CLIAMS:1. A computer implemented method for determining human activities from smart meter data, the method comprising:
receiving an aggregate consumption data from a plurality of smart meters;
determining consumption data, by disaggregating the aggregate consumption data, corresponding to one or more appliances;
generating a plurality of template feature vectors for each activity with respect to plurality of historical data;
generating a plurality of test vectors from the consumption data of the one or more appliances;
comparing the plurality of test vectors with the plurality of template feature vectors corresponding to a plurality of pre-defined activities;
generating a set of candidate activities from said comparison;
determining a plurality of human activities based on the set of candidate activities and contextual information related to the plurality of pre-defined activities.
2. The method as claimed in claim 1, wherein the plurality of template feature vectors correspond to an n-dimensional vector, and wherein each field in the n-dimensional vector stores at least one attribute of at least one appliance used for determining human activities.
3. The method as claimed in claim 2, wherein the dimensionality of the n-dimensional vector is equal to the total number of appliances used for determining human activities.
4. The method as claimed in claim 2, wherein the dimensionality of the n-dimensional vector is extended to a tuple for storing additional information about the corresponding appliance used for determining human activities.
5. The method as claimed in claim 1, wherein the method further comprises assigning a weight-age to each of said appliance.
6. The method as claimed in claim 1, wherein the method further comprises updating the plurality of template feature vectors by combining a past template feature vector and a current template feature vector by means of a weighted average.
7. The method as claimed in claim 5, wherein the weight-age assigned to each of said appliance represents the importance of each of said appliance for each of said activity.
8. The method as claimed in claim 1, wherein the consumption data includes time sequence information corresponding to usage of each of said appliances.
9. The method as claimed in claim 1, wherein generating the plurality of test vectors further comprises duration-based windowing on the consumption data of each of said appliance.
10. The method as claimed in claim 1, wherein comparing the plurality of test vectors with the plurality of template feature vectors is based on an innerproduct metric.
11. The method as claimed in claim 1, wherein the contextual information includes time-of-day information of each of said activity and duration of each of said activity.
12. An activity detection system (202) for determining human activities from smart meter data, the system (202) comprising:
a processor (212); and
a load disaggregation module (220) coupled to the processor (212), wherein the load disaggregation module (220) is configured to,
receive an aggregate consumption data from a plurality of smart meters; and
determine consumption data, from the aggregate consumption data, corresponding to one or more appliances; and
an activity training module (222) coupled to the processor (212), wherein the activity training module (222) is configured to,
generate a plurality of template feature vectors for each activity with respect to plurality of historical data, wherein the plurality of template feature vectors correspond to an n-dimensional vector; and
a recognition module (224) coupled to the processor (212), wherein the recognition module (224) is configured to,
generate a plurality of test vectors from the consumption data of the one or more appliances; and
compare the plurality of test vectors with the plurality of template feature vectors corresponding to a plurality of pre-defined activities; and
generate a set of candidate activities from said comparison;
determine a plurality of human activities based on the set of candidate activities and a contextual information related to said plurality of pre-defined activities; and
an updation module (226) coupled to the processor (212), wherein the updation module (226) is configured to,
update the plurality of template feature vectors by combining a past template feature vector and a current template feature vector by means of a weighted average.
13. The system (202) as claimed in claim 12, wherein the each field of the n-dimensional vector is further configured to store at least one attributes of at least one appliance used for the determining human activities.
14. The system (202) as claimed in claim 13, wherein the each field of the n-dimensional vector is extended to a tuple to store additional information of the at least one appliance used for determining human activities.
15. The system (202) as claimed in claim 12, wherein the updation module (226) is further configured to assign a weight-age to each of said appliance, and wherein the assigned weight-age represents the importance of each of said appliance for each of said activity.
16. The system (202) as claimed in claim 12, wherein the consumption data corresponding to each of said appliance is further configured to include time sequence information corresponding to usage of each of said appliance.
17. The system (202) as claimed in claim 12, wherein the recognition module (224) is further configured to generate the plurality of test vectors based on duration-based windowing.
18. The system (202) as claimed in claim 12, wherein the recognition module (224) is further configured to compare the plurality of test vector with the plurality of template feature vectors based on an innerproduct metric.
