Energy efficiency service system for AI based analysis on usage patterns of facilities and method of the same

Abstract

An artificial intelligence-based energy efficiency service system and method based on facility usage pattern analysis are disclosed, which provide an energy efficiency service by analyzing the usage patterns of a user's facility based on artificial intelligence. An artificial intelligence-based energy efficiency service system based on facility usage pattern analysis according to an embodiment of the present invention includes: a user facility information receiving unit configured to receive first actual energy data regarding a user facility being used by a user from at least one user terminal associated with a user; and a user facility operation information analysis unit configured to analyze the operating rate and operation pattern per unit time of the user facility based on the first actual energy data regarding the user facility. The user facility operation information analysis unit is configured to analyze the type of the user facility based on the first actual energy data by a first artificial intelligence model and determine a second artificial intelligence model corresponding to the analyzed type of the user facility; and to analyze the operating rate and operation pattern per unit time of the user facility by analyzing the first actual energy data by the second artificial intelligence model.

Description

Technology Field

[0001] The present invention relates to an energy efficiency service system and method, and more specifically, to an AI-based facility usage pattern analysis-based energy efficiency service system and method that provides energy efficiency services by analyzing the usage patterns of a user's facilities based on artificial intelligence. Background Technology

[0002] Currently, driven by carbon neutrality policies, there is a trend of shifting the social atmosphere from a focus on energy production to improving energy efficiency. Consequently, the adoption and operation of Factory Energy Management Systems (FEMS) to monitor and efficiently manage energy usage in factories is increasing, and the demand for FEMS is gradually rising due to recent recurring power shortages.

[0003] However, in the conventional energy efficiency market, data collection, diagnosis, consulting, and efficiency processes are conducted as separate businesses. Furthermore, the costs for implementing hardware and software for energy efficiency, such as power meters and network construction, are high, and the installation and operation of FEMS incur excessive costs amounting to hundreds of millions of won due to the expenses for installing efficiency facilities and consulting.

[0004] In particular, as the proportion of energy consumed by compressors in industrial energy consumption increases, interest in improving energy efficiency is growing in the compressor market. Compressors can be broadly classified into FSD (Fixed Speed ​​Drive) compressors and VSD (Variable Speed ​​Drive) compressors based on their inverter type. Since VSDs offer the advantage of reducing power consumption compared to FSDs, the demand and sales of VSD compressors are showing a continuous upward trend.

[0005] Nevertheless, users of existing FSD compressors often feel burdened by the additional costs involved in switching to VSD compressors. This is because confidence in energy efficiency is low due to the lack of quantitative quantification regarding the Return on Investment (ROI) for energy efficiency. Consequently, there is a growing need and demand for the establishment of AI (Artificial Intelligence) systems for diagnosis, monitoring, and energy reduction to measure energy efficiency. The problem to be solved

[0006] The present invention aims to provide an AI-based energy efficiency service system and method based on facility usage pattern analysis that analyzes the usage patterns of user facilities based on artificial intelligence and provides energy efficiency services.

[0007] Furthermore, the present invention aims to provide an AI-based energy efficiency service system and method based on equipment usage pattern analysis that can quantitatively analyze equipment operating rates and operating patterns by AI based on actual energy data of users.

[0008] Furthermore, the present invention aims to provide an AI-based energy efficiency service system and method based on equipment usage pattern analysis that can match equipment users and equipment suppliers based on equipment operating rates and operating patterns analyzed by AI.

[0009] Furthermore, the present invention aims to provide an AI-based energy efficiency service system and method based on equipment usage pattern analysis that mediates the sharing of equipment-related information and the equipment purchasing process based on information analyzed by artificial intelligence. means of solving the problem

[0010] An artificial intelligence-based energy efficiency service system based on facility usage pattern analysis according to an embodiment of the present invention comprises: a user facility information receiving unit configured to receive first measured energy data regarding a user facility being used by a user from at least one user terminal related to a user; and a user facility operation information analysis unit configured to analyze the operating rate and operation pattern per unit time of the user facility based on the first measured energy data regarding the user facility.

[0011] The above-mentioned consumer facility operation information analysis unit is configured to analyze the type of the consumer facility based on the first measured energy data using a first artificial intelligence model, determine a second artificial intelligence model corresponding to the analyzed type of the consumer facility, and analyze the operating rate and operating pattern per unit time of the consumer facility by analyzing the first measured energy data using the second artificial intelligence model.

[0012] The first artificial intelligence model may be an artificial intelligence model based on a Gaussian mixture model (GMM). The first artificial intelligence model may be configured to extract a representative waveform of the first measured energy data; and to classify the control scheme of the consumer facility by performing clustering on the representative waveform by comparing the representative waveform with a plurality of reference waveforms set by the GMM-based artificial intelligence model.

[0013] The above first artificial intelligence model can be configured to classify the control scheme of the user facility by determining the similarity between waveforms through analyzing the distance distribution between the representative waveform and the reference waveform based on dynamic time warping by the above GMM-based artificial intelligence model.

[0014] The above-mentioned consumer equipment may include a compressor. The above-mentioned consumer equipment operation information analysis unit may be configured to measure power consumption based on the unit-hour operation rate and operation pattern of the compressor by the above-mentioned second artificial intelligence model.

[0015] An artificial intelligence-based facility usage pattern analysis-based energy efficiency service system according to an embodiment of the present invention may further include: a supplier facility information receiving unit configured to receive supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals; an energy efficiency analysis unit configured to analyze supplier facilities related to the energy efficiency of the consumer facilities based on the operating rate and operating pattern of the consumer facilities; a supplier facility information transmission unit configured to transmit facility information of one or more first supplier facilities related to the energy efficiency of the consumer facilities and analyzed energy efficiency information to the consumer terminals; and a supplier facility supply request unit configured to request the supply of the second supplier facilities to the supplier terminals when a second supplier facility among the one or more first supplier facilities is selected by the consumer terminals.

[0016] The above energy efficiency analysis unit may be configured to analyze the operation pattern by determining the operating time, non-operating time, and operation termination time of the consumer facility based on the energy measured values ​​over time of the first measured energy data, and to calculate the operating rate according to the ratio of operating time per unit time.

[0017] The energy efficiency analysis unit described above may be configured to predict operating time energy efficiency information based on a value obtained by weighting the operating rate to the difference between the first operating time energy measured value occurring during the operating time among the first measured energy data and the second operating time energy predicted value occurring during the operating time of the supplier's facility; predict non-operating time energy efficiency information based on a value obtained by weighting the ratio of the non-operating time per unit time to the difference between the first non-operating time energy measured value occurring during the non-operating time among the first measured energy data and the second non-operating time energy predicted value occurring during the non-operating time of the supplier's facility; and generate energy efficiency information for the consumer's facility based on the operating time energy efficiency information and the non-operating time energy efficiency information.

