Greenhouses: heating, vents and screens work against each other

Heating, vents and screens work against each other

What happens. In many greenhouses the heating, the vents and the energy screens are controlled by separate rules. The heating warms the air while the vents open to shed humidity, and the screens open and close on their own schedule.

What it costs. Heating is the largest cost in many greenhouses, and heat that leaves through open vents is paid for twice. Several growers cut their cultivation or gave up when costs went up (source).

Why the usual fix fails. Retuning the rules by hand helps for the season it was done in. A climate computer reacts to the temperature it sees now, not to the weather coming in an hour.

How a fixed-price remote diagnosis checks it. From climate logs (inside and outside temperature, humidity, light, vent and screen positions, heating) I check how much heat is lost to the conflict between the actuators. The fixed-price remote diagnosis tells you whether predictive climate control with the weather forecast would cut energy and keep the crop conditions.

Climate controllers of greenhouse makers are simple

What happens. A greenhouse maker supplies a climate computer with setpoints for temperature and humidity and simple rules for the vents and screens. It works, but it was not designed to look ahead.

What it costs. The grower pays the difference as energy: heating when the sun is about to warm the house anyway, or venting after the cold has already come in.

Why the usual fix fails. Adding more rules makes the computer harder to understand and tune, and each customer wants a different set. A sophisticated controller built into one product does not travel to the next.

How a fixed-price remote diagnosis checks it. I take logs of a typical installation and check what a predictive controller on top of the climate computer would do with the same sensors and actuators. The fixed-price remote diagnosis tells you what gain to expect for your customers, and whether the controller can run on your existing hardware.

Control algorithms for horticulture automation are hard to staff

What happens. Companies that build automation for greenhouses and indoor farms need control engineers for climate, irrigation and lighting. The skills are scarce, and horticulture is not where most control graduates look.

What it costs. Product development slows, and the controller that ships is the simple version. Growers get less than the equipment could deliver.

Why the usual fix fails. A software team can build the interface and the data logging, but the control logic needs knowledge of dynamic models and optimisation. Hiring on the job market takes longer than the product cycle.

How a fixed-price remote diagnosis checks it. I review one of your controllers with its logs, as an outside check. The fixed-price remote diagnosis tells you where a predictive climate controller would help, how much of it your team could build on its own afterwards, and what to start with.

Motion control and robots for greenhouses are hard to staff

What happens. Greenhouse technology firms build moving equipment: harvesting robots, trolleys on rails, gantries, spray booms. These machines work in a damp, dusty environment with plants that are never the same twice.

What it costs. The development teams need motion-control and robot control skills that are rare, and roles stay open. Prototypes work in the lab but stumble in the crop.

Why the usual fix fails. Adding sensors fixes some failures but adds cost to every machine. Tuning each machine on site is not a business model.

How a fixed-price remote diagnosis checks it. I review a machine’s motion data and controller from a trial run. The fixed-price remote diagnosis tells you where a better motion planner or controller would help, and what the next test in a real crop should measure.