Classifications

• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
  • G01D4/00—Tariff metering apparatus
  • G01D4/002—Remote reading of utility meters
  • G01D4/004—Remote reading of utility meters to a fixed location

Definitions

• the present disclosure relates to the field of human activity monitoring, more particularly to the field of non-intrusive human activity monitoring.
• United States Patent Publication US20120053472 discloses a comprehensive solution for safety monitoring amongst old and/or sick people. It discloses the use of electricity, water and gas meters to monitor activities of an individual within a household. It also helps the individual in acquiring required help through direct correspondence with paramedics. In this process of direct correspondence, the disclosure focuses on sharing the acquired data with the paramedics, which might not necessarily be the primary recipient of the data.
• Another United States Patent US 8340831 discloses the use of Mixed Integer Programming on the operational data collected via a number of sensors to estimate the corresponding household activity. This disclosure uses non-intrusive methods for activity monitoring, by utilizing data obtained only with the help of electricity meters. However, this disclosure does not talk about analyzing the risk metrics.
• US20120158618 discloses the system for non-intrusive space monitoring. This system uses sensors which when attached to the processor provide signals, demonstrative of the electric supply signatures. This disclosure suggests the use of utilities like as gas, water and telephone to monitor human activity and human signature identification and verification. But the sensors mentioned in the disclosure may be intrusive in nature.
• An object of the present disclosure is to provide an activity monitoring system that does not intrude on the privacy of an individual.
• Another object of the present disclosure is to provide an activity monitoring system that eliminates health risks associated in monitoring the individual. Yet, another object of the present disclosure is to provide an activity monitoring system that does not interfere in the day-to-day activities of the individual.
• Another object of the present disclosure is to provide an activity monitoring system that does not instill a sense of fear in the individual being monitored.
• Another object of the present disclosure is to provide an activity monitoring system which requires less capital expenditure.
• Another object of the present disclosure is to provide an activity monitoring system which requires minimum maintenance requirements.
• Another object of the present disclosure is to provide an activity monitoring system which can aid in monitoring and helping patients with specific cognitive disabilities.
• Yet another object of the present disclosure is to provide an activity monitoring system which works efficiently and reliably.
• the present disclosure envisages a system for non-intrusive monitoring and determining activity patterns with the help of utility meters provided in a facility.
• the system for non-intrusive monitoring comprises:
• a communicator adapted to receive and transmit consumption signals from at least one utility meter; ii. a processor adapted to receive and process a plurality of consumption signals received via the communicator and further comprising a pattern determinator adapted to determine activity patterns in relation to predetermined time intervals; and iii. a repository adapted to store the activity patterns.
• the utility meter is selected from the group consisting of an electricity meter, a gas meter, a water meter and a heat meter.
• the system includes a comparator adapted to compare currently determined activity patterns with stored activity patterns to detect an anomaly.
• the processor further includes: i. a disaggregation module adapted to individually disaggregate the consumption signals to obtain data points in relation to time, wherein the data points are further plotted over a period of time to obtain activity patterns corresponding to an individual utility; ii. a data fusion module adapted to extract features from the activity patterns, the features including but not limited to start and stop times of utility usage, duration and consumption of utility usage, maximum and minimum utility consumption over a period of time; and iii. a template creation module adapted to create activity patterns denoting human activities in relation to the extracted features.
• the comparator comprises an alarm device adapted to emit an alarm in the event of an anomaly.
• the communicator includes a communication link adapted to communicate with insurance and utility service providers in the event of the anomaly. Additionally, the system includes a transmitter adapted to transmit activity patterns and detected anomalies to predetermined devices.
• a method for non- intrusively monitoring and determining activity patterns with the help of utility meters provided in a facility comprising the following steps:
• the method further includes the step of generating an alarm in the event of an anomaly.
• the method further includes the step of transmitting the activity patterns and detected anomalies to the pre-determined devices.
• the method includes the step of converting the activity patterns into a risk metric and communicating the risk metric to insurance and utility service providers.
• FIG. 1 illustrates the schematic of an activity monitoring system in accordance with an embodiment of the present disclosure.
• Figure 2 illustrates the schematic of an activity monitoring system for providing the risk metric to the utility and insurance service providers.
• Figure 1 illustrates the system for activity monitoring 100.
• an electric smart meter and water meter installed in a facility is used to monitor human activity in that facility.
• the activity monitoring system 100 of the present embodiment acquires its electric meter data 10 from an electric smart meter and its water meter data 20 from a water smart meter.
• the electric meter data 10 and the water meter data 20 is provided to a data acquisition module 30 having multiple interfaces for the smart water meters and the electricity meters.
• the data acquisition module 30 processes the received data to generate a pre-processed data by performing scaling and synchronization on the received data. This pre-processed data is then provided to a disaggregation and data- fusion module 40.
• the disaggregation modules 42 and 44 use a back-and-forth method to improve the disaggregation accuracy of electricity and water. These disaggregation modules 42 and 44 perform individual disaggregation of both electricity and water in the first step, with limited use of the information from the other meter. In further steps, the disaggregated results of water are used to improve the electricity disaggregation and vice versa. These steps are then repeated to achieve equilibrium in the results. Once the results are obtained, the data-fusion module 46 extracts various features that are useful for developing a template.
• the features to be extracted include start and stop times of appliance usage, duration and consumption of appliance usage, start and stop times of fixture usage, consumption from fixture usage, maximum consumption and its time of usage, water and electricity consumptions over different times of day, peak power consumption and time of usage, maximal water flow rate and time of usage.
