Description:FIELD OF INVENTION
The present invention relates to a system and method for detecting anomalies in the energy meter. More specifically present invention relates to the detection of various kinds of anomalies in the images captured during the process of installation.
BACKGROUND OF THE INVENTION
Imagining the world without any use of electricity is very difficult these days. Electricity is an essential aspect of modern life, as it powers almost every device and system we rely on, from powering our homes and businesses to transportation and manufacturing. Without energy, our modern society could not function efficiently, and we would struggle to maintain our quality of life. An energy meter is an important device used to measure the amount of electrical energy consumed by a household, business, or industry. It helps to track the energy usage, and monitor the energy bills, which helps consumers to be more conscious of their energy consumption and take steps to reduce it. Meter Installation is an essential component in the functioning of an electric utility. New meter installation for a customer involves standard procedures which must be followed by the engineer who is responsible for the installation. In past, cases have come up where standard procedure was not followed and meter was left with missing seals and other defects. To check for these anomalies, daily manual check of previous day’s installation is Traditional methods of detecting energy consumption anomalies in electricity meters relied on manual inspection or periodic readings by utility staff. . This manual checking takes lot of time and requires manpower. Manual checking is done on day to day basis for ensuring that all standard procedures were followed by the engineer.
CN109934821B discloses The invention provides a method and a system for detecting part defects, wherein the method comprises the following steps: acquiring a detection image of a detected part shot by image acquisition equipment from a preset shooting angle; inputting the detection image into a defect positioning model trained in advance, and determining the position of a defect on the part according to a detection result output by the defect positioning model; the category of the identified defect is determined by a defect classification model trained in advance. The scheme provided by the invention can be used for automatically detecting the surface defects of the parts.
Existing Invention takes a lot of time and depends on human intelligence and focus. Also, the existing invention has a higher probability of human error. Hence, there is a need for the present invention.
OBJECTIVE OF THE INVENTION
The main objective of the present invention is to detect anomalies in the energy meter.
Another objective of the present invention is to identify the anomalies in the images of the energy meter captured during the installation process.
Yet another objective of the present invention is to input to develop the machine-learning model to identify the kind of anomalies in the installation of electrical equipment.
Yet another objective of the present invention is to get rid of the manual meter checking process and thus saving time and manpower.
Yet another objective of the present invention is to improve the accuracy of detecting even minute anomalies that may go unnoticed during manual inspections.
Yet another objective of the present invention is to achieve cost efficiency by automating the inspection process.
Further objectives, advantages, and features of the present invention will become apparent from the detailed description provided herein below, in which various embodiments of the disclosed invention are illustrated.
SUMMARY OF THE PRESENT INVENTION
The present invention relates to a system for the detection of anomalies in the installation of the energy meter. The present invention includes a main server, an engineer’s device, a display unit, and an energy meter. The main server includes a database unit and a system processing unit. The database unit stores computer-readable instructions, machine-learning models, and image databases. The system processing unit executes computer-readable instructions to input the image database into a machine-learning model for detecting anomalies in the installation of energy meters. The engineer’s device is connected to the system processing unit of the main server. The display unit is connected to the system processing unit of the main server. The energy meter is the object that needs to be detected for anomalies. The system processing unit executes computer-readable instructions to collect a plurality of images of the energy meter through the engineer’s device. The database unit stores a plurality of images that are subsequent to the plurality of the images of the energy meter. The system processing unit inputs the plurality of image datasets into the machine-learning model in order to train a machine-learning model for detecting different categories of anomalies in the installation of energy meters. The display unit displays the images along with the anomalies. In an embodiment, the present invention detects different categories of anomalies in the installation of energy meters selected from improper sealing, missing stickers, and improper cable connection.
The main advantage of the present invention is to detect anomalies in the energy meter.
Another advantage of the present invention is that the present invention identifies the anomalies in the images of the energy meter captured during the installation process.
Yet another advantage of the present invention is that the present invention has trained the machine-learning model to identify all kinds of anomalies in the installation of the energy meter.
Yet another advantage of the present invention is that the present invention has eliminated the manual meter-checking process, thus saving time and manpower.
