WO2019202494A1

WO2019202494A1 - A system and a method for predicting failure risk of appliances

Info

Publication number: WO2019202494A1
Application number: PCT/IB2019/053118
Authority: WO
Inventors: Ashish Sharma; Ajit WADEKAR
Original assignee: Ashish Sharma; Wadekar Ajit
Priority date: 2018-04-16
Filing date: 2019-04-16
Publication date: 2019-10-24

Abstract

The present subject matter describes a system and a method for predicting failure risk of appliances. The system 101 comprises a user device 104, one or more power sockets 103 connected to appliances, a local server 105, a cloud server 101, processor 301 and a memory 303. The one or more power sockets 103 comprises a data monitoring unit 203-1 and a conversion module 203-2. The data monitoring unit 203-1 monitors data from the appliances. The conversion module 203-2 converts the data into standard data. The local server 105 receives the standard data from the one or more power sockets 103. A translation module 106 of the local server 105 translates the standard data into processable data. The system 101 comprises receiving the processable data from the local server 105, analyzing the processable data to generate feedback and transmitting it to the user device 104, thereby predicting failure risk.

Description

Title of invention:

A SYSTEM AND A METHOD FOR PREDICTING FAILURE RISK OF APPLIANCES

CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY

The present application claims priority from Indian provisional patent application ho.201821010840, filed on April 16, 2018.

TECHNICAL FIELD

The present subject matter described herein, in general, relates to a system and a method for predicting failure risk of appliances.

BACKGROUND

Electrical/electronic appliances more specifically household appliances have now become an indispensable part of everyone’s life. Due to regular use of these appliances, they become prone to failure and brings inconvenience to users' lives. Recently, there is no system to predict failure risk of the appliances in advance (i.e. before they actually stop working). The appliances would fail without notice suddenly which would leave the user stranded without the appliance for days as appliances would break down, then the repair company would come, uninstall the heavy duty appliance, then take it for repair which is very time consuming and difficult to deal with.

Further, even if the device predicts the failure risk of the appliance, the devices are not portable and compact. Rather, they are very large in size which are very difficult and time consuming to install. Further, once such devices are installed near any particular appliance, then it is very difficult to further uninstall it and again reinstall with any other appliance. Such devices also consume lot of space. Further, if the same device is installed or connected with one appliance and then with another appliance, the differentiation between two appliances, by monitoring their characteristics, becomes difficult and the prediction may not happen efficiently.

Therefore, there is a long-felt need for a system and a compact device for predicting failure risk of appliances in advance. Further, if the device is previously connected with one appliance and later it is connected with another appliance, then the system should be capable of differentiation between the characteristics of the two or more appliances and predicting failure risk (if any) accordingly. The system should also predict remaining useful life of the appliances. SUMMARY

The summary is provided to introduce the concepts related to a system and a method for predicting failure risk of appliances and the concepts are further described in the detail description. This summary is not intended to identify essential features of the claimed subject matter nor it is intended to use in determining or limiting the scope of claimed subject matter.

In one embodiment, a system for predicting failure risk of appliances is disclosed. The system may comprise a user device, one or more power sockets connected to one or more appliances. The one or more power sockets may comprise a data monitoring unit and a conversion module. The data monitoring unit may monitor and collect data from one or more appliances. Further, the conversion module may convert the data into standard data. Further, the system may comprise a local server asynchronously connected to the one or more power sockets. The local server may receive the standard data from the one or more power sockets. Further, a translation module of the local server may translate the standard data into processable data. The system may further comprise a cloud server, a processor and a memory coupled with the processor. The processor may be configured to execute a plurality of programmed instructions stored in the memory. The processor may execute a programmed instruction for receiving, the processable data from the local server. The processor may further execute a programmed instruction for analyzing, the processable data in order to generate feedback. The processor may further execute a programmed instruction for transmitting, the feedback to the user device 104, thereby predicting failure risk of the appliances.

In another embodiment, a method for predicting failure risk of appliances is disclosed. The method may comprise receiving, via a processor, processable data from a local server. A translation module of the local server may be configured to translate standard data into the processable data. Further, the standard data may be is received from a conversion module of one or more power sockets. Further, the conversion module may be configured to convert data, received from a data monitoring unit of the one or more power sockets, into the standard data. Further, the data monitoring unit may be configured to monitor and collect data from one or more appliances. The method may further comprise analyzing, via the processor, the processable data in order to generate feedback. The method may further comprise transmitting, via the processor, the feedback to a user device, thereby predicting failure risk of the appliances.

