Theme of this thesis is practical realization of so-called Island model which is one of way of parallel genetic programming. First part is theoretical. This part is describing terms of genetic programming, age-layered population structure and island model. In second part of thesis is described realization of island model in Java language.

Critical aspects of spacecraft missions, such as component organization, control algorithms, and trajectories, can be optimized using a variety of algorithms or solvers. Each solver has intrinsic strengths and weaknesses when applied to a given optimization problem. One way to mitigate limitations is to combine different solvers in an island model that allows these algorithms to share solutions. The program Spacecraft Trajectory Optimization Suite (STOpS) is an island model suite of heterogeneous and homogeneous Evolutionary Algorithms (EA) that analyze interplanetary trajectories for multiple gravity assist (MGA) missions. One limitation of STOpS and other spacecraft trajectory optimization programs (GMAT and Pygmo/Pagmo) is that they require a defined encounter body sequence to produce a constant length set of design variables. Early phase trajectory design would benefit from the ability to consider problems with an undefined encounter sequence as it would provide a set of diverse trajectories -- some of which might not have been considered during mission planning. The Hybrid Optimal Control Problem (HOCP) and the concept of hidden genes are explored with the most common EA, the Genetic Algorithm (GA), to compare how the methods perform with a Variable Size Design Space (VSDS). Test problems are altered so that the input to the cost function (the object being optimized) contains a set of continuous variables whose length depends on a corresponding set of discrete variables (e.g. the number of planet encounters determines the number of transfer time variables). Initial testing with a scalable problem (Branin's function) indicates that even though the HOCP consistently converges on an optimal solution, the expensive run time (due to algorithm collaboration) would only escalate in an island model system. The hidden gene mechanism only changes how the GA decodes variables, thus it does not impact run time and operates effectively in the island model. A Hidden Gene Genetic Algorithm ( HGGA) is tested with a simplified Mariner 10 (EVM) problem to determine the best parameter settings to use in an island model with the GTOP Cassini 1 (EVVEJS) problem. For an island model with all GAs there is improved performance when the different base algorithm settings are used. Similar to previous work, the model benefits from migration of solutions and using multiple algorithms or islands. For spacecraft trajectory optimization programs that have an unconstrained fly-by sequence, the design variable limits have the largest impact on the results. When the number of potential fly-by sequences is too large it prevents the solver from converging on an optimal solution. This work demonstrates HGGA is effective in the STOpS environment as well as with GTOP problems. Thus the hidden gene mechanism can be extended to other EAs with members containing design variables that function similarly. It is shown that the tuning of the HGGA is dependent on the specific constraints of the spacecraft trajectory problem at hand, thus there is no need to further explore the general capabilities of the algorithm.

The information presented in the thesis is a continuation of the Spacecraft Trajectory Optimization Suite (STOpS). This suite was originally designed and developed by Timothy Fitzgerald and further developed by Shane Sheehan, both graduate students at California Polytechnic State University, San Luis Obispo. Spacecraft utilizing low-thrust transfers are becoming more and more common due to their efficiency on interplanetary trajectories, and as such, finding the most optimal trajectory between two planets is something of interest. The version of STOpS presented in this thesis uses Multiple Gravity-Assist Low-Thrust (MGALT) trajectories paired with the island model paradigm to accomplish this goal. The island model utilizes four different global search algorithms: a Genetic Algorithm, Differential Evolution, Particle Swarm Optimization, and Monotonic Basin Hopping. The first three algorithms were featured in the initial version of STOpS written by Fitzgerald [1], and were subsequently modified by Sheehan [2] to work with a low-thrust adaptation of STOpS. For this work, Monotonic Basin Hopping was added to aid the suite with the MGALT trajectory search. Monotonic Basin Hopping was successfully validated against four different test functions which had been used to validate the other three algorithms. The purpose of this validation was to ensure Monotonic Basin Hopping would work as intended, ensuring it would work in cooperation with the other three algorithms to produce a near optimal solution. After verifying the addition of Monotonic Basin Hopping, all four algorithms were used in the island model paradigm to verify MGALT STOpS’ ability to solve two known orbital transfer problem. The first verification case involved an Earth to Mars transfer with fixed thruster parameters and a predetermined time of flight. The second verification case involved a transfer from Earth to Jupiter via a Mars gravity assist; two different versions of the verification case were solved against trajectories produced by industry optimization software, the Satellite Tour Design Program Low-Thrust Gravity Assist and the Gravity Assisted Low-thrust Local Optimization Program. In the first verification case, MGALT STOpS successfully validated the Earth to Mars trajectory problem and found results agreeable to literature. In the second verification case, MGALT STOpS was partially successful in validating the Earth to Mars to Jupiter trajectory problems, and found results similar to literature. The final software produced for this work is a trajectory optimization suite implemented in MATLAB, which can solve interplanetary low-thrust trajectories with or without the inclusion of gravity assists.

Evolution frequently plays out over ecological timescales. Local adaptation under the joint action of evolutionary and ecological processes frequently leads to novel outcomes, as is evidenced by the theoretical work on adaptation at species' borders. However, to date this body of work does not have a theory for the effect of stochastic processes on local adaptation. The primary goal of this thesis is to show that demographic and genetic fluctuations can significantly impact upon local adaptation. In addition, the effect of polygenic evolution is also analysed. Specifically, three types of models are considered. First a deterministic mainland-island, subject to hard directional selection, maladaptive gene flow and density regulation is solved for two different trait architectures: an explicit multilocus trait and a quantitative trait. The maladaptive and adaptive steady states can be bistable. This depends on the underlying architecture of the trait, as well as locus number and ploidy. Sourcesink structure can emerge, accompanied by a novel, upper critical threshold above which maladaptation occurs. The most favourable condition for local adaptation occurs for few loci and low migration. Second, a stochastic version of the mainland-island model is analysed as a diffusion process. This is the central premise of the thesis and is explored by examining properties of the stationary distributions of both trait architectures, and the first-passage properties of the single locus case. It is found that across a range of migration rates that depend on locus number and migrant polymorphism, local adaptation may be reversed or escape from maladaptation becomes possible at varying transition rates. The diffusion model is compared to a similar discrete model. The continuous model is in good qualitative agreement with the discrete model. Third, the stochastic model is generalised to the infinite island model, which evolves deterministically. Under deterministic dynamics a range of equilibria are possible, depending on whether habitat size varies or is fixed. Multilocus dynamics restrict the conditions for polymorphism. Stochastic dynamics can have potentially detrimental consequences for the persistence of the island population when drift is strong. The relevance of the stochastic model to border populations is discussed. Although the diffusion process imposes severe constraints on the permissible parameter ranges, it is still able to provide a good qualitative understanding of the impact demographic and genetic fluctuations have on local adaptation.

The Mountain Hemlock (MH) zone includes all subalpine forests along British Columbia’s coast. It occurs at elevations where most precipitation falls as snow and the growing season is less than 4 months long. The zone includes the continuous forest of the forested subzones and the tree islands of the parkland subzones (Figure 1). Old-growth stands are populated by mountain hemlock, Pacific silver fir, and Alaska yellow-cedar, and are among the least-disturbed ecosystems in the world. Canopy trees grow slowly and are commonly older than 600 years, while some Alaska yellow-cedars may be up to 2000 years old. Understanding regeneration patterns in the MH zone has become increasingly important as logging continues towards higher elevations of the zone where snowpacks are deeper.

No description available.