Yet another advantage of the present invention is that the present invention has improved the accuracy of detecting even minute anomalies that may go unnoticed during manual inspections.
Yet another advantage of the present invention is that the present invention has achieved cost efficiency by automating the meter inspection process with image analytics, which reduces labour costs, speeds up the process, and reduced errors, thereby improving accuracy.
Further objectives, advantages, and features of the present invention will become apparent from the detailed description provided herein below, in which various embodiments of the disclosed invention are illustrated.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings are incorporated in and constitute a part of this specification to provide a further understanding of the invention. The drawings illustrate one embodiment of the invention and together with the description, serve to explain the principles of the invention.
Fig.1 illustrates the line diagram of the present invention.
Fig. 2 illustrates a flow chart of the method of detecting anomalies in the energy meter.
Fig. 3 illustrates a flow chart for the method of development and training of the object detection model.
DETAILED DESCRIPTION OF THE INVENTION
While this invention is susceptible to embodiment in many different forms, there is shown in the drawings and will herein be described in detail specific embodiments, with the understanding that the present disclosure of such embodiments is to be considered as an example of the principles and not intended to limit the invention to the specific embodiments shown and described. In the description below, like reference numerals are used to describe the same, similar or corresponding parts in the several views of the drawings. This detailed description defines the meaning of the terms used herein and specifically describes embodiments in order for those skilled in the art to practice the invention.
Definition
The terms “a” or “an”, as used herein, are defined as one or as more than one. The term “plurality”, as used herein, is defined as two or more than two. The term “another”, as used herein, is defined as at least a second or more. The terms “including” and/or “having”, as used herein, are defined as comprising (i.e., open language). The term “coupled”, as used herein, is defined as connected, although not necessarily directly, and not necessarily mechanically. The term “comprising” is not intended to limit inventions to only claiming the present invention with such comprising language. Any invention using the term comprising could be separated into one or more claims using “consisting” or “consisting of” claim language and is so intended. The term “comprising” is used interchangeably used by the terms “having” or “containing”. Reference throughout this document to “one embodiment”, “certain embodiments”, “an embodiment”, “another embodiment”, and “yet another embodiment” or similar terms means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of such phrases or in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics are combined in any suitable manner in one or more embodiments without limitation. The term “or” as used herein is to be interpreted as an inclusive or meaning any one or any combination. Therefore, “A, B or C” means any of the following: “A; B; C; A and B; A and C; B and C; A, B and C”. An exception to this definition will occur only when a combination of elements, functions, steps or acts are in some way inherently mutually exclusive.
As used herein, the term "one or more" generally refers to, but is not limited to, the singular as well as the plural form of the term.
The drawings featured in the figures are for the purpose of illustrating certain convenient embodiments of the present invention and are not to be considered as limitations thereto. The term “means” preceding a present participle of an operation indicates a desired function for which there are one or more embodiments, i.e., one or more methods, devices, or apparatuses for achieving the desired function and that one skilled in the art could select from these or their equivalent in view of the disclosure herein and use of the term “means” is not intended to be limiting.
Fig.1 illustrates the line diagram of the system (100) for the detection of anomalies in the images captured during the installation process of an energy meter. The system (100) includes a main server (102), an engineer’s device (108), a display unit (110), and an energy meter (112). The main server (102) includes a database unit (104) and a system processing unit(106). An engineer’s device (108) is connected to the system processing unit(106) of the main server (102). The display unit (110) is connected to the system processing unit (106) of the main server (102).

Fig. 2 illustrates a flow chart for the method of detecting anomalies in the meter images captured during installation. In step (114), the images of the energy meter (112) are captured at different stages of the installation procedure by an engineer’s device (108). In step (116), the database unit (104) stores a plurality of images of the energy meter (112), captured by an engineer’s device (108). In step (118), the system processing unit (106) runs a trained object detection machine-learning model to detect the anomalies in the captured images In step (120), The display unit (110) displays the images along with the anomalies.