19. The system (202) as claimed in claim 12, wherein the recognition module (224) is further configured to include a time-of-day information of each of said activity and a duration of the each of said activity for each of said contextual information.
20. A non-transitory computer-readable medium having embodied thereon a computer readable program code for executing a method for determining human activities from smart meter data, the method comprising:
receiving an aggregate consumption data from a plurality of smart meters;
determining consumption data, from the aggregate consumption data, corresponding to one or more appliances;
generating a plurality of template feature vectors for each activity with respect to plurality of historical data;
generating a plurality of test vectors from the consumption data of the one or more appliances;
comparing the plurality of test vectors with the plurality of template feature vectors corresponding to a plurality of pre-defined activities;
generating a set of candidate activities from said comparison;
determining a plurality of human activities based on the set of candidate activities and contextual information related to the plurality of pre-defined activities. ,TagSPECI:TECHNICAL FIELD
The present subject matter relates, in general, to determining human activities through consumption of resources and, in particular, to determining human activity using smart meters.
DEFINITIONS OF TERMS USED IN THE COMPLETE SPECIFICATION
The expression ‘activities’ used hereinafter in the complete specification refers to activities in daily living as performed by humans, wherein, the complete specification particularly focuses on the daily human activities that consume various utilities including electricity, gas and water.
The expression ‘template feature vectors’ used hereinafter in the complete specification refers to novel representation of each activity in terms of a fixed-dimensional vector, and each field in the vector stores the attributes of an appliance used in the human activity.
The expression ‘test vectors’ used hereinafter in the complete specification refers to a set of activities which are generated from appliance data using duration-based windowing.
The expression ‘candidate activities’ used hereinafter in the complete specification refers to a set of activities which are selected from used set of appliances.
The expression ‘duration-based windowing’ used hereinafter in the complete specification refers to detecting a particular activity by scanning the particular activity related vector through the entire appliance time series for any given day.
The expression ‘innerproduct metric’ used hereinafter n the complete specification refers to the similarity sequence between two activities by correlating the amount of correlation per field.
The expression ‘load disaggregation’ used hereinafter n the complete specification refers to a module which disaggregates the given aggregated data from a smart meter such as an electrical meter and provides a consumption data of a particular appliance.
These definitions are in addition to those expressed in the art.
BACKGROUND
Resources, such as gas, electricity, and water are generally provided by utility companies around the world to households, businesses, and other consumers. The utility companies typically charge the consumers based on an amount of resources consumed by a consumer. For this, utility companies may deploy a meter, which functions as a measurement device to provide information pertaining to consumption of a particular resource. In other words, the meter may determine the amount of resources consumed, for example, an electricity meter determines amount of electricity consumed, a water meter measures amount of water utilized, and a gas meter that measures amount of gas burnt.
Conventionally, a meter such as a smart meter is an advanced digital electric meter located at the consumer’s structure or site of distribution of the resources. The smart meter may be a digital electric, water, or gas meter that records consumption in intervals and communicates the information via a communications network back to a utility company for monitoring and billing purposes (e.g., telemetering). The consumer’s structure may be, for example, the consumer’s home or office. The meter may be owned by the utility company and may be installed in a standard meter box. Thus, utility companies gauge consumption using the meters and bill their consumer’s appropriately. Accordingly, periodic reading in a predetermined cycle, such as monthly, quarterly, and half yearly of the meter may be performed to determine the resource consumption and to bill the consumer for the amount consumed. Further, the smart meter may also suspend the delivery of services provided under any condition.
Further, in an average household, there are numerous appliances that are used for daily activities that consume energy and natural resources. It is difficult to identify specific activities carried out by the individuals as multiple different appliances consume similar energy and/or natural resources. Various data sources like smart meters or sensors are usually installed in the households or buildings in order to monitor human activities. However, the deployment of sensors for capturing the activities or the audio and video are intrusive and privacy invasive. In the present context, the human activity detection using smart meter data provides possible alternatives to the issues related to the privacy invasion coming out from the deployment of sensors.
However, the conventional methods of activity detection using smart meter are based on the methodologies such as Rule-based methods, Model-based methods have some limitations. More specifically, the Rule-based methods are too specific where each activity is defined in terms of various appliances involved in that activity. Normally, a new set of rule need to be formulated for every new activity, thus such methods lacks the ability to generalize. Similarly, the Model-based methods need large amount of data for accurate estimation of parameters. In such Model-based methods, the training process, recognition of activities, and update process are computationally complex.
