WO2021176459A1 - Système et procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive - Google Patents

Système et procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive Download PDF

Info

Publication number
WO2021176459A1
WO2021176459A1 PCT/IN2020/050611 IN2020050611W WO2021176459A1 WO 2021176459 A1 WO2021176459 A1 WO 2021176459A1 IN 2020050611 W IN2020050611 W IN 2020050611W WO 2021176459 A1 WO2021176459 A1 WO 2021176459A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
appliance
power
energy
server
Prior art date
Application number
PCT/IN2020/050611
Other languages
English (en)
Inventor
Sajil PEETHAMBARAN
Manoj KRISHNAN M
Vivek KUTHANAZHI
Jim Joseph John
Ravi Menon
Original Assignee
Peethambaran Sajil
Krishnan M Manoj
Kuthanazhi Vivek
Jim Joseph John
Ravi Menon
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 Peethambaran Sajil, Krishnan M Manoj, Kuthanazhi Vivek, Jim Joseph John, Ravi Menon filed Critical Peethambaran Sajil
Publication of WO2021176459A1 publication Critical patent/WO2021176459A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply

Definitions

  • the present invention to the field of power monitoring and data analytics in the area of power management.
  • the present invention is generally directed to systems and methods of non-intrusive appliance load monitoring (“NIALM”).
  • NIALM non-intrusive appliance load monitoring
  • the present invention is directed to demand side management for commercial/industrial and residential electric energy usage, utilizing software analytics on complete load profile data to disaggregate into individual appliances and loads.
  • Appliance load monitoring is an effective way to communicate to users the amount of energy usage required by various appliances. Presenting users with such information in an understandable format allows users to take appropriate action to actively reduce total energy consumption. Moreover, providing itemized information per specific appliance also permits users to determine if acquiring a new or replacement appliance (for example, through purchase, lease, or rental) would reduce energy costs sufficient to validate the price of purchase, lease, or rental.
  • NIALM enables the breakdown of electricity usage for a property without entering the property or applying any sub-metering devices on the individual appliances/devices/loads inside the property. Further, basic techniques for performing NIALM, teaches generating and using appliance load signatures to extract information for individual loads from whole property load profile data measured by the utility meter.
  • the information extracted from the utility meter may comprise: power consumption; times when the appliance/load was turned on and off; and appliance/load health.
  • NIALM various techniques used to define load signatures and run pattern recognition algorithm on the load profile of the property under inspection.
  • a software analysis is performed on past data collected. Therefore such prior art techniques may be useful in breaking down the energy usage or itemizing the electric energy bill post-consumption, but fail to provide near real-time information that may immediately empower users to modify their energy usage.
  • appliances such as heating or air conditioning for which usage is based upon immediate conditions
  • data of previous usage may provide limited assistance in modifying present behavior and usage.
  • prior art techniques and methodologies may provide users with some basic information regarding their power consumption — but fail to provide the user with any additional advice or counseling as to how to effectively use the information to reduce energy consumption. Rather, the user is left with the notion that he or she should simply use particular appliances less often. This information is relatively meaningless with regard to appliances that users generally must use for example, refrigerators, electric ranges, washing machines, dryers, etc.
  • the time of energy usage may dictate the cost of such usage. For example, during peak energy usage times, utility companies may charge increased rates than during low usage times. Merely changing the time of day a particular appliance is used may result in significant cost savings.
  • An aspect of the present invention is to address at least the above- mentioned problems and/or disadvantages and to provide at least the advantages described below.
  • the present invention relates to a method of energy disaggregation using non-intrusive load monitoring in a premises.
  • the method comprising: retrieving a plurality of raw energy data in a sequence and send to a server for every 30 second duration over a network, the plurality of raw energy data including a plurality of electrical and non-electrical parameters, wherein each data has a encoded value corresponding to power consumption of the premises, analyzing the parameters at the server to detect specific patterns or signatures and to map to a particular class of appliance and their events with the corresponding time stamp of occurrence, classifying each unique event signature to a particular class of appliance, and the activity of the event is further categorized as a turn on or turn off event, wherein the sampling rate of one second allows for separation of turn-on operation of each equipment, and analyzing all the detected appliances in real-time to calculate the contribution of each appliance to the aggregate power and storing the power consumption values of the premises in a database.
