It is easier to create a project for a commercial organization than for a government organization. Over the past year and a half, we have implemented more than twenty LoRa tasks, but we will remember this one for a long time. Because here we had to work with a conservative system.
In this article I will tell you how we have simplified the management of city lighting and made it more accurate in relation to daylight hours. I will praise us and scold our national identity. I will also share why we abandoned wires in favor of a radio network and how another unemployed engineer appeared in the world.
First, I'll tell you what we did. Then - how we did it and what difficulties we overcame.
We have created a smart urban lighting control system in a regional city. It works via LoRaWAN. Commands are sent to the radio module to turn the light on and off. We used class C devices because the system has constant power.
Just in case, let me remind you that the class C radio module is permanently on the air, waiting for a server command.
We have a schedule for sending commands and a mechanism for reporting errors. There is also a check of the functionality of the radio module itself.
That's all. Here questions may arise: what did you do that was so revolutionary? The city lights worked without you: they came on in the evening and went out in the morning. What is the value of the project?
Counter question: have you noticed that city lighting does not always turn on on time? It can be quite dark outside, but the streetlights are not on. This is especially noticeable during transition periods, when daylight hours are actively decreasing or increasing. In the Ural region this is noticeable in October-November.
So we smoothly move on to the difficulties and features of the project.
Our experience, or how to improve urban lighting management
The customer is a government organization.
The lighting control system operates on a chain principle. This is when there are lamp posts with a common power supply. There can be from several to several dozen such pillars in one chain. It depends on the scale of the site.
Each circuit has its own control cabinet; it contains an electric meter and an on/off relay with the main power supply. I can’t attach a photo of the cabinet because the customer forbade it to be shown. Honestly, he looks so-so.
During the daytime there is no power on the poles. Therefore, it is impossible to install a light sensor or a personal relay on each lamp.
Total: we have an outdated chain-link system for city lighting, which needs to be improved and “modernized”.
Here are the obvious disadvantages of such a system:
1) A timer is used to regulate the time the lights are turned on and off.
But the device cannot keep up with the daylight hours. The engineer brings it in manually. He does this not every day, but with a given frequency. Accordingly, there is always an error.
2) In such a system there is no notification of breakdowns. Something went wrong, and the customer did not receive a prompt message. And this is very critical. Because such violations can result in significant fines and penalties. Still, it's an urban issue.
3) There is no automatic correction of energy consumption based on daylight hours. Hence the situation when it is already dark outside and the lights are not on.
4) There is no information about abnormal energy consumption indicating the area.
Someone has connected to the lamp, stealing energy, but the customer does not see it. By the way, such precedents often occur in regional cities with private buildings.
It is difficult to talk to a government customer. Because he is already accustomed to a system that seems to work, but he would like it to be better. At the same time, we need to make sure that it is easy to manage and that local craftsmen can handle it. You can’t invite specialists from the regional center every time.
And yet - it should be cheap and last a long time.
What we did:
1) Instead of wires, a radio network was used. This allowed us to stay within budget and make the system universal.
The control cabinet may be located in the center of an industrial zone or at the entrance to a city - running a wire to it is expensive and difficult, and not always possible. The radio network copes with the tasks perfectly, works stably and is cheaper for the customer.
2) To control the system, we used SI-12 radio modules from Vega. They have control contacts on which we put a power supply relay.
3) We screwed the survey onto the electric meter in the box. There is consumption - the lights are on, there is no consumption - they are turned off.
The survey provides information about the correct operation of the power relay. If it jams, we'll see it.
4) Calculate the average consumption - median consumption. For this we had technical parameters and the number of users.
This is how we were able to obtain information about anomalies. If the consumption is below average, then some of the lights have burned out. If it is higher than average, then someone has connected to the network and is stealing electricity.
5) We made an interface for lighting control. While it is “raw”, we are testing it and most likely will finalize it.
In the interface you can:
1. Add equipment of the “control cabinet” type with a specific address
2. See the state of the cabinet (on - off)
3. Set a schedule for him
4. Tie an electric meter to the cabinet
5. Manually turn on/off the lighting system.
This is necessary for repairs. Engineers work during the day, and the lights are turned off during this time. But the dispatcher will be able to turn them on from the remote control. In this case, you won’t have to disrupt the switching and go into the closet.
6. See the logs of a specific cabinet. They contain data on switching on and off, type (scheduled or manual), and status of operations.
Now the customer does not need to send an engineer to manually adjust the timer. We have improved system management, making it simpler, more stable and clearer. We honestly don’t know what the engineer will do now. But we hope that more serious tasks will be found for him.
The system is currently in testing. Therefore, I will be grateful for practical advice and questions.
We will continue to work on the project. There are thoughts about installing full-fledged controllers in cabinets. The schedule will be stored in their memory, so they will be able to control the lighting without radio communication.
We will also configure the smooth switching on of the lights. This is when, with the onset of twilight, city lighting works at 30 percent. The darker it gets on the street, the brighter the streetlights burn.
There are already ready-made systems for this. They are based on DALI or 0-10 lighting control protocols. In them, you can assign an address to each lamp and control it separately. But the infrastructure of many Russian cities is not ready for this. Upgrading a street lighting system is expensive, and no one is in a hurry to do it.
We are developing our own system that will work in a similar way. More on this in the following articles.
Archive of previous articles:
→ → → → → → → → →
Source: habr.com