In yet another embodiment, a non-transitory computer readable medium storing program for predicting failure risk of the appliances is disclosed. The program may include a programmed instruction for receiving, processable data from a local server. A translation module of the local server may be configured to translate standard data into the processable data. Further, the standard data may be is received from a conversion module of one or more power sockets. Further, the conversion module may be configured to convert data, received from a data monitoring unit of the one or more power sockets, into the standard data. Further, the data monitoring unit may be configured to monitor and collect data from one or more appliances. The program may include programmed instruction for analyzing, the processable data in order to generate feedback. The program may include programmed instruction for transmitting, the feedback to a user device, thereby predicting failure risk of the appliances.

BRIEF DESCRIPTION OF DRAWINGS

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 refer like features and components.

Figure 1 illustrates a network implementation 100 of a system 101 for predicting failure risk of appliances, in accordance with an embodiment of a present subject matter.

Figure 2 illustrates block diagram 200 representing components of the power sockets, in accordance with the embodiment of the present subject matter.

Figure 3 illustrates a system 101 and its components, in accordance with an embodiment of a present subject matter.

Figure 4 illustrates a process flow diagram 400 indicating a method for predicting failure risk of appliances, in accordance with the embodiment of the present subject matter.

DETAILED DESCRIPTION

Reference throughout the specification to“various embodiments,”“some embodiments,”“one embodiment,” or“an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. Thus, appearances of the phrases “in various embodiments,” “in some embodiments,” “in one embodiment,” or“in an embodiment” in places throughout the specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments.

Figure 1 illustrates a network implementation 100 of a system 101 for predicting failure risk of appliances, in accordance with an embodiment of a present subject matter. The system 101 may be implemented as a cloud server (hereinafter the system 101 is interchangeably referred as cloud server 101). In an embodiment, the cloud server 101 may be connected to a user device 104 over a network 102. It may be understood that the cloud server 101 may be accessed by multiple users through one or more user devices 104-1,104-2,104-3... 104-h, collectively referred to as user device 104 hereinafter, or user 104, or applications residing on the user device 104. The user 104 may be any person, machine, software, automated computer program, a robot or a combination thereof.

In an embodiment, though the present subject matter is explained considering that the system 101 is implemented as a cloud server, it may be understood that the system 101 may also be implemented in a variety of user devices, such as a but are not limited to, a portable computer, a personal digital assistant, a handheld device, a mobile, a laptop computer, a desktop computer, a notebook, a workstation, a mainframe computer, and the like. In one embodiment, the system 101 may be implemented in a cloud-computing environment. In an embodiment, the network 102 may be a wireless network, a wired network or a combination thereof. The network 102 can be accessed by the user device 104 using wired or wireless network connectivity means including updated communications technology.

In one embodiment, the system 101 may comprise one or more power sockets 103 and a local server 105. The local server 105 may comprise a translation module 106. The one or more power sockets 103 may be communicatively coupled with the local server 105 via short range communication such as Zigbee, bluetooth, etc.

Referring to figure 2, illustrates block diagram 200 representing components of the power sockets, in accordance with the embodiment of the present subject matter.

The one or more power sockets 103 may be connected to the one or more appliances. The power socket 103 may be connected to one appliance at one instance and the same power socket 103 may be connected to another appliance at some other instance. The one or more appliance may be such as but are not limited to an air conditioner, a laptop, a computer, a geyser, etc. The one or more power sockets 103 may comprise a non-isolated power supply 201, a controller 202. The controller 202 may be bi-directionally connected with a data monitoring unit 203-1, a conversion module 203-2, a Real Time Clock (RTC) 203-3, a local memory chip 203-4, a relay 203-5, one or more current measuring components 203-6 and a transceiver 203-7. Further, the relay 203-5 may be bidirectionally connected with the one or more current measuring components 203-6. The one or more current measuring components 203-6 may comprise a current transformer, a shunt resistor, a hall effect sensor, etc. The one or more current measuring components 203-6 may be bi-directionally connected with the data monitoring unit 203-1 by using two wires in a twisted manner, in order to measure current. The shunt resistor may handle current up to 20 Amp without excessive heating. Further, selection of components for the shunt resistor may be based on current, voltage, wattage, etc. of resistor. The one or more power sockets 103 as well as the data collected from the respective power sockets 103 may be allotted with a unique sequence number in order to uniquely identify the data of the corresponding power socket 103. The current measuring components 203-6 may be placed on the relay 203-5 between contact of the relay 203- 5 and metal so that they may be connected in series with the appliances.