Hence, there is a need for a system that provides the non-intrusive and privacy preserving detection of human activities using smart meter installed in a household or in similar places.
OBJECTS
An object of the present disclosure is to provide non-intrusive and privacy preserving determination of human activities from smart meter installed in a household or in similar places.
Another object of the present disclosure is to detect events in order to detect human activities associated with the occurrence of individual consumption events.
Still another object of the present disclosure is to provide representation of human activities in terms of smart meter data.
Still another object of the present disclosure is to extract and use the contextual information for human activity detection.
Still another object of the present disclosure is to detect anomalies and/or deviations from expected set of activities.
Still another object of the present disclosure is to provide non-intrusive monitoring of human activities in case of elderly individuals.
SUMMARY
This summary is provided to introduce concepts related to determining human activities from smart meter data. This summary is neither intended to identify essential features of the claimed subject matter nor is it intended for use in determining or limiting the scope of the present disclosure.
In an embodiment, method(s) and system(s) to non-intrusively determining human activity using the smart meter data is disclosed. The method includes receiving an aggregate consumption data from multiple smart meters installed in a household. The method, further, includes determining consumption data corresponding to various appliances from the received aggregated consumption data. Additionally, the method includes generating various template feature vectors corresponding to every activity with the help of historical data. Further, the method includes generating multiple test vectors from the determined consumption data of the multiple appliances. Once the test vectors and the template feature vectors are generated, the method, further, includes comparing the test vectors with template vectors to generate candidate activities. Subsequently, the method includes determining the various human activities using the generated candidate activities and contextual information related to the set of activities.
BRIEF DESCRIPTION OF THE FIGURES
The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the drawings to reference like features and modules.
Fig. 1 illustrates an exemplary layout of non-intrusive load monitoring modules, according to an implementation of the present disclosure.
Fig. 2 illustrates a network implementing an activity detection system for determining human activities from smart meter data, according to an implementation of the present disclosure.
Fig. 3 illustrates a method for determining human activities from smart meter data, according to an implementation of the present disclosure.
It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.
DETAILED DESCRIPTION
The present disclosure relates to a system and a method for determining human activities from smart meter data non-intrusively.
Unless specifically stated otherwise as apparent from the following discussions, it is to be appreciated that throughout the present disclosure, discussions utilizing terms such as “receiving” or “determining” or “generating” or “comparing” or the like, refer to the action and processes of a computer system, or similar electronic activity detection device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices.
The systems and methods are not limited to the specific embodiments described herein. In addition, modules of each system and each method can be practiced independently and separately from other modules and methods described herein. Each module and method can be used in combination with other modules and other methods.
According to an implementation, the present subject matter discloses a system and method for determining human activities using a smart meter data whereas the smart meter may be located at the consumer’s structure such as a household. The smart meter data consist of aggregate values of resources consumed such as electricity, water and gas. The disaggregation of the smart meter data provides the consumption due to individual appliances. Further, the data related to the usage of appliances such as timing of usage, and sequence in which appliances are used, are analysed to infer the human activities non-intrusively.
In another implementation, smart meters data are used to generate a representation for each activity. In other words, the activity is described in terms of appliances used and their timings such as start and end timings of appliance usage. Further, the generation of representation is same for all kind of activities, therefore, the approach is adaptable to include new human activities. Further to the representation of each of the activity, a template feature vectors generated for each of the activity. Additionally, a set of contextual information are also extracted from each activity data.
In another implementation, the determining of human activities from appliance data includes the steps of constructing of a test feature vectors from appliance data. The constructing of test feature vectors includes the mining of time series of number of occurrences of appliances. Thereafter, determining the weight-age for each appliance where it is used. For example, the weight-age for an appliance may be calculated by providing the information regarding the number of time an appliance is used divided by the number of time a set of activities being performed. Once the test feature vectors are calculated, the template feature vectors are compared with the test feature vectors. Further, based on the comparison, the similarity sequence for each activity is generated. The similarity sequences along with the contextual information are hypothesised to determine a set of activities being performed non-intrusively.