  • an energy monitoring system comprising: a plurality of data collection unit including a communication gateway for recording and transferring the recorded data to a remote server, the server including a processor and a memory, the processor is configured to execute a method of energy disaggregation using non-intrusive load monitoring in a premises, by retrieving a plurality of raw energy data in a sequence and send to a server for every 30 second duration over a network, the plurality of raw energy data including a plurality of electrical and non-electrical parameters, wherein each data has a encoded value corresponding to power consumption of the premises, analyzing the parameters at the server to detect specific patterns or signatures and to map to a particular class of appliance and their events with the corresponding time stamp of occurrence, classifying each unique event signature to a particular class of appliance, and the activity of the event is further categorized as a turn on or turn off event, wherein the sampling rate of one second allows for separation of turn-on operation of each equipment; and analyzing all the detected appliances in real-time
  • FIG. 1 is a diagram of a non-intrusive load monitoring system that measures electrical power delivered to multiple devices at a site.
  • FIG. 2 shows a flow chart of a method of energy disaggregation using non- intrusive load monitoring in a premises, in accordance with one embodiment of the present invention.
  • FIG. 3 shows an energy monitoring system implementing the method of FIG. 2 in accordance with one embodiment of the present invention.
  • FIG. 1 depicts an exemplary system for non-intrusive load monitoring (NILM) of the electrical power consumption of a site.
  • an electrical power source 104 supplies electrical power to a site 108 through an electrical line 106.
  • the electrical power source 104 includes any source of electrical power such as a commercial power grid or an electrical generator.
  • the electrical power source 104 provides an alternating current (AC) electrical power signal to the site 108 through the line 106.
  • the electrical power source 104 provides direct current (DC) power to the site 108.
  • the site 108 is any location that includes multiple appliances or electrical devices 112 that each consume electrical power supplied by the single electrical line 106. Each electrical device places an electrical load on the electrical power supplied through the electrical line 106.
  • a site is a single building that houses various devices that consume electrical power, such as a residence, commercial building, or industrial facility.
  • the magnitude of the electrical power signal delivered through the electrical line 106 varies as one or more of the devices 112 are activated and/or deactivated.
  • the electrical power supplied to the site 108 is monitored with an electrical power meter 116 and a NILM system 120.
  • the electrical power meter 116 is operatively connected to the electrical line 106 externally to the site 108 to enable the electrical power meter 116 to measure the total electrical power delivered to the site 108.
  • the electrical power meter 116 samples the magnitude of the electrical power signal in the electrical line 106 at regular intervals over time.
  • the electrical power meter 116 samples the magnitude of electrical power at a given sampling frequency of 300 Hz. Electrical systems in Europe and other parts of the world operate with 50 Hz AC signals, and the selected sampling frequency can be adjusted to accommodate both DC signals and AC signals operating over a wide range of frequencies. In other embodiments, the electrical power meter samples the electrical power signal with a frequency of at least two times the highest expected frequency of changes in the magnitude of the electrical power signal due to activation or deactivation of one or more of the devices 112.
  • the NILM system 120 includes a processor 124, memory 128, and user interface 132.
  • the NILM system 120 is operatively coupled to the electrical power meter 116 to receive data corresponding to the magnitude of the electrical power signal in the electrical line 106.
  • the data typically include measurements of both active and reactive power that are delivered to the site 108. One or both of the electrical voltage and current are measured to identify the power delivered through the electrical line 106.
  • the processor 124 is a microprocessor, microcontroller, or other digital computing device that performs stored instructions that are retrieved from the memory 128.
  • the processor 124 receives data corresponding to the measured electrical power from the electrical power meter 116.
  • the processor 124 During operation, the processor 124 generates output corresponding to the measured electrical power consumption of one or more of the devices 112 at the site 108 through the user interface 132.
  • the user interface 132 includes various output devices, such as display screens, audio output devices, and network communication devices, which enable an operator to monitor the electrical power usage of the devices 112 at the site 108.
  • the user interface 132 also includes input devices, such as keypads, touchscreen inputs, and network communication devices, which enable the operator to configure the NILM system 120 and send commands to the processor 124.
  • the NILM system During operation, the NILM system generates information corresponding to the electrical power usage of individual devices 112 at the site 108.
  • the NILM system 120 only receives aggregate power consumption data from the electrical power meter 116.
