Apparatus for supporting carbon reduction of video device using lstm time-series analysis and explainable artificial intelligence and method thereof
The carbon reduction support device uses AI LSTM and XAI to learn normal consumption patterns and adaptively control imaging device parameters, addressing inefficiencies in conventional methods by reducing power and carbon emissions through early detection and adaptive control.
Patent Information
- Application Number
- KR1020260072570
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2026-04-22
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2046-04-22
AI Technical Summary
Conventional video device management methods fail to detect early deviations from normal power consumption patterns, leading to increased energy costs and carbon emissions, and lack sufficient grounds for determining the cause of power consumption variations due to device aging, environmental changes, or operational conditions.
A carbon reduction support device using AI LSTM time series analysis and explainable AI (XAI) learns normal consumption patterns from power, environmental, and operational data to detect abnormalities and adaptively control imaging device parameters, reducing power consumption and emissions.
Early detection of abnormal power consumption allows for active parameter adjustments, reducing unnecessary energy use and emissions, enhancing device stability and management efficiency while providing clear explanations for the causes of increased consumption.
Smart Images

Figure 112026049113011-PAT00003_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a carbon reduction support device for an imaging device using artificial intelligence LSTM time series analysis and explainable XAI, and more specifically, to a carbon reduction support device for an imaging device using artificial intelligence LSTM time series analysis and explainable XAI that learns normal consumption patterns based on power data, environmental data, and operation data collected from an imaging device, and controls operation parameters of the imaging device using the same to reduce power consumption and carbon emissions. Background Technology
[0002] With the recent expansion of public safety, facility security, industrial site monitoring, and smart city development, various video devices, including CCTVs, are being installed and operated in large quantities. Since these video devices typically operate continuously for long periods or 24 hours, they consume a significant amount of cumulative power, which can lead to increased energy costs and carbon emissions. In particular, in environments where multiple video devices are installed in a distributed manner, it is difficult to continuously monitor the power status or inefficient operating parameters of individual devices.
[0003] Conventional video device management methods generally relied on a reactive approach of identifying and addressing issues after they occurred, such as device failure, overpower, leakage, communication failure, or degradation of video quality. However, this method has limitations in that it is difficult to detect early signs that the device is deviating from normal ranges in advance, appropriate response may be delayed while power waste continues, and in some cases, it may even lead to a risk of fire due to overload or leakage.
[0004] In addition, the power consumption of a video device can vary depending not only on whether the power is simply turned on or off, but also on operating conditions such as resolution, frame rate (FPS), whether video analysis functions are performed, network bandwidth usage, and the operating status of the cooling fan, as well as environmental conditions such as ambient temperature, humidity, and fine dust concentration. For example, in high-temperature environments, power consumption may increase due to increased device load, and in high-humidity environments, the failure rate or error rate may increase. In environments with high fine dust concentration, the operation of the cooling fan needs to be considered in terms of device protection and suppression of contamination spread.
[0005] However, conventional technologies have often been limited to simple threshold-based alarms or fixed-set value controls, failing to adequately reflect differences in installation environments across devices and time-series power consumption characteristics. Furthermore, since they cannot clearly explain to administrators whether the increase in power consumption is attributable to device aging, increased thermal load, increased communication load, environmental changes, or increased image processing load, there was a problem of insufficient grounds for determining why specific controls are necessary in actual operation.
[0006] Therefore, there is an increasing need for technology capable of reducing power consumption and carbon emissions by learning normal consumption patterns for each device based on power, environmental, and operational data collected from imaging devices, detecting abnormal conditions using deviations from normal patterns, analyzing the causes of increased power consumption or carbon emissions using Explainable Artificial Intelligence (XAI), and adaptively controlling the operational parameters of imaging devices according to the analysis results.
