WO2023238159A1 - Système de disjoncteur triphasé pour prédiction de défauts dans des appareils électriques à l'aide d'une intelligence artificielle - Google Patents

Système de disjoncteur triphasé pour prédiction de défauts dans des appareils électriques à l'aide d'une intelligence artificielle Download PDF

Info

Publication number
WO2023238159A1
WO2023238159A1 PCT/IN2023/050546 IN2023050546W WO2023238159A1 WO 2023238159 A1 WO2023238159 A1 WO 2023238159A1 IN 2023050546 W IN2023050546 W IN 2023050546W WO 2023238159 A1 WO2023238159 A1 WO 2023238159A1
Authority
WO
WIPO (PCT)
Prior art keywords
electrical appliance
phase
load
electrical
energy consumption
Prior art date
Application number
PCT/IN2023/050546
Other languages
English (en)
Inventor
Kaushik Bose
Rohan Gupta
Soumya BHATTACHARYA
Prasad PALEKAR
Meet Patel
Original Assignee
Sustainable Reference Analytics Private Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sustainable Reference Analytics Private Limited filed Critical Sustainable Reference Analytics Private Limited
Publication of WO2023238159A1 publication Critical patent/WO2023238159A1/fr

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00Testing or monitoring of control systems or parts thereof
    • G05B23/02Electric testing or monitoring
    • G05B23/0205Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0259Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
    • G05B23/0283Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/25Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
    • G01R19/2513Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02HEMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
    • H02H1/00Details of emergency protective circuit arrangements
    • H02H1/0092Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02HEMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
    • H02H3/00Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
    • H02H3/08Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current
    • H02H3/083Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current for three-phase systems
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05BCONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00Program-control systems
    • G05B2219/20Pc systems
    • G05B2219/26Pc applications
    • G05B2219/2639Energy management, use maximum of cheap power, keep peak load low

