Method and system for managing errors in charging station

Abstract

A system (200) and method (100) for managing errors in a charging station. The method (100) includes receiving, by a processor (204), charger error logs and meter value data associated with the charging station and determining, by the processor (204), at least one error occurred in the charging station based on the charger error logs and the meter value data. The method also includes determining, by the processor (204), an error code associated with the at least one error using a first machine learning (ML) model. The error code indicates an error category of a plurality of error categories. The method also includes mapping, by the processor (204), the error code to a corrective action stored in a diagnostic database (210) and automatically applying, by the processor (204), the mapped corrective action to rectify the at least one error in the charging station.

Description

[0001] 202413527

[0002] METHOD AND SYSTEM FOR MANAGING ERRORS IN CHARGING STATION

[0003] DESCRIPTION

[0004] The present disclosure relates generally to managing errors in a charging station, and more specifically to a method and system for managing errors in the charging station by automatically applying corrective actions based on machine learning (ML) models and also predicting possibility of similar errors in other charging stations.

[0005] Electrical Vehicle (EV) charging stations are steadily increasing everyday as electric vehicles become largely used throughout the globe. EV's are being accepted as a valuable component of auto industry. As a result, charging stations can be found in private residences, public dedicated charging stations and even at commercial and residential buildings such as offices and condominiums. EVs store electricity in rechargeable batteries that power electric motors. These rechargeable batteries are replenished by plugging them to the charging stations that is connected to a grid. The charging stations are continuously in communication with electric vehicles and EV charging management software over the internet. This generates operational data about charger operation, OCPP logs, and meter values during a transaction in a cloud backend.

[0006] There can be various instances of errors occurring at the charging stations. To diagnose an error condition and make the charging stations operational, a field technician must rely on resources such as personal experience, hardware logs, backend logs, and communication with backend teams, etc. The field technician must balance such unprecedented priorities among attending periodic scheduled maintenance tasks as a part of their job. There is a high risk associated with violating SLA (service lease agreement) when one or more of resources become unavailable to the field technician in a timely manner. With a rising population of electric vehicles on the road and growing network of charging station, it is important for businesses to focus on ways to reduce the risk of such operational downtime.

[0007] Currently, there does not exist a holistic solution that extends to multiple charger models or types which can troubleshoot errors at the charging stations without human intervention. Further, there 202413527 also exists no preventive mechanism to predict the occurrence of errors which has occurred in one charging station and has been addressed by the field technician, in another similar charging station.

[0008] In light of the above, there exists a need to have a comprehensive solution which reduces the operational downtime and avoid human intervention to address the errors in the charging stations. Also, the learning and experience that is gained from addressing an error at one charging station must be used to address similar issues occurring in similar charging stations elsewhere.

[0009] The above-mentioned challenges are addressed by the proposed solution by identifying an error from normal operational condition of the charging station and classifying between different error categories and types. Then the solution includes automatically using multi-class classification machine learning model to map the error to the diagnostic procedure sequence stored in a diagnostic database. Further, the solution includes initiating a self-healing activity (automatically detect the corrective action and rectify the error without any manual intervention) and verify if the issue is fixed. For errors that cannot be fixed, the solution includes cascading the diagnosis to a field personnel with relevant information to fix the problem and procure any necessary spares and equipment at the time of visit. The diagnostic history is recorded to improve future occurrences of the error in similar hardware located elsewhere. Prognosticate and predict the occurrence of error events by monitoring operational parameters using artificial intelligence and predictive machine learning model. Issue alerts and prognostic advice to users based on the operational pattern of the charging station.

[0010] Therefore, the proposed solution aims to improve the uptime, durability, stability, and reliability of the chargers with an AI / ML based cloud solution that monitors the logs and meter values over the OCPP network. It is designed to act as a holistic solution that can cater to multiple charger models / types and even across chargers manufactured. In the event of an error the proposed solution will automatically detect the corrective action and rectify the error without any manual intervention. It also provides prognostic advice to alert users about possible breakdowns to improve the reliability and durability of the hardware. 202413527

[0011] The object of the present disclosure is achieved by a computer-implemented method for managing errors in a charging station. The method includes receiving charger error logs and meter value data associated with the charging station and determining at least one error occurred in the charging station based on the charger error logs and the meter value data. The method includes determining an error code associated with the at least one error using a first machine learning (ML) model. The error code indicates an error category of a plurality of error categories. The method also includes mapping the error code to a corrective action stored in a diagnostic database and automatically applying the mapped corrective action to rectify the at least one error in the charging station.

[0012] In one or more embodiments, the method further includes determining an outcome of the application of the corrective action to rectify a plurality of errors in the charging station and storing the outcome of the application of the corrective action to rectify the plurality of errors, in the diagnostic database.

