Embodiments implement non-intrusive load monitoring using a novel learning scheme. A trained machine learning model configured to disaggregate device energy usage from household energy usage can be stored, where the machine learning model is trained to predict energy usage for a target device from household energy usage. Household energy usage over a period of time can be received, where the household energy usage includes energy consumed by the target device and energy consumed by a plurality of other devices. Using the trained machine learning model, energy usage for the target device over the period of time can be predicted based on the received household energy usage.

A monitoring system that is configured to monitor a property is disclosed. In one aspect, the monitoring system includes a sensor that is located at the property and that is configured to generate sensor data. The monitoring system further includes a voltage sensor that is configured to generate voltage data by measuring voltage at an electrical outlet located at the property. The monitoring system further includes a monitor control unit that is configured to receive the sensor data; receive the voltage data; determine an action of an electrical device that is located in the property or that is located at a neighboring property in a vicinity of the property; determine whether the electrical device is located at the property or at the neighboring property in the vicinity of the property; and perform a monitoring system action.

A system is provided for tracking, in a distributed water infrastructure, water usage by category. The system may comprise at least one processor configured to receive from at least one sensor associated with the distributed water infrastructure signals indicative of water usage in the distributed water infrastructure. The system may, based on the signals indicative of water usage, construct a plurality of profiles. The system may assign each profile to one of a plurality of water usage categories. The system may collect, from the at least one sensor, signals indicative of water usage for substantially all water delivered through the distributed water infrastructure in a time period. The system may construct a plurality of water usage profiles in the aggregate, encompassing substantially all water delivered through the distributed water infrastructure in the time period. The system may assign each constructed water usage profile to one of the plurality of water usage categories. The system may output, for display, water usage for the time period for each of the plurality of water usage categories.

A system is provided for tracking usage of a plurality of water appliances in a distributed water infrastructure. The system may comprise at least one processor configured to receive, from a location in the distributed water infrastructure upstream of the plurality of water appliances, historical water usage measurements; determine from the historical water usage measurements at least one unique water usage signature associated with each of the plurality of water appliances; receive, from the location in the distributed water infrastructure upstream of the plurality of water appliances, current water usage measurements; determine from the current water usage measurements at least one current water usage signature; compare the current water usage signature with at least one of the unique water usage signatures stored in memory to determine a match; and, based on the signature match, ascertain an identifier of a water appliance in current use.

A method of disaggregating an energy usage signal of a usage area includes steps of providing a gateway device for an aggregate energy usage signal of a usage area, installing a user application on a user computing device to display information from the gateway device, receiving a plurality of inputs, and determining the energy usage of the individual electrically powered devices in the usage area based on the aggregate energy usage signal and based on the plurality of inputs.

A system is provided for determining whether a specific water appliance is malfunctioning. The system may comprise at least one processor configured to detect, from at least one sensor in a distributed water infrastructure upstream of the plurality of water appliances, a plurality of normal water usage profiles; associate at least one of the plurality of profiles with each of the plurality of appliances; detect at least one current water usage profile; and compare the at least one current profile with at least one of the stored profiles to determine a corresponding identity of an associated appliance and to determine if a deviation exists between the stored profile for the identified appliance and the at least one current profile. The deviation may be reflective of a potential malfunction in the associated appliance. The system may initiate remedial action if the deviation, reflective of a potential malfunction, is determined.

This disclosure relates generally to method and system for low sampling rate electrical load disaggregation. At low sampling rates, disaggregation of energy load is challenging due to unavailability of events and signatures of the constituent loads. The disclosed energy disaggregation technique receives aggregated load data from a utility meter and sequentially obtains training data for determining disaggregated energy load at low sampling rate. Dictionaries are used to characterize the different loads in terms of power values and time of operation. The obtained dictionary coefficients are treated as graph signals and graph smoothness is used for propagating the coefficients from the training phase to the test phase by formulating an optimization model. The derivation of the optimization model identifies the load of interest and estimate their power consumption based on optimization model constraints. This method achieves accuracy greater than 70% for the loads of interest at low sampling rates.

A method for operating a power consumption metering system and a power consumption metering system are disclosed. In an embodiment a method include measuring, by a sensor deployed at a monitored site, high speed power consumption values over time to obtain a high speed value pattern of power consumption with a resolution of more than 1000 values per second, determining one or more harmonics of the high speed value pattern, measuring, by the sensor, low speed power consumption values over time to obtain a low speed value pattern of the power consumption with a resolution of less than 100 values per second, providing the harmonics and the low speed value pattern to a cloud based data processing system and identifying a status of a power consumer of the monitored site dependent on the measured harmonics and the low speed value pattern.

A load estimating device measures a voltage and a current supplied to a plurality of loads connected with a power supply, and obtains feature amounts of the plurality of loads from measurement values of the voltage and the current. A storage device stores a feature amount of each combination of two or more loads in advance. The load estimating device estimates what the plurality of loads connected with the power supply device are, on the basis of the obtained feature amounts and the feature amounts stored in the storage device. The feature amount includes a combination of an apparent power and a power factor.

A system is provided for detecting abnormal consumption in one portion of a distributed water infrastructure while normal water usage occurs in another portion of the distributed water infrastructure. The system may comprise at least one processor. The system may receive from at least one sensor associated with the distributed water infrastructure indications of regular water usage. The system may determine, from a plurality of indications received over a time period, a plurality of baseline water usage profiles. The system may receive from the at least one sensor a current water usage profile. The system may compare the current water usage profile with the plurality of baseline water usage profiles. The system may determine an abnormal water consumption based on the comparison between the current water usage profile and the plurality of baseline water usage profiles. The system may generate an abnormal water consumption signal when abnormal water consumption is determined.