Definitions

  • the embodiments herein generally relate to the field of circuit breakers, and more specifically to a three-phase circuit breaker system for predicting faults in electrical appliances using artificial intelligence for controlling the energy consumption of electrical appliances. Also, the three-phase circuit breaker system is integrated into a smart miniature circuit (Smart MCB).
  • Smart MCB smart miniature circuit
  • the existing conventional circuit breaker has the circuit breaking capability upon detecting the abnormal electrical values.
  • One of the existing energy management solutions include providing a utility meter with the circuit breaker panel for monitoring the energy consumption for very limited electrical parameters like units consumed on a monthly basis which are displayed on the electricity bill, but still, the user does not have the clarity in detail for the operational state of switchgear and appliance in the house. Under such a situation, the user installs another energy monitoring meter/device/multiple sensors connected at one unit on the main line giving detailed insights on the basic electrical consumption. But for such installation, the user needs to hire an experienced technician to do the retro-fitting job. For such installation, the user may have to either shift, replace or make additional arrangements to get the monitoring device fitted in the existing breaker/distribution panel.
  • Another existing energy management solution includes installing an energy monitoring device which includes a plurality of sensors to capture the electrical consumption like smart plugs that don't need any retro-fits but have to be installed in proximity to the individual appliance and far away from the circuit breaker panel. Such sensors/devices are connected to electrical outlets where the loads are connected. This group of energy monitoring requires multiple such sensors to be installed at every outlet.
  • Such devices have limitations on electrical ratings for normal appliance use only and cannot be connected to heavy appliances or the appliances directly drawing the power from the breaker panel. Installation of multiple devices may lead to additional costs and time to set up every plug. Also, such multiple devices need to be connected to Wi-Fi which may reduce the bandwidth for other devices. In case the devices are connected to a local network, then accessing the information would be difficult if the user is out of the network radius/range or far away from the monitoring devices.
  • the conventional circuit breakers are not capable of utilizing the data for the amount of current flowing through the breaker or the electrical events occurring on the circuit breaker. The conventional circuit breakers are made/manufactured to perform the circuit breaking/tripping functionality only upon detecting an abnormal change in the current values.
  • the conventional circuit breakers are completely inefficient or underutilized when it comes to detection of appliances connected to an individual phase. There is always a need to install a separate energy monitoring system in identifying the irregularity of the electrical load connected, and the health status of the appliance connected to each phase. The user becomes aware of the uneven load condition, and health of the appliances only when the appliance level or complete electrical breakdown occurs or an increase in electricity bills. To have clarity on the appliance health status or the electrical wiring user needs to get the electrical audit done which includes an additional expenditure for auditing and servicing of the faulty electrical system/appliances. Slightly advanced circuit breakers are equipped with Wi-Fi- enabled electrical relay circuitry categorized under an automation system to protect and control the on and off state of the electrical circuit breaker.
  • relay-equipped circuit breakers in automation systems
  • the relay could give false / misfire under various electrical conditions that could break the circuit.
  • the harmonics can impact the operation of a protective relay and the reason for tripping the circuit could be inexplicit.
  • appliance users are unable to monitor the state of appliances and their internal components even after the MCB is in a live (conducting) state all the time. Symptoms, if any, become observable only when there is a breakdown or a critical issue. Load distribution inside the homes is uneven most of the time which can lead to major hazards like phase imbalance in voltage which Inverter and UPS have low efficiency from unbalance voltage and creates more harmonic current in the power system, appliance damage, phase blackouts. Any large singlephase load, or a number of small loads connected to only one phase causes more current to flow from that particular phase causing voltage drop on line. Unbalancing increases I2R losses in the cable.
  • Unbalanced current and nonlinear loads produce increased neutral-ground voltage and neutral current in the neutral conductor.
  • the high neutral voltage causes significant disruption to the operation of microprocessor-based equipment and fragmentation of hard disks which eventually has detrimental effects on appliance performance.
  • the loss in the neutral wire can be considerable and may result in overloading due to the unbalanced loads and the zerosequence currents from non-linear loads.
  • Nonlinear loads do not draw current sinusoidally from the utility. Examples of non-linear loads include VFDs, LED lighting, photocopiers, computers, uninterruptible power supplies, televisions, and the majority of electronics that include a power supply.
  • Nonlinear loads generate odd harmonics.
  • the troublesome harmonics for loads are the 3rd and odd multiples of the 3rd (9th, 15th, etc.). These harmonics are called “triplens' ' and because the 3 -phase triplen harmonics are all in phase with each other. They will add rather than cancel on the neutral conductor of a 3 -phase 4-wire system. This can overload the neutral conductor.
  • An important way to check for harmonic currents is to measure the current in the neutral of a 3 -phase 4-wire system. If the neutral current is considerably higher than the value predicted from the imbalance in the phase currents, there is a good possibility of a heavy presence of triplen harmonics.
  • Some of the problems that may be encountered due to a high level of harmonics are: Premature failure and reduced lifespan of devices often occur when overheating is present, such as: Overheating of transformers, cables, circuit breakers and fuses, unstable operation of sensitive electronics that require a pure sinusoidal AC waveform, flickering lights. Additionally, triplen harmonics cause circulating currents on the delta winding of a delta-wye transformer configuration. When current triplen harmonics on the neutral of a 3 -phase 4-wire system reach the transformer, they are reflected to the delta-connected primary where they circulate. The result is transformer heating similar to that produced by an unbalanced 3-phase current. Economic issues caused by harmonic distortion include the upfront cost of either sizing equipment to handle harmonics, or investing in harmonic mitigation in the first place, Secondly, day-to-day added costs due to the inefficiency of the system and costs due to premature failure of equipment.
  • a three-phase circuit breaker system for predicting faults in electrical appliances using artificial intelligence for controlling the energy consumption of electrical appliances.
  • the three-phase circuit breaker system includes a smart miniature circuit breaker (smart MCB) configured to obtain phase-level electric consumption data associated with a household environment at a predetermined interval of time.
  • the three-phase circuit breaker system includes a fault -prediction server communicatively connected to the smart miniature circuit breaker.
  • the fault prediction server includes a memory that includes a set of instructions, and a processor.
  • the processor executes the set of instructions and is configured to (i) receive the phase level electric consumption data of at least one electrical appliance from the smart MCB in the household environment, (ii) identify the at least one electrical appliance that operate in specific phase by filtering the received phase-level electric consumption data based on a historic phase-level electric consumption pattern of the at least one electrical appliance, (iii) determine load signatures that are relevant to the at least one electrical appliance that is identified in each phase by comparing an electrical behavior, or characteristic of the at least one electrical appliance with a historic electrical behaviors or characteristics associated with the at least one electrical appliance, (iv) predict, by a deep learning-based nonintrusive load monitoring (NILM) model, a value of an energy consumption of the at least one electrical appliance that is identified in the specific phase by correlating the load signatures that are relevant to the at least electrical appliance with a historical data.
  • NILM deep learning-based nonintrusive load monitoring
  • the historical data includes one or more ground truth value of active power and reactive power of the at least one electrical appliance, (v) process the predicted value of the energy consumption of the at least one electrical appliance by (a) applying a predefined minimum detection threshold value on the predicted value of the energy consumption of the at least one electrical appliance to determine whether the predicted value of the energy consumption of the at least one electrical appliance falls below the predefined minimum detection threshold value of the at least one electrical appliance, (b) remove additional load signatures from the load signatures that are relevant to the at least electrical appliance if the predicted value of energy consumption of the at least electrical appliance falls above the predefined minimum detection threshold value, and (c) adjust the predicted value of the energy consumption of the at least one electrical appliance based on the predefined minimum detection threshold value of the of the at least one electrical appliance if the predicted value of the load signatures that is remaining in the at least electrical appliance falls above a predefined time based-threshold value of the at least one electrical appliance, and (d) predict faults in the at least one electrical appliance that is identified in the specific phase based on
  • the predefined time based-threshold value is applied on predicted value of the load signatures that is remaining in the at least electrical appliance to determine whether the predicted value of the load signatures that is remaining in the at least electrical appliance falls below the predefined time based-threshold value.
