioxide levels, lighting levels, regional occupancy, and energy 15 Analysis of consumption data; weather forecasts, electricity tariffs and grid heating, cooling and ventilation devices according to their carbon emission values temperature settings, fan and air damper operating rates, lighting automatic levels, motorized blind or sunshade positions By determining this, energy costs and carbon emissions can be reduced, and the indoor environment can be improved. air quality and user comfort are protected, and communication interruptions are eliminated. Safe operating settings with predefined limits for temperature and air quality. This ensures that building equipment can be continuously monitored. Around these fundamental concepts, the subject of the meeting is “Digital Twin-Based Building Management 25”. It is possible to develop a wide variety of applications related to the System (1)”, and the invention This cannot be limited to the examples described here, but is primarily stated in the claims. It is like that.

Claims

11 REQUESTS 1. In smart buildings, temperature, humidity, carbon dioxide levels, and lighting levels are monitored. By collecting occupancy and energy consumption data of the building via 5G Digital twin for 5 different sections to meet their climate control and lighting needs. to determine and adjust the settings in the relevant building in terms of energy cost, carbon emissions and enabling adjustment according to users' thermal comfort; - Temperature, humidity, and carbon dioxide levels of sections in the form of buildings, blocks, floors, or rooms. quantity, lighting level, occupancy rate, energy consumption, thermal comfort, Display of alarm and operating setting information; authorized user 10 approval, rejection and manual control decisions by at least one electronic device that provides an interface for entering operating settings device (2), - Temperature, relative humidity, and carbon dioxide levels in different parts of the building, lighting level, air speed, occupancy rate, and door or window status 15 To measure at least one of the related data points; the measurement value is given by the sensor ID, measurement to generate it using information such as time, building area, unit of measurement and measurement quality. at least one configured environmental sensor (3), - by analyzing the data received from the environmental sensor (3) and the external server (D) Digital display of building temperature, energy needs and thermal comfort status on a zone basis. On the twin, determine the energy settings for climate control and lighting. to determine the cost based on carbon emissions and PMV and PPD values. at least one optimization module configured (4), - User permissions and manual settings entered via electronic device (2), Measurement data received from the environmental sensor (3) is optimized by the optimization module (4) 25 the prediction and operating settings, building characteristics, model created by its parameters are configured to store connection interruption logs. a small database (5), − electronic device (2), environmental sensor (3), optimization module (4), database (5) and to exchange data with the external server (D); measurement data 30 to organize and prioritize, run the optimization module (4), 12 Converting the created work settings into control commands, communication. at least one edge configured to maintain building control in the event of a power outage. server (6), − Historical building use, environmental measurement and energy data stored in the database (5) LSTM 5 used in the optimization module (4) using consumption data training the Long Short-Term Memory (LST) network, the trained transmit model information to the edge server (6) and energy consumption, carbon To create periodic analyses and reports on emissions and thermal comfort. a system characterized by a cloud server (7) configured for (1).

2. Using any remote communication protocol, connect with the edge server (6) to communicate and exchange data through that communication smartphones, tablet computers, desktop computers or configured accordingly electronic device (2) characterized by a portable computer-like device A system like the one in Request 1 (1). 15 3. Temperature and relative humidity of sections in the form of buildings, blocks, floors or rooms, carbon dioxide levels, lighting levels, occupancy rates, energy consumption, PMV and PPD values, alarm information, connection status, and air conditioning. Lighting operating settings are 20 according to user authorization. an electronic device configured to provide an interface that allows its display A system like the one in Claim 1 or 2 characterized by (2) (1).

4. By being located in the usage areas of the building in the form of rooms, floors or blocks. temperature, relative humidity, carbon dioxide level, lighting level, air speed with 25 data on whether a door or window is open or closed, at least obtaining one and using radar or infrared methods in the relevant area Detecting individuals without identifying them and determining the regional population count. characterized by the environmental sensor (3) configured to create a system like any of the above requests (1). 30 13 5. Determine the environmental conditions and occupancy rate of the building's location. by measuring at time intervals and recording the value and unit of measurement of each measurement, the time it was carried out, the person's identification information, and the building where the measurement was taken. It is configured to generate measurement data that includes the region. Environmental The sensor (3) will transmit the measurement data it produces to the 5G NPN infrastructure if connected to it. Via RedCap or NB-IoT connection; with BLE, Zigbee or Modbus In terms of communication, it is a network that transfers these connections to the 5G NPN infrastructure. to transmit through the gateway for processing within the target system from the above requests characterized by the configured environmental sensor (3) a system like any of them (1). 10 6. The region-specific temperature, relative humidity, and carbon dioxide levels provided to him / her, lighting level, air speed, occupancy, door or window status, energy Using consumption and climate control data, indoor air, building RC thermal 15 measures heat transfer between the building mass, walls, windows, and the external environment. to calculate with the model and the heat resistance, heat capacity and Shading factor values ​​are determined using the EKF method based on actual measurement data. characterized by the optimization module (4) configured to update a system like any of the above-mentioned requests (1).

7. Using a trained LSTM network, retrieve historical occupancy, day, time, and weather conditions. regional occupancy and indoor temperature data for future time intervals estimating the gain, comparing these estimates with RC thermal model results. by combining them, determining the digital twin status of each area of ​​the building, weather temperature, mean radiant temperature, air velocity, relative humidity, clothing insulation, and 25 Calculating PMV and PPD values ​​based on user activity level and digital twin status, weather forecast, electricity tariff and grid carbon emissions Heating or cooling temperature setting by evaluating the factor using the MPC method, air damper opening, fan speed, lighting level and, if available Optimization configured to create the position of the sunshade or curtain 30 14 as in any of the above requests characterized by module (4) a system (1).