[0007] The technology forming the background of the present invention is disclosed in Korean Registered Patent No. 10-2146964 (published August 21, 2020). The problem to be solved
[0008] The present invention aims to solve these problems by providing a carbon reduction support device and method for an imaging device using artificial intelligence LSTM time series analysis and explainable Artificial Intelligence (XAI), which learns normal consumption patterns based on power data, environmental data, and operation data collected from an imaging device, and controls operation parameters of the imaging device using this to reduce power consumption and carbon emissions. means of solving the problem
[0009] According to an embodiment of the present invention, a device for reducing power consumption and carbon emissions of an imaging device comprises: a data collection unit that collects power data, environmental data, and operational data from an imaging device; and an analysis unit that uses the collected power data, environmental data, and operational data to generate a normal consumption pattern of the imaging device using an unsupervised learning-based LSTM (Long Short-Term Memory) model, and calculates a deviation between real-time collected data and the normal consumption pattern to detect an abnormal state.
[0010] The system includes: a control unit that receives the deviation between the real-time collected data and the normal consumption pattern, the cause information, and the environmental data, selects one of a plurality of control combinations, and controls the operating parameters of the imaging device according to the selected control combination; and a providing unit that, regarding the abnormal state, provides the power data, the environmental data, and the operating data to an administrator through explainable artificial intelligence (XAI) regarding the causes contributing to increased power consumption or increased carbon emissions.
[0011] The above environmental data may include at least one of temperature, humidity, and fine dust concentration.
[0012] The above operating parameters may include at least one of resolution, frame rate (FPS), level of video analysis function, network bandwidth, and cooling fan operation.
[0013] The control unit above can reduce at least one of the resolution, frame rate (FPS), and image analysis function level of the image device when the temperature is above a preset reference value.
[0014] The control unit may restrict some functions of the imaging device or switch to a low-power mode in response to the possibility of failure or an increase in the error rate when the humidity is above a preset reference value.
[0015] The control unit above may stop the continuous operation of the cooling fan or switch to intermittent or low-speed operation when the fine dust concentration is above a preset threshold value, and may maintain or resume the operation of the cooling fan when the internal temperature of the imaging device is above a protection threshold value.
[0016] Based on the analysis results, the control unit classifies the cause of increased power consumption into one of an increase in thermal load, an increase in communication load, and the need to suppress the diffusion of fine dust, and can select a different combination of operating parameters according to the classified cause.
[0017] The above control unit can adjust the frame rate (FPS) and cooling fan operation when the temperature or internal temperature is above a first reference value, adjust the resolution and network bandwidth when the network load is above a second reference value, and adjust the cooling fan operation and frame rate (FPS) when the fine dust concentration is above a third reference value.
[0018] The control unit can restore the operating parameters of the imaging device to normal operating parameters when the environment data and the deviation are each maintained for a predetermined time or longer after returning to a preset normal range.
[0019] The above-mentioned providing unit calculates the State of Health (SOH) of the imaging device based on the power data, the environment data, and the operation data, and the above-mentioned control unit can select a control combination with a high power consumption reduction intensity when the state of health is smaller than a preset reference value.
[0020] According to another embodiment of the present invention, a method for reducing carbon emissions of an imaging device, performed by a carbon reduction support device for an imaging device, comprises: collecting power data, environmental data, and operational data from the imaging device; generating a normal consumption pattern of the imaging device using an unsupervised learning-based LSTM (Long Short-Term Memory) model using the collected power data, environmental data, and operational data; detecting an abnormal state by calculating a deviation between real-time collected data and the normal consumption pattern; receiving the deviation between the real-time collected data and the normal consumption pattern, the analysis result, and the environmental data, and selecting one of a plurality of control combinations; controlling operational parameters of the imaging device according to the selected control combination; and regarding the abnormal state, providing to an administrator through explainable artificial intelligence (XAI) the cause that the power data, environmental data, and operational data contribute to an increase in power consumption or carbon emissions. Effects of the invention
[0021] According to the present invention, abnormal power consumption of an imaging device can be detected early and operating parameters can be actively adjusted according to the cause, thereby effectively reducing unnecessary power consumption and carbon emissions.
[0022] In addition, by flexibly responding to various environmental changes and operating conditions, device overload, failures, and performance degradation can be mitigated, and operating parameters can be appropriately restored upon recovery to a normal state, thereby improving the stable operation and management efficiency of the imaging device. Furthermore, by achieving both device protection and energy savings through differential control based on the status of the imaging device, the operational reliability and maintenance efficiency of the imaging device can be enhanced. Brief explanation of the drawing
[0023] FIG. 1 is a schematic diagram showing the interoperability relationship between a carbon reduction support device and an imaging device according to an embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of a carbon reduction support device of an imaging device according to an embodiment of the present invention. FIG. 3 is a flowchart illustrating a carbon reduction support method for an imaging device according to an embodiment of the present invention. Specific details for implementing the invention
[0024] Then, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily implement the invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals.