[0018] The above-mentioned user facility information receiving unit can receive second actual energy data related to the second supplier facility from the user terminal after the user facility is replaced with the second supplier facility. The artificial intelligence-based facility usage pattern analysis-based energy efficiency service system according to an embodiment of the present invention may further include a settlement unit configured to calculate an actual energy efficiency value by comparing the first actual energy data and the second actual energy data, and to calculate the cost of the energy efficiency service based on the actual energy efficiency value.

[0019] An energy efficiency service method based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention comprises: a step of receiving first actual energy data regarding a user facility being used by a user from at least one user terminal related to the user by a user facility information receiving unit; and a step of analyzing the operating rate and operating pattern per unit time of the user facility based on the first actual energy data regarding the user facility by a user facility operation information analysis unit.

[0020] The above-mentioned analyzing step includes: a step of analyzing the type of the consumer facility based on the first measured energy data by a first artificial intelligence model and determining a second artificial intelligence model corresponding to the analyzed type of the consumer facility; and a step of analyzing the operating rate and operating pattern per unit time of the consumer facility by analyzing the first measured energy data by the second artificial intelligence model.

[0021] The step of determining the second artificial intelligence model may include: a step of extracting a representative waveform of the first measured energy data; and a step of classifying the control scheme of the consumer facility by performing clustering on the representative waveform by comparing the representative waveform with a plurality of reference waveforms set by the GMM-based artificial intelligence model.

[0022] The above-mentioned classification step may include a step of classifying the control scheme of the user facility by determining the similarity between the waveforms through analyzing the distance distribution between the representative waveform and the reference waveform based on dynamic time warping by the GMM-based first artificial intelligence model.

[0023] The above classification step comprises: a step of normalizing the first measured energy data and removing outliers from the first measured energy data based on a set value set in relation to the consumer facility; a step of applying a time window having a set time interval to the first measured energy data and calculating the number of waveform peaks in the time interval within the time window while sliding the time window along the time axis to search for a time interval for extracting a representative waveform; a step of comparing the number of waveform peaks corresponding within the time window with a preset range of peak counts, and extracting data from the time interval corresponding to the time window that satisfies the preset number of waveform peaks among the first measured energy data to extract a first representative waveform; and a step of generating a second representative waveform by removing noise data in the first representative waveform where the current harmonic distortion rate deviates from a preset reference range for current harmonic distortion rate or the voltage harmonic distortion rate deviates from a preset reference range for voltage harmonic distortion rate. The method may include a step of classifying the control scheme of the consumer facility by analyzing the distance distribution between the second representative waveform and reference waveforms, which are basic waveforms extracted from actual energy data collected for various facilities, and determining the similarity between the second representative waveform and the reference waveforms.

[0024] The above control scheme may include at least two of various control schemes, including an intake air volume control method, a loading-unloading control method, an automatic dual control method, an internal space variable control method, a variable speed control method, and a blow-off control method.

[0025] The above-mentioned analysis step may include a step of measuring power consumption based on the operating rate and operating pattern per unit time of the compressor by the above-mentioned second artificial intelligence model.

[0026] An artificial intelligence-based energy efficiency service method based on facility usage pattern analysis according to an embodiment of the present invention may further include: receiving multiple supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals by a supplier facility information receiving unit; analyzing supplier facilities related to the energy efficiency of the consumer facilities based on the operating rate and operating pattern of the consumer facilities by an energy efficiency analysis unit; transmitting facility information of one or more first supplier facilities related to the energy efficiency of the consumer facilities and the analyzed energy efficiency information to the consumer terminal by a supplier facility information transmission unit; and requesting the supply of the second supplier facilities to the supplier terminal by a supplier facility supply request unit when the second supplier facilities among the one or more first supplier facilities are selected by the consumer terminal.

[0027] The above-mentioned analysis step may include the step of analyzing the operation pattern by determining the operating time, non-operating time, and operation termination time of the consumer facility based on the energy measured values ​​over time of the first measured energy data by the consumer facility operation information analysis unit, and calculating the operating rate according to the ratio of operating time per unit time.

[0028] An energy efficiency service method based on an artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention may further include: a step of predicting operating time energy efficiency information by the energy efficiency analysis unit based on a value obtained by weighting the operating rate to the difference between a first operating time energy measurement value occurring during the operating time among the first actual energy data and a second operating time energy prediction value occurring during the operating time of the supplier facility; a step of predicting non-operating time energy efficiency information by the energy efficiency analysis unit based on a value obtained by weighting the ratio of the non-operating time per unit time to the difference between a first non-operating time energy measurement value occurring during the non-operating time among the first actual energy data and a second non-operating time energy prediction value occurring during the non-operating time of the supplier facility; and a step of generating energy efficiency information for the consumer facility based on the operating time energy efficiency information and the non-operating time energy efficiency information by the energy efficiency analysis unit.

[0029] An energy efficiency service method based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention may further include: a step of receiving second actual energy data related to the second supplier facility from the consumer terminal by the consumer facility information receiving unit after the consumer facility is replaced with the second supplier facility; and a step of calculating an actual energy efficiency value by comparing the first actual energy data and the second actual energy data by the settlement unit, and calculating a cost for the energy efficiency service based on the actual energy efficiency value. Effects of the invention

[0030] According to an embodiment of the present invention, an artificial intelligence-based facility usage pattern analysis-based energy efficiency service system and method are provided, which analyze the usage patterns of user facilities based on artificial intelligence and provide an energy efficiency service.

[0031] In addition, according to an embodiment of the present invention, an AI-based energy efficiency service system and method based on equipment usage pattern analysis are provided, which quantitatively analyzes equipment operating rates and operating patterns by AI based on actual energy data of consumers.

[0032] In addition, according to an embodiment of the present invention, an artificial intelligence-based energy efficiency service system and method based on equipment usage pattern analysis that matches equipment users and equipment suppliers based on equipment operating rates and operating patterns analyzed by artificial intelligence are provided.

