KR20210041806A - METHOD FOR MANAGING SMART BUILDING USING IoT SENSOR AND ARTIFICIAL INTELIGENCE - Google Patents

Abstract

The present invention relates to a smart building management method using an IoT sensor and artificial intelligence. The smart building management method using an IoT sensor and artificial intelligence comprises the steps of: obtaining data about the building collected by one or more IoT sensors (S110); pre-processing the obtained data (S120); obtaining one or more user feedback (S130); pre-processing the obtained user feedback (S140); inputting the pre-processed data and user feedback into an artificial intelligence model (S150); obtaining data for control of the building facility from an output of the artificial intelligence model (S160); controlling the building facility based on the optained data (S170); and updating the artificial intelligence model by obtaining user feedback according to the control of the facility (S180).

Description

Smart building management method using IoT sensor and artificial intelligence {METHOD FOR MANAGING SMART BUILDING USING IoT SENSOR AND ARTIFICIAL INTELIGENCE}

The present invention relates to a smart building management method using an IoT sensor and artificial intelligence.

As artificial intelligence technology emerges and develops, automation systems in various fields are emerging. This is to build a more efficient and systematic system while minimizing the consumption of human resources.With the emergence of various IoT sensor modules that explored more sensitive and advanced technologies, an autonomous system by artificial intelligence was built to manage various facilities in buildings. The technology that can be done is being embodied.

On the other hand, the operation of a building maintenance facility and various power consumption facilities is a complex process that requires a wide variety of factors to be considered, and it is difficult for the facility manager to react in real time or to establish and execute a long-term building management plan.

Building management includes factors related to the external environment such as temperature, humidity, season, sunlight, and climate conditions outside the building, temperature inside the building, humidity, air quality, volatile organic compounds (VOC), lighting, power consumption, and facilities. A wide variety of variables must be considered, including factors related to the internal environment such as the number of users, floating population, or the size and location of windows. It is very difficult for humans to directly infer and adjust individual causal relationships as to how factors affect the construction of the final internal environment.

Moreover, VOC is a group of substances well known as the cause of sick building syndrome (SBS), and its concentration in the interior space of the building is controlled by air conditioning technology (空氣調和技術, HVAC: Heating, Ventilation, & Air Conditioning). Although it may be possible, the facility manager resides, and it is practically difficult to perform as much ventilation as needed where necessary by taking into account humidity and temperature through real-time monitoring.

Due to these practical difficulties, the majority of buildings are managed inefficiently and consume electricity inefficiently, resulting in a huge social cost. Not only in terms of cost, but also in terms of the sustainability of society as a whole, energy problems have spurred. There is something to add.

Therefore, we establish a building operation strategy that efficiently manages facilities through minimal input from facility managers, analyzes the demands and reactions of building users, and minimizes the cost and energy consumption required for facility operation while maintaining comfortable facilities inside the building. It can be said that there is a social request for building a building management system that can be done.

In response to the need for minimization of cost and energy consumption in relation to the development of artificial intelligence and IoT technology and building management, the technical idea disclosed in the present invention has been created.

Registered Patent Publication No. 10-1838554, 2018.03.08

The problem to be solved by the present invention is to provide a smart building management method using an IoT sensor and artificial intelligence.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

A smart building management method using an IoT sensor and artificial intelligence according to an aspect of the present invention for solving the above-described problem is the step of acquiring data on a building, collected by one or more IoT sensors (S110), the collection Preprocessing the obtained data (S120), obtaining one or more user feedback (S130), preprocessing the obtained user feedback (S140), inputting the preprocessed data and user feedback into an artificial intelligence model (S150), acquiring data for controlling the building facility from the output of the artificial intelligence model (S160), controlling the building facility based on the acquired data (S170), and controlling the facility And updating the artificial intelligence model by acquiring user feedback (S180).

In addition, the step (S180), the step of converting the user's feedback obtained in the step (S180) to a parameter based on fuzzy logic (S210), the step of inputting the converted parameter to the artificial intelligence model (S220) ), re-acquiring data for controlling the building facilities based on the output value of the artificial intelligence model (S230), and controlling the facilities of the building using the re-acquired data (S240). I can.

In addition, the step (S120) may include a step (S310) of classifying the data on the building, collected by the one or more IoT sensors, into data on the external environment and data on the internal environment, the step (S140) may include converting the obtained user feedback into a parameter based on fuzzy logic (S320), and in the step (S160), the output of the artificial intelligence is , The user's optimal preference state learned from the data on the external environment and the user feedback converted to the parameter, and the internal environment changed to the optimal preference state, learned from the data on the internal environment and the user's optimal preference state. It may contain control values related to facility control to be used.

In addition, the step (S120), the step of pre-processing the data obtained in the step (S110) (S410), the step of inputting the pre-processed data in the step (S410) to a pre-learned second artificial intelligence model (S420). ) And predicting the future power consumption of the building from the output of the second artificial intelligence model (S430).

In addition, the step (S160), obtaining a plurality of candidate data for controlling the building equipment (S510), calculating the power consumption when the building equipment is controlled according to each of the plurality of candidate data (S520), estimating the state of the building when the building equipment is controlled according to each of the plurality of candidate data (S530), and based on the state of the building and power consumption according to each of the plurality of candidate data The step of determining one final data (S540) may be further included.