[0025] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other components interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.
[0026] Embodiments of the present invention are described in detail with reference to the attached drawings so that those skilled in the art can easily implement the invention.
[0027] FIG. 1 is a schematic diagram showing the interoperability relationship between a carbon reduction support device and an imaging device according to an embodiment of the present invention.
[0028] Referring to FIG. 1, a carbon reduction support device (100) according to an embodiment of the present invention is connected to an imaging device (200) and collects power data, environmental data, and operation data from the imaging device (200), learns normal consumption patterns based on this, analyzes the cause of increased power consumption or increased carbon emissions, and controls the operation parameters of the imaging device (200) according to the analysis results.
[0029] The video device (200) may include various types of video acquisition devices such as CCTV, mobile camera, network camera, edge camera, etc., and the carbon reduction support device (100) can reduce power consumption and reduce carbon emissions by adjusting the resolution, frame rate (FPS), level of video analysis function, network bandwidth, and cooling fan operation of such video device (200).
[0030] FIG. 2 is a block diagram showing the configuration of a carbon reduction support device of an imaging device according to an embodiment of the present invention.
[0031] Referring to FIG. 2, a carbon reduction support device (100) of an imaging device according to an embodiment of the present invention may include a data collection unit (110), an analysis unit (120), a control unit (130), and a providing unit (140).
[0032] The data collection unit (110) can collect power data, environmental data, and operation data from the video device (200). Here, the environmental data may include at least one of temperature, humidity, and fine dust concentration, and the operation data may include at least one of resolution, frame rate (FPS), video analysis function level, network bandwidth, and cooling fan operation.
[0033] The analysis unit (120) can use collected power data, environmental data, and operation data to generate a normal consumption pattern of the imaging device (200) using an unsupervised learning-based LSTM (Long Short-Term Memory) model, and can detect an abnormal state by calculating the deviation between the real-time collected data and the normal consumption pattern.
[0034] The control unit (130) receives the deviation between real-time collected data and normal consumption patterns, cause information and environmental data, selects one of a plurality of control combinations, and can control the operation parameters of the imaging device (200) according to the selected control combination.
[0035] For example, the control unit (130) can reduce at least one of the resolution, frame rate (FPS), and image analysis function level when the temperature is above a preset threshold value, can limit some functions or switch to a low-power mode when the humidity is above a preset threshold value, and can stop the continuous operation of the cooling fan or switch to intermittent operation or low-speed operation when the fine dust concentration is above a preset threshold value.
[0036] Additionally, the control unit (130) can maintain or resume the operation of the cooling fan when the internal temperature is above a protection threshold value, and can select a control combination with a high power consumption reduction intensity when the state health is lower than a preset threshold value. Furthermore, the control unit (130) can restore the operating parameters of the imaging device (200) to normal operating parameters when the environment data and the deviation are each maintained for a predetermined time after returning to a preset normal range.
[0037] In addition, the control unit (130) can calculate the State of Health (SOH) of the imaging device (200) based on power data, environmental data, and operation data.
[0038] The providing unit (140) can provide the manager with explanationable artificial intelligence (XAI) regarding the cause of the increase in power consumption or carbon emissions in abnormal conditions, power data, environmental data, and operation data.
[0039] The providing unit (140) can provide such analysis results to the manager in the form of cause information, and, if necessary, can also provide the contribution, priority, or explanatory information of each cause.
[0040] The operation process of the present invention will be explained in more detail below, focusing on each step of FIG. 3.
[0041] FIG. 3 is a flowchart illustrating a carbon reduction support method for an imaging device according to an embodiment of the present invention.
[0042] First, the data collection unit (110) collects power data, environment data, and operation data from the video device (200) (S310).
[0043] Here, power data may include voltage, current, power consumption, active power, reactive power, power ripple, harmonic components, and instantaneous load change amount. Environmental data may include at least one of temperature, humidity, and fine dust concentration, and operational data may include resolution, frame rate (FPS), video analysis function level, network bandwidth, bitrate, and cooling fan operation status.