  • the processor is configured to receive the phase-level electric consumption data of the at least electrical appliance by (i) analyzing a phase-wise load distribution of the at least one electrical appliance and (ii) identifying, using a wattage value for each phase, minimum and maximum load values of the at least one electrical appliance on in specific phase over the predetermined interval of time under one or more operating load conditions based on the phase-wise load distribution of the at least one electrical appliance.
  • the one or more operating load conditions includes zero load, partial load, and full load operating conditions.
  • the processor is configured to remove the additional load signatures that are not relevant to the at least one electrical appliance that is identified in the
  • the processor is configured to discard the prediction value of the energy consumption of the at least one electrical appliance if the prediction value falls below the predefined minimum detection threshold value.
  • the processor is configured to classify the at least one electrical appliance that is identified in the specific phase based on the determined load signatures of the at least one electrical appliance.
  • the additional load signatures are removed based on a historic load signature of the at least one electrical appliance.
  • the three-phase breaker system includes (i) at least one an input terminal, or an output terminal to connect an electric supply and the at least one electric appliance, (ii) a current sensor or shunt resistor positioned at the at least one input terminal or the output terminal and is integrated to the smart MCB to enable measuring of the phase-wise electric load distribution of the at least electric appliance, the phase wise load distribution is measured by monitoring a flow electric current using the smart MCB.
  • a method of three-phase circuit breaker system for predicting faults in electrical appliances using artificial intelligence for controlling an energy consumption of the electrical appliances.
  • the method includes receiving a phase level electric consumption data of at least one electrical appliance from a smart miniature circuit breaker, (smart MCB) in the household environment.
  • the method includes identifying the at least one electrical appliance that operates in a specific phase by filtering the received phase-level electric consumption data based on a historic phase-level electric consumption patterns of the at least one electrical appliance.
  • the method includes determining load signatures that are relevant to the at least one electrical appliance that is identified in the specific phase by comparing an electrical behavior, or characteristic of the at least one electrical appliance with a historic electrical behaviors or characteristics associated with the at least one electrical appliance.
  • the method includes predicting, by a deep learning-based nonintrusive load monitoring (NILM) model, a value of an energy consumption of the at least one electrical appliance that is identified in the specific phase by correlating the load signatures that are relevant to the at least electrical appliance with a historical data, the historical data include one or more ground truth value of active power and reactive power of the at least one electrical appliance.
  • NILM deep learning-based nonintrusive load monitoring
  • the method includes processing the predicted value of the energy consumption of the at least one electrical appliance by (i) applying a predefined minimum detection threshold value on the predicted value of the energy consumption of the at least one electrical appliance to determine whether the predicted value of the energy consumption of the at least one electrical appliance falls below the predefined minimum detection threshold value of the at least one electrical appliance (ii) removing additional load signatures from the load signatures that are not relevant to the at least electrical appliance based on historic load signatures of the at least one electrical appliance, if the predicted value of energy consumption of the at least electrical appliance falls above the predefined minimum detection threshold value , (iii) adjusting the predicted value of the energy consumption of the at least one electrical appliance based on the predefined minimum detection threshold value of the at least one electrical appliance if the predicted value of the load signatures that is remaining in the at least electrical appliance falls above a predefined time based-threshold value of the at least one electrical appliance, and (iv) predicting the faults in the at least one electrical appliance that is identified in the specific phase based on the adjusted predicted
  • the predefined time based-threshold value is applied on the predicted value of the load signatures that is remaining in the at least electrical appliance to determine whether the predicted value of the load signatures that is remaining in the at least electrical appliance falls below the predefined time based- threshold value.
  • the method includes receiving the phase-level electric consumption data of the at least electrical appliance by (i) analyzing a phase-wise load distribution of the at least one electrical appliance and (ii) identifying, using a wattage value for each phase, minimum and maximum load values of the at least one electrical appliance on the specific phase over the predetermined interval of time under one or more operating load conditions based on the phase-wise load distribution of the at least one electrical appliance.
  • FIG. 1 is a block diagram of a three-phase circuit breaker system for performing a predictive and failure analysis of electrical appliances using an artificial intelligence -enabled three-phase circuit breaker system according to some embodiments herein;
  • FIG. 2 is a block diagram that illustrates a fault prediction server of FIG. 1 of the three- phase circuit breaker system according to some embodiments herein; -
  • FIGS.3A and 3B are a graphical representation of an output of the phase filtering module of FIG.2 according to some embodiments herein;
  • FIGS. 4A and 4B are a graphical representation of an output of the event generation module of FIG.2 according to some embodiments herein;
  • FIGS. 5 is a graphical representation of an output of a deep learning based non-intrusive load monitoring (NILM) model of FIG.2 according to some embodiments herein;
  • NILM non-intrusive load monitoring
  • FIGS. 6 is a graphical representation of an output of a thresholding module of FIG.2 according to some embodiments herein;
  • FIGS. 7 is a graphical representation of an output of a time -based filtering module of FIG.2 according to some embodiments herein;
  • FIGS. 8 is a graphical representation of an output of an energy re-correction module of FIG.2 according to some embodiments herein;
  • FIG. 9 illustrates an artificial intelligence-enabled three-phase circuit breaker with an external Current Transformer (CT) / Potential Transformer (PT) having suspended sensors according to some embodiments herein;
  • CT Current Transformer
  • PT Potential Transformer
  • FIG. 10 illustrates an artificial intelligence-enabled three-phase circuit breaker with suspended external clip-on sensors according to some embodiments herein;
  • FIG. 11 illustrates functionalities of the artificial intelligence -enabled three-phase circuit breaker system of FIG.l according to some embodiments herein;
  • FIG. 12A-B illustrates a small miniature circuit monitors a flow of electricity in the three-phase circuit breaker system lOOof FIG.l according to some embodiments herein;
  • FIG. 13 illustrates a hardware wireframe of the artificial intelligence enabled three- phase circuit breaker system of FIG.1 according to some embodiments herein;
  • FIG. 14 illustrates a user interface view of the artificial intelligence enabled three-phase circuit breaker system of FIG.l installed inside a panel indicating various parameters in a mobile application according to an embodiment herein;
  • FIG. 15 illustrates a user interface view of an analysis feature of the artificial intelligence enabled three-phase circuit breaker system of FIG.l in a mobile application according to some embodiments herein;
  • FIG. 16 illustrates the artificial intelligence-enabled three-phase circuit breaker system of FIG.l capable of operating on a single-phase supply according to an embodiment herein;
  • FIG. 17 illustrates the artificial intelligence-enabled three-phase circuit breaker system of FIG.l capable of operating on a three-phase supply according to an embodiment herein;
  • FIG. 18 illustrates the artificial intelligence-enabled three-phase circuit breaker system of FIG.16 having the provision for external as well as embedded PCB antenna for single -phase supply according to an embodiment herein;
  • FIG. 19 illustrates the artificial intelligence enabled three-phase circuit breaker system of FIG.17 having the provision for external as well as embedded PCB antenna for a three-phase supply according to an embodiment herein;
  • FIG. 20A-B illustrates an energy monitoring circuit or smart MCB of the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG.l according to an embodiment herein;
  • FIG. 21 is a hardware implementation of the artificial intelligence-enabled three-phase circuit breaker system of FIG.l in accordance with the embodiments herein;
  • FIGS. 22A-B illustrate a flow diagram of a method of performing a predictive and failure analysis of electrical appliances using an artificial intelligence-enabled three-phase circuit breaker system according to some embodiments herein;
  • FIG. 23 is a schematic diagram of a computer architecture in accordance with the embodiments herein.
  • An object of the invention is to provide a system with embedded functionality inside the three-phase circuit breaker allowing automated monitoring, and artificial intelligence- enabled controlling of energy consumption.
  • Another objective of the invention is to improve the accuracy of deep learning based non-intrusive monitoring (NILM) model and reduce false detection by performing preprocessing of data that includes searching for appliances in a specific phase and using appliance-specific behavior to only relevant signatures to the deep learning based non-intrusive monitoring (NILM) model for prediction and post-processing of data using thresholding, timebased processing and energy consumed re-correction techniques to improve the deep learning based non-intrusive monitoring (NILM) model accuracy after the prediction is completed by the deep learning based non-intrusive monitoring (NILM) model.