[0013] In one or more embodiments, the method further includes receiving outcomes of the application of the corrective action to rectify the plurality of errors from the diagnostic database and determining a co-relation between outcomes of the application of the corrective action to rectify the plurality of errors, the charger error logs and the meter value data. Then the method includes training a second machine learning (ML) model based on the co-relation.

[0014] In one or more embodiments, the method further includes clustering a set of charging stations based on a plurality of similarities and monitoring a plurality of operational parameters associated with the set of charging stations. The method includes predicting occurrence of the at least one error of the plurality of errors in the set of charging stations using the second ML model.

[0015] In one or more embodiments, the method further includes recommending the at least one corrective action to rectify the at least one error in the charging station and displaying the recommended at least one corrective action on a screen of the charging station.

[0016] In one or more embodiments, the first ML model is a multi-class classification model, and the second ML model is a prediction model. 202413527

[0017] In one or more embodiments, determining that the at least one error in the charging station persists after application of the mapped corrective action and sending an indication to at least one personnel to address the at least one error in the charging station.

[0018] In one or more embodiments, the determination of the first ML model includes determining a type of the charging station and determining historic data associated with the charging station from a cloud server and generating the first ML model using the historic data associated with the charging station. Further, the method includes storing the first ML model in the diagnostic database.

[0019] The object of the present disclosure is also achieved by a system for managing errors in a charging station. The system includes a processor and a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, configures the processor to receive charger error logs and meter value data associated with the charging station and determine at least one error occurred in the charging station based on the charger error logs and the meter value data. Further, the processor is configured to determine an error code associated with the at least one error using a first machine learning (ML) model. The error code indicates an error category of a plurality of error categories. Further, the processor is also configured to map the error code to a corrective action stored in a diagnostic database and automatically apply the mapped corrective action to rectify the at least one error in the charging station.

[0020] The object of the present disclosure is further achieved by a computer program code which, when executed by a processor, causes the processor to carry out steps of the aforementioned method.

[0021] The object of the present disclosure is further achieved by a computer program product comprising computer program code which, when executed by a processor, causes the processor to carry out steps of the aforementioned method. 202413527

[0022] Still other aspects, features, and advantages of the disclosure are readily apparent from the following detailed description, simply by illustrating a number of particular embodiments and implementations, including the best mode contemplated for carrying out the disclosure. The disclosure is also capable of other and different embodiments, and its several details may be modified in various obvious respects, all without departing from the scope of the disclosure. Accordingly, the drawings and description are to be regarded as illustrative in nature, and not as restrictive.

[0023] A more complete appreciation of the present disclosure and many of the attendant aspects thereof will be readily obtained as the same becomes better understood by reference to the following description when considered in connection with the accompanying drawings:

[0024] FIG 1 is a flowchart representation of a computer-implemented method for managing errors in a charging station, in accordance with one or more embodiments of the present disclosure;

[0025] FIG 2 is a block diagram representation of a system for managing errors in the charging station, in accordance with one or more embodiments of the present disclosure;

[0026] FIG 3A is an overview of signalling between various entities of the system, in accordance with one or more embodiments of the present disclosure;

[0027] FIG 3B is an overview of a method for applying a corrective action to rectify an error in the charging station, in accordance with one or more embodiments of the present disclosure;

[0028] FIG 3C is an overview of a method for predicting occurrence of the error in another charging station, in accordance with one or more embodiments of the present disclosure;

[0029] FIG 4 is a flowchart representation of overall the method of deploying an ML model and managing the errors in the charging station, in accordance with one or more embodiments of the present disclosure; and

[0030] FIG 5 is a flowchart representation illustrating an example error in the charging station being managed, in accordance with one or more embodiments of the present disclosure. 202413527

[0031] Various embodiments are described with reference to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for the purpose of explanation, numerous specific details are set forth in order to provide thorough understanding of one or more embodiments. It may be evident that such embodiments may be practiced without these specific details.

[0032] Examples of a method, a system, and a computer-program product for managing errors in a charging station are disclosed herein. In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It is apparent, however, to one skilled in the art that the embodiments of the disclosure may be practiced without these specific details or with an equivalent arrangement. In other instances, well-known structures and devices are shown in block diagram form in order to avoid unnecessarily obscuring the embodiments of the disclosure.

[0033] Conventional methods of addressing errant charging stations may require field technicians to visit the charging stations installation site to diagnose it, determine an appropriate correction for the diagnosed error and make the charging stations operational. This requires a lot of human resource and if the error is very minute then it’s a waste of the resource’s time and expertise. Unlike to the conventional methods and systems, the proposed solution diagnoses the error and automatically applies a corrective action to help the charging stations self-heal without any manual intervention.