In another implementation, the present disclosure provides direction towards updation and/or modification of features and/or parameters associated with different activities. The template feature vectors being updated by combing a past template feature vector and a current feature vector by means of a weighted average. The weight-age is decided on the basis of number of instances of an activity used to compute said past template feature vector and said current template feature vector. Further, the weight-age assigned to each of the appliance represents the importance of said appliance for each of said activity being detected.
In another implementation, the template feature vectors correspond to a fixed-dimensional vector such as n-dimensional vector, and each field in the n-dimensional vector stores at least one attributes of an appliance used for the determining human activities. Further, the dimensionality of the n-dimensional vector may be equal to the total number of appliances used for determining human activities. Furthermore, the dimensionality of the n-dimensional vector may be extended to a tuple for storing additional information about the corresponding appliance used for determining human activities.
Although the description herein is with reference to a smart meter, it would understood that the systems and methods may be implemented for other applications, albeit with a few variations, where the actual resource consumption can be predicted based on multiple parameters, as will be understood by a person skilled in the art.
Throughout the description and claims of this complete specification, the word “comprise” and variations of the word, such as “comprising” and “comprises,” means “including but not limited to,” and is not intended to exclude, for example, other additives, components, integers or steps. “Exemplary” means “an example of” and is not intended to convey an indication of a preferred or ideal embodiment. “Such as” is not used in a restrictive sense, but for explanatory purposes.
For a firmware and/or software implementation, the methodologies can be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. Any machine readable medium tangibly embodying instructions can be used in implementing the methodologies described herein. For example, software codes and programs can be stored in a memory and executed by a processing unit.
In another firmware and/or software implementation, the functions may be stored as one or more instructions or code on a non-transitory computer-readable medium. Examples include computer-readable media encoded with a data structure and computer-readable media encoded with a computer program. The computer-readable media may take the form of an article of manufacturer. The computer-readable media includes physical computer storage media. A storage medium may be any available medium that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer; disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blue-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
It should be noted that the description merely illustrates the principles of the present subject matter. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described herein, embody the principles of the present subject matter and are included within its spirit and scope. Furthermore, all examples recited herein are principally intended expressly to be only for pedagogical purposes to aid the reader in understanding the principles of the invention and the concepts contributed by the inventor(s) to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the invention, as well as specific examples thereof, are intended to encompass equivalents thereof.
The manner, in which the systems and methods for determining human activities from smart meter data shall be implemented, has been explained in details with respect to the Fig. 1, 2 and 3. While aspects of described systems and methods determining human activities from smart meter data can be implemented in any number of different activity detection systems, utility environments, and/or configurations, the embodiments are described in the context of the following exemplary system(s).
Fig. 1 illustrates an exemplary layout of non-intrusive load monitoring modules, according to an implementation of the present disclosure. In the present exemplary layout, the smart meters such as an electrical meter, a water meter, and a gas meter are installed to measure the resource utilization by their respective appliances. For example, electricity loads consumed by a microwave oven or a refrigerator or a table lamp are measured aggregately by the electrical meter. Likewise, water loads consumed by a shower equipment or a dishwasher are measured by the water meter and gas loads consumed by the cooking equipment or a room heater are measured by the gas meter. Subsequently, the aggregated loads preferably in a digital format are provided to their respective load disaggregation module to disaggregate the aggregated loads data to determine the consumption by the individual appliances with their contextual information. For example, the aggregated electrical load data is provided to an electrical load disaggregation to determine the consumption of electricity by the microwave oven or the refrigerator on a given time and date. Similarly, the aggregated water load data is provided to a water load disaggregation to determine the consumption of water by the dishwasher on a given time and date and the aggregated gas load data is provided to a gas load disaggregation to determine the consumption of gas by the cooking equipment or the room heater on a given time and date. Subsequent to the disaggregation of aggregated data, the data related to usage of the appliances such as timing of usage and sequence in which the appliances are used, may analysed to infer the possible human activities.
Fig. 2 illustrates a network implementing an activity detection system 202 for determining human activities from smart meter data. The activity detection system 202 can be implemented as a variety of communication devices, such as a laptop computer, a notebook, a workstation, a mainframe computer, a server and the like. The activity detection system 202 described herein, can also be implemented in any network environment comprising a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc. Further, the activity detection system 202 may be implemented inside smart meter devices 204.