  • NILM deep learning based non-intrusive monitoring
  • Another objective of the invention is to make full utilization of the three-phase circuit breaker system and the data for an amount of current flowing through the three-phase circuit breaker system and the event occurring on the three-phase circuit breaker system which indirectly helps in the optimization of a loss occurring in an electrical systems or appliances and directly helps in reducing the number of faulty events occurring on the electrical system.
  • FIGS. 1 through 23 where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments.
  • FIG. 1 is a block diagram that illustrates a three-phase circuit breaker system 100 for predicting faults in electrical appliances using artificial intelligence for controlling an energy consumption of the electrical appliances according to some embodiments herein.
  • the three- phase circuit breaker system 100 includes a smart miniature circuit board (MCB) 102, a network 104, and a user device 106, a fault prediction server 108, and a deep-learning based non-intrusive monitoring (NILM) model 110.
  • the fault prediction server 108 enables to provide a smart diagnosis with advanced preventive and failure analysis of the electrical systems and appliances on a residential level.
  • the fault prediction server 108 obtains an electric consumption data of the electrical appliances from the smart MCB 102 through the network 104.
  • the network 104 may be a Wi-Fi network, a Bluetooth network, or a cellular network.
  • the fault prediction server 108 identifies the specific electrical appliance that operates in a specific phase by filtering the received phase-level electric consumption data based on a historic phase-level consumption pattern of different electrical appliance (i.e the fault prediction server 108 is configured to search for appliances in the specific phase that decreases an inaccuracy in appliance energy detection up to 66% and compute required up to 66%.
  • the fault prediction server 108 determines load signatures that are relevant to the specific electrical appliance that is identified in a specific phase.
  • the fault prediction server 108 compares an electrical behavior, or characteristic of the specific electrical appliance with historic electrical behaviors or characteristics associated with the specific electrical appliances.
  • the fault prediction server 108 is configured to send only relevant signatures to the deep-learning based non-intrusive load monitoring (NILM) model. 110 for prediction of health status or fault of electric appliances.
  • the fault prediction server 108 is configured to predict the energy consumption of the specific appliance using the deep-learning based non-intrusive load monitoring (NILM) model 110.
  • the fault prediction server 108 predicts a value of an energy consumption of the specific electrical appliance that is identified in the specific phase using the deep learning-based nonintrusive load monitoring (NILM) model 110.
  • the deep learningbased nonintrusive load monitoring (NILM) model 110 correlates the load signatures that are relevant to the specific electrical appliance with a historical data of different electrical appliances.
  • the fault prediction server 108 processes the predicted value of the energy consumption of the specific electrical appliance.
  • the fault prediction server 108 applies a predefined minimum detection threshold value on the predicted value of the energy consumption of the at least one electrical appliance to determine whether the predicted value of the energy consumption falls of the at least one electrical appliance below the predefined minimum detection threshold value of the at least one electrical appliance, to remove noise from the predicted value of the energy consumption of the at least one electrical appliance.
  • the fault prediction server 108 removes additional load signatures from the load signatures that are not relevant to the specific electrical appliance based on historic load signatures of the at least one electrical appliance if the predicted value of energy consumption of the specific appliance falls above the predefined minimum detection threshold value.
  • the fault prediction server 108 adjusts the predicted value of the energy consumption of the specific electrical based on the predefined minimum detection threshold value of the specific electrical appliance if the predicted value of the load signatures that is remaining in the at least electrical appliance falls above a predefined time based-threshold value of the at least one electrical appliance.
  • the fault prediction server 108 predicts faults in the at least one electrical appliance that is identified in the specific phase based on a historic predicted value of the energy consumption of the electrical appliances.
  • the predefined time based-threshold value is applied on the predicted value of the load signatures that is remaining in the at least electrical appliance to determine whether the predicted value of the load signatures that is remaining in the at least electrical appliance falls below the predefined time based-threshold value.
  • the fault prediction server 108 is configured to remove additional signatures classified incorrectly using time and frequency-based logic.
  • the fault prediction server 108 is configured to use the received load signatures of the electrical appliance to extrapolate the output by the deep learning-based nonintrusive load monitoring (NILM) model 110 to improve the accuracy of the deep learningbased non-intrusive load monitoring (NILM) model 110.
  • the fault prediction server 108 is configured to predict the health status/ fault specific of the electric appliances and communicate it through the user device 106.
  • the artificial intelligence-enabled three-phase circuit breaker system 100 is configured to record faults and controls through a microcontroller and further communicates to a remote entity for an action.
  • the three-phase circuit breaker system 100 is configured to record the event of the electrical appliances and further communicate to the remote entity for the action.
  • the microcontroller may be communicated to the three-phase circuit breaker system 100 based on the action from the remote entity.
  • the action from the remote entity is communicated to one or more artificial intelligence-enabled three-phase circuit breaker systems.
  • FIG. 2 is a block diagram that illustrates a fault prediction server 108 of FIG. 1 of the three-phase circuit breaker system 100 according to some embodiments herein.
  • the fault prediction server 108 includes a database 202, a phase level electric consumption receiving module 204, a phase filtering module 206, an event generator module 208, an energy consumption value predicting module 210, a thresholding module 212, a time -based filtering module 214, an energy re-correction module 216, and a final prediction module 218.
  • the phase level electric consumption receiving module 204 is configured to obtain electric consumption data of an electric appliance from a smart MCB associated with a residential house.
  • the phase filtering module 206 is configured to identify an electrical appliance that operates in a specific phase by filtering the received phase-level electric consumption data based on a historic phaselevel consumption pattern of different electrical appliances.
  • the event generator module 208 is configured to determine load signatures that are relevant to the electrical appliance that is identified in the specific phase by comparing an electrical behavior, or characteristic of the at least one electrical appliance with historic electrical behaviors or characteristics associated with the at least one electrical appliance.
  • the energy consumption value prediction module 210 is configured to predict a value of an energy consumption of the specific electrical appliance that is identified in the specific phase by correlating the load signatures that are relevant to the electrical appliance with a historical data.
  • the historical data includes one or more ground truth value of active power and reactive power of the at least one electrical appliance.
  • the thresholding module 212 applies a predefined minimum detection threshold value on the predicted value of the energy consumption of the at least one electrical appliance to determine whether the predicted value of the energy consumption of the at least one electrical appliance falls below the predefined minimum detection threshold value of the at least one electrical appliance, to remove noise from the predicted value of the energy consumption of the at least one electrical appliance.
  • the time -based filtering module 214 is configured to remove additional load signatures from the load signatures that are not relevant to the electrical appliance based on historic load signatures of the at least one electrical appliance, if the predicted value of energy consumption of the specific electrical appliance falls above the predefined minimum detection threshold value.
  • the energy re-correction module 216 adjust the predicted value of the energy consumption of the at least one electrical appliance based on the predefined minimum detection threshold value of difference electrical appliance if the predicted value of the load signatures that is remaining in the at least electrical appliance falls above a predefined time based-threshold value of the at least one electrical appliance.
  • the final prediction module 218 predict faults in the at least one electrical appliance that is identified in the specific phase based on the adjusted predicted value of the energy consumption of the at least one electrical appliance.
  • the predefined time based-threshold value is applied on the predicted value of the load signatures that is remaining in the at least electrical appliance to determine whether the predicted value of the load signatures that is remaining in the at least electrical appliance falls below the predefined time based-threshold value.
  • FIGS. 3 A and 3B are graphical representations of an output of the phase filtering module 206 of FIG.2 according to some embodiments herein.
  • the graphical representation of the phase level electric consumption data of the electrical appliance before a process of phase filtering is shown in Fig.3A.
  • the graphical representation of the electric consumption data of the electrical appliance after the process of phase filtering is shown in Fig.3B.