In one embodiment, the size of the non-isolated power supply 201 is very small due to which the form factor of the one or more power sockets 103 decreases. For example, the form factor of the one or more power sockets 103 is 36mm x 21 mm x l6mm). Further, the non-isolated power supply may provide necessary current at lowest wattage. Further, a half-wave rectifier may be used in order to achieve neutral and ground common line which may be needed for data monitoring. Further, a MOSFET based switcher may be used for increasing efficiency of the system 101. The power factor of the non-isolated power supply 201 may reach at its peak at around 0.7 to 0.75 when connected at full load, due to which the efficiency of the system 101 may remain high when completely ON and in the full load mode.

In one embodiment, brass plates may be used inside the one or more power sockets 103 which may provide low resistance to current and thereby causing less heating in metal parts.

In one embodiment, the RF circuit/transceiver 203-7 and the controller 202 may be CC2538 based SOC (System-on-Chip) which may have integrated Zigbee RF with ARM processor. The ARM processor may be used to perform multiple tasks on single chip such as communication with the data monitoring unit 203-1, the RTC 203-3, the local memory chip 203-4, etc. In one embodiment, in order to increase the efficiency of the system 101, filters may be added with the non-isolated power supply and the 4-layer designing may be achieved. Further, in order to achieve lesser form factor, batteries may be kept separated.

In one embodiment, in order to reduce network traffic and to collect the data timely, the data for pre-defined time (say every 1 hour) may be collected and processed further and transmitted to the local server 105 in small chunks/packets (say chunks/packets of 5 minutes). If the heavy load appliance is connected, the data may be read multiple times from the data monitoring unit 203-1 and may be summed or averaged (depending upon the one or more parameters) up to the pre defined time.

In one embodiment, the relay 203-5 may be an automotive grade relay, which may work up to 20 Amp’s at 230 Volts AC at just 5V DC.

In one embodiment, the data monitoring unit 203-1 may monitor and collect data from the one or more appliances. The data monitoring unit 203-1 may be ADE7953. The data monitoring unit 203- 1 may be of very high resolution (having error rate less than 0.5%) which may measure current for low wattage appliance as well. The data monitoring unit 203-1 may monitor and collect the data only when the one or more power sockets 103 is ON and the appliance is connected to the one or more power sockets 103. Further, the RTC 203-3 may get exact time for collecting the data. The data may comprise numerical values related to one or more parameters.

In an exemplary embodiment, the data for the voltage is 216468 and the data for the current is 5863805. Here, the data for the voltage i.e. 216468 may be pre-calibrated and may be found equivalent to 240V. Similarly, the data for the current i.e. 5863805 may be found equivalent to 0.16 Amp.

Now, the value of voltage for 1 unit data = 240/216468 = 0.001

Therefore, 1 unit = 0.00 IV

Similarly, the value of current for 1 unit data = 0.16/5863805 = 2.7 x 10 ⁸ Amp.

Therefore, 1 unit = 2.7 x 10 ⁸ Amp.

In one embodiment, the one or more parameters may comprise such as but are not limited to RMS voltage, RMS current, active power, reactive power, apparent power, active energy, reactive energy, apparent energy, power factor, peak current, peak voltage, voltage sag, over current, over voltage, no load etc.

The conversion module 203-2 may convert the data into standard data. The standard data is derived by calculating hexadecimal or decimal values corresponding to the numerical values of the one or more parameters.

In an exemplary embodiment, the data for voltage = 220000 and current = 14814814.

The standard data for the voltage = 220000 x 0.001 = 220V; and

The standard data for the current = 14814814 x (2.7 x 10 ⁸) = 0.4 Amp.

In one embodiment, the local server 105 may be asynchronously connected to the one or more power sockets 103 which may perform multiprocessing without any delay. Due above Asynchronous connection, different process may be run on the local server 105 such as but are not limited to Led Rings control, monitoring of network Service, gateway application etc.

The transceiver 203-7 may transmit the standard data to the local server 105 in real time.

In one embodiment, if the transmission of the standard data from the one or more power sockets 103 to the local server 105 fails or gets interrupted for certain time duration, the local memory chip 203-4 may store the standard data for a pre-defined time period or until the standard data is successfully transmitted. The pre-defined time period may be 60 days.

In one embodiment, the local memory chip 203-4 may have 40 million life cycle. Operation of the local memory chip 203-4 may work in paging i.e. every page has 256 Byte, in that manner, around 4096 pages may be available, which may store lMega byte of the standard data. Further, a page management algorithm may be used which may save number of read write cycle in initial 4 bytes of every page. The pages that may be corrupted and the address at which the standard data should be written or read from may be stored on first page of memory, which may again has its own read/write cycle maintained at its initial 4 bytes. Due to this, memory corruption may not happen, and the standard data may remain consistent.