[0034] Conventional methods of determining the abnormal behavior in the charging stations or equipment is done using video and device status obtained from an extra hardware installed in the charging pile. However, the requirement of the additional hardware makes the solution expensive and cumbersome. Also, if the additional hardware malfunctions, then there is no way the error of the charging station is determined and addressed. Unlike to the conventional methods and systems the proposed method does not require installation of any additional hardware and can handle the errors by automatically applying the corrective actions to the errors determined in the charging stations.

[0035] Unlike to the conventional methods and systems the proposed method includes learning the errors and the solutions associated with the errors by machine learning models. Then using the 202413527 learning to predict possible similar errors in other charging stations along with the solutions. Hence the proposed solution also includes prognosticating and predicting the occurrence of errors.

[0036] Referring now to FIG 1 , illustrated is a flowchart of a method (as represented by reference numeral 100) for managing errors in a charging station, in accordance with an embodiment of the present disclosure. As used herein, managing errors in the charging station refers to a process of identifying errors which has occurred in the charging station based on certain parameters of the charging station and automatically applying corrective actions to rectify the identified error in the charging station using a machine learning (ML) model. Further, the error management also includes using the corrective action to train the ML model and predict the possibility of a similar error in another charging station. For example, if the error determined at the charging station is a rise in temperature beyond a certain threshold, then the proposed solution includes determining an automatically applying an appropriate solution to reduce the temperature to a level which is below the said threshold. Further, the circumstances which led to the rise in the temperature and the manner in which the temperature is reduced is used to train the ML model. This ML model is then used to predict the possibility of a similar rise in the temperature under similar circumstances in some other charging stations. The prediction can be displayed on the charging station so that necessary actions might be taken in due course of time to avoid the occurrence of the error.

[0037] This approach aims to enhance the efficiency at which the errors that occur in the charging stations are handled without much delay and without manual intervention. Also, the use of the ML models to predict the errors in other charging stations based on the learning from the error management in one of the charging station makes the proposed solution even stronger to efficiently handle the errors at the charging station and associated infrastructure.

[0038] Referring to FIG 2, illustrated is a block diagram of a system 200 for managing errors in a charging station, in accordance with one or more embodiments of the present disclosure. The terms system 200 and charging management system 200 may be used interchangeable throughout the specification document and refers to the same entity. It may be appreciated that the system 200 described herein may be implemented in various forms of hardware, software, firmware, special purpose processors, ora combination thereof. One or more of the present embodiments may take 202413527 a form of a computer program product comprising program modules accessible from computer- usable or computer-readable medium storing program code for use by or in connection with one or more computers, processors, or instruction execution system. For the purpose of this description, a computer-usable or computer-readable medium may be any apparatus that may contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The medium may be electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation mediums in and of themselves as signal carriers are not included in the definition of physical computer-readable medium include a semiconductor or solid state memory, magnetic tape, a removable computer diskette, random access memory (RAM), a read only memory (ROM), a rigid magnetic disk and optical disk such as compact disk read-only memory (CD-ROM), compact disk read / write, and digital versatile disc (DVD). Both processors and program code for implementing each aspect of the technology may be centralized or distributed (or a combination thereof) as known to those skilled in the art. Further, it may be noted that terms charger, charging station, Electric Vehicle Supply Equipment (EVSE), charging dock, all refer to an equipment associated with delivering charge to an electric vehicle (EV) which is connected to the charger through a connector of the charger.

[0039] In an example, the system 200 may be embodied as a computer-program product 200 programmed for performing the said purpose. The system 200 may be incorporated in one or more physical packages (e.g., chips). By way of example, a physical package includes an arrangement of one or more materials, components, and / or wires on a structural assembly (e.g., a baseboard) to provide one or more characteristics such as physical strength, conservation of size, and / or limitation of electrical interaction. It is contemplated that in certain embodiments the computing device may be implemented in a single chip. As illustrated, the system 200 includes a communication mechanism such as a bus 202 for passing information among the components of the system 200. The system 200 includes a processor 204 and a memory 206. Herein, the memory 206 is communicatively coupled to the processor 204. In an example, the memory 206 may be embodied as a computer readable medium on which program code sections of a computer program are saved, the program code sections being loadable into and / or executable in a system to make the system 200 execute the steps for performing the said purpose. 202413527

[0040] Generally, as used herein, the term “processor” refers to a computational element that is operable to respond to and processes instructions that drive the system 200. Optionally, the processor includes, but is not limited to, a microprocessor, a microcontroller, a complex instruction set computing (CISC) microprocessor, a reduced instruction set (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, or any other type of processing circuit. Furthermore, the term “processor” may refer to one or more individual processors, processing devices and various elements associated with a processing device that may be shared by other processing devices. Additionally, the one or more individual processors, processing devices and elements are arranged in various architectures for responding to and processing the instructions that drive the system 200.

