Environmental Sensing for Black Soldier Fly larvae

Context

The environmental sensor for black soldier fly larvae was created to uncover the ideal conditions for maximum breeding and hence higher output of larvae to be used as animal feed. The Sustainable Living Lab (SL2) in Singapore has created this prototype from scratch and implemented it on the Raitong Organics Farm in Northeastern Thailand. The entire process of conceptualizing, designing, iterative development, testing and implementation spanned close to a month, including five days spent on the Raitong Organics Farm itself.

Problem Statement

Black soldier flies are known to be efficient converters of organic materials, from food waste to manure, into insect biomass, with a conversion rate of about 20% in ideal conditions. The black soldier fly larvae are composed of 42% crude protein and 29% fat and are higher in saturated fats than most insects. This makes them highly valuable as animal feed due to their high nutritional content. Moreover, these black soldier flies help to close nutrient loops, reducing pollution and costs. These flies are also proven to be non-transmitters of disease-causing pathogens, making them safe to breed for animal consumption. Unfortunately, the yield of black soldier fly larvae on the Raitong farm has been rather low, around 10 to 15%. The environmental sensor aims to circumvent this problem by helping to analyse the appropriate conditions required for maximum breeding of the black soldier flies.

Factors Being Monitored

For this project, we will be monitoring the following:

• Ambient air temperature

• Ambient humidity

• Light intensity

• Moisture content of the food waste

We would like to find out what combination of environmental conditions will result in the greatest growth of the black soldier flies.

Experiment Plan

The black soldier fly breeding container is located within the piggery. One side is closer to the sun light and thus we expect it to be hotter and less damp as well. The box is also made with three different channels where the larvae can climb up and out to the containers when they are about to pupate and need dryer conditions. These larvae are self harvesting!

The plan is to monitor and compare the different environmental conditions and the resulting harvest from the two halves of the breeding box. This way we will understand how environmental conditions affect the BSF larvae's growth.

Prototype Development

There are three sensors with four sensor reading outputs, namely temperature (deg Celsius) and humidity (%) from the DHT22 sensor, light level measured in Lux by the BH1750 light sensor, and finally, the moisture content measured in percentage (%) by the capacitive soil moisture sensor.

There is also a microSD card module with a Real Time Clock (RTC) that will log the sensor data in regular intervals of 30 minutes. The DS3231 RTC will help include a date time stamp into a text file in the microSD card each time the four sensor values are logged.

Early Results

We did not manage to do proper testing of the setup nor collect data as we had to fly back to Singapore.

Future Improvements

1. Improve the robustness of the setup:-

• We had some reliability issues with the soil moisture sensor because the exposed electronics on the sensor cannot get wet or get exposed to heat. So we have to create a protective waterproof casing for it to work properly.

• We also have to ensure proper environment protection for the enclosure, to ensure no moisture or dust gets into the box.

• We have to ensure proper wiring methods to ensure circuit connectivity. This means no breadboard and soldering the circuit onto a perforated board or making a printed circuit board.

• We have to stress test the setup over a period of time to check that the system is stable and can log data consistently.

2. Adding more sensors

• We can explore adding more sensors, such as a pH sensor.

3. Data collection and analysis

• We have to collect data over the course of weeks, months or even years testing for different environment conditions and then analyse the data to reveal correlations with improved breeding outcomes. This can then be shared online publicly for research purposes.

4. Adding Internet of Things capability

• The current setup requires manual removal of the microSD card from the box which will disrupt the sensor data logging. The microSD card has to be manually read by a computer and the data extracted for analysis. A more ideal setup would include automatic upload of the data into the cloud without the need for the user to do any kind of physical process. This can be achieved by using a combination of LoRa radio and WiFi technology.