Fig. 3 illustrates a flow chart for the method of development and training of an object detection model. In step (122), a system processing unit (106) creates an image dataset by classifying different types of anomalies and developing a separate object detection model for each type of anomaly. In step (124) an engineer’s device (112) captures the images of the energy meter with anomalies and without anomalies. In step (126), the system processing unit (106) label each image and mark the region of interest to train an object detection model; In step (128), the System processing unit (106) train and evaluate the object detection model by splitting the dataset into 90% training and 10% testing model. In step (130) the System processing unit (106), pre-process the collected images by resizing and removing noises. In step (132), the system processing unit (106) evaluates the trained object detection model on the basis of some metrics and exports the trained models for object detection.
The present invention relates to a system for the detection of anomalies in the installation of the energy meter. The present invention includes a main server, an engineer’s device, a display unit, and an energy meter The main server includes a database unit and a system processing unit. In an embodiment, the main server is including, but is not limited to, a desktop computer, a laptop, a tablet, a smartphone, and a mobile phone. The database unit stores computer-readable instructions, machine-learning models, and image databases. The system processing unit executes computer-readable instructions to input the image database into a machine-learning model for detecting anomalies in the installation of energy meters. The engineer’s device is connected to the system processing unit of the main server. In an embodiment, the engineer’s device is selected from a smartphone or a mobile phone. The display unit is connected to the system processing unit of the main server. The energy meter is the object that needs to be detected for anomalies. The system processing unit executes computer-readable instructions to collect a plurality of images of the energy meter through the engineer’s device. The database unit stores a plurality of images that are subsequent to the plurality of the images of the energy meter. The system processing unit inputs the plurality of image datasets into the machine-learning model in order to train a machine-learning model for detecting different categories of anomalies in the installation of the energy meter. The display unit displays the images along with the anomalies. In an embodiment, the present invention detects different categories of anomalies in the installation of energy meters selected from improper sealing, missing stickers, and improper cable connection.

In an embodiment, the present invention relates to a method of detection of anomalies in the energy meter installation, the method includes:
a system processing unit of the main server is connected to an engineer’s device and a display unit;
engineers follow the standard operating procedure for the installation of energy meters;
an engineer’s device captures the images of the energy meter in different stages of the installation procedure;
an engineer’s device sends the captured images to the main server for captured images to be stored in the database unit;
a system processing unit executes computer-readable instructions to pre-process captured images of the energy meter that includes resizing and denoising;
a system processing unit executes computer-readable instructions that run a trained object detection machine-learning model that generates a confidence score based on the detection of the anomalies in the captured images;
anomalies with a confidence score greater than the threshold are accepted whereas anomalies with a threshold score less than the threshold are rejected.
the system processing unit executes computer-readable instructions to prepare a report of each rejected anomaly against the installation number of the energy meter;
and a display unit displays the report and the engineer is instructed to remove anomalies in the installation.
In an embodiment, a method of development and training of an object detection model, the method having
a system processing unit executes computer-readable instructions to classify the images of different types of anomalies, from the past cases of meter installation;
a system processing unit executes computer-readable instructions to build a separate object detection model for each type of anomaly;
an engineer’s device executes computer-readable instructions to collect the images of energy meter with anomalies and without anomalies;
a system processing unit executes computer-readable instructions to create a dataset for object detection model by repeatedly collecting images for all the classified anomalies;
a system processing unit executes computer-readable instructions to label each image and mark the region of interest in each image;
a System processing unit executes computer-readable instructions to train an object detection model;
a System processing unit executes computer-readable instructions to train and evaluate the object detection model by splitting the dataset into 90% training and 10% testing model;
a System processing unit executes computer-readable instructions to convert image dataset into TF Record to maximize the input/output operations and efficacy of storage;
a System processing unit executes computer-readable instructions to pre-process the collected images by resizing and removing noises;
Tuning a combination of hyper parameters of the model to maximize the model accuracy and minimize a loss function;
a system processing unit executes computer-readable instructions to train object detection models on the collected image dataset;
a system processing unit executes computer-readable instructions to evaluate the trained object detection model on the basis of some metrics and export the trained models for object detection.
In an embodiment, the dataset includes a variety of image sample files of energy meters with anomalies present and images without anomalies.
In an embodiment, the machine-learning model is being trained to detect the anomalies in the images captured at the time of installation is ssdmobnet model.