[0033] In addition, according to an embodiment of the present invention, an artificial intelligence-based energy efficiency service system and method based on equipment usage pattern analysis are provided, which facilitates the sharing of equipment-related information and the process of purchasing equipment based on information analyzed by artificial intelligence. Brief explanation of the drawing

[0034] FIG. 1a is a configuration diagram of an energy efficiency service system based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. FIG. 1b is an example of a representative waveform extracted from first measured energy data according to an embodiment of the present invention. FIG. 1c is a diagram showing the clustering results of representative waveforms extracted according to an embodiment of the present invention. Figure 1d is a conceptual diagram showing a method for calculating similarity between waveforms. FIG. 2 is a configuration diagram of an energy management server that constitutes an energy efficiency service system based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. FIG. 3 is a configuration diagram of a consumer facility operation information analysis unit and an energy efficiency analysis unit constituting an energy management system according to an embodiment of the present invention. FIG. 4 is a flowchart illustrating an energy efficiency service method based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. FIG. 5 is an example of first measured energy data of a consumer facility to explain the operation pattern analysis process of the consumer facility operation information analysis unit constituting an energy management system according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating the settlement process of an energy management method according to an embodiment of the present invention. FIG. 7 is a diagram illustrating the settlement process of an energy management method according to an embodiment of the present invention, and is an example of second actual energy data of a consumer facility after being replaced with a supplier facility. FIG. 8 is an illustrative diagram for explaining a measurement data transmission method according to an embodiment of the present invention. Specific details for implementing the invention

[0035] The advantages and features of the present invention and the methods for achieving them will become clear by referring 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 various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components.

[0036] In this specification, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. As used in this specification, "module," "part," and "model" refer to a unit that processes at least one function or operation, and may refer to hardware components such as, for example, software, an FPGA, or one or more processors. In describing embodiments of the present invention, if it is determined that a detailed description of related known functions or known configurations may unnecessarily obscure the essence of the present invention, such detailed description is omitted.

[0037] An energy efficiency service system and method based on an artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention receives first actual energy data regarding a user facility currently in use by a user from at least one user terminal associated with the user, and analyzes the operating rate and operating pattern per unit time of the user facility based on the first actual energy data regarding the user facility for the energy efficiency of the user facility. In the present invention, the operating rate and operating pattern of the user facility can be analyzed by a process of determining a second artificial intelligence model corresponding to the type of user facility by analyzing the type of user facility based on the first actual energy data using a first artificial intelligence model, and by analyzing the first actual energy data using the second artificial intelligence model determined according to the analyzed type of user facility to analyze the operating rate and operating pattern per unit time of the user facility. According to an embodiment of the present invention, by determining different artificial intelligence models according to the type of user facility and analyzing the operating rate and operating pattern per unit time of the user facility, a service for energy efficiency tailored to the type of user facility can be provided.

[0038] FIG. 1a is a configuration diagram of an energy efficiency service system based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. Referring to FIG. 1a, the energy efficiency service system (10) according to an embodiment of the present invention may include at least one user terminal (100) and an energy management server (300).

[0039] The user terminal (110, 120, 130) is a user terminal, for example, of a factory or building, and can be configured to request various energy-related services from the energy management server (300), such as energy efficiency of the equipment, energy monitoring, search for supplier information and / or supplier matching.

[0040] When the energy efficiency service system (10) is configured to provide services such as searching for supplier information or matching consumers and suppliers, the energy efficiency service system (10) may further include at least one supplier terminal (200). The supplier terminal (210, 220, 230) is, for example, a terminal of a supplier that supplies various facilities used by consumers such as factories or buildings, and may be configured to provide information related to the facilities provided by the supplier to the energy management server (300) and to request energy-related services such as searching for consumer information and matching consumers.

[0041] The energy management server (300) can provide various energy efficiency services, such as analyzing the operating rate and operating pattern per unit of time of the user's equipment to improve the energy efficiency of the equipment used by the user, and additionally matching the user with the supplier, allowing the user to search for supplier equipment information, or allowing the supplier to search for user equipment information.

[0042] To this end, the energy management server (300) can receive first actual energy data regarding the user equipment (140) used by the user in a factory, building, etc. from a user terminal (100) related to the user. The energy management server (300) can automatically analyze the type of the user equipment (140) based on the first actual energy data obtained through remote collection by the first artificial intelligence model and the second artificial intelligence model, and analyze the operating rate and operating pattern per unit time.

[0043] As such, the energy management server (300) according to the embodiment of the present invention analyzes the operating rate and operating pattern per unit time for the user facility (140) using the first artificial intelligence model and the second artificial intelligence model, and thus can provide an optimal energy efficiency plan by accurately analyzing the operating rate and operating pattern per unit time for the user facility (140) without separately receiving user facility information related to the user facility (140).

[0044] The energy management server (300) can analyze the type of consumer facility (140) based on the first actual energy data by the first artificial intelligence model, and determine a second artificial intelligence model corresponding to the analyzed type of consumer facility (140) in order to derive an accurate operating rate and operating pattern per unit time for the consumer facility (140).

[0045] The second artificial intelligence model is an artificial intelligence model for analyzing the operating rate and operating pattern per unit time of the consumer facility (140) from the first measured energy data related to the consumer facility (140), and may include a data processing / analysis algorithm for the first measured energy data and various parameters for data processing / analysis. The energy management server (300) can analyze the operating rate and operating pattern per unit time of the consumer facility by analyzing the first measured energy data using the second artificial intelligence model.

[0046] In an embodiment of the present invention, the first artificial intelligence model may be a Gaussian mixture model (GMM)-based artificial intelligence model. In one embodiment, the first artificial intelligence model may classify the control scheme of the consumer facility (140) by extracting a representative waveform of the first measured energy data and performing clustering on the representative waveform by comparing the representative waveform with a plurality of reference waveforms set by the GMM-based artificial intelligence model.

[0047] In an embodiment, the first artificial intelligence model first normalizes the first measured energy data to have a set normalization range, such as 0 to 1, and can remove outliers from the first measured energy data based on set criteria related to consumer facilities, such as compressor capacity.

[0048] In addition, the first artificial intelligence model can apply a time window having a set time interval to the first measured energy data and calculate the number of waveform peaks in the time interval within the time window while sliding the time window along the time axis to search for a time interval for extracting a representative waveform.

[0049] The first artificial intelligence model can extract data from the time interval corresponding to the time window satisfying the set number of waveform peaks within the time window as a representative waveform when the number of waveform peaks within the time window satisfies a preset range of peaks (e.g., 3 or more, 4 or fewer, or 4 or more, 5 or fewer, etc.) among the first measured energy data.

[0050] In addition, the first artificial intelligence model can remove the noise data when, for example, the current harmonic distortion rate deviates from the preset current harmonic distortion rate reference range (e.g., 20% or more) or the voltage harmonic distortion rate deviates from the preset voltage harmonic distortion rate reference range (e.g., 5% or more).