  • the process phase filtering includes removing unwanted phases from a three-phase raw input data or the phase level electric consumption data of electrical appliances that are received the smart MCB 102 based on a historic phase-level consumption pattern of the at least one electrical appliance or depending on the electrical appliance phase information available in the database 202 and identify a specific electrical appliance that operates in each phase.
  • phase filtering increases an accuracy by up to 28.56%. Also, as the specific electrical appliance is searched for only in the specific phase, it decreases the compute required by up to 66% which enables users with a more accurate appliance breakdown.
  • the phase filtering module 206 decreases the inaccuracy in appliance energy detection by up to 66% by searching for the electrical appliance in the specific phase performed.
  • a sample data for the R phase received from the meter circuit the sample data include but not limited to a meter ID is 3637, a time is 1685519149, an energy is 2075.66, an active Power is 1999.8, a Reactive Power is 178.5, an apparent Power: is 239.5, a voltage is 242.01, a current is 1.59, a power Factor (PF) is 0.642, or a frequency is 49.99.
  • an operating threshold is 1 if the Active Power (AP) > 500, for Washing Machine, the operating threshold is 1 if the AP > 200 and for the Geyser, the operating threshold is 1 if AP > 1700.
  • the AC and washing machine would be triggered based on the operating threshold condition since the active power value is 1999.8.
  • the phase filtering module 206 filters the phase level electric consumption data of the different appliances, for example, the AC was triggered in the R phase in the database 202, the AC phase is also specified as the R phase. Therefore, the AC will be processed in the R phase.
  • the washing machine was triggered in the R phase, but in the database 202, the washing machine is specified in the Y phase. The washing machine will be discarded from further processing because the phase mentioned in the database 202 does not match the triggered phase.
  • the Geyser was triggered in the R phase according to threshold operating conditions, even though the Geyser phase is not specified in the database 202, and the Geyser will be processed in the R phase.
  • the user also provides information regarding the different electrical appliances includes specifying which appliances are present, their characteristics, and any relevant details or preferences associated with them. For example, the user might indicate that they have the air conditioner (AC), the washing machine (WM), and a geyser. The user may also provide specific operating threshold conditions for each appliance, such as the minimum active power (AP) required to trigger the different electrical appliances.
  • AC air conditioner
  • WM washing machine
  • AP minimum active power
  • the table 1 represents the accuracy improvement possible in various scenarios after passing the phase-level electric consumption data of the electrical appliances through Phase Filtering module 206.
  • the phase filtering module 206 with a classification Accuracy of 70% is considered for reference. All accuracy improvement is observed relative to this model only.
  • Formula: 1. % Accuracy Improvement (Total data points - Total Wrong Classified data points by model) *100/Total data points.
  • Total Wrong Classified data points by model (Total data points received by the model after Phase Filtering * (100 - Model Accuracy))
  • FIGS. 4A and 4B are graphical representations of an output of the event generation module 208 of FIG. 2 according to some embodiments herein.
  • the graphical representation of the output of the phase filtering module 206 is shown in Fig.4A which is input to the event generation module 208.
  • the graphical representation of the electric consumption data of the specific electrical appliance after the process of phase filtering is shown in Fig.4B.
  • the event generation uses thresholds to identify whether a particular change classifies as a given appliance and removes irrelevant signatures. In the event generation module 208, the thresholds are used to determine whether a specific change or measurement qualifies as the specific electrical appliance.
  • NILM deep- learning-based nonintrusive load monitoring
  • the table 2 represents the accuracy improvement possible in various scenarios after passing the relevant load signature through the event generator module 208.
  • the event generator module 208 with a classification accuracy of 70% is considered for reference. All accuracy improvement is observed relative to this model only.
  • Formula: 1. % Accuracy Improvement (Total data points - Total Wrong Classified data points by model) *100/Total data points.
  • Total Wrong Classified data points by model (Total data points received by the model after Event Generator * (100 - Model Accuracy)).
  • FIGS. 5 is a graphical representation of an output of the deep-learning-based nonintrusive load monitoring (NILM) model 110 of FIG.2 according to some embodiments herein.
  • the graphical representation of the output of the deep-learning-based non-intrusive load monitoring (NIEM) model 110 shows an expected value of an energy consumption of the at least one electrical appliance of the deep-learning-based non-intrusive load monitoring (NIEM) model 110 at 502.
  • the graphical representation of the output of the deep- learning-based non-intrusive load monitoring (NILM) model shows an actual (predicted value) value of energy consumption of the at least one electrical appliance of the deep-learning-based non-intrusive load monitoring (NILM) model 110 at 504.
  • FIGS. 6 is a graphical representation of an output of a thresholding module 212 of FIG.2 according to some embodiments herein.
  • the graphical representation of the thresholding module 212 shows the expected value of energy consumption of the at least one electrical appliance at 602 after noise removal from the predicted value of energy consumption of the at least one electrical appliance.
  • the graphical representation of the thresholding module 212 shows the actual value of energy consumption of the at least one electrical appliance at 604 after noise removal from the predicted value of energy consumption of the at least one electrical appliance.
  • the table 3 represents the accuracy improvement possible in various scenarios after passing the predicted value of energy consumption of the at least one electrical appliance through the thresholding module 212.
  • the thresholding module 212 with the classification accuracy of 70% is considered for reference. All accuracy improvement is observed relative to this model only.
  • % Accuracy Improvement (Total data points - Total Wrong Classified data points by model) *100/Total data points.
  • Total Wrong Classified data points by model (Total data points *(100 - Model Accuracy) * % of False events Left after the Thresholding).
  • FIGS. 7 is a graphical representation of an output of the time-based filtering module 214 of FIG.2 according to some embodiments herein.
  • the graphical representation of the timebased filtering module 214 shows the expected value of energy consumption of the at least one electrical appliance at 702 after removing an additional load signature from the predicted value of energy consumption of the at least one electrical appliance.
  • the graphical representation of the time -based filtering module 214 shows the actual value of energy consumption of the at least one electrical appliance at 704 after removing the additional load signatures from the predicted value of energy consumption of the at least one electrical appliance.
  • the table 4 represents the accuracy improvement possible in various scenarios after passing the predicted value of energy consumption of the at least one electrical appliance through the time -based filtering module 214.
  • the time -based filtering module 214 with a classification accuracy of 70% is considered for reference. All accuracy improvement is observed relative to this model only.
  • Formula: 1. % Accuracy Improvement (Total data points - Total wrong classified data points by the time-based filtering module) *100/Total data points.
  • Total Wrong Classified data points by the time -based filtering module (Total data points *(100 - Model Accuracy) * % of False events Left after the Time-Based Filtering).
  • FIGS. 8 is a graphical representation of an output of the energy re-correction module 216 of FIG.2 according to some embodiments herein.
  • the graphical representation of the energy re-correction module 216 shows the expected value of energy consumption of the at least one electrical appliance at 802 after the predicted value of energy consumption of the at least one electrical appliance is adjusted.
  • the graphical representation of the energy recorrection module 216 shows the actual value of energy consumption of the electrical appliance at 804 after the predicted value of energy consumption of the at least one electrical appliance is adjusted.
  • the energy re-correction module 216 uses the predicted value of energy consumption of the at least one electrical appliance that is received by the deep-learning-based NILM model 110 to extrapolate its output thus further improving the accuracy of the deep learning model based NILM model 110 and help in increasing the accuracy with which appliances are detected up to 25%. It enables users with a more accurate appliance breakdown.
  • the table 5 represents the accuracy improvement possible in various scenarios after passing the predicted value of energy consumption of the at least one electrical appliance through the energy re-correction module 216.
  • the deep learning-based NILM model 110 with a regression Accuracy of 30% is considered for reference. All accuracy improvement is observed relative to the deep learning based NILM model 110 only.
  • the table 6 represents the impact of various modules on classification and regression Accuracy.
  • the phase filtering module 206, the event generator module 208, the thresholding module 212, and the time -based filtering module 214 are used to improve the classification accuracy only. While the energy re-correction module 216 affects only regression accuracy.
  • the deep-learning-based NILM model 110 with the classification accuracy of 70% is considered for reference. All accuracy improvement is observed relative to this model only.
  • Final % Accuracy Improvement (Total data points - Total Wrong Classified data points by model) * 100/Total data points.
  • Total Wrong Classified data points by model (Total data points left after Phase Filtering * Total data points left after Event Generation* (100 - Model Accuracy) * % of False events left after the thresholding * % of False events left after time -based filtering).