  • the NILM system 120 identifies the activation, deactivation, and operating power of each of the different devices 112 on the site from changes in the total electrical power consumed by the site 108 over time.
  • the memory 128 stores data corresponding to the changes in the waveform of the electrical power signal supplied to the site 108 when one of the devices 112 is either activated or deactivated.
  • the NILM system 120 initiates operation with no a priori knowledge of the characteristics of the devices 112 in use at the site 108.
  • An operator enters the number of devices 112 that are present at the site 108 through the user interface 132, and the NILM system 120 identifies signatures corresponding to each of the devices 112 through a recording and deconvolution process that is described below.
  • FIG. 2 shows a flow chart of a method of energy disaggregation using non- intrusive load monitoring in a premises, in accordance with one embodiment of the present invention.
  • the method retrieves a plurality of raw energy data in a sequence and send to a server or a central processing unit for every 30 second duration over a network.
  • the retrieving of data readings are sampled every second which works on 30 second sample windows.
  • the plurality of raw energy data including a plurality of electrical and non-electrical parameters, where each data has an encoded value corresponding to power consumption of the premises.
  • the electrical parameter including the power, voltage, and power factor values, and the non-electrical parameters including time between start and end of operation, working cycle and time of usage of the appliance.
  • the method which uses non-electrical parameters like working duration (time between start and end of operation), working cycle (say for example time between two compressor cut offs), time of usage etc. to tag the equipment, in addition to electrical parameters.
  • the method analyzes the parameters at the server or the central processing unit to detect specific patterns or signatures and to map to a particular class of appliance and their events with the corresponding time stamp of occurrence.
  • the step of analyzing the retrieved raw data are stored in a JSON format for easy of data interoperability in the process, and the step of mapping of the detected events includes the power signatures and their timestamp of occurrence envisaged, and further aids the mapped signatures and events to the dataset of appliances.
  • the mapping including comparing all features including event type, the power, power factor, phase, time of occurrence and frequency of operation to identify the appliance, wherein the output data is in a new JSON format.
  • the method classifies each unique event signature to a particular class of appliance, and the activity of the event is further categorized as a turn on or turn off event, the sampling rate of one second allows for separation of turn-on operation of each equipment
  • the method analyzes all the detected appliances in real-time to calculate the contribution of each appliance to the aggregate power and storing the power consumption values of the premises in a database.
  • the method may further displays the live “working” equipment in order to facilitate the customers /users to understand which equipment is working at that very instant real time with an accuracy levels of 96% in real-time energy disaggregation.
  • the method updates by recording, if any unknown signature or event is detected, in the database by taking feedback from the previously recorded user inputs in order to check the validity of the detected signature. Further, the method also predicts the appliance type of an event from previously recorded datasets, wherein step of predicting by reading all the data sets recorded previously for a particular appliance type by checking through all the events collected from the premises including characteristics of the appliances, and analyzes the success rate of prediction using the current tuning parameters, and make corrections to improve the accuracy of predictions. The method also combines historical data from a particular device location with the live data and uses the same in tuning parameters for effective prediction. In an example, the method uses readings sampled every second which works on 30 second sample windows.
  • the method engages historical data from previous data block in case there is a start event towards the end of the previous block. Say for example, if there is a rise (or fall) in current at the 28th sample of a block, the method will not be able to predict the appliance / equipment with 2 more additional samples. So this data will be combined with the next window and reconstructed to identify the equipment.
  • the method also receives the feedback information for processing which will help in the analysis to identify and fix errors, the error percentage is less than 1 %.
  • the feedback system may work on the past 1 hour’s data and does one more level of analysis to identify and fix any errors.
  • the method also alerts the user in a real-time which compares the usage parameters and electrical parameters, wherein alerting the user is based on the threshold set by the user, where if there is a deviation from the regular usage pattern on a particular day, if the always ON standby load is higher than usual, if power factor or voltage drops beyond a particular level, if the usage of a particular equipment is high on a particular day.
  • the method uses a unique sampling frequency of 1 reading per phase per second. When compared to other solution, which use 1 reading per minute (under sampling) or several readings per second (over sampling). Oversampling leads to a larger data chunk to process.