In an embodiment, the present invention relates to a system for the detection of anomalies in the installation of the energy meter. The present invention includes one or more main servers, one or more engineer’s devices, one or more display units, and one or more energy meters. The one or more main servers include one or more database units and one or more system processing units. In an embodiment, one or more main servers are including, but are not limited to, a desktop computer, a laptop, a tablet, a smartphone, and a mobile phone. The one or more database units store computer-readable instructions, machine-learning models and image databases. The system processing unit executes computer-readable instructions to input image database into a machine-learning model for detecting anomalies in the installation of one or more energy meters. The one or more engineer’s devices are connected to the system processing unit of the one or more main servers. The one or more display units are connected to the system processing unit of the one or more main servers. The one or more energy meters are the object that needs to be detected for anomalies. The system processing unit executes computer-readable instructions to collect a plurality of images of the energy meter through the one or more engineer’s devices. The one or more database units store plurality of images that are subsequent to the plurality of the images of the energy meter. The system processing unit input the plurality of image datasets into the machine-learning model in order to train a machine-learning model for detecting different categories of anomalies in the installation of one or more energy meters. The display unit displays the images along with the anomalies. In an embodiment, the present invention detects different categories of anomalies in the installation of energy meter selected from improper sealing, missing sticker, and improper cable connection.
In an embodiment, the one or more engineer’s devices are selected from a smartphone or a mobile phone.
In an embodiment, the present invention relates to a method of detection of anomalies in the energy meter installation, the method includes:
a system processing unit of the one or more main servers are connected to one or more engineers’ devices and a display unit ;
engineers follow the standard operating procedure for the installation of energy meters;
the one or more engineers’ devices, capture the images of the energy meter in different stages of the installation procedure ;
the one or more engineers’ devices send the captured images to one or more main servers for captured images to be stored in one or more database units;
a system processing unit executes computer-readable instructions to pre-process captured images of the energy meter that includes resizing and denoising;
a system processing unit executes computer-readable instructions that run a trained object detection machine-learning model that generates a confidence score based on the detection of the anomalies in the captured images;
anomalies with a confidence score greater than the threshold are accepted whereas anomalies with a threshold score less than the threshold are rejected;
the system processing unit executes computer-readable instructions to prepare a report of each rejected anomaly against the installation number of the energy meter;
and a display unit displays the report and the engineer is instructed to remove anomalies in the installation.

In an embodiment, a method of development and training of an object detection model, the method having
a system processing unit executes computer-readable instructions to classify the images of different types of anomalies, from past cases of meter installation;
a system processing unit executes computer-readable instructions to build a separate object detection model for each type of anomaly;
the one or more engineers’ devices collect the images of the energy meter with anomalies and without anomalies;
a system processing unit create a dataset for object detection model by repeatedly collecting images for all the classified anomalies;
a system processing unit label each image and mark the region of interest in each image;
a system processing unit executes computer-readable instructions to train an object detection model;
a system processing unit executes computer-readable instructions to train and evaluate the object detection model by splitting the dataset into 90% training and 10% testing model;
a System processing unit executes computer-readable instructions to convert image dataset into TF Record to maximize the input/output operations and efficacy of storage;
a System processing unit executes computer-readable instructions to pre-process the collected images by resizing and removing noises;
Tuning a combination of hyperparameter of the model to maximize the model accuracy and minimize a loss function;
a system processing unit executes computer-readable instructions to train object detection model on the collected image dataset;
a system processing unit executes computer-readable instructions to evaluate the trained object detection model on the basis of some metrics and exporting the trained models for object detection.
In an embodiment, the dataset includes a variety of image samples files of energy meter with anomalies present and images without anomalies.
In an embodiment, the machine-learning model is being trained to detect the anomalies in the images captured at time of installation is ssdmobnet model.