[0041] Herein, the memory 206 may be volatile memory and / or non-volatile memory. The memory 206 may be coupled for communication with the processor 204. The processor 204 may execute instructions and / or code stored in the memory 206. A variety of computer-readable storage media may be stored in and accessed from the memory 206. The memory 206 may include any suitable elements for storing data and machine-readable instructions, such as read only memory, random access memory, erasable programmable read only memory, electrically erasable programmable read only memory, a hard drive, a removable media drive for handling compact disks, digital video disks, diskettes, magnetic tape cartridges, memory cards, and the like.

[0042] In particular, the processor 204 has connectivity to the bus 202 to execute instructions and process information stored in the memory 206. The processor 204 may include one or more processing cores with each core configured to perform independently. A multi-core processor enables multiprocessing within a single physical package. Examples of a multi-core processor include two, four, eight, or greater numbers of processing cores. Alternatively, or in addition, the processor 204 may include one or more microprocessors configured in tandem via the bus 202 to enable independent execution of instructions, pipelining, and multithreading. The processor 204 may also be accompanied with one or more specialized components to perform certain processing functions and tasks such as one or more digital signal processors (DSP), and / or one or more application-specific integrated circuits (ASIC). Other specialized components to aid in performing the inventive functions described herein include one or more field programmable gate 202413527 arrays (FPGA) (not shown), one or more controllers (not shown), or one or more other specialpurpose computer chips.

[0043] The system 200 may further include an interface 208, such as a communication interface (with the said terms being interchangeably used) which may enable the system 200 to communicate with other systems for receiving and transmitting information. The communication interface 208 may include a medium (e.g., a communication channel) through which the system 200 communicates with other system. Examples of the communication interface 208 may include, but are not limited to, a communication channel in a computer cluster, a Local Area Communication channel (LAN), a cellular communication channel, a wireless sensor communication channel (WSN), a cloud communication channel, a Metropolitan Area Communication channel (MAN), and / or the Internet. Optionally, the communication interface 208 may include one or more of a wired connection, a wireless network, cellular networks such as 2G, 3G, 4G, 5G mobile networks, and a Zigbee connection.

[0044] The system 200 also includes a diagnostic database 210. As used herein, the diagnostic database 210 is an organized collection of structured data, typically stored in a computer system and designed to be easily accessed, managed, and updated. The diagnostic database 210 may be in form of a central repository of information that can be queried, analysed, and processed to support various applications and business processes. In the system 200, the diagnostic database 210 provides mechanisms for storing, retrieving, updating, and deleting data, and typically includes features such as data validation, security, backup and recovery, and data modelling. The diagnostic database 210 here includes outcomes of application of the corrective action to rectify multiple errors. The diagnostic database 210 may be designed using relational or non-relational database management systems, depending on the specific requirements and preferences of the system 200. The first ML model and the second ML model are also stored in the diagnostic database 210.

[0045] The system 200 further includes an input device 212 and an output device 214. The input device 212 may take various forms depending on the specific application of the system 200. In an example, the input device 212 may include one or more of a keyboard, a mouse, a touchscreen display, a microphone, a camera, or any other hardware component that enables the user to 202413527 interact with the system 200. Further, the output device 214 may be in the form of a display. It is to be understood that, when reference is made in the present disclosure to the term “display” this refers generically either to a display screen on its own or to a screen and an associated housing, drive circuitry and possibly a physical supporting structure, of which all, or part of is provided for displaying information.

[0046] In the present system 200, the processor 204 and accompanying components have connectivity to the memory 206 via the bus 202. The memory 206 includes both dynamic memory (e.g., RAM, magnetic disk, writable optical disk, etc.) and static memory (e.g., ROM, CD-ROM, etc.) for storing executable instructions that when executed perform the method steps described herein for managing errors in the charging station. In particular, the memory 206 includes an error management module 216 to perform steps for managing errors in the charging station. Also, in the system 200, the memory 206 may be configured to store the data associated with or generated by the execution of the inventive steps.

[0047] Referring to FIGS 1 and 2 in combination, the various steps of the method 100 as described hereinafter may be executed in the system 200, or specifically in the processor 204 of the system 200, for - managing errors in a charging station. For purposes of the present disclosure, managing errors in the charging station in the present method 100 is embodied as a mapping algorithm which maps corrective actions to rectify the errors in the charging station, and automatically applying a corresponding action for a specific error at the charging station. The method 100 also embodies as a predictive algorithm which predicts occurrence of errors in different charging stations using the second ML model, which is trained using the mapping of the corrective actions. It may be appreciated that although the method 100 is illustrated and described as a sequence of steps, it may be contemplated that various embodiments of the method 100 may be performed in any order or a combination and need not include all of the illustrated steps.