  • the sampling rate of one second allows for separation of TURN-ON operation of each equipment. For example, there are chances that 2 equipment (A and B) are turned ON during the same second, which leads to a cumulative signature that does not belong to either A or B.
  • the present invention method uses another level of prediction where it guesses the combination of equipment (based on previously recorded signatures from that premise) and waits for the next TURN OFF operation to detect these unclassified working equipment.
  • an AC and Microwave are turned ON at the same second. Now the cumulative load will be a combination of AC and microwave loads. Flowever, the microwave may be turned OFF earlier than the AC (or vice-versa). The method uses this turn OFF to predict the equipment that was turned ON earlier.
  • the method or system has a real-time alert mechanism which compares the usage parameters, electrical parameters to alert the user. For example, if there is a deviation from the regular usage pattern, it alerts the user. Say, if the usage between 9 AM-10 PM has been high on a particular day, if the always ON standby load is higher than usual, it will alert. If power factor or voltage drops beyond a particular level, it alerts. If the usage of a particular equipment is high on a particular day. For example, AC has been run for 8 hours on a day, instead of 3 hours. If the consumption of a particular equipment is high. For example, AC ran only for 3 hours (as usual), but the current drawn was high. The method or system allows the users where they can set custom thresholds to alert them when load / usage goes nears these threshold values.
  • Figure 3 shows an energy monitoring system implementing the method of
  • FIG. 2 in accordance with one embodiment of the present invention.
  • the energy monitoring systems has two major components (1) Data collection unit, which is the hardware module that has the physical presence at the premise. This module has an inbuilt communication gateway to transfer the recorded data to the remote server for further analytics.
  • the other is (2) Software Analytical solution which is the software that resides on a remote server. The purpose of this software is to dissect the raw data from the data collection unit into meaningful information in a manner that is easily presentable to the user.
  • These modules function in conjunction with each other and do not have a standalone functional existence.
  • the software delivers a detailed report on the energy consumption trends by disaggregating the usage data of a premise and detecting the appliance activity at the line it monitors.
  • the claimed module encompasses the basic Non-intrusive Load Monitoring (NILM) concepts with several improvements in prediction that contribute to improved precision and consistency of results.
  • NILM Non-intrusive Load Monitoring
  • the data collection unit gathers the raw information that is fed into the software solution.
  • This unit resides at the premise of the consumer and records power values from the distribution line at specific intervals.
  • the unit comprises of current and voltage measurement circuits, a timer module to record the timestamp of the captured data, a communication gateway that can connect to the server through the internet. In the unlikely event of disruption in data connectivity, the unit must store the recorded values in its internal memory. The data can then be restored onto the server once the connection is re-established. This interval between successive data recordings is set as 1 sec (i.e. a frequency of 1 Hertz) for optimal performance of the algorithm.
  • the sampled values of 30 seconds are scrambled and uploaded to the server using MQTT protocol.
  • the server could be a virtual machine or a physical server at any location.
  • the Software Analytical solution analyze the appliance signature and accumulating information on appliance activity. There is a significant amount of learning about the user behavior - drawing patterns about the consumer and his appliance usage activities. By leveraging its self-learning capability, algorithm can predict user behavior about energy consumption at a very granular level. This could further help in improving the convenience and efficiency in their day-to-day activities.
  • the algorithm can be split into two parts. The first one scans through real-time energy data to identify patterns that can be mapped to a class of appliance. It helps identify the active appliance, calculates energy bills and records the user activities. This is a one parse system to identify the algorithm performance and appliance events, giving user a real time experience.
  • the second half of the Algorithm focuses on the intricacies, refining the collected data and suggesting improvements that helps improve overall efficiency of the first half. This can also be tagged as a feedback mechanism. It detects unmapped appliance events or power surges from its larger, draws user attention to such triggers, takes user inputs to further analyze the logs generated in the first half. This process helps improve the prediction accuracy and to help fine tune the parameters for efficient appliance detection.
  • the data collection unit includes an input and summing / protection circuit which is the input power module of the hardware unit.
  • the DC power required to run the hardware is derived from this point.