, Claims:WE CLAIM
1. A system(100) for detection of anomalies in the installation of the energy meter (112), the system(100) comprising:
an at least one main server(102), the at least one main server(102) having:
an at least one database unit (104), the at least one database unit(104) stores computer-readable instructions, machine-learning model and image database, and
an at least one system processing unit(106), the at least one system processing unit(106) executes computer-readable instructions to input the image database into a machine-learning model for detecting anomalies in the installation of an at least one energy meter (112);
an at least one engineer’s device (108), the at least one engineer’s device (108) is connected to the system processing unit (106) of the at least one main server (102); and
an at least one display unit (110), the at least one display unit (110) is connected to the system processing unit (106) of the at least one main server (102), Wherein, the display unit (110) displays the images along with the anomalies;

wherein, the system processing unit(106) executes computer-readable instructions to collect a plurality of images of the energy meter (112) through at least one engineer’s device(108) and the at least one database unit(104) stores a plurality of images that are subsequent to the plurality of the images of the energy meter (112),
wherein, the engineer’s device(108) captures the images of the energy meter in different stages of the installation procedure, and sends the captured images to the at least one main server(102),
wherein, the system processing unit(106) executes computer-readable instructions to pre-process captured images of the energy meter(112),
wherein, the system processing unit(106) system processing unit(106) executes computer-readable instructions that run a trained object detection machine-learning model that generates a confidence score based on the detection of the anomalies in the captured images,
wherein, anomalies with a confidence score greater than the threshold are accepted whereas anomalies with a threshold score less than the threshold are rejected,
wherein, the system processing unit(106) executes computer-readable instructions to prepare a report of each rejected anomaly against the installation number of the energy meter,
wherein, the system processing unit(106) inputs the plurality of image datasets into the machine-learning model in order to train a machine-learning model for detecting anomalies in the installation of an energy meter (112).
2. The system (100) as claimed in claim 1, wherein the at least one main server (102) is selected from a desktop computer, a laptop, a tablet, a smartphone, a mobile phone.
3. The engineer’s device (108) as claimed in claim 1, wherein the at least one engineer’s device (108) is selected from a smartphone or a mobile phone.
4. The dataset as claimed in claim 1, wherein the dataset includes a variety of image samples files of energy meter (112) with anomalies present and images without anomalies.
5. The system (100) as claimed in claim 1, where a method for detection of anomalies in the energy meter (112) installed, the method comprising:

a system processing unit(106), connected to an engineer’s device (108) and a display unit (110),
engineers follow the standard operating procedure for the installation of energy meters;
an engineer’s device(108) captures the images of the energy meter in different stages of the installation procedure,
an engineer’s device(108) sends the captured images to the at least one main server(102) for captured images to be stored in the database unit(104);
a system processing unit(106) executes computer-readable instructions to pre-process captured images of the energy meter(112) that includes resizing and denoising;
a system processing unit(106) executes computer-readable instructions that run a trained object detection machine-learning model that generates a confidence score based on the detection of the anomalies in the captured images;
anomalies with a confidence score greater than the threshold are accepted whereas anomalies with a threshold score less than the threshold are rejected.
the system processing unit executes computer-readable instructions to prepare a report of each rejected anomaly against the installation number of the energy meter;
and a display unit(110) displays the report and the engineer is instructed to remove anomalies in the installation.
6. The method as claimed in claim 5, wherein a method of development and training of an object detection model, the method comprising
a system processing unit(106) executes computer-readable instructions to classify the images of multiple anomalies, from the past cases of meter installation;
a system processing unit(106) to build a separate object detection model for each type of anomaly;
an engineer’s device (108) to collect the images of the energy meter (112) with anomalies and without anomalies;
a system processing unit(106) to create a dataset for an object detection model by repeatedly collecting images for all the classified anomalies;
a system processing unit(106) to label each image and mark the region of interest;
a System processing unit (106) executes computer-readable instructions to train an object detection model;
a System processing unit (106) to train and evaluate the object detection model by splitting the dataset into 90% training and 10% testing model;
a System processing unit (106) to convert image dataset into TF Record to maximize the input/output operations and efficacy of storage;
a System processing unit (106) to pre-process the collected images by resizing and removing noises;
Tuning a combination of hyper parameter of the model to maximize the model accuracy and minimize a loss function;
a system processing unit(106) to train object detection model on the collected image dataset; and
a system processing unit(106) to evaluate the trained object detection model on the basis of some metrics and exporting the trained models for object detection.
7. The method as claimed in claim 5, wherein the machine-learning model is being trained to detect the anomalies in the images captured at time of installation is ssdmobnet model