In addition, receiving a facility control input from the first user (S610), controlling the facility of the building according to the control input of the first user (S620), and the artificial It may further include the step of updating the intelligence model (S630).

In addition, the data on the facility collected in the step (S110) is VOC (Volatile Organic Compunds) concentration, fine dust concentration, carbon dioxide concentration, comprehensive air quality, lighting, natural light, indoor temperature, outdoor temperature, humidity, vibration, building It is characterized in that at least one of an external climate condition, a season, power consumption of a facility inside a building, the number of users inside the facility, or a movement inside the facility, and in the step (S170), the controlled facility is HVAC (Heating, Ventilation and Air). Conditioning) facilities, air conditioning facilities, heating facilities, elevators, security systems, lighting facilities, air purifiers, windows, shading facilities, door shielding facilities, humidifiers, dehumidifiers, facilities for relocating internal facilities, firefighting facilities, or home appliances It may be characterized in that more than one.

An apparatus according to an aspect of the present invention for solving the above-described problem includes a memory for storing one or more instructions, and a processor for executing the one or more instructions stored in the memory, wherein the processor executes the one or more instructions. , Performs a smart building management method using an IoT sensor and artificial intelligence according to the disclosed embodiment.

A computer program according to an aspect of the present invention for solving the above-described problems is combined with a computer that is hardware, and a record that can be read from a computer to perform a smart building management method using an IoT sensor and artificial intelligence according to the disclosed embodiment. It is stored on the medium.

Other specific details of the present invention are included in the detailed description and drawings.

By the technology disclosed in the present invention, data on the internal environment of the building or the environment inside and outside the building that affects the power consumption can be easily collected, and efficient without a complicated understanding of the relationship between the factors related to the building management. Building management can be done.

In addition, even if the building's facility manager performs only minimal manipulation or control, the artificial intelligence that is self-updated by autonomous environmental adaptation provides various optimization scenarios to the building user or facility manager, and collects user's feedback to make this fuzzy logic. By processing by, more flexible control can be achieved.

By converting the user's verbal feedback into parameters by fuzzy logic, artificial intelligence learns the internal environment that provides optimal comfort compared to the external environment experienced by actual users, and can set it as a standard or purpose state, and to maintain the state. It has the effect of minimizing energy consumption, increasing user productivity, improving building operation, and contributing to the sustainability of society.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a flowchart illustrating a process of acquiring data for managing a building using artificial intelligence.
2 is a flowchart illustrating a process of obtaining user feedback and processing it using fuzzy logic.
3 is a flowchart illustrating a process of obtaining a control value of a facility for changing an internal environment of a building by using collected sensor data and user feedback.
4 is a flowchart illustrating a process of predicting a building's power consumption using sensor data.
5 is a flowchart illustrating a process of acquiring a plurality of candidate data for building management and determining one final data.
6 is a flowchart illustrating a process of receiving an input on facility control from a user and updating artificial intelligence through it.
7 is a block diagram of an apparatus according to an exemplary embodiment.
8 is a diagram showing a process of obtaining an optimal preference state of a user and deriving a control value through it.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used herein, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and "and/or" includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used with meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

The term "unit" or "module" used in the specification refers to a hardware component such as software, FPGA or ASIC, and the "unit" or "module" performs certain roles. However, "unit" or "module" is not meant to be limited to software or hardware. The “unit” or “module” may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example, "sub" or "module" refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, Includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables. Components and functions provided within "sub" or "modules" may be combined into a smaller number of components and "sub" or "modules" or into additional components and "sub" or "modules". Can be further separated.

In the present specification, a computer means all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in the corresponding hardware device according to embodiments. For example, the computer may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and application running on each device, and is not limited thereto.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Each of the steps described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed by different devices according to embodiments.

Buildings referred to in the present specification may refer to various types of buildings, such as large buildings used by a large number of people, as well as general residential buildings made up of one or more members.

The membership function (or membership function) used for fuzzification referred to in the present specification may be defined in various forms such as a triangular shape, a trapezoidal shape, or an arbitrary curved shape, but is not limited thereto.

The fuzzy reasoning referred to in the present specification may mean one of Maldini-type fuzzy inference and Sugeno-type fuzzy inference, but is not limited thereto.

The defuzzification referred to in the present specification mainly refers to the center of gravity method (Cetroid Technique), but a method such as a mean of maximum may be variously employed, but is not limited thereto.

1 is a flowchart illustrating a process of acquiring data for managing a building using artificial intelligence.

In step S110, the computer acquires data on the building, collected by one or more IoT sensors.

In one embodiment, the at least one IoT sensor described above may employ one or more of conventional wireless communication technologies, and transmit the collected data to a computer using this.

In an embodiment, the above-described building data may include data related to the exterior of the building and data related to the interior of the building.

In addition, the above-described data on the outside of the building include temperature, humidity, weather conditions, wind speed, air component ratio, fine dust concentration, illuminance, sunlight irradiation angle, sunlight, movement of external objects, presence of toxic components, or building It may include weather forecasts for areas where there is.

In addition, the above-described data on the inside of the building include temperature, humidity, air composition ratio, fine dust concentration, illumination, number of users inside the building, usage time by space division inside the building, occupancy by space division inside the building or building The usage pattern of internal facilities may be included.

In step S120, the computer preprocesses the collected data.

In one embodiment, the above-described pre-processing may mean processing the collected data into a form for input into an artificial intelligence model.