  • the voltage divider circuit delivers the stepped down voltage values to the measuring circuit. Since this module directly accesses the AC power line, there is a provision to protect the rest of the device from undesirable or accidental surges.
  • the input power module has 4 wires connecting it to the main supply - one for each phase and a neutral line. Further, includes a measuring circuit module which handles power measurement.
  • the input Current Transformers (CTs) that are clipped across the phase lines / wires feed the necessary input to this module.
  • This module also receives the voltage signal from the Input circuit.
  • the major component of this module is the metering IC and related circuit.
  • the metering IC is capable to measuring voltages, active and reactive power.
  • the output from this module is fed to the microprocessor over Serial port / SPI.
  • This is a low power circuit and does not include any protection circuits.
  • the unit has a Wi-Fi module/ microprocessor which retrieves the electrical data from the metering / measuring circuit over SPI bus, combines it with relevant information of the present time before it is uploaded to the server for further data processing.
  • the Real-time Clock circuit serves the purpose of providing inputs for the current date and time or epoch value.
  • the microprocessor has Wi-Fi support, and is capable of uploading the recorded data to the cloud during fixed pre-programmed intervals.
  • the recorded data has to be stored in the EEPROM of the processor.
  • the data from the EEPROM will be uploaded once the network connection is reestablished.
  • the unit has a GSM modem which acts as the communication gateway to bring in additional functionality of a Wi-Fi hotspot to connect any future peripherals.
  • the modem supports either 2G/3G/4G/LTE data connections.
  • the modem accepts simultaneous connections from at least 10 devices.
  • the data collection unit will have 4 probes coming out from the main body. 3 probes are intended to measure current across each phase of the power supply.
  • the fourth wire acts as the input voltage detector.
  • This probe also acts as the powering support for the hardware. It can monitor current up to 100 A on each phase (for a 3-phase connection). Also, it captures power info of the distribution board it is attached to.
  • each node will have 3 Current Transformers (CTs) connected to it, and these CTs have to clipped across the 3 incoming phase wires of the distribution board being monitored.
  • CTs Current Transformers
  • the unit has to be powered separately, through the power cables on the device.
  • the power cables (as mentioned above) also serve the purpose of measuring line voltages.
  • the unit also uses inbuilt GSM modem for connectivity to cloud.
  • the hardware can also be configured to use available Wi-Fi networks on the premise.
  • the hardware has inbuilt ROM of 4 MB, and is capable of storing up to 2 days of energy data which will be published to cloud when the connection is re established.
  • the unit at a premise do not communicate with each other, and will only communicate with the cloud server. They may be placed on different floors or far apart on the same premise / building.
  • the system includes a central processing system 310
  • the one or more of data collection unit 340 at a premise are independent with each other, and capable of communicating to the server independently.
  • the central processing system 310 is preferably located on one or more cloud-based servers.
  • the energy monitoring device or the data collection unit 340 including a communication gateway for recording and transferring the recorded data to a remote server over a network 330, the server including a processor and a memory.
  • the device or unit has 4 probes coming out from the main body, wherein the 3 probes are intended to measure current across each phase of the power supply, and the fourth probe acts as the input voltage detector, and wherein the fourth probe also acts as the powering support for the hardware.
  • the device or unit monitors current upto 400 A on each phase (for a 3-phase connection) and also captures power information of the distribution board it is attached to.
  • the unit or device is capable of, in case of data disconnections due to network issues, the memory of the unit stores energy data which will be published to cloud when the connection is re-established.
  • the device or unit 340 may be connected to the system via the network as depicted in figure, for example via the Internet through a wired or wireless connection.
  • the unit comprises of current and voltage measurement circuits, a timer module to record the timestamp of the captured data, a communication gateway that can connect to the server through the internet.
  • the interval between successive data recordings at the data collection unit is set as 1 sec (i.e. a frequency of 1 Hertz) for optimal performance, where the sampled values of 30 seconds are scrambled and uploaded to the server using MQTT protocol.
  • the processor analyze the appliance signature and accumulating information on appliance activity, and also learns about the user behavior drawing patterns about the consumer and his appliance usage activities, with the self-learning capability, the processor is capable of predicting the user behavior about energy consumption at a very granular level which facilitate in improving the convenience and efficiency in their day-to-day activities. Further, the processor scans through real-time energy data to identify patterns which is mapped to a class of appliance in order to identify the active appliance, calculates energy bills and records the user activities thereby providing performance and appliance events in a real time.