  • the table 7 represents the final regression accuracy improvement possible in various scenarios after passing the predicted value of energy consumption of the at least one electrical appliance through the energy re-correction module 216.
  • the deep learning based NILM model 110 with the regression Accuracy of 30% is considered for reference. All accuracy improvement is observed relative to this model only.
  • FIG. 9 illustrates an artificial intelligence-enabled three-phase circuit breaker with an external current Transformer (CT) / Potential Transformer (PT) having suspended sensors according to some embodiments herein.
  • the suspended sensors 9A, 9B, and 9C include a Rogowski coil work with a de supply that produces a voltage that is proportional to the rate of change of the current enclosed by the coil-loop.
  • FIG. 10 illustrates an artificial intelligence-enabled three-phase circuit breaker system 100 with suspended external clip-on sensors according to some embodiments herein.
  • the external clip-on sensors 10A, 10B, and 10C measure a load current or the electric consumption data of the electrical appliance in the specific phase.
  • the suspended sensors may be toroidal- type current sensors.
  • FIG. 11 illustrates functionalities of the artificial intelligence -enabled three-phase circuit breaker system 100 according to some embodiments herein.
  • 1102 represents the three phases four wire incomer supply to the smart MCB 102
  • 1104 represents a monitoring and communication module that can be either on left or right side of the MCB
  • 1106 represents the smart MCB 102
  • 1108 represents a neutral link from load and the power or electric supply
  • 1110 represents the phase levels of the smart MCB 102 and the supply going to phase level loads
  • 1112 represents a neutral incomer and outgoing for load
  • 1114 represents a din rail mount arrangement for the smart MCB 102.
  • FIG. 12A-B illustrates a smart miniature circuit 102 for monitoring a flow of electricity in the three-phase circuit breaker system 100 according to some embodiments herein.
  • the smart miniature circuit 102 includes a microcontroller and antenna 1202, a wireless communication circuit 1204, a low voltage power supply module 1206, and an energy metering circuit 1208.
  • the microcontroller 1204 is a central processing unit of the smart miniature circuit and executes instructions and controls an overall operation.
  • the antenna is used for wireless communication, allowing the smart miniature circuit 102 to exchange information with external devices.
  • the smart miniature circuit 102 monitors the flow of electricity after the flow of electricity of the specific appliance input to the Smart MCB 102.
  • the low-voltage power supply module 1206 provides a necessary electrical power to operate the smart miniature circuit.
  • the energy metering circuit 1208 is responsible for measuring and monitoring the electrical energy consumption. It accurately measures parameters like voltage, current, and power factor to determine the amount of energy consumed by the connected load.
  • the wireless communication circuit 1204 facilitates the transmission and reception of data wirelessly. It enables the smart miniature circuit 102 to communicate with other devices or systems, such as a central monitoring station or a smartphone app. This wireless connectivity allows for remote monitoring, control, and data exchange.
  • An exemplary wireless communication circuit 1204 is shown in FIG. 12B.
  • the wireless communication circuit may be through multiple communication platforms and protocols including Wi-Fi routers, Bluetooth, internet, cellular, and laptop/ mobile phones.
  • FIG. 13 illustrates a hardware wireframe of the artificial intelligence-enabled three- phase circuit breaker system 100 of FIG.l according to some embodiments herein.
  • 1302 represents a 4-pole circuit breaker
  • 1304 represents a controller module
  • 1306 represents a sensor coil at B phase
  • 1308 represents Neutral out terminal of the Artificial Intelligence enabled three-phase circuit breaker
  • 1310 represents PCB with Wi-Fi chip, measuring and controller circuit
  • additional electronic circuitry 1312 represents external antenna
  • 1314 represents external antenna connector
  • 1316 represents PCB antenna
  • 1318 represents out terminal of Y phase
  • 1320 represents sensor coil at R phase
  • 1322 represents input terminal for R phase
  • 1324 represents sensor coil at Y phase
  • 1326 represents input terminal for B phase
  • 1328 represents sensor coil at Neutral phase
  • 1330 represents LED Indicators and reset button
  • 1332 represents Din rail mount.
  • the hardware wireframe of FIG. 9 is included in a mobile platform or the user device 106 enabling the user to perceive multiple data points from
  • FIG. 14 illustrates a user interface of view of the artificial intelligence enabled three- phase circuit breaker system of FIG.1 installed inside a panel indicating various parameters in a mobile application according to an embodiment herein.
  • the various parameters include 1 - Voltage (V) of individual live phase, 2 - Current (A) carried in each phase, 3 - Appliances detected in each phase (E.g.- AC, Geyser, etc.), 4 - Neutral current (A) value, 5 - Phase overload alert (denoted with exclamation icon), 6 - appliance health status on individual phase (denoted with dislike icon for unhealthiness of appliance).
  • FIG. 15 illustrates a user interface view 1500A-C of an analysis feature of the artificial intelligence-enabled three-phase circuit breaker system of FIG.1 in a mobile platform or the user device 106 according to some embodiments herein.
  • An air conditioner has been diagnosed for non-ideal consumption.
  • the user interface view 1500A shows “breakdown alert” of the electrical appliance to the user device 106.
  • the user interface view 1500B shows “air conditioner”, “washing machine”, and “geyser” to the user device 106 when the user device 106 is associated with a user clicking the “breakdown alert”.
  • the user interface view 1500C shows an analysis report of the “air conditioner” to the user device 106 if the user selects “air conditioner” in the user interface at 1500B.
  • FIG. 16 illustrates the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG.2 capable of operating on a single -phase supply according to an embodiment herein.
  • the three-phase circuit breaker system 100 includes a tripping circuit 1602, the energy metering circuit 1208, current sensor 1604, a memory 1606, a communication processor 1608, a central processing unit 1204, an antenna circuit 1610, a voltage sensor 1612, FED and push button 1614, communication components 1616, and a power supply 1618.
  • the one end of the tripping circuit 1602 is connected to a neutral, load, and another of the tripping circuit 1602 is connected to the neutral, and a line.
  • These components make up an advanced three-phase circuit breaker system 100 with the artificial intelligence capabilities.
  • the tripping circuit 1602 interrupts the flow of electricity in case of abnormal or excessive current, ensuring safety and protection against electrical faults.
  • the energy metering circuit 1208 accurately measures and monitors the consumption of electrical energy, providing valuable information for energy management.
  • the current sensor 1604 detects and measures the electric current flow and enables precise monitoring and control of electrical parameters.
  • the voltage sensor 1612 detects and measures the electrical voltage, providing information about the voltage level and ensuring proper voltage regulation.
  • the LED and push button 1614 provide visual indicators and allowing the users to interact with the three-phase circuit breaker system 100.
  • FIG. 17 illustrates the artificial intelligence-enabled three-phase circuit breaker system of FIG.2 capable of operating on a three-phase supply according to an embodiment herein.
  • the three-phase circuit breaker system includes a tripping circuit 1602, the energy metering circuit 1208, a current sensor 1604, a memory 1606, a communication processor 1608, a central processing unit 1204, an antenna circuit 1610, a voltage sensor 1612, FED and push button 1614, communication components 1616, and a power supply 1618.
  • the three-phase circuit breaker system 100 can effectively detect abnormal current conditions, measure energy consumption, communicate with external devices or systems, and provide essential protection and control functions for each phase of the electrical appliance.
  • the three-phase circuit breaker system 100 maintains the stability, reliability, and safety of the power supply 1618.
  • FIG. 18 illustrates the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG.16 having the provision for external as well as embedded PCB antenna for singlephase supply according to an embodiment herein.
  • the provision for external as well as embedded PCB antenna for single-phase supply means that the three-phase breaker circuit system is designed to accommodate different types of antennas for wireless communication.
  • the inclusion of the external antenna allows for flexibility in terms of antenna placement and positioning.
  • the external antennas enable the three-phase circuit breaker system to have better signal reception and transmission.
  • the external antenna can be connected to the three-phase circuit system through suitable connectors or interfaces.
  • the three-phase circuit breaker system also incorporates embedded PCB antennas, which are integrated directly into the circuit board of the three-phase circuit system.
  • FIG. 19 illustrates the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG.17 having the provision for external as well as embedded PCB antenna for a three- phase supply according to an embodiment herein.
  • the provision for external as well as embedded PCB antenna for the three-phase supply means that the three-phase breaker circuit system is designed to accommodate different types of antennas for wireless communication.
  • the three-phase breaker circuit system 100 enables reliable wireless communication for monitoring the flow of electricity in the three-phase supply scenarios.
  • FIG. 20A-B illustrates an energy monitoring circuit or smart MCB 102 of the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG.l according to an embodiment herein.
  • the three-phase circuit breaker system 100 includes an input terminal, or an output terminal 2002 to connect an electric supply, and the electric appliance or load.
  • a current sensor, or shunt resistor 2006 is positioned in the input terminal or the output terminal.