[0051] FIG. 1b is an example of representative waveforms extracted from first measured energy data according to an embodiment of the present invention. FIG. 1c is a diagram showing the clustering results of representative waveforms extracted according to an embodiment of the present invention. FIG. 1d is a conceptual diagram showing a method for calculating similarity between waveforms.

[0052] Referring to FIGS. 1a to 1d, the first artificial intelligence model can classify the control scheme of the consumer facility (140) by determining similarity between waveforms through dynamic time warping based on a GMM-based artificial intelligence model, by analyzing the distance distribution between a representative waveform extracted from the first actual energy data and a preset reference waveform (a statistically processed waveform of energy data acquired for various facilities) (Shapelet s1, s2).

[0053] The reference waveform (Shapelet s1, s2) may be a basic waveform extracted from actual energy data collected for equipment such as a specific compressor. In the example illustrated in FIG. 1c, the types of consumer equipment are classified into four types (Class 1, 2, 3, 4). The horizontal and vertical axes of FIG. 1c represent the distance d(x, s1) between the representative waveform extracted for consumer equipment and the first reference waveform (Shapelet s1), and the distance d(x, s2) between the representative waveform extracted for consumer equipment and the second reference waveform (Shapelet s2), respectively.

[0054] At this time, since the lengths of the two waveforms are different, the similarity between the two waveforms can be analyzed in a high dimension using the kernel trick to determine the type of user equipment (140) corresponding to the representative waveform, for example, the control scheme. The similarity between the waveforms can be analyzed using a Dynamic Time Warping Matching algorithm based on an encoder-decoder supervised learning algorithm or a Gaussian mixture model unsupervised learning algorithm, as exemplified in FIG. 1d, and other algorithms such as the Euclidean Matching algorithm may also be used.

[0055] The control scheme of the consumer facility (140) may include at least two of various control schemes, for example, an inlet modulation method, a loading / unloading control method, an auto dual control method, a variable displacement control method, a variable speed control method (motor speed control method), and a blow-off control method using a blow-off valve in a centrifugal compressor, and one or more control schemes corresponding to the consumer facility (140) may be determined from among these various control schemes.

[0056] In one embodiment, the consumer facility (140) may include a compressor. The energy management server (300) can measure the power consumption of the consumer facility (140) based on the unit time operation rate and operation pattern of the compressor corresponding to the consumer facility (140) by the second artificial intelligence model.

[0057] FIG. 2 is a configuration diagram of an energy management server that constitutes an energy efficiency service system based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. Referring to FIG. 1a and FIG. 2, the energy management server (300) may include a supplier facility information receiving unit (310), a consumer facility information receiving unit (320), a consumer facility operation information analysis unit (330), an energy efficiency analysis unit (340), a supplier facility information transmission unit (350), a supplier facility supply request unit (360), a settlement unit (370), and a control unit (380).

[0058] In one embodiment, the energy management server (300) is provided as a cloud energy management server to reduce data collection and communication costs, so that it can collect data on a cloud basis from a meter (150) of a consumer terminal (100) connected to a consumer facility (140), or receive consumer facility information and supplier facility information in another way. The meter (150) may include a remote terminal unit (RTU) for electricity meters, a mobile energy meter, and / or SCADA data.

[0059] The consumer facility information may include actual energy data (first actual energy data) measured for the consumer facility (140). In addition, the consumer facility information may further include facility information such as the specifications, type, and year of manufacture of the consumer facility (140) in addition to the actual energy data.

[0060] When additional information about the equipment of the consumer facility (140) is provided along with the actual energy data of the consumer facility (140), the energy management server (300) can identify the type of the consumer facility (140) based on the information about the consumer facility (140) and analyze the energy efficiency information of the consumer facility (140).

[0061] In contrast, for consumer facilities (140) for which facility information of the consumer facilities (140) is not provided, the energy management server (300) can identify the type of consumer facilities (140) using the first artificial intelligence model and the second artificial intelligence model described above and analyze energy efficiency information. Examples of consumer facilities (140) include compressors, compressors, motors, inverters, pumps, extruders, etc., but are not limited thereto.

[0062] When the energy management server (300) additionally provides services such as consumer-supplier matching, the energy management server (300) may receive supplier equipment information related to supplier equipment supplied by each supplier from a plurality of supply terminals (200). The supplier equipment information may include equipment information such as specifications, type, year of manufacture, rated voltage / current during operation, and power consumption during non-operation of the supplier equipment supplied by the supplier.

[0063] The supplier facility information receiving unit (310) may be configured to receive multiple supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals (200).

[0064] The user facility information receiving unit (320) can receive first actual energy data regarding the user facility (140) currently being used by the user from at least one user terminal (100) associated with the user. The user facility information receiving unit (320) may additionally receive user facility information in addition to the first actual energy data.

[0065] The consumer facility operation information analysis unit (330) may be configured to analyze the operation rate and operation pattern per unit time of the consumer facility based on the first measured energy data regarding the consumer facility (140). The consumer facility operation information analysis unit (330) may analyze the type of the consumer facility (140) based on the first measured energy data by the first artificial intelligence model and determine a second artificial intelligence model corresponding to the analyzed type of the consumer facility (140).

[0066] In an embodiment of the present invention, the first artificial intelligence model for analyzing the type of consumer facility (140) may be a Gaussian mixture model-based artificial intelligence model. The input data of the first artificial intelligence model is actual energy data of the consumer facility, and the output data may be the type of the consumer facility.

[0067] In addition to GMM-based AI models, the first AI model can also be implemented as supervised AI models such as Support Vector Machine (SVM), Long Short-Term Memory (LSTM), and LSTM-CNN / RNN (LSTM Convolutional / Recurrent Neural Network), as well as unsupervised AI models such as Variational Autoencoder, Adversarial Autoencoder, and MAD-GAN (Multivariate Anomaly Detection with Generative Adversarial Network).

[0068] The first artificial intelligence model of the consumer facility operation information analysis unit (330) can classify the control scheme of the consumer facility by extracting a representative waveform of the first measured energy data and comparing the extracted representative waveform with a set number of reference waveforms and a GMM-based artificial intelligence model to perform clustering on the representative waveform.

[0069] In an embodiment, the first artificial intelligence model can classify the control scheme of a consumer facility by determining the similarity between waveforms through an analysis of the distance distribution between a representative waveform extracted from actual energy data of the consumer facility based on dynamic time warping by a GMM-based artificial intelligence model and reference waveforms collected for each of the various types of consumer facilities.