In one embodiment, the translation module 106 may translate the standard data into processable data. The processable data may be derived by translating the standard data into machine readable format. For example: the processable data obtained from the standard data = {voltage: 220, current:

0.4}. Further, the processable data may be transmitted to the cloud server 101 via the network 102.

In one embodiment, if the transmission of the processable data from the local server 105 to the cloud server 101 fails or gets interrupted for certain time duration, the local server 105 may store the processable data for pre-defined time period or until the processable data is transmitted successfully.

Referring to figure 3, a system 101 and its components, in accordance with an embodiment of a present subject matter. The system 101 may comprise at least one processor 301, an input/output (I/O) interface 302, a data monitoring unit 203-1, and a memory 303. In one embodiment, the at least one processor 301 is configured to fetch and execute computer-readable instructions stored in the memory 303. In one embodiment, the I/O interface 302 may include a variety of software and hardware interfaces, for example, a web interface, a Graphical User Interface (GUI), and the like. The I/O interface 302 may allow the user device 104 to interact with the cloud server 101. Further, the I/O interface 302 may enable the user device 104 to communicate with other computing devices. The I/O interface 302 can facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite.

In one embodiment, the I/O interface 302 is an interaction platform that facilitates interaction between the user device 104 and the system 101. The I/O interface 302 may allow commands for a command line interface or GUI which may enable a user to create, modify and delete either of data, metadata, program, logic, algorithm, parameters associated with encryption method, encryption program and encryption language. In one embodiment, the memory 303 may 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 memory cards. The memory 303 may comprise modules 304 and data 308.

In one embodiment, the modules 304 may comprise routines, programs, objects, components, data structure, etc., which performs particular tasks, functions or implement abstract data types. The modules 304 may further include a data receiving module 305, a data analyzing module 306 and a data transmitting module 307. The data 308 may comprise a centralized repository 309 and other data 310.

In one embodiment, the data receiving module 305 may receive the processable data from the local server 105. The data analyzing module 306 may analyze the processable data in order to generate feedback. The data analyzing module 306 may classify the type of different appliances connected with each one of the one or more power sockets at different time period. The classification may be done by analyzing and identifying signature of each of the appliances based upon the one or more parameters. The data analyzing module 306 may classify the type of different appliances by using a machine learning technique. This may be achieved by fitting multilabel classification algorithms such as but are not limited to logistic regression, neural network, decision tree, random forest, support vector machine etc. Such models may run at pre-defined frequency and may update the type of the appliance.

In one embodiment, in order to analyze and identify signatures of different appliances, graph may be plotted using the values of the processable data for small interval (say 5 minutes chunks).

In one embodiment, the data analyzing module 306 may analyze the processable data using machine learning technique, in order to generate feedback. Algorithms such as linear regression, neural network, binary classification algorithm such as logistic regression, Bayes point machine, boosted decision trees etc. may be used to analyze the processable data so that the failure risk of the appliances may be predicted.

In one embodiment, the data transmitting module 307 may transmit the classification of the type of different appliances to the local server 105 in real time. Further, the data transmitting module 307 may transmit the feedback to the user device 104 in real time. The alert may be generated on the user device 104. The feedback may provide prediction about the failure risk of the appliances. Further, the system 101 may also predict remaining useful life of the one or more appliances based upon the frequency of usage and the one or more parameters.

In one embodiment, in absence of network connectivity, the system 101 may be configured to predict the failure risk and the remaining useful life of the one or more appliances in the local server 105. In one embodiment, the local server 105 may send and receive the data or commands in JSON data exchange format. The JSON may be converted to Zigbee packets and vice versa. Further due to low data rate as compared to Wi-Fi, the system may provide a buffering mechanism which may cache plurality of simultaneous consecutive data or command and send/transmit it further. The plurality of simultaneous consecutive data in the cache may be sent one by one, efficiently and without dropping out.

At step 401, the data receiving module 305 may receive the processable data from the local server 105.

In one embodiment, the data monitoring unit 203-1 may monitor and collect data from the one or more appliances. The conversion module 203-2 may convert the data into standard data. The transceiver 203-7 may transmit the standard data to the local server 105. The translation module 106 may translate the standard data into the processable data.

At step 402, the data analyzing module 306 may analyze the processable data using machine learning technique, in order to generate feedback.

