Jan Corfixen Sørensen

An agent-based extensible climate control system for sustainable greenhouse production

The slow adoption pace of new control strategies for sustainable greenhouse climate control by industrial growers, is mainly due to the complexity of identifying and resolving potentially conflicting climate control requirements. In this paper, we present a multi-agent-based climate control system that allows new control strategies to be adopted without any need to identify or resolve conflicts beforehand. This is achieved by representing the climate control requirements as separate agents. Identifying and solving conflicts then becomes a negotiation problem among agents sharing the same controlled environment. Negotiation is done using a novel multi-objective negotiation protocol that uses a generic algorithm to find an optimized solution within the search space. The multi-agent-based control system has been empirically evaluated in an ornamental floriculture research facility in Denmark. The evaluation showed that it is realistic to implement the climate control requirements as individual agents, thereby opening greenhouse climate control systems for integration of independently produced control strategies.

Enhancing State-of-the-Art Multi-Objective Optimization Algorithms by Applying Domain Specific Operators

To solve dynamic multi-optimization problems, optimization algorithms are required to converge quickly in response to changes in the environment without reducing the diversity of the found solutions. Most Multi-Objective Evolutionary Algorithms (MOEAs) are designed to solve static multi-objective optimization problems where the environment does not change dynamically. For that reason, the requirement for convergence in static optimization problems is not as time-critical as for dynamic optimization problems. Most MOEAs use generic variables and operators that scale to static multi-objective optimization problems. Problems emerge when the algorithms can not converge fast enough, due to scalability issues introduced by using too generic operators. This paper presents an evolutionary algorithm CONTROLEUM-GA that uses domain specific variables and operators to solve a real dynamic greenhouse climate control problem. The domain specific operators only encode existing knowledge about the environment. A comprehensive comparative study is provided to evaluate the results of applying the CONTROLEUM-GA compared to NSGAII, ϵ-NSGAII and ϵ-MOEA. Experimental results demonstrate clear improvements in convergence time without compromising the quality of the found solutions compared to other state-of-art algorithms.

DynaGrow – Multi-Objective Optimization for Energy Cost-efficient Control of Supplemental Light in Greenhouses

The Danish greenhouse horticulture industry utilized 0.8 % of the total national electricity consumption in 2009 and it is estimated that 75 % of this is used for supplemental lighting. The increase in energy prices is a challenge for growers, and need to be addressed by utilizing supplemental light at low prices without compromising the growth and quality of the crop. Optimization of such multiple conflicting objectives requires advanced strategies that are currently not supported in existing greenhouse climate control systems. It is costly to incorporate advanced optimization functionality into existing systems as the software is not designed for such changes. The growers can not afford to buy new systems or new hardware to address the changing objectives. DynaGrow is build on top of existing climate computers to utilize existing infrastructure. The greenhouse climate control problem is characterized by non-linearity , stochasticity, non-convexity, high dimension of decision variables and an uncertain dynamic environment. Together, these mathematical properties are handled by applying a Multi-Objective Evolutionary Algorithm (MOEA) for discovering and exploiting critical trade-offs when optimizing the greenhouse climate. To formulate advanced objectives, DynaGrow integrates local climate data, electricity energy price forecasts and outdoor weather forecasts. In spring 2015, one greenhouse experiment was executed to evaluate the effects of DynaGrow. The experiment was run as three treatments in three identical greenhouse compartments. One treatment was controlled by a standard control system and the other three treatments were controlled by different DynaGrow configurations. A number of different plant species and batches were grown in the three treatments over a season. The results from DynaGrow treatment demonstrated that it was clearly possible to produce a number of different sales-ready plant species and at the same time optimize the utility of supplemental light at low electricity prices without compromising product quality.

