What are the Benefits of Applying AGEDB’s System to Your Energy Infrastructure?
Failure Analysis and Monitoring System
Background
One of the long-standing concerns of many energy companies around the world is how to supply energy stably. Countless relay/measurement equipment exists in the process of supplying energy produced by companies to users. However, many companies have difficulties in supplying energy stably due to failures in these equipment and are still investing enormous time and money in responding to failures. Distributing digitized measurement equipment to collect and analyze data is one of the efforts to reduce these costs. The question is whether you are ready to respond quickly if any one piece of equipment fails in the large-scale infrastructure needed to supply energy to hundreds of thousands of users. In particular, if there is no way to quickly identify the failed equipment layer and the equipment in that layer that has a problem, the company’s facility management costs increase and it is difficult to accurately measure users’ energy consumption.
Challenge
‘D Energy Company’, for example, struggled to manage failures in the operation of installed digital metering equipment. There are several tiers of equipment between the home meter and the ‘Company A’ computer room. When you press the microwave oven button at home, the power usage information is sent to the modem layer that transmits/receives data in the residential complex, the data control unit layer installed in public facilities or underground to control the flow of data, and the enterprise it is collected to the corporate server through the ‘front-end processor’ layer that exists in the problem is that this process takes too long to determine exactly which piece of equipment has failed.
If a failure occurs in the data control unit that manages the data of 10 regions, the repair personnel can be dispatched to the nearest branch from where this unit is installed. However, if it is not possible to quickly determine whether the failed equipment is A or a modem in one of the 10 regions under A, then it is inevitable to invest more human and material resources to respond to the failure.
Company D had a large amount of log data but did not have a big data system to store and manage it.
Since all log data containing equipment network information was stored in a relational database, the performance to process huge network data was very poor. As a result, many unnecessary costs were incurred, such as delays in repairs due to not knowing exactly which equipment had failed, and energy usage not being properly measured at the time of failure.
Solution
In this situation, AGEDB can provide four solutions.
AGEDB is a product that can use graph query modeling along with the existing relational model and is a solution product that has the advantage of querying and exploring RDB because SQL syntax is supported [1].
(1) AGEDB stores all log data in the Hadoop ecosystem (Apache Hadoop) and then builds a system that manages the data in graph form using Bitnine’s AgensGraph, the graph database of AGEDB. Each device, in reality, is represented as a single ‘vertex’ in the graph model, and the relationship between devices is expressed as an ‘edge’, forming a ‘graph’ where multiple devices are connected to each other.
AGEDB stores unstructured log data in Hadoop and then models it into a graph structure, and equipment data can be viewed and modified on the graph.
(2) AGEDB provides a platform to visualize equipment networks.
This platform provides visualization that allows you to understand the overall structure and status of your infrastructure at a glance, as well as check the status of your equipment. Administrators can monitor the equipment status by region and section, failure response status, statistical values, etc. while viewing the dashboard, and can also directly view log data by selecting each device on the graph.
(3) AGEDB has built a system to analyze failures in detail based on these platforms.
With this scheme, you can immediately find and analyze devices that are exceeding acceptable temperatures or slowing your network too much.
In the end, it is possible to organize the status and causes of failures and inquire when necessary, and to quickly respond to failures by analyzing what types of failures exist, which failures must be dealt with first, and what are the repeated failure patterns.
In addition, by learning and using failure patterns in an artificial intelligence model, it provides a basis for “preventing” failures rather than simply “managing” them.
(4) AGEDB has established a system to manage equipment history. This system manages and records all activities required to maintain and manage a large number of equipment.
In addition, it provides a function to view the repair history, repair details, and parts replacement history of each equipment in graph by linking with the graph model.
All equipment data is coded to support the systematic management of equipment by the person in charge of equipment operation.
Benefits
The core of the solution provided by AGEDB to the company is a platform that manages physically and complexly connected equipment in the real world as it is in the real world using a graph model. The first thing I noticed when introducing the solution was the difference in ‘performance’.
Companies can find equipment with problems in a large equipment network and quickly query the equipment’s status and repair history.
The changes brought about by graph models do not end here.
GDB models are easy to visualize and easy to understand.
The company’s equipment manager can understand the overall structure of the infrastructure at a glance and grasp the status intuitively through the visualization platform.
In addition, since the status of individual equipment can be inquired with one-click, there is no longer a need to go through a table join process to retrieve data after a long time when curious about the status of a specific equipment.
AGE DB’s failure analysis system provides a cornerstone for more advanced analysis beyond cost reduction through efficient failure response.
Reference