  • the three-phase circuit breaker system is integrated into the smart MCB 102 to enable measuring the phase-wise electric load distribution of the electric appliance.
  • the current sense resistors or shunt resistors 2006 are provided as an input to the energy monitoring circuit breaker in addition to rectification, filter, and amplification circuits.
  • the relation between voltage and current, according to Ohm’s law enables measuring amperes delivered to a load 2004.
  • FIG. 21 is a hardware implementation of the artificial intelligence-enabled three-phase circuit breaker system 100 of FIG. 1 in accordance with the embodiments herein.
  • the artificial intelligence-enabled three-phase circuit breaker system 100 uses an I2C communication protocol to transfer data to the microcontroller for performing a control action bit by bit along a single wire using SDA (Serial Data) - line for the master and slave to send and receive data and SCL (Serial Clock) line that carries the clock signal.
  • SDA Serial Data
  • SCL Serial Clock
  • FIGS. 22A-B are a flow diagram of a method of performing a predictive and failure analysis of electrical appliances using an artificial intelligence-enabled three-phase circuit breaker system according to some embodiments herein.
  • the method includes receiving a phase level electric consumption data of at least one electrical appliance from a smart miniature circuit breaker, (smart MCB) 102 in the household environment.
  • the method includes identifying the at least one electrical appliance that operates in a specific phase by filtering the received phase-level electric consumption data based on a historic phase-level electric consumption pattern of the at least one electrical appliance.
  • the method includes determining load signatures that are relevant to the at least one electrical appliance that is identified in a specific phase by comparing an electrical behavior, or characteristic of the at least one electrical appliance with a historic electrical behaviors or characteristics associated with the at least one electrical appliance.
  • the method includes predicting, by a deep learning-based nonintrusive load monitoring (NILM) model, a value of an energy consumption of the at least one electrical appliance that is identified in the specific phase by correlating the load signatures that are relevant to the at least electrical appliance with a historical data, the historical data include one or more ground truth value of active power and reactive power of the at least one electrical appliance.
  • NILM deep learning-based nonintrusive load monitoring
  • the method includes processing the predicted value of the energy consumption of the at least one electrical appliance by (i) applying a predefined minimum detection threshold value on the predicted value of the energy consumption of the at least one electrical appliance to determine whether the predicted value of the energy consumption falls below the predefined minimum detection threshold value of the at least one electrical appliance, (ii) removing additional load signatures from the load signatures that are not relevant to the at least electrical appliance based on historic load signatures of the at least one electrical appliance if the predicted value of energy consumption of the at least electrical appliance falls above the predefined minimum detection threshold value of the at least one electrical appliance, (iii) adjusting the predicted value of the energy consumption of the at least one electrical appliance based on the predefined minimum detection threshold value of the at least one electrical appliance if the predicted value of the load signatures that is remaining in the at least electrical appliance falls above a predefined time based-threshold value of the at least one electrical appliance, and (iv) predicting the faults in the at least one electrical appliance that is identified in the specific phase
  • the predefined time based-threshold value is applied on the predicted value of the load signatures that is remaining in the at least electrical appliance to determine whether the predicted value of the load signatures that is remaining in the at least electrical appliance falls below the predefined time based-threshold value.
  • the phase-level electric consumption data from the three-phase circuit breaker system 100 helps in analyzing a phase- wise load distribution data.
  • the three-phase circuit breaker system 100 identifies and captures minimum and maximum load values in real-time over a certain period on each phase using wattage values for all phases.
  • the three-phase circuit breaker system 100 captures a peak and bottom wattage values of all phases over time under various load operating conditions for analyzing an unbalance in all phases.
  • the three-phase circuit breaker system 100 calculates the % loss and its results into monetary values.
  • the three- phase circuit breaker system 100 monitors and compute, analyze, identify, and captures the minimum and maximum current values in real-time over a certain period on three-phase using wattage and current values for all phases from the circuit breaker.
  • the three-phase circuit breaker system 100 captures the peak and bottom wattage and current values of all phases over time under various load operating conditions (like zero load, partial load, and full load conditions) for analyzing the losses due to unbalanced conditions.
  • the three-phase circuit breaker system 100 calculates 100 % abnormality in neutral currents.
  • the three-phase circuit breaker system 100 utilizes electrical parameters fetched from the hardware and the unbalanced load computation in real-time over a specific period that helps in determining the losses and the neutral current under normal and abnormal conditions.
  • the three-phase circuit breaker system 100 identifies key appliances generating harmonics.
  • the three-phase circuit breaker system 100 utilizes the electrical phase values from the smart MCB (102), more specifically the neutral values under normal and abnormal conditions, variable load conditions that helps in identifying the generation of harmonics.
  • the electrical values captured in real-time over a period are analyzed with the appliance signatures or the load signatures for the respective time slot of data captured.
  • the three-phase circuit breaker system 100 utilizes health monitoring algorithms to capture the electrical values and signatures of the appliances that are compared with the reference / ideal values, unexpected changes in the appliance signatures are captured over time which help in early identifying the breakdown symptoms of the appliances.
  • the lowest consumption point of the entire day is determined.
  • the lowest consumption point represents the minimum watts of energy consumed throughout the day, which is the sum of energy consumed by all the appliances which remain ON throughout the day.
  • the lowest consumption Watt is then interpolated to get the energy consumed by always active appliances like cameras, doorbell, Wi-Fi, etc.
  • Equation 1C - % unbalance (Maximum deviation from the average value/ average values) * 100.
  • the three-phase circuit breaker system 100 provides the users with more knowledge and transparency on the overall electrical consumption of their appliances and the health state of the key appliance without deploying additional/multiple sensors inside the circuit breaker panel, or on electrical outlets/sockets, without any need for a separate audit to check appliance heath.
  • the appliance goes faulty or works abnormally, the user has no idea where the fault occurred and what are the possible causes and solutions for the incident.
  • the technician may service the appliance by refilling the coolant which might not be the correct solution for which the user has to pay the service charge.
  • the users will be able to access the data remotely and take necessary control actions to save on electricity and other expenditures needed to rectify the appliance faults.
  • the three-phase circuit breaker system 100 avoids retro-fits.
  • the three-phase circuit breaker system 100 is capable of handling multiple functionalities and gives the user an option to seamlessly fit the hardware without needing to make additional space for the installation of an energy monitor.
  • the three-phase circuit breaker system 100 may be implemented using a static type of current sensors/hall effect IC / shunt-based measurement along with additional components like voltage stabilization, filters, pullup resistors, and supply component.
  • the desired digital output converted is equivalent to the sensed current value which is then given to the microcontroller.
  • FIG. 23 A representative hardware environment for practicing the embodiments herein is depicted in FIG. 23, with reference to FIGS. 1 through 22A-B.
  • This schematic drawing illustrates a hardware configuration of a server 108/computer system/ mobile device in accordance with the embodiments herein.
  • the mobile device includes at least one processing device 10 and a cryptographic processor 11.
  • the special -purpose CPU 10 and the cryptographic processor (CP) 11 may be interconnected via system bus 14 to various devices such as a random access memory (RAM) 15, read-only memory (ROM) 16, and an input/output (RO) adapter 17.
  • the RO adapter 17 can connect to peripheral devices, such as disk units 12 and tape drives 13, or other program storage devices that are readable by the system.
  • the mobile device can read the inventive instructions on the program storage devices and follow these instructions to execute the methodology of the embodiments herein.
  • the mobile device further includes a user interface adapter 20 that connects a keyboard 18, mouse 19, speaker 25, microphone 23, and/or other user interface devices such as a touch screen device (not shown) to the bus 14 to gather user input.
  • a communication adapter 21 connects the bus 14 to a data processing network 26, and a display adapter 22 connects the bus 14 to a display device 24, which provides a graphical user interface (GUI) 30 of the output data in accordance with the embodiments herein, or which may be embodied as an output device such as a monitor, printer, or transmitter, for example.
  • GUI graphical user interface
  • a transceiver 27, a signal comparator 28, and a signal converter 29 may be connected with the bus 14 for processing, transmission, receipt, comparison, and conversion of electric or electronic signals.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Power Engineering (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Automation & Control Theory (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)