For example, a computer can convert values that cannot be represented as numbers into numeric codes. In addition, the computer can fill the missing measurement value with the median value, the average value of a specific section, the mode, the corresponding value by regression analysis, or the predicted value by pattern estimation, and the like, and can delete meaningless values.

In addition, the computer can delete abnormal measurement values due to malfunction of the sensor. For example, a measured value in a range that is physically difficult to occur, such as when the external temperature at a point in time is measured at 200 degrees Celsius, can be treated as an error, deleted, and replaced with an estimated appropriate value.

In addition, the computer can create a new variable by using the existing measured value. For example, in order to determine the internal condition of a building, a computer can perform a specific operation using temperature and humidity as variables and use the variable derived as a result of this operation as a new parameter for determining the internal condition of the building.

In addition, the computer can obtain more appropriate variables by using calculation methods such as log function, square root, standard deviation, variance, and average distribution for specific variables in order to determine the internal state of the building.

In addition, the computer can create a dataset for learning artificial intelligence by unifying time zones and selectively collecting only specific variables. In addition, when a specific variable is controlled, the computer can create a data set by collecting changes in other variables. For example, a computer collects the values of the remaining variables when the temperature is 25 degrees Celsius, and writes the data in a form for input into an artificial intelligence model so that it can learn the pattern of changes in the other variables when the temperature inside the building is 25 degrees Celsius. It can be processed. This processed data can be processed in a controlled state of variables other than temperature, as well as setting a specific time period or creating a data set in which one or more variables are controlled.

In this way, computers can transform the data into a form that the AI model can learn for input into the AI model. This may be a form of calculating one or more variables in association with each other, or may be a form of changing one variable into another factor through an operation.

In step S130, the computer acquires one or more user feedbacks.

In an embodiment, the computer may recognize a voice through an IoT sensor to obtain user feedback, provide an application capable of receiving user feedback through a user's device, or additionally have a display device for feedback input.

In an embodiment, the user feedback may be in the form of a verbal input by voice, a character string input, direct facility control by the user, or a score form with a specific upper limit value as a perfect score.

For example, if the user adjusts the desired internal temperature to 20 degrees while the computer adjusts the internal temperature to 25 degrees, the computer can obtain the feedback by determining that the user prefers 20 degrees to the 25 degrees in the corresponding condition.

And, if the user input to stop the operation of the temperature control device is received as soon as the actual temperature reaches 23 degrees while the internal temperature is adjusted to a control value of 20 degrees, the preferred feedback for the above 20 degrees is 23 degrees. Can be replaced by preference for feedback.

In an embodiment, the computer may request user feedback through the user's device. For example, it is possible to request and obtain feedback on the internal environment of a building or facility control of a computer through a request to install an application on the user's device.

In step S140, the computer preprocesses the obtained user feedback.

In an embodiment, the computer may convert the above-described verbal user feedback voice into a character string, and in this process may additionally utilize a separate artificial intelligence model capable of converting the voice into text.

In an embodiment, the computer may utilize fuzzy logic in the process of preprocessing user feedback.

For example, in the process of performing fuzzification on temperature, a computer is'very cold','cold','cool','suitable','warm','hot','very hot', etc. You can set various sets of and approximate their membership functions in the shape of triangles, trapezoids, bells, and normal distributions.

For example, 26 degrees Celsius may have a membership degree of 0.3 in'suitable' and 0.7 in'warm'. If the computer determines the internal temperature as exemplified above based on the fuzzy logic, it can provide a'slightly warm' temperature for the actual user to experience.

In addition, the use of this fuzzy logic is to allow the machine to flexibly input linguistic expressions about the senses where the boundaries of humans are unclear. Different membership functions can be used in consideration of the perceived temperature.

For example, when the computer determines the season as early winter, a degree of membership of 20 degrees Celsius may be changed from'cool' to'slightly cold'.

In addition, the computer can utilize fuzzy logic in the process of processing data on the external or internal environment.

In an embodiment, the computer may determine the season based on the input date and time information, and the determination of the season may be made based on fuzzy logic.

For example, the computer may determine September 25 as'end of September','late summer','early autumn', or'a combination of 0.4 late summer and 0.6 early autumn'.

In addition, the computer may change the standard of the range of parameters of the internal environment, which is judged to have high comfort inside the building, based on this seasonal judgment. Also, the computer may change the standard of the above-described internal environment parameter range and change the membership function of the corresponding parameter. Here, the parameter means a number representing the state of the internal environment such as temperature and humidity.

The criterion of this parameter range refers to the user's optimal preference state, which will be described later, and is a value determined based on the feedback of users who used the existing building, but it can be changed by acquiring feedback from new users. to be.

In addition, based on the above-described seasonal judgment, when it is judged as'midsummer' or'midwinter', the computer is set to a temperature lower than the comfort standard in consideration of the access time and body temperature or sensory temperature of the person entering the room after outdoor activities. The corresponding thermostat can be controlled to cool or to heat at a high temperature.

For example, in the case of a building user entering the room after outdoor activities in the middle of summer, even if the appropriate temperature for the user staying inside the building is determined to be 25 degrees, the computer is outdoors based on the judgment based on the season and the external temperature and humidity. The air conditioner can be operated 18 degrees in advance based on the user entering the room, and the temperature control device can be operated to gradually achieve a comfortable environment from the time the user enters the room.