[0048] In embodiments of the present disclosure, at step 101 , the method 100 includes receiving charger error logs and meter value data associated with the charging station. The charger error logs are files that records errors that occur when the charging station is in an operative mode. The charger error logs may include for example but not limited to information about the error that occurred, when the error occurred (i.e. , a timestamp), how critical was the error, and a cause of the error. 202413527

[0049] The meter value data associated with the charging station can include for example but not limited to total energy consumed by an EV during charging, charging voltage, current, power factor, charging time, charging patterns associated with the charging station. Here, the charger error logs, and meter value data are received by the system 200 from a cloud server through an appropriate communication protocol like for example but not limited to Open Charge Point Protocol (OCPP). The OCPP is a communication standard protocol to facilitate interoperability between Electric vehicles (EV) charging stations and a charging management system.

[0050] In embodiments of the present disclosure, at step 102, the method 100 includes determining an error occurred in the charging station based on the charger error logs and the meter value data. Here, the system 200 performs certain pre-processing procedure on the received data and analyses the processed data. Based on the analysis the system 200 determines the error that has occurred in the charging station. The errors can be for example but not limited to charger connector fault, slow charging speed, disconnected wires, over-temperature protection (OTP), over-voltage protection (OVP), under-voltage protection (UVP), etc.

[0051] In embodiments of the present disclosure, at step 103, the method 100 includes determining an error code associated with the error using a first machine learning (ML) model. The error code indicates an error category. The first ML model is a multi-class classification model which predicts discrete class labels for each input. The error codes can be for example, E00 for restart required, 004 for Over-temperature protection (OTP), 005 for Over-voltage protection (OVP), and 006 for Under-voltage protection (UVP), etc.

[0052] Here, the multi-class classification model determination includes determining a type of the charging station and historic data associated with the charging station from a cloud server 302. Further, the multi-class classification model is generated based on the historic data associated with the charging station. Here, for different types of the charging station the multi-class classification model used may be different, as the error types associated will be different. The different types of charging stations are for example but not limited to Level 1 charging station for slow charging, Level 2 charging station for fast charging and Level 3 charging station for rapid charging. Further, the multi-class classification model once generated is stored in the diagnostic 202413527 database 210. Therefore, the diagnostic database 210 includes multiple multi-class classification models for various types of the charging stations.

[0053] In embodiments of the present disclosure, at step 104, the method 100 includes mapping the error code to a corrective action stored in the diagnostic database 210. An example, error codes and corrective action mapping table is provided in table. 1 .

[0054] Table. 1

[0055] The mapping is stored in the diagnostic database 210. As and when the system 200 determines an error in the charging station, the error code and the mapping of the corrective action is stored in the diagnostic database 210.

[0056] In embodiments of the present disclosure, at step 105, the method 100 includes automatically applying the mapped corrective action to rectify the error in the charging station. The corrective action corresponding to the error code is then automatically retrieved and applied to the charging station to correct the error. Further, the system 200 also verifies if the error is rectified and stores the outcome of the applied corrective action at the diagnostic database 210.

[0057] Therefore, the proposed solution is a self-healing technique for the charging station before the intervention of the field personnel. Also, the field personnel are brought in to address the error only when the error is not being fixed after multiple attempts to do so. Hence the proposed solution is not only a cost-effective solution but ensures effective resource utilization.

[0058] In embodiments of the present disclosure, further includes receiving the outcomes related data from the diagnostic database 210 and determining a co-relation between the outcomes, the charger error logs and the meter value data. Further, the co-relation is sued to train a second machine learning (ML) model. The second ML model is a prediction model. Here, the co-relation 202413527 can be for example but not limited to an outcome for an error which occurred at a specific charger is related to the meter data of that charger. Here, it may be noted that the first ML model and the second ML model are provided sequentially as an embodiment and can be combined into a single ML model performing both the classification of the errors and the prediction of the errors. And hence the depiction of two separate ML models is not a limitation on the scope of the proposed solution.

[0059] Further, the proposed solution includes clustering a set of charging stations based on their similarities and monitoring their operational parameters. Then the system 200 predicts the occurrence of the error in any of the set of charging stations using the prediction model. Here, the operational parameters can be for example but not limited to scheduling charging, revenue collection, live tracking of charger use, load balancing, etc. Then the system 200 recommends the corrective action to rectify the error in any of the charging stations and displays the recommended corrective action on a screen or user interface (Ul) of the charging station or on a Ul of an application running on a mobile device of a charger operator. Therefore, the proposed solution not only self-heals one charging station but learns the corrective action applied for bringing about the self-healing and then uses the learning to predict the possibility of the similar error in other charging stations.

