Project screenshot

LVCloud is an Internet of Things (IoT) application that monitors and takes care of the grid. Check out how we have combined new technologies, such as machine learning, into a sophisticated app that can detect and solve problems with electricity networks before anyone knew they existed.

Ruby on Rails, Vue.js, MQTT
The need

Taking care of the grid

When electricity networks malfunction, this can cause serious problems for our society and this is not a science fiction scenario anymore. You could have laughed about it a couple of decades ago, but nowadays we are too dependent on the grid. It needs to be monitored so that specialists can intervene and solve problems quickly before power is lost to businesses and homes

Hardware-wise the grid is monitored by auto reclosing devices, which are sophisticated circuit breakers. But they need sophisticated software to operate correctly.

First project image
The solution

Monitor the grid like a pro

The reclosing device's main job is to restore the electricity supply within the area it operates in. When something goes wrong, the circuit breaker will open and therefore isolate that part of the electricity network where the fault is present. However, using the in-built intelligence of the auto-reclosing circuit breaker, it will test to see if the fault is still present and if not, it will close and restore supplies to customers.

It's XXI century so the recloser is much more sophisticated and it also enables recording of critical information about the grid's state.

Now this is where Naturaily steps in. We have created software that collects information from reclosers and analyzes it to find traces of malfunction or physical damage. Any type of non-obvious problems like too high humidity caused by various incidents or cables in a poor state of repair. Think of “Minority Report” class software for finding electricity network pre-faults.

Other devices exist: TDRs (Time-domain reflectometers), ATLMs (ALVIN Transformer Load Monitors). We gather data from all of them. And also enable managing them with our software.

In addition to a web based dashboard, engineers are notified via email or text.

Second project image
The tools

Ruby-based microservices

The software is created for large scale: hundreds of thousands of devices and millions of events. But the number of devices is not the biggest problem. The main requirement was that the application works in real time! To handle this, we have designed Ruby-based microservices architecture and wrote the most critical parts in faster, more concurrent language - Elixir. Apart from scalability, we wanted to bring XXI century to this solution and implemented machine learning algorithms to find patterns and detect prefaults.

We use Docker for one-click super convenient server creation process for each end customer. It’s critical as our solution is more PaaS-like solution where each end customer has its own hardware set up.

MongoDB and Redis are used to store and work with data. As our data model must be quite elastic and write speed is crucial, MongoDB is a great choice. MQTT is the standard communication protocol between end devices and our system.

To put it shortly: heaven of internet of things development stack.

ruby vue mqtt
Third project image
The results

Watch Netflix peacefully

We are kidding, of course. This has much more impact for the stability of the network and your life. Charging your Tesla can overload the grid, so the sooner maintenance people get information about potential malfunction, the better, because you won't even know a problem existed.

Fourth project image Fifth project image Sixth project image
blue-wave

Contact us