Abstract

L'invention concerne un procédé d'un système de disjoncteur triphasé visant à prédire des défauts dans des appareils électriques à l'aide d'une intelligence artificielle pour maîtriser une consommation d'énergie des appareils électriques. Le procédé comprend les étapes consistant à (ii) recevoir des données de consommation électrique au niveau des phases d'un appareil électrique en provenance d'un disjoncteur miniature intelligent, (ii) identifier un appareil électrique qui fonctionne dans une phase spécifique, (iii) déterminer des signatures de charge qui sont pertinentes pour l'appareil électrique qui est identifié dans une phase spécifique, (iv) faire prédire, par un modèle NILM basé sur un apprentissage profond, une valeur d'une consommation d'énergie de l'appareil électrique qui est identifié dans la phase spécifique (v) traiter la valeur prédite de la consommation d'énergie (a) en éliminant un bruit de la valeur prédite de la consommation d'énergie (b) en éliminant des signatures de charge supplémentaires (c) en ajustant la valeur prédite de la consommation d'énergie, et (d) en prédisant les défauts dans l'appareil électrique dans la phase spécifique.
PCT/IN2023/050546 2022-06-10 2023-06-10 Système de disjoncteur triphasé pour prédiction de défauts dans des appareils électriques à l'aide d'une intelligence artificielle WO2023238159A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IN202221033420 2022-06-10
IN202221033420 2022-06-10

Publications (1)

Publication Number Publication Date
WO2023238159A1 true WO2023238159A1 (fr) 2023-12-14

Family

ID=89117978

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IN2023/050546 WO2023238159A1 (fr) 2022-06-10 2023-06-10 Système de disjoncteur triphasé pour prédiction de défauts dans des appareils électriques à l'aide d'une intelligence artificielle

Country Status (1)

Country Link
WO (1) WO2023238159A1 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190064234A1 (en) * 2016-03-03 2019-02-28 Ecocentric Group Limited System and method for assessing electrical energy or power characteristics or features of electrical devices
US20220109997A1 (en) * 2017-02-28 2022-04-07 Leviton Manufacturing Co., Inc. Communication enabled circuit breakers

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190064234A1 (en) * 2016-03-03 2019-02-28 Ecocentric Group Limited System and method for assessing electrical energy or power characteristics or features of electrical devices
US20220109997A1 (en) * 2017-02-28 2022-04-07 Leviton Manufacturing Co., Inc. Communication enabled circuit breakers

Similar Documents

Publication Publication Date Title
US11552500B2 (en) Systems and methods for managing electrical loads
CN107209245B (zh) 能量计量系统及其校准方法
US8310370B1 (en) Smart circuit breaker with integrated energy management interface
JP2008261826A (ja) 電気量監視装置および電気量監視システム
JP6290868B2 (ja) 電力監視システムおよび電力監視方法
JP7249431B2 (ja) 電力消費行為のための検出方法、装置、電子機器および記憶媒体
CN103503262B (zh) 用于监控差动保护系统中电流互感器的方法和装置
WO2019241313A2 (fr) Moniteur de dispositif volt-var
JP2012198033A (ja) 相判別装置
US20170292999A1 (en) Transformer monitoring and data analysis systems and methods
CN103869181A (zh) 监控装置及其识别电器装置的方法
KR20200089473A (ko) 사용정보를 제공할 수 있는 세대분전반
CN104520724A (zh) 测量中性点对地电压的系统和方法
KR101277141B1 (ko) 전자식 전력량계와 이의 배전계통 감시방법
WO2023238159A1 (fr) Système de disjoncteur triphasé pour prédiction de défauts dans des appareils électriques à l'aide d'une intelligence artificielle
US20220200281A1 (en) Systems and methods for evaluating electrical phasors to identify, assess, and mitigate power quality issues
EP3292417B1 (fr) Analyse d'un circuit de puissance
KR101311490B1 (ko) 스마트 아웃렛을 이용한 아이티 장비 모니터링 시스템 및 모니터링 방법
CN208158090U (zh) 一种用电保护器
KR101674402B1 (ko) 손실전력측정 알고리즘 및 이를 이용한 손실전력 측정장치
CN117289043A (zh) 检测由电能质量事件引起的掉落负载的方法和系统
KR101868390B1 (ko) 전력량계, 전력량계 관리 장치 및 전력량 측정 방법
Stanca et al. Test Bench for Functionality and Reliability of Power Protection Devices.
KR101715778B1 (ko) 시간차 동기 알고리즘 기반 송배전 계통 손실전력 측정 방법
Zakaria et al. Smart Energy Monitoring System

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23819412

Country of ref document: EP

Kind code of ref document: A1