In one embodiment, the computer may obtain information on the body temperature of the user of the building to be managed. The body temperature to provide information on the body temperature can be detected by taking a thermal image based on infrared rays based on the IoT sensor, or by an IoT temperature sensor attached to a refrigerator handle or a front door handle. It can be detected by a variety of devices or methods that are not limited by means and methods.

In addition, the computer can acquire information about the body temperature of the building user and operate the device inside the building based on this.

For example, if a building user's body temperature is higher than usual and it is determined that he has a cold, the appropriate temperature or humidity can be set higher, the ventilation function of the air conditioner can be weakly adjusted, and the humidity can be adjusted.

In addition, if the building user is a female, if it is detected that the user's basal body temperature is higher than usual, the computer determines that the user is in the high temperature of the menstrual cycle and increases the user's comfort according to the value set by the user. For this, you can change the belonging function of various parameters.

Fuzzy theories such as fuzzification, fuzzy reasoning, and defuzzification have reached a sufficient degree of maturity, and, of course, various techniques that can be utilized by a person skilled in the art in the above-described embodiments can be used. The technical idea disclosed in is not limited to the above-described embodiment.

In step S150, the computer inputs the preprocessed data and user feedback into the artificial intelligence model.

In an embodiment, the artificial intelligence model described above may be one or more different artificial intelligence models, and may be artificial intelligence models having different purposes having different input values and output values.

For example, the artificial intelligence model described above may be one artificial intelligence that comprehensively judges the state of the internal environment, but adjusts one or more of parameters such as temperature, humidity, or internal VOC (Volatile Organic Compounds) concentration. It may be an artificial intelligence model for

In step S160, the computer obtains data for controlling the building equipment from the output of the artificial intelligence model.

In one embodiment, the above-described data for the control of building facilities are the control values of individual facilities, weights considering the influence between facilities and facilities, power consumption, and the impact on one or more internal environmental parameters when operating a specific facility. , An optimal preference state of one or more current building users, a difference value between the optimal preference state and the current internal environment, a target time required to reach the optimal preferred state, or a control value of the facility in consideration of the required time.

For example, assuming that the current temperature is 30 degrees Celsius, the user's optimal preference is 25 degrees Celsius, and the target time required is 5 minutes, considering the user's body temperature and outside temperature, the computer has a target time required of 20 minutes. It is possible to satisfy the target required time by operating the strength of the temperature control device more strongly than in the case.

And, as in the above-described case, when a facility is operated with a stronger output in consideration of the target time required, the control value may be differently adjusted or other facilities may be controlled together in consideration of the power consumption.

For example, if there is a need to drop the temperature quickly, if the outside temperature is lower than the inside temperature, the computer opens a window or operates the ventilation system first to lower the inside temperature partly due to the inflow of outside air, and then turn on the cooling facility. You can make it work.

In step S170, the computer controls the building equipment based on the acquired data.

In one embodiment, the computer may transmit various control signals to a facility connected to the computer using wireless communication or wired communication.

The above-described building facilities include ventilation and air conditioning facilities including HVAC (Heating, Ventilation and Air Conditioning), motors for opening and closing windows, motors for moving awnings or curtains, dehumidifiers, elevators, lighting facilities, water facilities, regardless of type A maintenance facility or security facility may be included.

In an embodiment, the computer may control the building equipment with a control value obtained by newly processing the acquired data.

In step S180, the computer updates the artificial intelligence model by obtaining user feedback according to the control of the facility.

In an embodiment, when the computer operates the facility at a point in time and the internal environment changes after a certain period of time due to this, the computer may request a user's feedback on this. In addition, the computer can fuzz the state in which the user does not give any feedback as a positive feedback. In addition, the computer may perform fuzzy processing of the user's feedback in consideration of a time interval between the time when the feedback is input and the time when the first control signal is transmitted.

For example, if a user's new control input is received within a very short time after the user is controlled by the computer, the narrower the time interval, the greater the difference between the control value entered by the user and the control value transmitted by the computer. , It can be determined that there is negative feedback.

2 is a flowchart illustrating a process of obtaining user feedback and processing it using fuzzy logic.

In step S210, the computer converts the user's feedback obtained in step S180 into a parameter based on fuzzy logic.

Here, the above-described parameter refers to a value obtained by converting user feedback into a quantified value.

In one embodiment, the computer may convert the user's verbal voice feedback into a character string, and input this into an artificial intelligence model to obtain a quantified output.

For example, if the user says'Oh, it looks a little cold', the computer approximates this as'a little cold' feedback, and the approximated feedback can be quantified by fuzzifying the approximated feedback.

Specifically, if the user's uttered voice ‘Oh, it’s a little cold’ is collected by the IoT sensor, and the computer converts it into a character string and inputs it into the artificial intelligence model, the artificial intelligence model can express the corresponding language expression. Received input, this can be quantified as'very cold','cold','a little cold','cool','suitable','warm','a little hot','hot','very hot', etc. It is possible to output which one of the expressions it belongs to. The computer may perform fuzzification, fuzzy reasoning, and defuzzification from the output of the artificial intelligence model to convert the user's verbal feedback into quantified values, that is, the above-described parameters.

In an embodiment, the computer may receive user feedback as a score in a form that does not need to be converted into a quantified parameter. In addition, when receiving user feedback as a score, the computer may distinguish and recognize the user who gives the user feedback, and may perform fuzzy processing of the user feedback considering that the criteria for giving the score may be different for each person.