[0044] First, the data collection unit (110) can collect the data at a predetermined interval, and can also collect additional data when events such as a sudden increase in power consumption, an increase in internal temperature, an increase in network load, or a change in operating parameters occur. In addition, the data collection unit (110) can improve the accuracy of subsequent learning and analysis steps by performing preprocessing such as time alignment, missing value correction, outlier removal, and normalization.
[0045] Next, the analysis unit (120) uses the collected power data, environment data, and operation data to generate a normal consumption pattern of the imaging device (200) using an unsupervised learning-based LSTM model (S320).
[0046] Here, the normal consumption pattern is not a simple average power consumption value or a fixed threshold, but may be a time-series reference pattern that reflects the correlation between the pattern of power consumption change over time and environmental and operating conditions.
[0047] For example, changes in power consumption due to increased resolution or FPS, increased load due to rising ambient or internal temperatures, and changes in processing load due to changes in network bandwidth usage may be reflected in normal consumption patterns.
[0048] Accordingly, normal standards corresponding to the characteristics of each device can be formed for multiple video devices (200) with different installation environments, operating conditions, and time zones. In addition, according to an embodiment of the present invention, since unsupervised learning can be performed using time-series data collected from normal operating parameters without relying on labeled failure data, it has excellent applicability to actual field conditions.
[0049] And, the analysis unit (120) calculates the deviation between real-time collected data and normal consumption patterns to detect an abnormal state (S330).
[0050] Here, the deviation between real-time collected data and normal consumption patterns can be calculated as the difference between predicted power consumption and actual power consumption, the degree of deviation from normal time series patterns, changes in correlations between multivariate features, or a combination thereof.
[0051] For example, if the actual power consumption increases excessively compared to the normal consumption pattern even under the same operating conditions, or if the power usage pattern changes abnormally despite similar environmental conditions such as temperature, humidity, and fine dust concentration, the analysis unit (120) can determine this as an abnormal state. In addition, by considering not only the absolute magnitude of the deviation but also the duration, repetition frequency, and slope of change, temporary fluctuations and actual abnormal states can be distinguished more precisely.
[0052] Next, the control unit (130) receives the deviation between real-time collected data and normal consumption patterns, analysis results, and environmental data, and selects one of a plurality of control combinations (S340).
[0053] First, the control unit (130) analyzes the causes of power data, environmental data, and operation data contributing to increased power consumption or increased carbon emissions regarding abnormal conditions.
[0054] The control unit (130) can quantitatively or qualitatively determine the degree of influence that each of the plurality of input variables has on the increase in power consumption.
[0055] For example, if a rise in temperature or internal temperature is determined to be the primary cause, it can be analyzed as an increase in thermal load; similarly, if an increase in network bandwidth occupancy, transmission volume, or communication delay is determined to be the primary cause, it can be analyzed as an increase in communication load. Furthermore, if there is a strong correlation between increased fine dust concentration and cooling fan operation, it can be analyzed as a need to suppress the diffusion of fine dust.
[0056] Additionally, the control unit (130) can calculate the State of Health (SOH) of the imaging device (200) based on power data, environmental data, and operation data. Here, SOH may be an indicator that reflects the degree of aging of the device, the frequency of repetition of abnormal conditions, the degree of accumulated heat stress, the persistence of power deviation, and the instability of operating conditions.
[0057] For example, if the deviation remains above a reference value for a long time, if repeated overheating occurs, or if the power usage pattern is consistently abnormal under the same conditions, the SOH may decrease. Such SOH information can be used as a criterion for determining the control strength when selecting a control combination later.
[0058] In this way, the control unit (130) performs an analysis of the cause of the abnormal state and selects one of a plurality of control combinations using the deviation between real-time collected data and normal consumption patterns, the analysis results, and environmental data.
[0059] Here, the control combination may be a method of coordinating multiple operational parameters together.
[0060] For example, the control unit (130) can select any one of a control combination that adjusts the frame rate (FPS) and cooling fan operation, a control combination that adjusts the resolution and network bandwidth, or a control combination that adjusts the cooling fan operation and the frame rate (FPS).