• the activity template creation module 50 trains a multitude of different models that can represent the activities based on the features extracted by the data-fusion module 46. The training is performed based on the features extracted on historical measurement data.
• the activity template creation module 50 trains a factor graph model that represents the activities.
• the activity template creation module 50 associates various human activities to feature vectors that are extracted in the data- fusion module 46.
• the template models provide a representation for the association from feature vectors to human activity labels.
• the activity template creation module 50 develops a consumption template based on the features extracted in the case of smart meters with low sampling rates. These consumption templates can also be represented using the template models.
• the activity template creation module 50 trains the model either periodically or in a continuous basis. In case of periodic approach, the models are trained once in every few days so that it is up-to-date with respect to the changed behavior of the inmates. In case of the continuous approach, the training will happen in near real-time to update the model parameters.
• a detection module 60 includes an activity detection module 62 and an anomaly detection module 64. These modules accept a predicted output from the trained model and compare it with the processed real-time measurement. This results in identifying different activities and their temporal characteristics.
• the detection module 60 develops a consumption template for the current data, it also provides a confidence measure on the activities detected to enable decision making.
• the anomaly detection module 64 raises an alarm in case of a marked difference in the predicted vs. actual output.
• the invention can aid in monitoring the elderly in a household.
• the anomaly detection module 64 is connected to a communication interface in the form of Short Message Service (SMS), email and the like to convey the anomalous behavior to a concerned person. This can in-turn be used by an individual who is monitoring the elderly to check upon them.
• SMS Short Message Service
• the system can aid in monitoring and helping patients with specific cognitive disabilities (e.g., memory loss due to old age, Alzheimer's).
• the anomaly detection module 64 can trigger a message (e.g. SMS) to the occupant to remind them of a missed activity.
• Figure 2 illustrates the schematic of an activity monitoring system for providing the risk metric to the utility and insurance service providers 200.
• the disaggregation framework present in the current disclosure allows for an approach to improve accuracy by allowing for intentional modification of loads or using explicit information about unique loads in the household. In one possible embodiment, this can be in the form of unique lighting fixture in different rooms.
• Such engineered loads 140 help in monitoring activities in case multiple people are present in a facility. Similar electrical loads in different places within a facility can be chosen such that their electrical ratings are different. For instance, in a house with three bathrooms, lights for individual bathrooms can be chosen in such a way that they have different power ratings.
• the outcome of load disaggregation can be used to identify the location of the loads within the house. This, in turn, helps in monitoring the activities associated with those electrical loads.
• the present disclosure proposes that the risk can be modeled as a function of parameters such as (a) power rating of appliance, (b) age of appliance, (c) ease of operation of appliance. Further, the present disclosure proposes to send a risk metric (computed based on the risk model) to the insurance and utility providers.
• the processing module 160 consists of a human activity monitoring module 164 and a home gateway module 162.
• the human activity monitoring module 164 involves detecting and monitoring the consumption of utilities and detecting the anomalies in the duration of operation, and consumption or impact on the electrical, water systems of the facility.
• the home gateway module 162 communicates only the information related to overall risk and power utilization and does not communicate any information about human activities. This allows the system to be sensitive to privacy of individuals.
• a risk metric is then extracted from the data provided by these modules and then communicated to the utility providers 170 and insurance 180 providers.
• the proposed system acquires the electricity and water smart meter data from the smart meters and yields disaggregated electricity and water usages.
• the electric usages include but are not limited to the use of microwave oven, geyser, clothes washer and dish washer.
• the disaggregation can be based on appliance/fixture characteristics or the consumption characteristics per appliance/fixture.
• the disaggregated water and electric usages are used to confirm the room/location in which the human activity is taking place. This can be aided by contextual information. For example, an oven being operational and kitchen tap being On' at a particular part of the day implies a cooking activity in the kitchen. Similarly a bathroom light being On' and a flush usage imply a toilet activity.
• Some of the electric appliances like clothes washer and dish washer also use water to perform the tasks of clothes or dish washing. Therefore, inputs from the water disaggregated usages help in accurate identification of water consuming electric appliance actually in use. Such inputs are essential when the electric appliances share the same waveform and electric load disaggregation is not sufficient.
• a rule-based approach is used to associate the usage to human activity. If the lights and oven in kitchen are operational during night and there are intermediary kitchen sink usages identified around the same time duration, the usages can be related to cooking.
• Templates of similar activities are formed using historical data from the facilities and the classifiers are trained with these templates or activity labels. These templates are formed by the process of disaggregation and data fusion/association. For example, a cooking activity might involve the use of oven, lights, and kitchen tap. The disaggregated electric activities are cross-checked with the list of disaggregated water activities to check if the electricity and the water used correspond to cooking. Repeated observations including the time of the day/ day of the week information will create a template for cooking activity. Similarly templates are formed for different

Landscapes

• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Alarm Systems (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=52394283

Family Applications (1)

Country Status (2)

Cited By (6)

Citations (5)

• 2014
  • 2014-12-31 IN IN570MU2014 patent/IN2014MU00570A/en unknown
  • 2014-12-31 WO PCT/IB2014/067437 patent/WO2015124972A1/en active Application Filing

Patent Citations (5)

Also Published As

Similar Documents

Legal Events