[0060] Therefore, the proposed solution aims to improve the uptime, durability, stability, and reliability of the chargers with an AI / ML based cloud solution that monitors the logs and meter values over the OCPP network. It is designed to act as a holistic solution that can cater to multiple charger models / types and even across chargers manufactured. In the event of an error the proposed solution will automatically detect the corrective action and rectify the error without any manual intervention. It also provides prognostic advice to alert users about possible breakdowns to improve the reliability and durability of the hardware.

[0061] FIG 3A is an overview of signalling between various entities of the system, in accordance with one or more embodiments of the present disclosure. Referring to the FIG. 3A, the system 200 is part of the charger 300A. At step 1 , the charger 300A is continuously in communication with the cloud backend 302 sending real-time data. The communication between the charger 300A and the cloud backend 302 is carried on through a compatible protocol such as for example but not 202413527 limited to Open Charge Point Protocol (OCPP). The OCPP is an application protocol for communication between Electric vehicle (EV) charging stations and a charging station network. At step 2, there is continuously streaming of data from the cloud backend 302 to the classification model 304.

[0062] The classification model 304 receives the meter value data 320A and the charger error logs 320B from the cloud backend 302; and determines the presence of the error in the charger 300A. The classification model 304 then identifies the error code associated with the type of error and the type of the charger. The classification model 304 then sends the error code to the diagnostic database 210, at 308 and in response receives the diagnostic procedure associated with the error code. At step 3, the classification model 304 sends remote commands 306 indicating the diagnostic procedure to be carried out at the charger 300A. At step 4, the commands are sent to the charger 300A, and the identified diagnostic procedure is automatically implemented on the charger 300A. Further, at step 5, the charger 300A sends the details of the outcome of the diagnostic procedure that was automatically executed on the charger 300A. At step 6, the outcome of the diagnostic procedure executed at the charger 300A is stored again at the diagnostic database 210 as part of the diagnostic history, at 310.

[0063] At step 7, it may be noted that various other chargers 300B also communicate with the cloud backend 302. Further, at step 8, the diagnostic history is also shared with the predictive model 312 along with the data from the cloud backend 302, at step 9. The predictive model 312 uses the diagnostic history and the data received from the cloud backend 302 to predict the possibility of the errors from the diagnostic history occurring in any other charger and at step 10, displaying the warning message on a user interface (Ul) 314 of the charger to the user 316.

[0064] FIG 3B is an overview of a method for applying a corrective action to rectify an error in the charging station, in accordance with one or more embodiments of the present disclosure. Referring to the FIG. 3B, the error management module 216 receives the meter value data 320A and the charger error logs 320B. Further, at step 322, the error management module 216 performs data pre-processing to the received data. The data pre-processing includes data quality analysis, data transformation and feature engineering techniques. 202413527

[0065] Then at 304 the processed data is provided to the multi-class classification model which identifies the error code related to the identified error at the charging station from the normal operational condition. The error code classifies between different error categories and types automatically based on the learning of the historical data by the multi-class classification model.

[0066] On receiving the processed data, the multi-class classification model maps the error code to the diagnostic procedure or sequence, at 308, to rectify the charging station, a program is created to rectify the error by initiating the troubleshooting activity at the charging station. Further, the error management module 216 also verifies if the error identified is fixed at the charging station. In case the error is not fixed even after a specific number of attempts then cascade the diagnosis to the field team with relevant information to fix the problem and procure any necessary spares and equipment at the time of visit.

[0067] Further, the error management module 216 stores the procedure executed to address the error in the diagnostic database 210, at 324. This will again be used by the classification model on the occurrence of the error in any other charging station.

[0068] FIG 3C is an overview of a method for predicting occurrence of the error in another charging station, in accordance with one or more embodiments of the present disclosure. Referring to the FIG. 30, the proposed solution includes predicting the occurrence of the error in other charging stations based on the learning attained from one charging station.

[0069] To be able to predict the occurrence of the error in another charging station the error management module 216 of the system 200 requires three major inputs which includes meter value data 330A, charger error logs 330B and history of the outcomes when the corrective action is applied to a specific charger to rectify the errors in it 330C. Here, the history is retrieved from the diagnostic database 210. The error management module 216, at step 332 then determines the co-relation between the meter value data 330A, the charger error logs 330B and the history 330C and clusters the data for similar hardware, location information, usage pattern, and sequence of events. Further, the co-relation is used to train the prediction model. 202413527

[0070] The proposed solution includes recording the corrective actions or the diagnostic history to improve future occurrences of the error in similar hardware located elsewhere. It also predicts the occurrence of error events by monitoring operational parameters using artificial intelligence and machine learning models.