[0061] More specifically, the control unit (130) classifies the cause of increased power consumption into one of the following based on the analysis results of the providing unit (140): increased heat load, increased communication load, and the need to suppress the spread of fine dust, and can select a different combination of operating parameters according to the classified cause.
[0062] For example, the control unit (130) can select a control combination that adjusts the frame rate (FPS) and cooling fan operation when the temperature or internal temperature is above a first reference value, can select a control combination that adjusts the resolution and network bandwidth when the network load is above a second reference value, and can select a control combination that adjusts the cooling fan operation and frame rate (FPS) when the fine dust concentration is above a third reference value.
[0063] In addition, the control unit (130) can select a control combination with a high power consumption reduction strength when the calculated SOH is smaller than a preset reference value.
[0064] Next, the control unit (130) controls the operation parameters of the image device (200) according to the selected control combination (S350).
[0065] Specifically, the control unit (130) can adjust at least one of the resolution, frame rate (FPS), level of video analysis function, network bandwidth, and cooling fan operation.
[0066] For example, when the temperature is above a preset reference value, the control unit (130) can reduce the image processing load and power consumption by reducing at least one of the resolution, FPS, and image analysis function level.
[0067] Additionally, the control unit (130) may restrict some functions or switch to a low-power mode in response to the possibility of failure or an increase in the error rate when the humidity is above a preset threshold value. Additionally, the control unit (130) may suppress the spread of fine dust by stopping the continuous operation of the cooling fan or switching to intermittent operation or low-speed operation when the fine dust concentration is above a preset threshold value. In this case, internal heat generation and power consumption may be reduced simultaneously by adjusting the frame rate (FPS) together. However, if the internal temperature is above a protection threshold value, the operation of the cooling fan may be maintained or resumed to protect the device.
[0068] Additionally, the control unit (130) can restore the operating parameters of the imaging device (200) to normal operating parameters when the deviation between the environmental data, the real-time collected data, and the normal consumption pattern is maintained for a predetermined period of time or longer after returning to a preset normal range.
[0069] For example, if the resolution and FPS are reduced due to a high temperature condition, and subsequently both the temperature and deviation return to the normal range and that condition is maintained for a certain period of time or longer, the control unit (130) can gradually restore the resolution, FPS, and image analysis function levels to their original state. Accordingly, the present invention is not limited to simple one-time power saving control, but can appropriately restore operational performance when the normal state is restored after adaptively responding to environmental changes.
[0070] Next, the providing unit (140) provides the manager with explanationable artificial intelligence (XAI) regarding the abnormal condition, power data, environmental data, and operational data, which causes the increase in power consumption or carbon emissions to increase (S360).
[0071] That is, regarding the abnormal conditions detected and analyzed by the analysis unit (120), the providing unit (140) can provide the manager with the cause of the increase in power consumption or carbon emissions caused by the power data, the environment data, and the operation data through Explainable Artificial Intelligence (XAI).
[0072] Here, the providing unit (140) not only notifies whether an abnormal state has occurred, but can also intuitively present to the manager the analysis results on how each input data affects the increase in current power consumption.
[0073] For example, the providing unit (140) may identify the increase in heat load due to temperature rise, the increase in communication load due to network load rise, the need for cooling fan control due to fine dust concentration rise, or the increase in processing load due to increased level of video analysis function as the main causes of the increase in power consumption and provide this to the manager.
[0074] At this time, the providing unit (140) can provide the contribution, priority, change trend, or degree of excess relative to a reference value of each cause in the form of text, graphs, color displays, warning messages, reports, or dashboards, thereby enabling the manager to easily understand the cause of the increase in power consumption of the current video device (200).
[0075] Additionally, the providing unit (140) can provide the manager with objective grounds for “why carbon should be reduced” through explainable artificial intelligence (XAI). For example, the providing unit (140) can provide the manager with information on how much the operating parameters of a specific video device (200) have deviated from the normal consumption pattern, how much additional power consumption or carbon emission increase is expected if the deviation continues, and how much savings are expected when operating parameters such as resolution, frame rate (FPS), network bandwidth, or cooling fan operation are adjusted.