Comparative study of evolutionary multi-objective optimization algorithms for a non-linear Greenhouse climate control problem

Non-trivial real world decision-making processes usually involve multiple parties having potentially conflicting interests over a set of issues. State-of-the-art multi-objective evolutionary algorithms (MOEA) are well known to solve this class of complex real-world problems. In this paper, we compare the performance of state-of-the-art multi-objective evolutionary algorithms to solve a non-linear multi-objective multi-issue optimization problem found in Greenhouse climate control [1]. The chosen algorithms in the study includes NSGAII, ε-NSGAII, ε-MOEA, PAES, PESAII and SPEAII. The performance of all aforementioned algorithms is assessed and compared using performance indicators to evaluate proximity, diversity and consistency. Our insights to this comparative study enhanced our understanding of MOEAs performance in order to solve a non-linear complex climate control problem. The empirical findings of this comparative study show that based on the performance indicators, three algorithms, ε-MOEA, ε-NSGAII and NSGAII outperform the other algorithms and provide high quality solution sets in an appropriate time.

DynaGrow: Next Generation Software for Multi-Objective and Energy Cost-Efficient Control of Supplemental Light in Greenhouses

It is not possible for growers to compromise product quality by saving energy but the increasing electricity prices challenge the growers economically. Optimization of such multiple conflicting goals requires advanced strategies that are currently not supported in existing greenhouse climate control systems. DynaGrow is built on top of the existing climate control computers and utilizes the existing hardware. By integrating with exiting hardware it is possibly to support advanced multi-objective optimization of climate parameters without investing in new hardware. Furthermore, DynaGrow integrates with local climate data, electricity price forecasts and outdoor weather forecasts, in order to formulate advanced control objectives. In September 2014 and February 2015 two greenhouse experiments were run to evaluate the effects of DynaGrow. By applying multi-objective optimization, it was possible to produce a number of different cultivars and save energy without compromising quality. The best energy savings were achieved in the February 2015 experiment where the contribution from natural light was limited.

An extensible component-based multi-objective evolutionary algorithm framework

The ability to easily modify the problem definition is currently missing in Multi-Objective Evolutionary Algorithms (MOEA). Existing MOEA frameworks do not support dynamic addition and extension of the problem formulation. The existing frameworks require a re-specification of the problem definition and recompilation of source code implementing the problem specification. The presented, Controleum framework is based on Dynamic Links and a component-based system to support dynamic reconfiguration of the problem formulation without any need for recompilation of source code. Four different experiments with different compositions of objectives from the horticulture domain are formulated based on a state of the art micro-climate simulator, electricity prices and weather forecasts. The experimental results demonstrate that the Controleum framework support dynamic reconfiguration of the problem formulation without compromising the composed objectives.

Satisficing-Based Approach to Resolve Feature Interactions in Control Systems

To handle the complexity of modern control systems there is an urgent need to develop features as independently developed units of extension. However, when independently developed features are later composed they become coupled through the shared environment resources. As a consequence, the system requirements may no longer be entailed when independent features try to control the same shared environment. Malfunctioning behavior as a consequence of feature interference is know in the literature as the feature interaction problem. This paper present an approach that uses designtime specification of independent requirements, in combination with a runtime arbitrator that search for feature interaction free programs which entail the system requirements. In case of conflicting requirements that can’t be satisfied simultaneously, the mechanism supports explanation of the interactions as a context sharing problem. We demonstrate our approach in a real-life control system for industrial pot plant cultivation in greenhouses and show that solutions are found for compatible requirements and that conflicts are identified and explained for incompatible requirements.

Using Agent Satisfiability to Identify and Explain Interactions among Independent Greenhouse Climate Control Requirements

The slow adoption pace of new control strategies for sustainable greenhouse climate control by industrial growers, is mainly due to the complexity of identifying and explaining potentially conflicts when integrating independently climate control requirements. In this paper, we show how the satisfiability of agents, implementing independent climate control requirements, can be used to identify and explain conflicting control interactions, which emerge because the agents share the same resources in the controlled environment. Potential conflicts due to unfulfilled climate control requirements correspond to low agent satisfiability. Low satisfiability indicates that an agent’s goal is conflicting with the proposed settings of the greenhouse climate. This allows us to explain to which degree independent climate control requirements are fulfilled by visualizing the satisfiability of the corresponding agents. We have evaluated our approach using real climate control data. The evaluation showed that it is possible to identify and explain conflicts among agents sharing the same controlled environment.