[0070] The user facility operation information analysis unit (330) is configured to analyze the first measured energy data by analyzing the unit time operation rate and operation pattern of the user facility using a second artificial intelligence model determined according to the type of the analyzed user facility (140).

[0071] The second artificial intelligence model of the user facility operation information analysis unit (330) can be pre-learned (supervised learning or unsupervised learning) based on learning data corresponding to the type of user facility (140). The input data of the second artificial intelligence model is the actual energy data of the user facility, and the output data is the operating rate and operating pattern per unit time of the facility.

[0072] The operating rate and / or operating pattern of the user facility (140) per unit time analyzed by the user facility operation information analysis unit (330) can be used for predicting energy consumption or power generation, analyzing energy efficiency, analyzing energy usage patterns, analyzing energy optimization scheduling, etc.

[0073] Through the above-described user facility operation information analysis unit (330), analysis data such as air pressure, discharge temperature, operating status, past operating history, total operating time, maintenance time, operating rate, operating pattern, power data, flow rate data, power status, number of motor operations, motor operation time, load operation time, etc. related to the user facility (140) can be obtained.

[0074] In addition, according to an embodiment of the present invention, the usage pattern, operation pattern, energy consumption, potential energy saving amount, and energy efficiency ROI of a user facility (140) such as a compressor can be analyzed or predicted with a high accuracy of about 95% or more (based on TPR, MAPE).

[0075] The energy efficiency analysis unit (340) may be configured to analyze the supplier's equipment related to the energy efficiency of the consumer's equipment (140) based on the operating rate and operating pattern of the consumer's equipment (140) analyzed by the first artificial intelligence model and the second artificial intelligence model of the consumer's equipment operation information analysis unit (330).

[0076] The supplier facility information transmission unit (350) may be configured to transmit facility information of one or more first supplier facilities related to energy efficiency of the consumer facility (140) and analyzed energy efficiency information to the consumer terminal (100).

[0077] The supplier facility supply request unit (360) may be configured to request the supply of the second supplier facility to the supplier terminal when the second supplier facility is selected among one or more first supplier facilities by the user terminal (100).

[0078] The settlement unit (370) can calculate the actual energy efficiency value based on the operating rate and operating pattern of the first actual energy data collected for the user's facility, and calculate the cost of the energy efficiency service based on the actual energy efficiency value.

[0079] The control unit (380) may include a processor that executes a program to provide services such as consumer-supplier matching, consumer / supplier information search, and equipment trading by controlling each component of the energy management server (300).

[0080] FIG. 3 is a configuration diagram of a user facility operation information analysis unit and an energy efficiency analysis unit constituting an energy management system according to an embodiment of the present invention. Referring to FIGS. 1a to 3, the user facility operation information analysis unit (330) may include an operation pattern analysis unit (332) and an operation rate calculation unit (334).

[0081] The operation pattern analysis unit (332) can analyze the operation pattern by determining the operation time, non-operation time, and operation end time of the consumer facility based on the energy measurement values ​​over time of the first actual energy data collected for the consumer facility (140) by the second artificial intelligence model. The operation rate calculation unit (334) can calculate the operation rate according to the ratio of the operation time per unit time.

[0082] The energy efficiency analysis unit (340) may include an operating time energy efficiency prediction unit (342), an non-operating time energy efficiency prediction unit (344), and an energy efficiency information generation unit (346). The operating time energy efficiency prediction unit (342) can predict operating time energy efficiency information based on a value obtained by weighting the operating rate to the difference between the first operating time energy measurement value generated during operating time among the first actual energy data collected for the consumer facility (140) and the second operating time energy prediction value generated during the operating time of the supplier facility.

[0083] The non-operating time energy efficiency prediction unit (344) can predict non-operating time energy efficiency information based on a value obtained by weighting the ratio of non-operating time per unit time to the difference between the first non-operating time energy measurement value generated during non-operating time among the first actual energy data collected for the consumer facility (140) and the second non-operating time energy prediction value generated during non-operating time of the supplier facility.

[0084] The energy efficiency information generation unit (346) can generate energy efficiency information for the consumer facility (140) based on operating time energy efficiency information and non-operating time energy efficiency information. At this time, the artificial intelligence model can analyze the power status (usage pattern) (Loading / Unloading / Stop) of the consumer facility (140) by utilizing an encoder-decoder artificial neural network using an LSTM artificial intelligence model.

[0085] In this way, through the analysis of equipment usage patterns (Loading / Unloading / Stop) based on actual measurement data, it is possible to analyze the Return on Investment (ROI) for energy efficiency improvements resulting from the replacement of existing low-efficiency consumer equipment (e.g., low-efficiency compressors, etc.) with high-efficiency consumer equipment (e.g., high-efficiency compressors, etc.).

[0086] Meanwhile, the consumer equipment information receiving unit (320) can additionally receive second actual energy data related to the second supplier equipment from the consumer terminal (100) after the consumer equipment (140) is replaced with the second supplier equipment (high-efficiency consumer equipment). The remotely collected second actual energy data can be used for energy efficiency verification.

[0087] The settlement unit (370) can calculate an energy efficiency measurement value by comparing the first actual energy data collected for the consumer facility (140) before replacement with the second actual energy data collected for the second supplier facility installed on the consumer side to replace the consumer facility (140), and calculate the cost of the energy efficiency service based on the calculated energy efficiency measurement value.

[0088] FIG. 4 is a flowchart illustrating an energy efficiency service method based on artificial intelligence-based facility usage pattern analysis according to an embodiment of the present invention. Referring to FIGS. 1a to 4, the user facility information receiving unit (320) of the energy management server (300) can receive first measured energy data regarding the user facility (140) currently being used by the user from at least one user terminal (100) (S12, S13).

[0089] The consumer facility operation information analysis unit (330) of the energy management server (300) can analyze the operation rate and operation pattern per unit time of the consumer facility (140) based on the first measured energy data regarding the consumer facility (140) (S14). At this time, if the consumer facility operation information analysis unit (330) has not received type information or facility information of the consumer facility (140), it can analyze the type of the consumer facility (140) based on the first measured energy data by the first artificial intelligence model and determine a second artificial intelligence model corresponding to the analyzed type of the consumer facility (140). The first artificial intelligence model for analyzing the type of the consumer facility (140) may be an artificial intelligence model based on a Gaussian mixture model (GMM).

[0090] The first artificial intelligence model of the consumer facility operation information analysis unit (330) can classify the control scheme of the consumer facility by extracting a representative waveform of the first measured energy data and comparing the extracted representative waveform with a set number of reference waveforms and a GMM-based artificial intelligence model to perform clustering on the representative waveform.