[0076] Accordingly, the manager can not only receive warning signals but also determine the necessity and expected effects of carbon reduction control based on objective data. As a result, the providing unit (140) supports the manager's decision-making regarding the optimization of the operation of the imaging device (200), reduces unnecessary energy consumption, and contributes to the construction of a smart energy city and the realization of carbon neutrality.
[0077] As such, according to an embodiment of the present invention, the power consumption pattern of the imaging device (200) is learned based on unsupervised learning, and the cause of the abnormal state is analyzed using an explainable artificial intelligence (XAI) method. Then, by combining and controlling a plurality of operating parameters according to the analysis results, it is possible to simultaneously achieve power consumption reduction, carbon emission reduction, device protection, and stable operation.
[0078] As such, according to an embodiment of the present invention, abnormal power consumption of an imaging device can be detected early and operating parameters can be actively adjusted according to the cause, thereby effectively reducing unnecessary power consumption and carbon emissions.
[0079] In addition, by flexibly responding to various environmental changes and operating conditions, device overload, failures, and performance degradation can be mitigated, and operating parameters can be appropriately restored upon recovery to a normal state, thereby improving the stable operation and management efficiency of the imaging device. Furthermore, by achieving both device protection and energy savings through differential control based on the status of the imaging device, the operational reliability and maintenance efficiency of the imaging device can be enhanced.
[0080] The present invention has been described with reference to the embodiments illustrated in the drawings, but this is merely illustrative, and those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of the present invention should be determined by the technical spirit of the appended claims. Explanation of the symbols
[0081] 100: Carbon reduction support device for imaging equipment 110: Data Collection Department 120: Analysis Department 130: Control unit 140: Providing unit 200: Video device
Claims
Claim 1 A carbon reduction support device for an imaging device, comprising: a data collection unit that collects power data, environmental data, and operational data from an imaging device; an analysis unit that uses the collected power data, environmental data, and operational data to generate a normal consumption pattern of the imaging device using an unsupervised learning-based LSTM (Long Short-Term Memory) model, and calculates a deviation between real-time collected data and the normal consumption pattern to detect an abnormal state; a control unit that receives the deviation between the real-time collected data and the normal consumption pattern and the environmental data, selects one of a plurality of control combinations, and controls the operational parameters of the imaging device according to the selected control combination; and a provision unit that, regarding the abnormal state, provides to an administrator, through explainable artificial intelligence (XAI), the causes of the power data, environmental data, and operational data contributing to increased power consumption or increased carbon emissions. Claim 2 A carbon reduction support device of an imaging device according to claim 1, wherein the environmental data includes at least one of temperature, humidity, and fine dust concentration. Claim 3 A carbon reduction support device for an imaging device according to claim 1, wherein the operating parameters include at least one of resolution, frame rate (FPS), image analysis function level, network bandwidth, and cooling fan operation. Claim 4 A carbon reduction support device for an imaging device according to claim 2, wherein the control unit reduces at least one of the resolution, frame rate (FPS), and image analysis function level of the imaging device when the temperature is above a preset reference value. Claim 5 A carbon reduction support device for an imaging device according to claim 2, wherein the control unit limits some functions of the imaging device or switches to a low-power mode in response to the possibility of failure or an increase in the error rate when humidity is above a preset reference value. Claim 6 A carbon reduction support device for an imaging device according to claim 2, wherein the control unit stops the continuous operation of the cooling fan or switches to intermittent operation or low-speed operation when the fine dust concentration is above a preset reference value, and maintains or resumes the operation of the cooling fan when the internal temperature of the imaging device is above a protection reference value. Claim 7 A carbon reduction support device for an imaging device according to claim 1, wherein the control unit classifies the cause of increased power consumption into one of increased heat load, increased communication load, and the need to suppress the diffusion of fine dust based on the analysis result, and selects a different combination of operating parameters according to the classified cause. Claim 8 A carbon reduction support device for an imaging device according to claim 6, wherein the control unit adjusts the frame rate (FPS) and cooling fan operation when the temperature or the internal temperature of the imaging device is greater than or equal to a first reference value, adjusts the resolution and network bandwidth when the network load is greater than or equal to a second reference value, and adjusts the cooling fan operation and frame rate (FPS) when the fine dust concentration is greater