  • the processor further focuses on the intricacies, refining the collected data and suggesting improvements that helps improve overall efficiency and further tagged as a feedback mechanism, where the processor further also detects unmapped appliance events or power surges from its larger, draws user attention to such triggers, takes user inputs to further analyze the logs generated in the first half, which helps in improving the prediction accuracy and to fine tune the parameters for efficient appliance detection.
  • the energy monitoring device or unit 340 may include an energy sensor (not shown in figure) and a clock (not shown in figure) receives and stores power series data.
  • the power series data is a data which directly relates to the rate of energy consumption over time of one or several devices or appliances that consume electrical power.
  • the data is obtained by frequent sampling of electrical power consumption. For example electrical power consumption may be sampled once every 10-30 seconds or several times a second, in accordance with the clock. The change in power consumption over a time can then be visualized on a graph as an energy signature.
  • the power series data typically comprises a time series of power (or alternatively energy) consumption values measured at regular time intervals, and specified in a suitable unit such as Watt if (average) electrical power is measured at each time interval or joule or kWh if total energy consumption over each time interval is measured. Time information may be implicit (due to constant measurement intervals) or each power/energy value may be associated with an explicit time value.
  • the energy monitoring device 340 may for example be a whole house power monitoring device, a smart meter or a smart plug.
  • the energy monitoring device may also be integrated into the appliance (shown as load which may include one or more appliance) being monitored.
  • a data log of the energy monitoring device also receives and stores identification information from an internal memory of the device.
  • This identification information may be uploaded by the user or the provider to the internal memory of the device before or during its installation via a user interface 320.
  • the identification information may include information in relation to the nature of the monitoring device and whether the power series data is aggregate data or relates to a single domestic electrical appliance.
  • the identification data may also include information in relation to the user, the house and the household such as the number of occupants in the household, the address of the house, the size of the house, and the socio-economic group of the household. This information may be input by the user via the user interface 320.
  • the location or address of the house may be obtained via a geo-location sensor connected to or part of the energy monitoring device 340.
  • the data log of the central processing system receives the stored power series data and the identification information from the data log of the metering device via a network 330.
  • the processor of the system carries out a disaggregation algorithm to obtain an isolated signature for the stored power series data.
  • the stored power series data may be aggregated power series data of, for example, the whole house (including other domestic appliances), or the power series data may relate to a single appliance for example wherein the metering device is a smart plug.
  • the energy monitoring device or data collection unit or meter which installed non-intrusively (not even like a meter, but as a plug n play device) at the points where these meters are installed.
  • the energy monitoring device provides all these values that a conventional meter can provide, through a web interface and a mobile application which solves the complexity in data consolidation and eases the task of monitoring energy usage.
  • the system including the energy monitoring device provides a summary report of usage from all individual equipment. Say for example, a conventional meter shows daily usage of 2000 kWh. In addition to the usage information (i.e.
  • the presently claimed energy monitoring device can also give the following details: For example, equipment 1 consumed 500 kWh with the times of usage (say during these times of the day), and equipment 2 consumed 260 kWh etc.
  • the claimed system does this without placing a sensor at equipment 1 or equipment 2.
  • It uses the principle of Energy Disaggregation as an approach to perform NILM (Non-intrusive Load Monitoring) algorithm.
  • NILM Non-intrusive Load Monitoring
  • the underlying principle is that each electrical equipment will have unique way of drawing power (current) when it starts working, works and stops working. This is referred to as the electrical signature of the equipment.
  • the energy monitoring device is cost-effective and not complex in installation, and avoids any manual consolidation of readings which eventually becomes difficult because of higher data points.
  • the benefits of analysis using the device is to identify where energy is being used in the premise, and also to identify if any of the equipment is drawing more power than its rated capacity.
  • the claimed system alerts user in such circumstances leading to timely maintenance this preventing downtime due to failure of equipment.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • General Health & Medical Sciences (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)