[0091] In the embodiment, the first artificial intelligence model can classify the control scheme of the consumer facility by determining the similarity between waveforms by analyzing the distance distribution between a representative waveform extracted from the actual energy data of the consumer facility based on dynamic time warping by a GMM-based artificial intelligence model and a reference waveform collected for each of the various types of consumer facilities. The consumer facility operation information analysis unit (330) can analyze the operation rate and operation pattern per unit time of the consumer facility by analyzing the first actual energy data by a second artificial intelligence model determined according to the type of the analyzed consumer facility (140).

[0092] FIG. 5 is an example of first measured energy data of a consumer facility to explain the operation pattern analysis process of the consumer facility operation information analysis unit constituting an energy management system according to an embodiment of the present invention. Referring to FIGS. 1a to 5, the operation pattern analysis unit (332) of the consumer facility operation information analysis unit (330) can analyze the operation pattern by determining the operation time (T1), non-operation time (T2), and operation end time (T3) of the consumer facility (140) by a second artificial intelligence model based on the energy measured values ​​over time of the first measured energy data (20) collected for the consumer facility (140). The operation rate calculation unit (334) of the consumer facility operation information analysis unit (330) can calculate the operation rate according to the ratio of the operation time (T1) per unit time (T0).

[0093] During the operating time (T1), the amount of power generated by the consumer facility (140) is equal to the operating power (P1). During the non-operating time (T2), the amount of power generated by the consumer facility (140) is equal to the non-operating power (P2) due to the motor idling of the consumer facility (140), which becomes a cause of unnecessary power consumption. The energy efficiency analysis unit (340) of the energy management server (300) can analyze the supplier facility related to the energy efficiency of the consumer facility (140) based on the operating rate and operating pattern of the consumer facility (140) (S15).

[0094] The operating time energy efficiency prediction unit (342) of the energy efficiency analysis unit (340) can predict operating time energy efficiency information based on the difference between the first operating time energy measurement value (P1) generated during the operating time (T1) among the first actual energy data (20) collected for the consumer facility (140) and the second operating time energy prediction value generated during the operating time of the supplier facility (200), with the operating rate weighted.

[0095] The non-operating time energy efficiency prediction unit (344) of the energy efficiency analysis unit (340) can predict non-operating time energy efficiency information based on a value obtained by weighting the ratio of non-operating time per unit time to the difference between the first non-operating time energy measurement value generated during non-operating time (T2) among the first actual energy data (20) collected for the consumer facility (140) and the second non-operating time energy prediction value generated during non-operating time of the supplier facility (200).

[0096] The energy efficiency information generation unit (346) is in the operating time energy efficiency prediction unit (342). Energy efficiency information for the user facility (140) can be generated based on the operating time energy efficiency information calculated by and the non-operating time energy efficiency information predicted by the non-operating time energy efficiency prediction unit (344).

[0097] Meanwhile, the energy efficiency analysis unit (340) can generate energy efficiency information by reflecting energy efficiency information based on seasonal deviation (change). For example, additional energy may be generated to cool the compressor in the summer, and energy efficiency information for the summer season, etc., can be predicted by correcting data measured in the winter season, etc.

[0098] Meanwhile, the supplier facility information receiving unit (310) of the energy management server (300) may receive supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals (200) (S11). The supplier facility information transmission unit (350) of the energy management server (300) may transmit facility information of one or more first supplier facilities related to energy efficiency of the consumer facility (140) and analyzed energy efficiency information to the consumer terminal (100) (S16). The supplier facility supply request unit (360) of the energy management server (300) may request the supply of the second supplier facility to the supplier terminal when the second supplier facility among one or more first supplier facilities is selected by the consumer terminal (100) (S17, S18, S19).

[0099] FIG. 6 is a flowchart for explaining the settlement process of an energy management method according to an embodiment of the present invention. The consumer facility information receiving unit (320) can receive second actual energy data related to the second supplier facility from the consumer terminal (100) after the consumer facility (140) is replaced with the second supplier facility (S21, S22).

[0100] The settlement unit (370) can calculate an energy efficiency measurement value by comparing the first actual energy data collected for the consumer facility (140) before replacement with the second actual energy data collected for the second supplier facility installed on the consumer side to replace the consumer facility (140), calculate the cost of the energy efficiency service based on the calculated energy efficiency measurement value, transmit evaluation information for the supplier facility to the supplier terminal, and transmit ROI-based settlement information to the consumer terminal (100) (S23, S24, S25).

[0101] FIG. 7 is a diagram illustrating the settlement process of an energy management method according to an embodiment of the present invention, and is an example of second measured energy data of a consumer facility after being replaced with a supplier facility. For example, as the inverter of the consumer facility is changed from a Fixed Speed ​​Drive (FSD) to a Variable Speed ​​Drive (VSD) corresponding to the supplier facility, the power consumption of the consumer facility may be reduced compared to before the facility replacement.

[0102] Referring to FIG. 7, the operation pattern analysis unit (332) of the consumer facility operation information analysis unit (330) can analyze the operation pattern by determining the operation time (T1), non-operation time (T2), and operation end time (T3) of the consumer facility (140) based on the energy measured over time of the second actual energy data (30) collected for the consumer facility (140) after replacement. The operation rate calculation unit (334) of the consumer facility operation information analysis unit (330) can calculate the operation rate according to the ratio of the operation time (T1) per unit time (T0).

[0103] After the equipment replacement, during the operating time (T1), the power consumption of the consumer equipment (140) is reduced to a second operating power (P3) compared to the first operating power (P1) before the equipment replacement. During the non-operating time (T2), the second non-operating power (P4) generated due to the motor idling of the consumer equipment (140) is significantly reduced compared to the non-operating power (P2) before the replacement.

[0104] Accordingly, the energy consumption of the consumer facility (140) can be reduced by the amount of reduced operating energy (40) during the operating time (T1), and the energy consumption of the consumer facility (140) can be reduced by the amount of reduced non-operating energy (50) during the non-operating time (T2). As previously explained, the energy efficiency analysis unit (340) of the energy management server (300) can analyze the supplier facility related to the energy efficiency of the consumer facility (140) based on the operating rate and operating pattern of the consumer facility (140) (S15).

[0105] The operating time energy efficiency prediction unit (342) of the energy efficiency analysis unit (340) can predict operating time energy efficiency information based on the difference between the first operating time energy measurement value (P1) generated during the operating time (T1) among the first actual energy data (20) collected for the consumer facility (140) and the second operating time energy prediction value generated during the operating time of the supplier facility (200), with the operating rate weighted.

