Data-driven decision support scheme for multi-area light environment control in greenhouse

The environmental parameters of greenhouses are coupled with each other and the spatial distribution is uneven, which makes the control of greenhouse complex. Light is a key factor in crop growth, and optimizing the light environment control of greenhouse is significance. A data-driven decision support scheme for controlling light environment in different areas of a greenhouse is proposed, which is expected to achieve accurate control of multiple areas, meet crop growth needs, and reduce light resource consumption. The measurement and control architecture of this scheme and the improved decision-making model for light environment control are introduced. Two experiments were conducted in a real strawberry greenhouse, including different areas at the same time and the same area at different times. A continuous experiment was conducted in an area for one day. Compared to light saturation control, for strawberries, an average of about 70% of light demand is saved in each light environment control decision on this day, at the cost of an average loss of 22% of photosynthetic rate. The scheme not only enriches the application of data-driven models, but also provides the design of a deployment method for precise light supplementation oriented to the actual crop situation, indirectly reducing the need for environmental management and the corresponding resource consumption of unevenly distributed environmental parameters in the greenhouse.

Automated summary

The environmental parameters of greenhouses are interrelated and unevenly distributed, making greenhouse control complex. Optimizing the light environment in greenhouses is significant for crop growth. A data-driven decision support scheme for controlling light environment in different areas of a greenhouse is proposed, aiming to achieve accurate control, meet crop growth needs, and reduce light resource consumption.