For example, a specific user may evaluate a very satisfied state as 100 points and a dissatisfied state as 70 points, but that and other users rate a satisfactory state as 90 points, and a dissatisfied state as 45 points. It can be evaluated by points. These numerical inputs are subject to the individual user's subjectivity, and each user can objectively evaluate scores for the same environmental condition differently.

For example, the computer generates a graph that takes an approximate value by normalizing the difference between the lowest score and the highest score of a specific first user and various feedback scores, and synthesizes the difference between the lowest score and the highest score of the second user and various feedback scores. A graph taking the approximate value normalized by is generated, and evaluation based on different criteria for each individual can be objectified by comparing the two graphs described above. This is to objectify the subjectively scored score as much as possible by measuring the degree to which the specific score corresponding to the user feedback is far from the average based on the standard deviation.

In an embodiment, the fuzzy logic technique employed to convert the above-described user feedback into a parameter may be achieved by employing all techniques commonly used by those skilled in the art.

In step S220, the computer inputs the converted parameter into the artificial intelligence model.

In an embodiment, the computer may pre-process the transformed parameters into different datasets and input them to the artificial intelligence model for training of artificial intelligence models for different purposes or outputting results.

In step S230, the computer again acquires data for controlling the building facility based on the output value of the artificial intelligence model.

This process is to allow the computer to calculate the control value reflecting the user feedback using artificial intelligence.

In an embodiment, the re-acquired data for facility control may have a control value that is not different from a control value that has been previously acquired and used as a result.

For example, if the user feedback is a feedback indicating that the current state is satisfied, or if the user feedback is input before a sufficient time has elapsed for the internal environment to change to correspond to the corresponding control value, the same control value may still be obtained. .

In an embodiment, the data for controlling the facility may be data for changing or updating an optimal preference state of one or more building users.

The above-described optimal preference state means an internal environmental state in which the sum of satisfaction points for the internal environment of one or more building users is maximized. When a new building user enters a building or an existing building user inputs new feedback, this optimal preference state may also be changed to a new state, and data for facility control may refer to a numerical value for this state.

In one embodiment, the above-described control data may be simply a control value for controlling a facility, and is newly adjusted in consideration of how the operation of each facility affects each other by simultaneously operating one or more different facilities. It can also be a control value.

For example, a computer may operate an air conditioner to control humidity, lower the temperature, and simultaneously operate a dehumidifier. If the control of such facilities is determined to have high power consumption, a modified control value may be provided to operate only the dehumidifier. I can.

As another example, there may be a case in which two or more users input different feedbacks on the same internal environmental state in the same indoor space. In this case, the computer recognizes the location where the user who inputs the feedback resides through the IoT sensor or receives input from the user, and controls the facility so that the sensory temperature at a specific location is adjusted by a method such as turning the wind direction of the air conditioner to the location. I can.

In addition, when the gap between the feedback experienced by two or more users cannot be narrowed through the facility control of the computer, the computer may propose a change in the space used by a specific user or install an additional shield.

In step S240, the computer controls the equipment of the building by using the acquired data again.

This is to control the facility by reflecting the fuzzy-processed user feedback.

3 is a flowchart illustrating a process of obtaining a control value of a facility for changing an internal environment of a building by using collected sensor data and user feedback.

In addition, FIG. 8 is a diagram showing a process of obtaining an optimal preference state of a user and deriving a control value through it.

In step S310, the computer classifies the data on the building, collected by the one or more IoT sensors, into data on the external environment and data on the internal environment.

This is because it is necessary to use the two separately, and the data on the external environment is to perform a building management method in consideration of how to operate the facilities of the building or the user's perceived temperature, etc., and the data on the internal environment is stored. Separation is to obtain a control value for controlling the internal facilities by calculating a difference value from the current internal environment and the internal environment that is the target of construction.

In addition, separating data on the external environment and the internal environment is not limited to the above-described purpose, and can be processed, edited, and utilized for various purposes.

The data on the external environment described above may include all factors that may affect the living environment inside and outside the building, such as wind direction, wind speed, sunrise and sunset time outside the building.

In step S320, the computer converts the acquired user feedback into parameters based on fuzzy logic.

Converting the user feedback into a quantified parameter in the above-described step (S320) is not different from the above-described method described by exemplifying the method in which other steps are performed, but for that purpose, the data related to the above-described internal environment. It may be further included to calculate an optimal preference state described later through comparison, combination, or analysis with.

In addition, in the step S160, the output of the artificial intelligence may be the user's optimal preference state learned from the data on the internal environment, the data on the external environment, and user feedback converted into the parameters.

In an embodiment, the above-described artificial intelligence may convey to a computer whether a user feels a sense of satisfaction or comfort and how much when a specific internal environment is created under conditions related to a specific external environment.

For example, on a rainy day outside the building, the indoor temperature is slightly higher than the outdoor temperature, and the humidity is much lower than the outdoors. The status can be output as the user's optimal preference status.

Here, the above-described'slightly high' temperature and'very low' humidity are converted into quantified values using a fuzzy logic in consideration of the user's feedback.

Specifically, the user's optimal preference state refers to the state of the internal environment in which the user feels the highest satisfaction, but the emotional expression of'satisfaction' in which human boundaries are unclear can be easily changed by various factors, so it is more flexible. It is designed to additionally consider data on the external environment in order to carry out the autonomous management method of the building.