[0106] The non-operating time energy efficiency prediction unit (344) of the energy efficiency analysis unit (340) can predict non-operating time energy efficiency information based on a value obtained by weighting the ratio of non-operating time per unit time to the difference between the first non-operating time energy measurement value generated during non-operating time (T2) among the first actual energy data (20) collected for the consumer facility (140) and the second non-operating time energy prediction value generated during non-operating time of the supplier facility (200).

[0107] The energy efficiency information generation unit (346) is in the operating time energy efficiency prediction unit (342). Energy efficiency information for the user facility (140) can be generated based on the operating time energy efficiency information calculated by and the non-operating time energy efficiency information predicted by the non-operating time energy efficiency prediction unit (344).

[0108] FIG. 8 is an illustrative diagram for explaining a measurement data transmission method according to an embodiment of the present invention. Referring to FIG. 8, measurement data (actual energy data) measured for a consumer facility (140) can be collected by an energy management server (300) according to a preset measurement data transmission cycle.

[0109] At this time, the measurement data transmission cycle can be varied based on the type of user facility (140) predicted for the user facility (140) and / or the state change cycle generated by analysis by statistical processing based on the data measured for the user facility (140) or by analysis by an AI model.

[0110] The energy management server (300) can adjust the transmission period of measurement data transmitted from the measuring instrument (150) so that it becomes smaller than the state change cycle (or state change prediction cycle) analyzed for the consumer facility (140). At this time, the energy management server (300) can set the transmission period of measurement data to be at least half of the state change cycle (or state change prediction cycle) analyzed for the consumer facility (140) so that data is not transmitted at a shorter period than necessary.

[0111] Additionally, the energy management server (300) can monitor the status change of the consumer facility (140) in real time by means of a meter (150), and can collect actual energy data transmitted in the form of an event from the meter (150) when a change in the power status of the consumer facility (140) is detected. Accordingly, by recognizing the status change of the consumer facility (140) in real time, the operation pattern or energy efficiency information regarding the consumer facility (140) can be analyzed with higher accuracy.

[0112] At least some of the configurations of the embodiments described above may be implemented as hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an Arithmetic Logic Unit (ALU), a Digital Signal Processor, a microcomputer, a Field Programmable Gate Array (FPGA), a Programmable Logic Unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions.

[0113] The processing unit may execute an operating system and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of understanding, the processing unit may be described as being used as a single unit, but a person of ordinary skill in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements.

[0114] For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as a parallel processor, are also possible. Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure the processing unit to operate as desired or instruct the processing unit independently or collectively.

[0115] Software and / or data may be embodied in any type of machine, component, physical device, virtual equipment, computer storage medium, or device so as to be interpreted by a processing device or to provide instructions or data to a processing device. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0116] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software.

[0117] Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operation of the embodiments, and vice versa.

[0118] Although the embodiments have been described above with reference to limited examples and drawings, those skilled in the art can make various modifications and variations from the description above. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents. Therefore, other implementations, other embodiments, and equivalents to the claims are also included within the scope of the claims set forth below. Explanation of the symbols

[0119] 100, 110, 120, 130: User terminal 140 : User Facilities 150 : Measuring instrument 200, 210, 220, 230 : Supplier terminal 300 : Energy Management Server 310 : Supplier Equipment Information Receiving Unit 320 : User Facility Information Receiving Unit 330 : User Facility Operation Information Analysis Department 332 : Operation Pattern Analysis Unit 334 : Operating Rate Calculation Section 340 : Energy Efficiency Analysis Department 342 : Operating Time Energy Efficiency Prediction Section 344 : Non-operating Time Energy Efficiency Prediction Unit 346 : Energy Efficiency Information Generation Unit 350 : Supplier Equipment Information Transmission Unit 360 : Supplier Facility Supply Request Section 370 : Settlement Department 380 : Control unit