In one implementation, the activity detection system 202 is connected to one or more smart meter devices 204-1, 204-2…204-N, individually and commonly hereinafter referred to as device(s) 204, and a database 208, through a network 206. The devices 204 may be implemented as, but are not limited to, hand-held devices, laptops or other portable computers, tablet computers, mobile phones, personal digital assistants (PDAs), Smartphone, and the like. The devices 204 may be located within the vicinity of the activity detection system 202 or may be located at different geographic location as compared to that of the activity detection system 202. Further, the devices 204 may themselves be located either within the vicinity of each other, or may be located at different geographic locations.
The network 206 may be a wireless or a wired network, or a combination thereof. The network 206 can be a collection of individual networks, interconnected with each other and functioning as a single large network (e.g., the internet or an intranet). The network 206 can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and such. The network 206 may either be a dedicated network or a shared network, which represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), etc., to communicate with each other.
The database 208 may be implemented as, but not limited to, enterprise database, remote database, local database, and the like. The database 208 may be located within the vicinity of the activity detection system 202 and devices 204 or may be located at different geographic location as compared to that of the activity detection system 202 and devices 204. Further, the database 208 may themselves be located either within the vicinity of each other, or may be located at different geographic locations. Furthermore, the database 208 may be implemented inside the device 204 or inside the activity detection system 202 and the database 208 may be implemented as a single database.
In one implementation, the activity detection system 202 includes processor(s) 212. The processor 212 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) is configured to fetch and execute computer-readable instructions stored in a memory.
The functions of the various elements shown in the figure, including any functional blocks labelled as “processor(s)”, may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Moreover, explicit use of the term “processor” should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read only memory (ROM) for storing software, random access memory (RAM), non-volatile storage. Other hardware, conventional and/or custom, may also be included.
Also, the activity detection system 202 includes interface(s) 210. The interfaces 210 may include a variety of software and hardware interfaces that allow the activity detection system 202 to interact with the entities of the network 206, or with each other. The interfaces 210 may facilitate multiple communications within a wide variety of networks and protocol types, including wire networks, for example, LAN, cable, etc., and wireless networks, for example, WLAN, cellular, satellite-based network, etc.
The activity detection system 202 may also include a memory 214. The memory 214 may be coupled to the processor 212. The memory 214 can include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.
Further, the activity detection system 202 may include module(s) 216 and data 218. The modules 216 may be coupled to the processors 212 and amongst other things, include routines, programs, objects, components, data structures, etc., which perform particular tasks or implement particular abstract data types. The modules 216 may also be implemented as, signal processor(s), state machine(s), logic circuitries, and/or any other device or component that manipulate signals based on operational instructions. Further, the modules 216 can be implemented in hardware, instructions executed by a processing unit / processor, or by a combination thereof. In another aspect of the present subject matter, the modules 216 may be machine-readable instructions (software) which, when executed by a processor/processing unit, perform any of the described functionalities.
In an implementation, the modules 216 may include a load disaggregation module 220, an activity training module 222, a recognition module 224, and other module(s) 226. The other module(s) 226 may include an updation module and programs or coded instructions that supplement applications or functions performed by the activity detection system 202. Further, the data 218 may include consumption data 228, activity data 230, and other data 232. The other data 232, amongst other things, may serve as a repository for storing data that is processed, received, or generated as a result of the execution of one or more modules in the modules 216. Although the data 218 is shown internal to the activity detection system 202, it may be understood that the data 218 can reside in an external repository, which may be coupled to activity detection system 202.
In one implementation, the activity detection system 202 may provide information related to human activities from smart meter data. The activity detection system 202 may include the load disaggregation module 220 to receive an aggregate consumption data from smart meters installed in a household. Further, the load disaggregation module 220 configured to determine a consumption data corresponding to individual appliance.
According to the present implementation, the activity detection system 202 may include the activity training module 222 to generate the template feature vectors for each activity with respect to plurality of historical data. Further, each of the template feature vectors correspond to a fixed-dimensional vectors such as n-dimensional vector, where the dimensionality of the n-dimensional vector may be equal to the total number of appliances used for the determining the activities.
