Robert Kok

Use of artificial neural networks in transient drainage design

The use of Artificial Neural Networks (ANN) was investigated as an alternative method to obtain solutions to the Boussinesq equation. First, 52,280 solutions were obtained with a numerical method, each one corresponding to a unique set of parameter values (the inputs). Each solution consisted of two water table level values (the training outputs) corresponding to 12- and 24-h delays. The input/output combinations were then divided into learning and recall sets. Next, 15 fully connected, feed-forward ANNs, having slightly different architectures, were each trained with the learning set for 50,000 cycles. Their performance was then evaluated with the recall set (i.e., the recall outputs obtained from these inputs were compared to the corresponding training outputs) and the most appropriate ANN selected for further training and testing. This “best” ANN was subjected to 21 different training regimes so as to determine the most appropriate one, the regimes consisting of various quantities of training with the learning set. Performance of the ANN at the various training levels was then again evaluated and the optimal regime determined. Finally, the “optimally trained” ANN was transformed into a C program and compiled and linked into a DOS executable file which was then used to calculate solutions to the Boussinesq equation for all 52,280 cases.

A generic, individual-based approach to modelling higher trophic levels in simulation of terrestrial ecosystems

In this article, a description is given of the manner in which higher trophic levels (animals) are represented in a generally configurable ecosystem model. The animals are modelled using an individual-based approach that is sufficiently generic to allow for the representation of organisms of different types of species via the specification of appropriate sets of parameter values. Animal behaviours and physiological functions are described with simple mechanistic rules that are derived from various assumptions about, for example, growth rates, metabolic requirements, digestion and assimilation of food, or gestation. The animals interact in a detailed, spatially explicit environment that consists of a terrain, an atmosphere, and various species of primary producers. The model has been implemented in simulation to explore population dynamics in multi-species ecosystems configured with two and three trophic levels. Sample simulation results are presented, together with a discussion of the effectiveness of the approach for the representation of animals in ecosystem modelling.

Carbon neutral electricity production by Synechocystis sp. PCC6803 in a microbial fuel cell

The aim of this work was to illustrate the use of photosynthetic microbes in a microbial fuel cell to produce electricity without the requirement of an external carbon source. This research here describes the use of a cyanobacterium Synechocystis PCC6803, to produce electricity without any net CO(2) production in a two-chambered MFC. Conditions for optimum electricity production were determined through standardizing operating parameters. A maximum power density of 6.7mWm(-3)(anode chamber volume) was achieved under high intensity lighting (10,000lux). Light intensity and wavelength directly affected electricity production, indicating the pivotal role played by photosynthesis. The maximum removal of CO(2) was 625mmolm(-3) over 20h under high intensity light. The results presented here will contribute to the understanding of how cyanobacteria can be exploited for the direct conversion of CO(2) to electric current.

The Production of Insects for Human Food

Abstract In this paper are discussed some aspects of a promising new sector of agriculture: the use of insects to convert low quality plant materials such as cellulose to structured and highly palatable animal flesh. The paper was written with the needs of spaceship crews in mind. An insect farm, based on the use of packed beds for organism incubation, growth and reproduction, is discussed and system design parameters are determined. The drugstore beetle ( Stegobiumpaniceum ) is investigated as an initially attractive organism. Several system configurations (batch, semi-continuous and continuous) are explored and the advantages and disadvantages of a variety of growth reactors (batch, plug flow and backmix) are examined. The design of a reproduction facility and the problems attendant to the maintenance of population synchrony are also considered. Calculations are presented for a facility to supply the animal protein for 100 people. From the results it is evident that the concept is feasible.

A generic primary producer model for use in ecosystem simulation

Abstract This article is a description of a primary producer (plant) model that has been developed as part of a larger ecosystem modelling project in the field of biosystems engineering. Plants in the ecosystem are represented with a generic model that can be configured to represent many different types of species, including herbs, bushes and trees. Each plant, or small collection of plants, is represented by a unique “object” with attributes and functions. Population level dynamics are therefore governed by and emerge from object-level rules that describe processes such as growth and reproduction. The plant objects are distributed over a spatially explicit terrain, and their behaviour is driven by irregular climatic forcing functions. The plant canopy is modelled with three layers of vertical differentiation, which affects inter-species competition for available radiant energy. In addition to competing with other plants for essential nutrients, sunlight and water, the plant objects may also be subject to predation by herbivores. The performance of the plant model is demonstrated with simulations based on ecosystems configured with small collections of species distributed about a 500×500 m2 terrain. Results show this modelling approach to be effective in depicting phenomena at the individual, population, and ecosystem-levels, including effects of herbivory and inter-species competition on plant growth, patterns of plant population dynamics and spatial distribution, and total system biomass.