For example, if the external environment is cold, the user can feel more comfort and satisfaction at slightly higher temperature and humidity than usual. In addition, such a feeling of satisfaction may be different for each user even in the internal environment of the same condition, and even for the same user, it may be different according to the condition or health status of that day or time.

Accordingly, in an embodiment, the above-described data on the internal environment may include information on the body temperature or health status of a user who is using the inside of the building. The user's body temperature may be collected by a thermal image sensing camera having a human scale, a door handle, a temperature sensor installed on a handle or a contact part of frequently used facilities such as a refrigerator.

In addition, the computer may request the user to input information on the health status of the building user through a separate display and input device or the user's device.

In one embodiment, when there are two or more building users, each user may provide a different satisfaction score for the same internal environment condition as a feedback. In this case, the computer is a value obtained by summing all the satisfaction scores of the plurality of building users. The internal environment can be created to achieve this maximum.

In addition, user feedback on the composition or satisfaction of the above-described internal environment is added up and reflected in the judgment of the computer, or it may be necessary to cyclically obtain user feedback several times and adjust new control values to be used for learning of the artificial intelligence model. have.

That is, in the above-described step, the computer can be said to be using the user feedback for two main purposes: obtaining a new control value and finding a state in which the user feels comfortable.

In step S330, the computer obtains a control value for facility control for changing the internal environment of the building from the obtained data about the user's optimal preference state and the internal environment.

In one embodiment, the computer acquires data on the current internal environment state, compares it with the user's optimal preference state obtained in the above-described step (S320), and, if there is a difference, the interior of the building in a direction to eliminate the difference. Calculate control values for controlling equipment.

In addition, when the computer operates one individual facility to obtain the above-described control value, the effect of the operation of such facility on various individual elements constituting the internal environmental state can be considered.

For example, if the building is cooled by operating the air conditioner, the air conditioner is operated to lower the temperature, but this may affect the humidity. In consideration of this effect, when the computer needs to lower the temperature and humidity of the internal environment at the same time, the control value of the facility that lowers the humidity such as a dehumidifier or ventilation facility is set to a lower value in consideration of the degree to which the humidity decreases by operating the air conditioner. Can be printed.

In addition, for example, if the temperature of the internal environment needs to be increased, if there is an effect of increasing humidity in the process of increasing the temperature, the computer will increase the temperature and operate the dehumidifier together to suppress the unnecessarily rising humidity. Control value can be output.

4 is a flowchart illustrating a process of predicting a building's power consumption using sensor data.

In step S410, the computer preprocesses the data obtained in step S110.

In an embodiment, the computer may separate data on the amount of power consumed while each of the facilities inside the building operates from the above-described building-related data.

In addition, the computer may transform, edit, process, or perform computational processing of the above-described power consumption data in a form suitable for input into a second artificial intelligence model to be described later.

For example, data on messenger consumption can be deleted as in the above-described preprocessing process, data that are estimated to be obscure or obvious errors, and empty data values due to deletion or omission of data collection are averaged, regression analysis, etc. It can be replaced with an estimated value, and only relevant data can be grouped and processed into a new dataset.

In step S420, the computer inputs the data pre-processed in step S410 to the pre-learned second artificial intelligence model.

In step S430, the computer predicts the future power consumption of the building from the output of the second artificial intelligence model.

In an embodiment, the computer may predict the power consumption at various time intervals that are preset or input from the user.

For example, the computer may predict power consumption for a day or for a period of various units such as a week, a month, and a year.

In an embodiment, the computer may change the prediction of the power consumption to a different value due to changes in the external environment.

For example, if a heavy rain falls on a hot summer day and the external temperature falls and affects the internal temperature, it can be predicted that the power consumption will decrease as the cooling equipment is operated less due to this effect.

In an embodiment, in predicting the amount of power consumption, the computer may make a prediction based on accumulated data over a continuous period of the past, and this period may be determined by a preset value or a user input.

For example, in order to predict the annual power consumption, the computer may collect power consumption data for the past three years and estimate the power consumption for the current year based on this.

In one embodiment, in order to predict the power consumption at a specific point in time, the computer collects data on the power consumption on a day having a value similar to the data on the internal and external environment of the building on the date or period to be predicted, and based on this, the power consumption amount Can be predicted.

For example, if the maximum temperature is 32°C and the minimum temperature is 25°C, and data on the forecast of rain with a rainfall of 40mm are acquired, all data related to the external environment most similar to the external environment are obtained as a label. By collecting data on the power consumption of the day, it is possible to predict the power consumption of the date.

In addition, if the external environment has little effect on the power consumption due to the characteristics of the building to be managed, the power consumption can be predicted in consideration of the number of users inside the building, and the number of users inside the building is the number of entrants passing through the entrance and exiting. It can be estimated by counting the number of characters. In addition, a motion sensor, a vibration sensor, etc. may be additionally provided to calculate the number of users inside the building, and the number of users may be estimated by allowing a computer to recognize or analyze an image of a CCTV photographing a point passing through the entrance. .

5 is a flowchart illustrating a process of acquiring a plurality of candidate data for building management and determining one final data.

In step S510, the computer acquires a plurality of candidate data for controlling the building equipment.