Abstract

L'invention porte sur un procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive dans des locaux. Dans un mode de réalisation, cela est accompli par la récupération d'une ou de plusieurs données d'énergie brute dans une séquence et leur envoi à un serveur pour chaque durée de 30 secondes sur un réseau, les données d'énergie brutes comprenant un ou plusieurs paramètres électriques et non électriques ; l'analyse des paramètres au niveau du serveur pour détecter des signatures ou des modèles particuliers et pour les mettre en correspondance avec une classe particulière d'appareils et leurs événements avec l'estampille temporelle correspondante d'apparition ; la classification de chaque signature d'événement unique dans une classe particulière d'appareils, et l'activité de l'événement étant en outre catégorisée en tant qu'événement de mise en marche ou d'arrêt ; l'analyse de tous les appareils détectés en temps réel pour calculer la contribution de chaque appareil à l'énergie globale et le stockage des valeurs de consommation d'énergie des locaux dans une base de données.
PCT/IN2020/050611 2020-03-05 2020-07-17 Système et procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive WO2021176459A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IN202041009508 2020-03-05
IN202041009508 2020-03-05

Publications (1)

Publication Number Publication Date
WO2021176459A1 true WO2021176459A1 (fr) 2021-09-10

Family

ID=77613131

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IN2020/050611 WO2021176459A1 (fr) 2020-03-05 2020-07-17 Système et procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive

Country Status (1)

Country Link
WO (1) WO2021176459A1 (fr)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115201615A (zh) * 2022-09-15 2022-10-18 之江实验室 基于物理约束神经网络的非侵入式负荷监测方法及装置
CN115327221A (zh) * 2022-10-13 2022-11-11 北京京仪北方仪器仪表有限公司 一种非侵入式无线电能计量系统及方法
CN116205544A (zh) * 2023-05-06 2023-06-02 山东卓文信息科技有限公司 基于深度神经网络和迁移学习的非侵入式负荷识别系统
CN116861318A (zh) * 2023-09-05 2023-10-10 国网浙江省电力有限公司余姚市供电公司 一种用户用电负荷分类方法、装置、设备及存储介质