than or equal to a third reference value. Claim 9 A carbon reduction support device for an imaging device according to claim 1, wherein the control unit restores the operating parameters of the imaging device to normal operating parameters when the environmental data and the deviation are each maintained for a predetermined time or longer after returning to a preset normal range. Claim 10 A carbon reduction support device for an imaging device according to claim 1, wherein the control unit calculates the State of Health (SOH) of the imaging device based on the power data, the environment data, and the operation data, and the control unit selects a control combination with a high power consumption reduction intensity when the State of Health is smaller than a preset reference value. Claim 11 A method for reducing carbon emissions of an imaging device, performed by a carbon reduction support device for an imaging device, comprising: a step of collecting power data, environmental data, and operational data from an imaging device; a step of generating a normal consumption pattern of the imaging device using an unsupervised learning-based LSTM (Long Short-Term Memory) model using the collected power data, environmental data, and operational data; a step of detecting an abnormal state by calculating a deviation between real-time collected data and the normal consumption pattern; a step of receiving the deviation between the real-time collected data and the normal consumption pattern, the detection result of the abnormal state, and the environmental data, and selecting one of a plurality of control combinations; a step of controlling the operational parameters of the imaging device according to the selected control combination; and a step of providing an administrator with, through explainable artificial intelligence (XAI), the cause of the power data, environmental data, and operational data contributing to an increase in power consumption or carbon emissions regarding the abnormal state. Claim 12 In claim 11, the step of collecting environmental data is a carbon reduction support method of an imaging device that collects at least one of temperature, humidity, and fine dust concentration. Claim 13 In claim 11, the step of controlling the operation parameters of the imaging device is a carbon reduction support method for an imaging device that controls at least one of resolution, frame rate (FPS), image analysis function level, network bandwidth, and cooling fan operation. Claim 14 A carbon reduction support method for an imaging device according to claim 11, wherein the step of controlling the operation parameters of the imaging device reduces at least one of the resolution, frame rate (FPS), and image analysis function level of the imaging device when the temperature is above a preset reference value. Claim 15 In claim 11, the step of controlling the operating parameters of the imaging device comprises restricting some functions of the imaging device or switching to a low-power mode in response to the possibility of failure or an increase in the error rate when humidity is above a preset reference value, thereby supporting carbon reduction of the imaging device. Claim 16 A carbon reduction support method for an imaging device according to claim 11, wherein the step of controlling the operation parameters of the imaging device comprises stopping the continuous operation of the cooling fan or switching to intermittent operation or low-speed operation when the fine dust concentration is above a preset reference value, and maintaining or resuming the operation of the cooling fan when the internal temperature of the imaging device is above a protection reference value. Claim 17 A carbon reduction support method for an imaging device according to claim 11, wherein the step of selecting one of the plurality of control combinations comprises classifying the cause of increased power consumption based on the analysis result into one of increased heat load, increased communication load, and the need to suppress the diffusion of fine dust, and selecting a different combination of operating parameters according to the classified cause. Claim 18 A method for supporting carbon reduction of an imaging device according to claim 16, wherein the step of controlling the operation parameters of the imaging device comprises adjusting the frame rate (FPS) and cooling fan operation when the temperature or the internal temperature of the imaging device is greater than or equal to a first reference value, adjusting the resolution and network bandwidth when the network load is greater than or equal to a second reference value, and adjusting the cooling fan operation and frame rate (FPS) when the fine dust concentration is greater than or equal to a third reference value. Claim 19 A carbon reduction support method for an imaging device according to claim 11, wherein the step of controlling the operating parameters of the imaging device is to restore the operating parameters of the imaging device to normal operating parameters when the environmental data and the deviation are each maintained for a predetermined time or longer after returning to a preset normal range. Claim 20 A method for supporting carbon reduction of an imaging device according to claim 11, wherein the step of selecting one of the plurality of control combinations comprises calculating the State of Health (SOH) of the imaging device based on the power data, the environment data, and the operation data, and selecting a control combination with a greater power consumption reduction intensity among the plurality of control combinations when the calculated State of Health is smaller than a preset reference value.
Citation Information
Patent Citations
The lighting control system using CCTV camera
KR100658403B1
Apparatus and method for cooling camera
WO2017111241A1