In an embodiment, the computer may acquire a plurality of candidate control values for different amounts of time or power consumed to create an internal environment serving as one purpose.

For example, in order to bring the interior temperature of a building from 25 degrees Celsius to 20 degrees Celsius, the computer can simulate a combination of various facility controls to reach the desired state. In addition to simply operating the cooling facility, it is possible to provide users with a variety of scenarios to reach the target state through more efficient power consumption by cooling the internal temperature faster using the ventilation facility or by controlling the humidity together. It is also possible to provide a control value for each required time to the user by setting the required time required to reach the terminal in various ways.

In step S520, the computer calculates a power consumption amount when the building equipment is controlled according to each of the plurality of candidate data.

In an embodiment, the computer may simulate and calculate the amount of power consumed by the building while maintaining the internal environment when the candidates of various control values are each executed as it is.

In an embodiment, when prioritizing specific parameters along with power consumption, the computer may provide the most efficient building facility control scenario for each parameter.

In step S530, the computer estimates the state of the building when the building equipment is controlled according to each of the plurality of candidate data.

In one embodiment, when the equipment inside the building is operated or controlled by the corresponding candidate control values, the computer shows how parameters representing various environmental conditions inside the building can change over time, such as a graph. It can be visualized and presented to the user.

In an embodiment, the computer may estimate a change pattern of the internal environment in which the influence of the computer is reflected in consideration of data on the external environment.

In step S540, the computer determines one final data based on the state of the building and power consumption according to each of the plurality of candidate data.

In one embodiment, the computer may request the user to select one control value set from among a plurality of candidate control values. You can provide information together.

In an embodiment, the computer may control and manage building facilities by selecting one of a plurality of candidate control values according to a preset priority, even if no input from a user is received.

For example, if the power consumption is set to a minimum, the computer selects a scenario with the minimum power consumption in the building and controls the building's internal facilities, which is premised to satisfy the optimal preferences of the building users. It should be.

That is, the computer can provide the control value for the change of the state of the internal environment of the building in consideration of the consumption of time and resources from various angles, to the user for each corresponding parameter according to the parameter having the first priority. The building is managed by selecting one of these or using the control values of the building equipment selected by the computer based on a preset priority.

6 is a flowchart illustrating a process of receiving an input on facility control from a user and updating artificial intelligence through it.

In step S610, the computer receives a facility control input from the first user.

In one embodiment, the above-described facility control input is regardless of its type, and in addition to electrically inputting the control signal, if there is a motor that moves the curtain or awning as illustrated above, it may be controlled by hand. have. However, in this case, the computer can convert an action that took place offline into an electrical signal through the sensor and receive it.

In step S620, the computer controls the facilities of the building according to the control input of the first user.

In step S630, the computer updates the artificial intelligence model based on the control input of the first user.

That is, this is to allow the computer to accept this as user feedback even if the user simply controls the facility without inputting a control signal of the facility as user feedback. This can help the building's users to be a long-term optimization of the equipment inside the building even if they do not act in consideration of the optimization of the building equipment separately and live daily.

In one embodiment, the data on the building collected in the step (S110) is VOC (Volatile Organic Compunds) concentration, fine dust concentration, carbon dioxide concentration, comprehensive air quality, lighting, natural light, indoor temperature, outdoor temperature, humidity. , Vibration, weather conditions outside the building, season, power consumption of facilities inside the building, the number of users inside the building, or movement inside the building.

In addition, in the step (S170), the controlled facilities are HVAC (Heating, Ventilation and Air Conditioning) facilities, air conditioning facilities, heating facilities, elevators, security systems, lighting facilities, air purifiers, windows, shading facilities, and door shielding facilities. , A humidifier, a dehumidifier, a facility for moving the location of an internal facility, a firefighting facility, or a home appliance.

In an embodiment, the above-described building-related data may further include an element that may have an influence on the internal environment condition of the building, and a sensor module for detecting this may be additionally provided.

In an embodiment, the computer may build a database of information on the power consumption of the building to be managed based on data collected from the IoT sensor. In addition, information from the established database can be pre-processed and transmitted to an artificial intelligence model, and through this, the artificial intelligence can predict the power consumption of the building to be managed.

For example, the process of pre-processing the information in the database described above is to extract data by determining the time point and end point of the period based on the changed time point such as members of building users or the use pattern of facilities, etc. It can mean processing to form data pairs.

This is because the amount of power consumption may change in the case of changes in the members of building users, changes in building usage patterns of building users, or new installations of internal facilities.

And, as the power consumption of the building is inevitably different depending on the change of season and temperature, the power consumption of last July was extracted from the database to predict the power consumption of July of this year, and the number of users or power consumption pattern, etc. Can reflect changes in

In addition, the computer can create data pairs in the form of extracting only data of a specific date or time period, not a continuous period, and transmitting it to the artificial intelligence.

For example, in order to predict the power consumption of a day reflecting today's weather, weather information that falls within a range similar to today's weather forecast or the power consumption of a day with data on the external environment is extracted from the database and delivered to artificial intelligence. You can create a data pair of the form to do.

Alternatively, the computer can predict a shorter range of power consumption. For example, if the weather changes rapidly unlike the forecast, it reflects information on sudden heavy rains to predict the pattern of changes in the external environment such as a drop in temperature and a rise in humidity, and the power consumption on a day that had a similar pattern. By extracting the data, you can estimate the amount of power consumed in the rest of the day.