[0071] Therefore, the proposed solution also includes a prognosticate and predict approach. It alerts or hints on breakdown or error conditions based on the historic data and lessons learnt to avoid the possibility of any future similar errors by actively taking corrective actions. Therefore, the proposed solution improves the durability, stability and reliability of the chargers.

[0072] FIG 4 is a flowchart representation of overall the method of deploying an ML model and managing the errors in the charging station, in accordance with one or more embodiments of the present disclosure. Referring to the FIG. 4, at step 402, the system 200 identifies the type of the charger and at step 404, determines if an appropriate multi-class classification model exists for the charger type in the diagnostic database 210. Further, on determining that the appropriate multi-class classification model exists, at step 406 the system 200 maps the multi-class classification model to the charger type.

[0073] On determining that the appropriate multi-class classification model does not exist for the charger type in the diagnostic database 210, then the system 200 at step 408 collects the data from the cloud backend 302 and at step 410 the system 200 creates the multi-class classification model using the historical data. Further, at step 412, the system 200 saves the multi-class classification model to the diagnostic database 210. Now each charger type has a multi-class classification model available in the diagnostic database 210.

[0074] At step 414, the system 200 receives the real-time meter value data from the charger and at step 416 receives the real-time OCPP logs. At step 418, the system 200 determines if there is an error at the charger based on the real-time meter value data and the real-time OCPP logs. On determining that there persists an error at the charger, at step 420 the system 200 reads the realtime meter value data and the real-time OCPP logs and at step 422 sends it to multi-class classification model. Further, the system 200 at step 424 maps the error code to the diagnostic procedure or corrective action associated with the specific error or the charger type. At step 426, 202413527 the system 200 automatically triggers self-healing procedure i.e., application of the mapped corrective action to rectify the error in the charger.

[0075] At step 430, the system 200 checks if the error still persists and if the number of attempts to rectify the error is less than three. On determining that the error persists, and the number of attempts is less than 3 at step 428, the system 200 tries applying the mapped corrective action. Further, at step 432, the system 200 determines that the error does not persist and is rectified. At step 432, if the system 200 determines that the number of attempts is greater than three then it automatically raises a customer ticket to notify a field engineer to manually rectify the error.

[0076] At step 434, the system 200 monitors the meter value, events, and other parameters and at step 436 the system 200 uses the predictive model for clustering and correlation of the received data. Further, at step 438 the system 200 predicts the outcomes for abnormality and at step 440 suggests predictive action to avoid future errors. It may be noted that at step 442 the system 200 avoids the error in future due to the notification of the possible errors in advance.

[0077] FIG 5 is a flowchart representation illustrating an example error in the charging station being managed, in accordance with one or more embodiments of the present disclosure. Referring to the FIG. 5, at step 502, the charging session at the charging station begins and at step 504 the system 200 begins streaming the data through the cloud server 302. At step 506, an error message “over temperature” appears on a screen of the charger and the error is identified. At step 508, multi-class classification model classifies the error into one of the existing error types based on the type of the charger. At step 510, the system 200 determines the corrective action which is mapped to the determined error type in the diagnostic database 210. At step 512, the system 200 implements the corrective action to rectify the error in the charger. The corrective action can be for example, pausing the operation of the charger. At step 514, the system 200 intimates the user through application Ul about the corrective action being implemented. At step 516, the system 200 determines the temperature value after the implementation of the corrective action at the charger. At step 518, the system 200 determines that the temperature is optimum and at step 520, it resumes operation on determining that the operation conditions are conducive. Further, the system 200 at step 522, records the corrective action as part of the diagnostic history in the diagnostic database 210. 202413527

[0078] Further, at step 524, if the system 200 determines that the temperature is still above the operating conditions even after the application of the corrective action, the system 200 increases pause time and restarts the charger. At step 526, the system 200 warns the user by displaying a warning message through the application Ul. At step 528, the system 200 then attaches ticket for regional service personnel with data evidence and spare part information to be addressed manually as the self-healing approach could not address the error and at step 530, the cycle ends, and the procedure ends at step 532.

[0079] Here, the system 200 tries multiple times to apply the corrective measures to the detected error before raising the ticket to allocate the same to the service personnel. Therefore, the service personnel are directed to manually fix the error at the charger only after the historically implemented corrective actions have been tried to solve the error.

[0080] Therefore, the proposed invention provides a unified solution that can monitor and handle different product types of the chargers using multiple classifier models. Also, the proposed solution predicts the future error occurrences based on shared history from similar hardware. As a result, the potential errors are predicted and handles much in advance thereby preventing the errors from occurring.

[0081] The proposed solution is cost and effort savings from frequent field visits by technicians and service personnel. Also, any charger which implements self-healing of the errors will provide an improved product perception among the customers and also impacting market image of the chargers.