At step 403, the data transmitting module 307 may transmit the feedback to the user device 104 in real time.

Although implementations for a system and a method for predicting failure risk of appliances have been described in language specific to structural features and/or methods, it is to be understood that the appended claims are not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed as examples of implementations for predicting failure risk of appliances. The embodiments, examples and alternatives of the preceding paragraphs or the description and drawings, including any of their various aspects or respective individual features, may be taken independently or in any combination. Features described in connection with one embodiment are applicable to all embodiments, unless such features are incompatible.

Claims

We Claim:

1. A system for predicting failure risk of appliances, the system comprising:

a user device 104; one or more power sockets 103 connected to one or more appliances, wherein the one or more power sockets 103 comprise: a data monitoring unit 203-1 configured to monitor and collect data from one or more appliances; and a conversion module 203-2 configured to convert the data into standard data; a local server 105 asynchronously connected to the one or more power sockets 103, wherein the local server 105 is configured to receive the standard data from the one or more power sockets 103; and wherein a translation module 106 of the local server 105 is configured to translate the standard data into proces sable data; a cloud server; a processor 301; and a memory 303 coupled with the processor 301, wherein the processor 301 is capable of executing programmed instructions stored in the memory 303 for: receiving, the processable data from the local server 105; analyzing, the processable data in order to generate feedback; and transmitting, the feedback to the user device 104, thereby predicting failure risk of the one or more appliances.

2. The system 101 of claim 1, wherein the one or more power sockets 103 comprise a non isolated power supply 201.

3. The system 101 of claim 1, wherein the one or more power sockets 103 further comprise a Real Time Clock (RTC) 203-3 in order to get exact time for collecting the data.

4. The system 101 of claim 1, wherein upon failure of transfer of the standard data to the translation module 106, the one or more power sockets 103 are configured to store the standard data in a local memory chip 203-4 for a pre-defined time period.

5. The system 101 of claim 1, wherein upon failure of transfer of the processable data from the local server 105 to the cloud server, the local server 105 is configured to store the processable data for a pre-defined period.

6. The system 101 of claim 1, wherein the one or more power sockets 103 and the data collected from the respective power sockets 103 are allotted with a unique sequence number in order to uniquely identify the data of the corresponding power socket 103.

7. The system 101 of claim 1, wherein the data monitoring unit 203-1 comprises one or more current measuring components 203-6, wherein the one or more current measuring components 203-6 comprises a current transformer, a shunt resistor and hall effect sensors.

8. The system 101 of claim 1, wherein the data comprises numerical values related to one or more parameters.

9. The system 101 of claim 8, wherein the one or more parameters comprises at least one of RMS voltage, RMS current, active power, reactive power, apparent power, active energy, reactive energy, apparent energy, power factor, peak current, peak voltage, voltage sag, over current, over voltage and no load.

10. The system 101 of claim 1, wherein the standard data is derived by calculating hexadecimal or decimal values corresponding to the numerical values of the one or more parameters.

11. The system 101 of claim 1, wherein the processable data is derived by translating the standard data into machine readable format.

12. The system 101 of claim 1, wherein the processor 301 is configured to analyze the processable data using a machine learning technique.

13. The system 101 of claim 1, wherein the processor 301 is configured to classify the type of the different appliances connected with each one of the one or more power sockets 103 at different time period, using a machine learning technique.

14. The system 101 of claim 13, wherein the classification is done by analyzing signature of each of the appliances based upon the one or more parameters.

15. The system 101 of claim 13, wherein the processor 301 is configured to update the local server 105 with the classification of the type of the different appliances.

16. The system 101 of claim 1, wherein the processor 301 is configured to predict remaining useful life of the one or more appliances based upon the frequency of usage and the one or more parameters.

17. The system 101 of claim 1, wherein the system 101 may be configured predict failure risk and the remaining useful life of the one or more appliances in the local server 105, if the network connectivity is not present.

18. A method for predicting failure risk of appliances, the method comprising:

receiving, via a processor 301, processable data from a local server 105, wherein a translation module 106 of the local server 105 is configured to translate standard data into the processable data; wherein the standard data is received from a conversion module 203-2 of one or more power sockets 103; wherein the conversion module 203-2 is configured to convert data, received from a data monitoring unit 203-1 of the one or more power sockets 103, into the standard data; and wherein the data monitoring unit 203-1 is configured to monitor and collect data from one or more appliances; analyzing, via the processor 301, the processable data in order to generate feedback; and transmitting, via the processor 301, the feedback to a user device 104, thereby predicting failure risk of the one or more appliances.