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130110621A1 (en) * 2011-02-04 2013-05-02 Myenersave, Inc. Systems and methods for improving the accuracy of appliance level disaggregation in non-intrusive appliance load monitoring techniques
IN2014CH00311A (fr) * 2014-01-24 2015-07-31 Robert Bosch Eng & Business Solutions Ltd

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130110621A1 (en) * 2011-02-04 2013-05-02 Myenersave, Inc. Systems and methods for improving the accuracy of appliance level disaggregation in non-intrusive appliance load monitoring techniques
IN2014CH00311A (fr) * 2014-01-24 2015-07-31 Robert Bosch Eng & Business Solutions Ltd

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115201615A (zh) * 2022-09-15 2022-10-18 之江实验室 基于物理约束神经网络的非侵入式负荷监测方法及装置
CN115201615B (zh) * 2022-09-15 2022-12-20 之江实验室 基于物理约束神经网络的非侵入式负荷监测方法及装置
CN115327221A (zh) * 2022-10-13 2022-11-11 北京京仪北方仪器仪表有限公司 一种非侵入式无线电能计量系统及方法
CN116205544A (zh) * 2023-05-06 2023-06-02 山东卓文信息科技有限公司 基于深度神经网络和迁移学习的非侵入式负荷识别系统
CN116205544B (zh) * 2023-05-06 2023-10-20 山东卓文信息科技有限公司 基于深度神经网络和迁移学习的非侵入式负荷识别系统
CN116861318A (zh) * 2023-09-05 2023-10-10 国网浙江省电力有限公司余姚市供电公司 一种用户用电负荷分类方法、装置、设备及存储介质
CN116861318B (zh) * 2023-09-05 2023-11-21 国网浙江省电力有限公司余姚市供电公司 一种用户用电负荷分类方法、装置、设备及存储介质

Similar Documents

Publication Publication Date Title
WO2021176459A1 (fr) Système et procédé de décomposition d'énergie utilisant une surveillance de charge non intrusive
US20200059424A1 (en) Appliance Network Connectivity Apparatus
Kolter et al. REDD: A public data set for energy disaggregation research
US8983670B2 (en) Energy consumption disaggregation system
EP2496913B1 (fr) Procédé et appareil de surveillance de consommation électrique
Mir et al. Review on smart electric metering system based on GSM/IOT
US9797935B2 (en) Device, arrangement and method for verifying the operation of electricity meter
EP2535997A2 (fr) Systèmes, procédés et appareil pour l'évaluation de la consommation de puissance de charge utilisant un compteur de puissance
US20150268281A1 (en) System and method for monitoring, analyzing and acting upon electricity patterns
KR101172602B1 (ko) 가전기기별 전력 관리를 위한 방법, 시스템 및 컴퓨터 판독 가능한 기록 매체
JP2012141307A (ja) 電力計を利用したエネルギー管理を提供するためのシステム、方法および装置
KR100622985B1 (ko) 수용가의 도전 및 누전 감시가 가능한 원격검침 시스템
KR101608328B1 (ko) 전력소비 패턴 분석을 통한 디바이스 인식장치 및 그 방법
CN109085454B (zh) 一种基于大数据分析的计量串户智能甄别方法
KR20130046693A (ko) 전력 관리 장치 및 그 제어 방법
Mayhorn et al. Characteristics and performance of existing load disaggregation technologies
Nardello et al. A low-cost smart sensor for non intrusive load monitoring applications
Geetha et al. Smart metering system
Santhosh et al. Iot based monitoring and optimizing of energy utilization of domestic and industrial loads
CN107250931B (zh) 能源服务推荐引擎
WO2018064244A1 (fr) Détection de ligne de transmission électrique
Reinhardt et al. Can smart plugs predict electric power consumption? A case study
CN111328375A (zh) 使用仪表数据解集来检测电力盗窃的系统和方法
Singh et al. Distribution Transformer Monitoring for smart grid in India
Kelati et al. Iot based appliances identification techniques with fog computing for e-health

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: 20923633

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20923633

Country of ref document: EP

Kind code of ref document: A1