[0082] While the present disclosure has been described in detail with reference to certain embodiments, it should be appreciated that the present disclosure is not limited to those embodiments. In view of the present disclosure, many modifications and variations would be present themselves, to those skilled in the art without departing from the scope of the various embodiments of the present disclosure, as described herein. The scope of the present disclosure is, therefore, indicated by the following claims rather than by the foregoing description. All changes, modifications, and 202413527 variations coming within the meaning and range of equivalency of the claims are to be considered within their scope.

[0083] 202413527

[0084] Reference Numerals method 100 step 101 step 102 step 103 step 104 step 105 system 200 bus 202 processor 204 memory 206 interface 208 database 210 input device 212 output device 214 error management module 216 chargers 300A-B cloud server 302 classification model 304 remote commands 306 step 308 predictive model 312 user interface (U I) 314 user 316 meter value data 320A charger error logs 320B step 322 step 324 step 330A-C step 332 step 402 202413527 step 404 step 406 step 408 step 410 step 412 step 414 step 416 step 418 step 420 step 422 step 424 step 426 step 428 step 430 step 432 step 434 step 436 step 438 step 440 step 442 step 502 step 504 step 506 step 508 step 510 step 512 step 514 step 516 step 518 step 520 step 522 step 524 202413527 step 526 step 528 step 530 step 532

Claims

202413527PATENTANSPRUCHE / PATENT CLAIMS1. A method (100) for managing errors in a charging station, the method (100) comprising: receiving, by a processor (204), charger error logs and meter value data associated with the charging station; determining, by the processor (204), an error code associated with at least one error in the charging station, using a first machine learning (ML) model, wherein the first ML model is a multiclass classification model; mapping, by the processor (204), the error code to a corrective action stored in a diagnostic database (210); and automatically applying, by the processor (204), the mapped corrective action to rectify the at least one error in the charging station.

2. The method (100) according to claim 1 , further comprising: determining, by the processor (204), an outcome of the application of the corrective action to rectify a plurality of errors in the charging station; and storing, by the processor (204), the outcome of the application of the corrective action to rectify the plurality of errors, in the diagnostic database (210).

3. The method (100) according to claim 2, further comprising: receiving, by the processor (204), outcomes of the application of the corrective action to rectify the plurality of errors from the diagnostic database (210); determining, by the processor (204), a co-relation between outcomes of the application of the corrective action to rectify the plurality of errors, the charger error logs and the meter value data; and training, by the processor (204), a second machine learning (ML) model based on the corelation.

4. The method (100) according to claim 2, wherein the second ML model is a prediction model.

5. The method (100) according to claim 3, further comprising: clustering, by the processor (204), a set of charging stations based on a plurality of similarities; monitoring, by the processor (204), a plurality of operational parameters associated with the set of charging stations; and24202413527 predicting, by the processor (204), occurrence of the at least one error of the plurality of errors in the set of charging stations using the second ML model.

6. The method (100) according to claim 5, further comprising: recommending, by the processor (204), the at least one corrective action to rectify the at least one error in the charging station; and displaying, by the processor (204), the recommended at least one corrective action on a screen of the charging station or an application on a mobile device.

7. The method (100) according to claim 1 , further comprising: determining, by the processor (204), that the at least one error in the charging station persists after application of the mapped corrective action; and sending, by the processor (204), an indication to at least one personnel to address the at least one error in the charging station.

8. The method (100) according to claim 1 , wherein the determination of the first ML model comprises: determining, by the processor (204), a type of the charging station; determining, by the processor (204), his4toric data associated with the charging station from a cloud server; generating, by the processor (204), the first ML model using the historic data associated with the charging station; and storing, by the processor (204), the first ML model in the diagnostic database (210).

9. The method (100) according to claim 1 , wherein the error code indicates an error category of a plurality of error categories.

10. The method (100) according to claim 1 , wherein the at least one error occurred in the charging station is determined based on the charger error logs and the meter value data.

11. A system (200) for managing errors in a charging station, the system (200) comprising: a processor (204); and a memory (206) coupled to the processor (204), wherein the memory (206) comprises instructions which, when executed by the processor (204), configures the processor (204) to: receive charger error logs and meter value data associated with the charging station;202413527 determine an error code associated with at least one error in the charging station, using a first machine learning (ML) model, wherein the first ML model is a multi-class classification model; map the error code to a corrective action stored in a diagnostic database (210); and automatically apply the mapped corrective action to rectify the at least one error in the charging station.

12. A computer program product, comprising computer program code which, when executed by a processor (204), cause the processor (204) to carry out the method (100) of one of the claims 1 to 10.

13. A computer-readable medium comprising a computer program product comprising computer program code which, when executed by a processor (204), cause the processor (204) to carry out the method (100) of one of the claims 1 to 10.

Images