Claims

Claim 1 A consumer facility information receiving unit configured to receive first actual energy data regarding consumer facilities being used by a consumer from at least one consumer terminal related to a consumer of a factory or building; a consumer facility operation information analysis unit configured to analyze the operating rate and operation pattern per unit time of the consumer facilities based on the first actual energy data regarding the consumer facilities; a supplier facility information receiving unit configured to receive multiple supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals; an energy efficiency analysis unit configured to analyze supplier facilities related to the energy efficiency of the consumer facilities based on the operating rate and operation pattern of the consumer facilities; and a supplier facility information transmission unit configured to transmit facility information of one or more first supplier facilities related to the energy efficiency of the consumer facilities and the analyzed energy efficiency information to the consumer terminal. The system includes a supplier facility supply request unit configured to request the supply of the second supplier facility to the supplier terminal when the second supplier facility among the one or more first supplier facilities is selected by the consumer terminal, and the consumer facility operation information analysis unit comprises: analyzing the type of the consumer facility based on the first measured energy data by a first artificial intelligence model and determining a second artificial intelligence model corresponding to the analyzed type of the consumer facility; and analyzing the operating rate and operating pattern per unit time of the consumer facility by analyzing the first measured energy data by the second artificial intelligence model, and the first artificial intelligence model comprises: extracting a representative waveform of the first measured energy data;An artificial intelligence-based energy efficiency service system based on facility usage pattern analysis, configured to classify the control scheme of the consumer facility by performing clustering on the representative waveform with a plurality of reference waveforms set by an artificial intelligence model based on a Gaussian mixture model (GMM), wherein the control scheme includes at least two of various control schemes including an intake air volume control method, a loading-unloading control method, an automatic dual control method, an internal space variable control method, a variable speed control method, and a blow-off control method. Claim 2 delete Claim 3 delete Claim 4 An energy efficiency service system based on facility usage pattern analysis according to claim 1, wherein the first artificial intelligence model is configured to classify the control scheme of the consumer facility by determining similarity between waveforms through analyzing the distance distribution between the representative waveform and the reference waveform based on dynamic time warping by the GMM-based artificial intelligence model. Claim 5 An artificial intelligence-based energy efficiency service system based on facility usage pattern analysis, wherein the consumer facility includes a compressor, and the consumer facility operation information analysis unit is configured to measure power consumption based on the unit time operation rate and operation pattern of the compressor by the second artificial intelligence model. Claim 6 delete Claim 7 An artificial intelligence-based energy efficiency service system based on equipment usage pattern analysis, wherein the energy efficiency analysis unit determines the operating time, non-operating time, and operation termination time of the consumer equipment based on the energy measured value over time of the first measured energy data, analyzes the operating pattern, and calculates the operating rate according to the ratio of the operating time per unit time. Claim 8 In claim 7, the energy efficiency analysis unit is configured to: predict operating time energy efficiency information based on a value obtained by weighting the operating rate to the difference between a first operating time energy actual value generated during the operating time among the first actual energy data and a second operating time energy predicted value generated during the operating time of the supplier facility; predict non-operating time energy efficiency information based on a value obtained by weighting the ratio of the non-operating time per unit time to the difference between a first non-operating time energy actual value generated during the non-operating time among the first actual energy data and a second non-operating time energy predicted value generated during the non-operating time of the supplier facility; and generate energy efficiency information of the consumer facility based on the operating time energy efficiency information and the non-operating time energy efficiency information, an artificial intelligence-based facility usage pattern analysis-based energy efficiency service system. Claim 9 The artificial intelligence-based facility usage pattern analysis-based energy efficiency service system according to claim 7, wherein the user facility information receiving unit is configured to receive second actual energy data related to the second supplier facility from the user terminal after the user facility is replaced with the second supplier facility, and further comprises a settlement unit configured to calculate an actual energy efficiency value by comparing the first actual energy data and the second actual energy data, and to calculate a cost according to the energy efficiency service based on the actual energy efficiency value. Claim 10 A step of receiving first actual energy data regarding a user facility being used by a user from at least one user terminal related to a user of a factory or building by a user facility information receiving unit; a step of analyzing the operating rate and operating pattern per unit time of the user facility based on the first actual energy data regarding the user facility by a user facility operation information analysis unit; a step of receiving multiple supplier facility information related to multiple different supplier facilities provided by multiple suppliers from multiple supplier terminals by a supplier facility information receiving unit; a step of analyzing supplier facilities related to the energy efficiency of the user facility based on the operating rate and operating pattern of the user facility by an energy efficiency analysis unit; and a step of transmitting facility information of one or more first supplier facilities related to the energy efficiency of the user facility and the analyzed energy efficiency information to the user terminal by a supplier facility information transmission unit. The method includes the step of requesting the supply of the second supplier facility to the supplier terminal when the second supplier facility is selected among the one or more first supplier facilities by the supplier facility supply request unit, and the step of analyzing by the consumer facility operation information analysis unit includes: the step of analyzing the type of the consumer facility based on the first measured energy data by the first artificial intelligence model and determining a second artificial intelligence model corresponding to the analyzed type of the consumer facility; and the step of analyzing the operating rate and operating pattern per unit time of the consumer facility by analyzing the first measured energy data by the second artificial intelligence model, and the step of determining the second artificial intelligence model includes: the step of extracting a representative waveform of the first measured energy data;An energy efficiency service method based on AI-based facility usage pattern analysis, comprising the step of classifying the control scheme of the consumer facility by performing clustering on the representative waveform by comparing the representative waveform with a plurality of reference waveforms set by an AI model based on GMM, wherein the control scheme includes at least two of various control schemes including an intake air volume control method, a loading-unloading control method, an automatic dual control method, an internal space variable control method, a variable speed control method, and a blow-off control method. Claim 11 delete Claim 12 In claim 10, the classifying step comprises: a step of classifying the control scheme of the consumer facility by determining similarity between waveforms through analysis of the distance distribution between the representative waveform and the reference waveform based on dynamic time warping by the GMM-based first artificial intelligence model. Claim 13 In claim 10, the classifying step comprises: a step of normalizing the first measured energy data and removing outliers from the first measured energy data based on a preset value in relation to the consumer facility; a step of applying a time window having a preset time interval to the first measured energy data and calculating the number of waveform peaks in the time interval within the time window while sliding the time window along the time axis to search for a time interval for extracting a representative waveform; a step of comparing the number of waveform peaks corresponding within the time window with a preset range of peak counts, and extracting the data of the time interval corresponding to the time window that satisfies the preset number of waveform peaks among the first measured energy data to extract a first representative waveform; and a step of generating a second representative waveform by removing noise data in the first representative waveform where the current harmonic distortion rate deviates from a preset reference range for current harmonic distortion rate or the voltage harmonic distortion rate deviates from a preset reference range for voltage harmonic distortion rate. An artificial intelligence-based energy efficiency service method based on facility usage pattern analysis, comprising the step of classifying the control scheme of the consumer facility by analyzing the distance distribution between the second representative waveform and reference waveforms, which are basic waveforms extracted from actual energy data collected for various facilities, and determining the similarity between the second representative waveform and the reference waveforms. Claim 14 A method for an energy efficiency service based on artificial intelligence-based facility usage pattern analysis, wherein, in claim 10, the consumer facility includes a compressor, and the analyzing step includes the step of measuring power consumption based on the operating rate and operating pattern per unit time of the compressor by the second artificial intelligence model. Claim 15 delete Claim 16 In claim 10, the analyzing step comprises: analyzing the operating pattern by determining the operating time, non-operating time, and operating termination time of the consumer facility based on the energy measured value according to time of the first measured energy data by the consumer facility operation information analysis unit, and calculating the operating rate according to the ratio of the operating time per unit time, an artificial intelligence-based facility usage pattern analysis-based energy efficiency service method. Claim 17 The method of claim 16 further comprises: a step of predicting operating time energy efficiency information by the energy efficiency analysis unit based on a value obtained by weighting the operating rate to the difference between a first operating time energy actual value occurring during the operating time among the first actual energy data and a second operating time energy predicted value occurring during the operating time of the supplier facility; a step of predicting non-operating time energy efficiency information by the energy efficiency analysis unit based on a value obtained by weighting the ratio of the non-operating time per unit time to the difference between a first non-operating time energy actual value occurring during the non-operating time among the first actual energy data and a second non-operating time energy predicted value occurring during the non-operating time of the supplier facility; and a step of generating energy efficiency information of the consumer facility based on the operating time energy efficiency information and the non-operating time energy efficiency information by the energy efficiency analysis unit. Claim 18 The method of claim 17 further comprises the steps of: receiving second actual energy data related to the second supplier facility from the consumer terminal by the consumer facility information receiving unit after the consumer facility is replaced with the second supplier facility; and calculating an actual energy efficiency value by comparing the first actual energy data and the second actual energy data by the settlement unit, and calculating a cost for an energy efficiency service based on the actual energy efficiency value. Claim 19 A computer-readable recording medium having a program recorded thereon for executing an artificial intelligence-based facility usage pattern analysis-based energy efficiency service method described in any one of claims 10, 12 to 14, and 16 to 18.

Images