In addition, the computer predicts the amount of power consumption based on the information provided by the artificial intelligence model, but may take into account the weather, the season, the outside temperature, the climate, or the facility usage pattern of the building user.

In an embodiment, the computer may predict power consumption based on an expected power consumption pattern for a specific period in the future at the request of a user or a facility manager.

7 is a block diagram of an apparatus according to an exemplary embodiment.

The processor 102 may include one or more cores (not shown) and a graphic processing unit (not shown) and/or a connection path (eg, a bus) for transmitting and receiving signals with other components. .

The processor 102 according to an embodiment executes one or more instructions stored in the memory 104 to perform the method described with reference to FIGS. 1 to 6 and 8.

For example, the processor 102 acquires new training data by executing one or more instructions stored in the memory, performs a test on the acquired new training data using the learned model, and labels the test result. Extracting first learning data in which the obtained information is obtained with an accuracy equal to or higher than a predetermined first reference value, deleting the extracted first learning data from the new learning data, and removing the new learning data from which the extracted learning data is deleted. The learned model can be retrained by using.

Meanwhile, the processor 102 temporarily and/or permanently stores a signal (or data) processed inside the processor 102, a RAM (Random Access Memory, not shown) and a ROM (Read-Only Memory). , Not shown) may further include. In addition, the processor 102 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, RAM, and ROM.

The memory 104 may store programs (one or more instructions) for processing and controlling the processor 102. Programs stored in the memory 104 may be divided into a plurality of modules according to functions.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Components of the present invention may be implemented as a program (or application) and stored in a medium to be executed by being combined with a computer that is hardware. Components of the present invention may be implemented as software programming or software elements, and similarly, embodiments include various algorithms implemented with a combination of data structures, processes, routines or other programming elements, including C, C++. , Java, assembler, or the like may be implemented in a programming or scripting language. Functional aspects can be implemented with an algorithm running on one or more processors.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You will be able to understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

Claims (9)

In the method executed by the computer,
Obtaining data on a building, collected by one or more IoT sensors (S110);
Pre-processing the acquired data (S120);
Obtaining one or more user feedback (S130);
Pre-processing the obtained user feedback (S140);
Inputting the preprocessed data and user feedback into an artificial intelligence model (S150);
Obtaining data for controlling the building equipment from the output of the artificial intelligence model (S160);
Controlling the building equipment based on the acquired data (S170); And
Updating the artificial intelligence model by obtaining user feedback according to the control of the facility (S180); Containing,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
The step (S180),
Converting the user's feedback obtained in step S180 into a parameter based on fuzzy logic (S210);
Inputting the converted parameter into the artificial intelligence model (S220);
Re-acquiring data for controlling the building facility based on the output value of the artificial intelligence model (S230); And
Controlling the equipment of the building by using the acquired data again (S240); Containing,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
The step (S120),
Classifying the data on the building, collected by the at least one IoT sensor, into data on an external environment and data on an internal environment (S310); Including,
The step (S140),
Converting the acquired user feedback into parameters based on fuzzy logic (S320); Including,
In the step (S160), the output of the artificial intelligence,
It is characterized in that the optimal preference state of the user learned from the data on the internal environment, the data on the external environment, and the user feedback converted into the parameter,
Obtaining (S330) a control value for facility control for changing the internal environment of the building from the obtained data about the user's optimal preference state and the internal environment; Containing,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
The step (S120),
Pre-processing the data obtained in the step (S110) (S410);
Inputting (S420) the data pre-processed in the step (S410) into a pre-trained second artificial intelligence model; And
Predicting future power consumption of the building from the output of the second artificial intelligence model (S430); Containing,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
The step (S160),
Obtaining a plurality of candidate data for controlling the building equipment (S510);
Calculating power consumption when the building equipment is controlled according to each of the plurality of candidate data (S520);
Estimating a state of the building when the building equipment is controlled according to each of the plurality of candidate data (S530); And
Determining one final data based on the state of the building and power consumption according to each of the plurality of candidate data (S540); Further comprising,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
Receiving a facility control input from a first user (S610);
Controlling the facilities of the building according to the control input of the first user (S620); And
Updating the artificial intelligence model based on the control input of the first user (S630); Further comprising,
Smart building management method using IoT sensors and artificial intelligence. The method of claim 1,
The data on the facility collected in the step (S110) are VOC (Volatile Organic Compunds) concentration, fine dust concentration, carbon dioxide concentration, comprehensive air quality, lighting, natural light, indoor temperature, outdoor temperature, humidity, vibration, outside the building. It is characterized in that it is at least one of climate conditions, seasons, power consumption of facilities inside the building, number of users inside the facility, or movement inside the facility,
In the step (S170), the controlled facilities are HVAC (Heating, Ventilation and Air Conditioning) facilities, air conditioning facilities, heating facilities, elevators, security systems, lighting facilities, air purifiers, windows, shading facilities, door shielding facilities, humidifiers. , A dehumidifier, a facility for moving the location of an internal facility, a firefighting facility, or a home appliance,
Smart building management method using IoT sensors and artificial intelligence. A memory for storing one or more instructions; And
A processor that executes the one or more instructions stored in the memory,
The processor executes the one or more instructions,
An apparatus for performing the method of claim 1. A computer program combined with a computer as hardware and stored in a recording medium readable by a computer to perform the method of claim 1.

Images