Use of an Object-Based Model to Represent Complex Features of Ecosystems

In this article, an ecosystem model that has been developed as an engineering tool is briefly described. Sample results from two simulations are then presented, and the model is examined with regards to its usefulness and applicability vis e vis the representation of complex features of ecosystems. The model is in many ways unique: First, its scope (i.e., the number of different types of ecosystem components that are included) is much broader than that of most other ecosystem models, and key processes are represented at relatively high spatial and temporal resolutions (10 metres and 10 minutes, respectively). Second, it is entirely object-based: every abiotic and biotic component in the system is represented as a distinct entity. Thus, each organism, or small group of organisms, is treated as an individual object that lives in a spatially explicit environment composed of cells arranged in a 2-D lattice. Third, the model is completely configurable, so that a wide range of ecosystem configurations and their corresponding initial conditions can be specified for simulation. Thus, both the biological composition (i.e., number and type of species, initial population sizes, etc.) and the environment (i.e., terrain and atmosphere) of an ecosystem can be specified. When implemented in simulation, configurations based on simple food webs exhibit sustained material cycling, non-random spatial variation and distribution of organisms over the terrain, and persistent, multi-trophic level population dynamics. It is argued that these phenomena are emergent, and are indicative of spatial and temporal self-organisation in the modelled ecosystems.

An analytical framework for the design of autonomous, enclosed agro-ecosystems

Abstract It is possible to make agricultural production units autonomous by giving them sufficiently sophisticated control systems. In this regard we are particularly interested in creating and enclosed unit in which a number of species co-exist and are produced simultaneously; we refer to this as an enclosed agro-ecosystem. Our intent is to create an enclosed agro-ecosystem which is autonomous having, independent decision-making power and authority over its internal workings as well as its interactions with the external world. Moreover, we want it to behave sagaciously, being both cognitive and conscious. We ascribe such functionality to the capacity of an entitys control system and the intent of this paper is to establish a framework and terminology to allow us to analyze the full range of control activities displayed by biological machines so that we might mimic their operation and add similar functionality to agro-ecosystems. Initially, we have done this with reference to a production greenhouse which might be made partially or fully autonomous with this approach. To establish a basis for the design of an appropriate control system a model of the enclosed agro-ecosystem is first elaborated. This is followed by a general description of a control system structure which could accomodate most types of commonly observable biological control activities; the terminology and the framework are developed within the context of that description.

The development of an in situ fermentation electrode calibrator

Abstract To allow for replacement and in situ cleaning and calibration of sensors in a fermenter, a probe calibrator was conceived, designed, developed and tested. The calibrator built functioned with a dissolved oxygen electrode but could, if somewhat modified, operate with many other types of sensors. The calibrator is operated by means of an accompanying control unit and is designed to be mounted on a fermenter headplate or through a sidewall penetration. Calibrator performance was tested during a series of fermentations.

Daily Average Temperatures: Modeling and Generation with a Fourier Transform Approach

A mathematical model has been developed to generate daily average temperatures (DATs) on a yearly basis for any climate. The model contains a complete description of the patterns and variations in the DATs for a given location using a set of 29 descriptor values, the magnitudes of which are determined using a Fourier frequency analysis of physical weather data. The model was tuned to imitate the climates at three Canadian sites — Montreal, Winnipeg, and Vancouver. Daily average temperatures were then synthesized for all three locations and these were compared with the physical weather data from which the descriptor values had been derived. A number of evaluation methods were used to assess the effectiveness of the model in reproducing realistic DAT values. First, the model’s ability to imitate patterns in the physical data was judged by comparing the distributions and statistics of the DATs at three granularities daily, weekly, and monthly. Second, variability in the data sets was compared by considering the pattern in the differences between consecutive daily average temperatures. Thirdly, a number of characterization measures, such as the heat units for a given crop (e.g., degree-days for corn), were calculated. Lastly, a set of DAT values from Halifax was used to tune the model, and its ability to predict temperature patterns for that climate was assessed.