8. Temperature, relative humidity, carbon dioxide amount obtained from the environmental sensor (3), lighting level, air speed, door or window status, and regional person 5 Measurement data regarding the number of people; electricity, natural gas and water consumption records; operational data related to heating, cooling, ventilation, and lighting processes; historical building usage and occupancy information and outside air quality data received from the external server (D). Electricity tariffs and grid based on temperature, humidity, and solar radiation forecasts. Data structured to keep track of carbon emission factor data 10 as in any of the above claims characterized by base (5) a system (1).

9. Regional occupancy, internal heat generated by optimization module (4) Air conditioning with gains, energy needs, PMV and PPD results, 15 ventilation, lighting, and blinds or curtains if available. Operating settings related to its location; as well as BMS alarm and event logs, user consents, manual interventions, and events occurring during connection interruptions. status logs and reconciliation performed after reconnection Data structured to store transactions with time and building area information. 20 as in any of the above claims characterized by base (5) a system (1).

10. Electronic device (2), environmental sensor (3), optimization module (4), database (5), to communicate with the cloud server (7) and the external server (D) and these 25 edge configured to exchange data over communication any of the above requests characterized by its server (6) such a system (1).

11. Environmental sensor 30 positioned near the building or the area where the building is located. (3), temperature, relative humidity transmitted via energy meters and BMS, carbon dioxide level, lighting level, air speed, occupancy, door or window status, energy consumption, air conditioning operating status, alarms and events. receiving data via 5G NPN; outside air temperature received from external server (D), Electricity tariffs and grid carbon footprint calculations based on humidity and solar radiation forecasts. Obtaining emission factor data, indoor and outdoor data for the relevant building 5 region, data source, measurement or data time, data value, unit of measurement, and quality. with the edge server (6) configured to organize according to its information a system like any of the above characterized demands (1).

12. The instant data received and the building characteristics kept in the database (5), past 10 usage data and model parameters and air within the target system forecast, electricity tariff and grid carbon emission factor data make available the optimization module (4); optimization module (4) by running the RC thermal model, trained LSTM network, EKF and MPC methods 15 to instantly store the created digital twin state in its memory. from the above requests characterized by the configured edge server (6) a system like any other (1).

13. Heating or cooling temperature generated by the optimization module (4) settings, air damper opening, fan speed, lighting level and availability 20 the target building's operating settings regarding the position of sunshades or curtains control commands that include region, execution time, and set value to convert; those commands include digital signature, physical work limit and more. with a timestamp that prevents the previously used command from being executed again. Authenticate via credentials and execute verified commands through the BMS 25 characterized by the edge server (6) configured to transmit a system like any of the above requests (1).

14. If it is determined that the 5G NPN connection has been interrupted, the system enters edge operating mode. pass; the last valid digital twin state and operating settings are stored in local memory 30 by maintaining the RC thermal model and the simplified MPC process and 16 based on temperature, carbon dioxide level, occupancy rate, and the last valid operating status. structured to implement defined secure local control rules any of the above requests characterized by the edge server (6) a system like one of them (1).

15. If the connection is re-established, locally during the connection interruption The status records created and the status records located at the center are compared with CRDT. Within this scope, the LWW-Register method includes timestamp, source ID, and Comparing Lamport's logical clock information; authorized user. According to manual interventions performed by 10, conflicting records to reconcile, transfer the reconciled situation to the database (5) and optimize to restart the module (4) via the current digital twin status from the above requests characterized by the configured edge server (6) a system like any other (1).

16. Communicating with electronic devices (2), database (5) and edge server (6) and to exchange data through this established communication. from the above requests characterized by the configured cloud server (7) a system like any other (1).

17. Historical building usage and occupancy information stored in the database (5) temperature, relative humidity, carbon dioxide levels, lighting level, energy consumption, data on air conditioning operational status and past outdoor conditions. to train the LSTM network used in the optimization module (4) 25 of the above requests characterized by the configured cloud server (7) a system like any other (1).

18. Historical occupancy profile, outside air temperature, humidity and solar radiation, type of day. and provide clock cycle data as input to the LSTM network; to advance the network Area-based occupancy rate, internal heat gain, and adjusted 30 for time intervals. with the prediction results to be able to estimate outside temperature and humidity values 17 The difference between the actual values ​​relates to MSE and occupancy classification. by evaluating the difference using the cross-entropy loss method, belonging to the LSTM network. Characterized by the cloud server (7) configured to update the weights a system like any of the above-mentioned requests (1).

19. An LSTM network retrained weekly using historical three-month data. transmit the trained weights to the edge server (6) and the edge server (6) Regional occupancy, internal temperature using the weights in the optimization module (4) generating local data and predictions regarding external environmental conditions 10 characterized by the cloud server (7) configured to provide a system like any of the above requests (1).

20. Historical energy consumption, electricity tariff, grid stored in the database (5). carbon emission factor, PMV, PPD, operating settings, alarm and event data. Energy consumption, evaluated in terms of specific periods and building areas, 15 Periodic analysis of energy cost, carbon emissions and thermal comfort and characterized by the cloud server (7) configured to generate reports a system like any of the above-mentioned requests (1). 25

Images