According to the present implementation, the activity detection system 202 may further include the recognition module 224 to generate the test vectors from the consumption data of the individual appliance. Further, the consumption data corresponding to the individual appliance may include time sequence information corresponding to the usage of the same appliance. Further, the test vectors generated by the recognition module 224 may be based on duration based windowing. Since each filed in the fixed-dimensional vectors stores the attributes of an appliance used in the human activity. Therefore, in an exemplary scenario, different activities are associated with different fixed-dimensional vectors and the recognition module 224 may be configured to select the fixed-dimensional vectors related to the same set of activities. In another exemplary scenario, each field in the fixed-dimensional vectors may be extended to be a tuple, to store additional information about the corresponding appliances used in the activity.
According to the present implementation, the recognition module 224 further configured to compare the test vectors with the template feature vectors. The comparison of the test vectors with the template feature vectors may be based on the innerproduct metrics. In an exemplary scenario, it may be based on cosine similarity. Further to the comparison, the recognition module 224 may be configured to generate a set of candidate activities. The generation of the candidate activities may be selected on the basis of only used set of appliances. For an example, there may be six appliances are used during a set of activities, thus the candidate activities selection may be based on the activities performed in that manner. In another example, the candidate activities may be all set of expected activities happened in a given scenario. Further, the contextual information may also help in selecting suitable candidate activities for the activity detection.
According to the present implementation, the recognition module 224 further configured to determine a set of human activities based on the set of candidate activities and contextual information related to the pre-defined activities. Further, the activity detection system 202 may also include the updation module 226 to update the template feature vector by combining a past template feature vector and a current feature vector by means of a weighted average. The weight-age may be decided on the basis of number of instances of an activity used to compute the past template feature vector and the current template feature vector. Further, the use of smart meters for human activity detection, as mentioned above, is non-intrusive and privacy preserving.
In another implementation, the updation module 226 may also assign a weight-age to each of the appliance, and the assigned weight-age represents the importance of each appliance for each activity.
It would be understood by any person skilled in the art that though the present subject matter is explained using a meter, it is merely an embodiment. Any application for the purpose of reading a number or character or strings or pictures or shapes off an instrument or meter, predicted utilizing some historical information or predefined criterion would not differ from the scope of the present subject matter.
Fig. 3 illustrates a method 300 for non-intrusively determining human activity using the smart meter data according to an embodiment of the present subject matter. The method 300 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, and modules, functions, which perform particular functions or implement particular abstract data types. The method 300 may also be practiced in a distributed computing environment where functions are performed by remote processing devices that are linked through a communication network. In a distributed computing environment, computer executable instructions may be located in both local and remote computer storage media, including memory storage devices.
The order in which the method 300 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 300, or alternative method. Additionally, individual blocks may be deleted from the method 300 without departing from the spirit and scope of the subject matter described herein. Furthermore, the method 300 can be implemented in any suitable hardware, software, firmware, or combination thereof. In an example, the method 300 may be implemented in a computing system, such as an activity detection system 202.
Referring to method 300, at bock 302, receiving an aggregate consumption data from smart meters installed in a household. In an implementation, the load disaggregation module 220 receives an aggregate consumption data from the smart meters.
At block 304, determining consumption data, from the aggregate consumption data, corresponding to the appliances. In an implementation, the load disaggregation module 220 determines consumption data, from the aggregate consumption data, corresponding to the appliances.
At block 306, generating template feature vectors corresponding to every activity with the help of historical data. In an implementation, the activity training module 222 generates template feature vectors corresponding to every activity with the help of historical data.
At block 308, generating multiple test vectors from the determined consumption data of the multiple appliances. In an implementation, the recognition module 224 generates the test vectors from the determined consumption data of the multiple appliances.
At block 310, comparing the test vectors with template feature vectors corresponding to pre-defined activities. In another implementation, the recognition module 224 compares the test vectors with template feature vectors corresponding to pre-defined activities.
At block 312, generating a set of candidate activities from the comparison done at block 310. In another implementation, the recognition module 224 generates the set of candidate activities from the comparison between the test vectors and the template feature vectors.
At block 314, determining the human activities using the generated candidate activities and contextual information related to the set of activities. In another implementation, the recognition module 224 determines the human activities using the generated candidate activities and contextual information related to the set of activities.
Although implementations for determining human activity using the smart meter data have been described in language specific to structural features and/or method, it is to be understood that the appended claims are not necessarily limited to the specific features or method described. Rather, the specific features and method are disclosed as exemplary implementations for determining human activity using the smart meter data.