arXiv subjects

Graham Jones

Publications and source records attributed to Graham Jones.

4 recordsLinked to original sources

arXiv2026

Real-time virtual circuits for plasma shape control via neural network emulators: integration and testing in the MAST-U PCS

The deployment of advanced, AI-enabled control algorithms in tokamak experiments requires robust integration with existing plasma control system (PCS) architectures and extensive pre-experimental validation. In this contribution, we describe the integration and testing of neural-network-emulated virtual circuits for plasma shape control within the MAST Upgrade (MAST-U) PCS environment. The neural network models predict the plasma shape using the plasma current, poloidal field coil currents, and plasma profile parameters. In this paper, we explain how they are deployed via a real-time C++ inference server that interfaces with the PCS, returning the shape prediction and its Jacobian, and how, from the latter, virtual circuit matrices and updated coil current requests are computed for real-time actuation. Emphasis is placed on the validation workflow and best practices adopted to ensure confidence in the proposed control framework prior to experimental deployment. This work demonstrates practical AI-based shape control components for fusion control systems, with direct relevance for upcoming MAST-U experiments and future devices.

physics.plasm-ph↗

arXiv2026

The Frequency of Solar Eclipses for a Given Place: A New Approach to a Classic Question

In a classic 1982 paper in this journal, Jean Meeus used a statistical approach for finding the mean frequency of a total and an annular eclipse of the Sun at a given place on the surface of the Earth. In this current paper we tackle the problem again, taking advantage of the much greater computing power now available. We obtain narrower estimates of once every 373 $\pm$ 7 years for a total eclipse, and once every 226 $\pm$ 4 years for an annular eclipse. In addition, we obtain a result of once every 2.59 $\pm$ 0.02 years for a partial eclipse. We provide further insights into the "latitude effect", showing that solar eclipses occur most frequently around the Arctic and Antarctic Circles. We also show how the gradual shift of aphelion and perihelion with respect to the seasons produces a $\sim$21,000-year cycle in the frequency of eclipses in the Northern and Southern Hemispheres.

astro-ph.SR↗

arXiv2012

Time to a single hybridization event in a group of species with unknown ancestral history

We consider a stochastic process for the generation of species which combines a Yule process with a simple model for hybridization between pairs of co-existent species. We assume that the origin of the process, when there was one species, occurred at an unknown time in the past, and we condition the process on producing n species via the Yule process and a single hybridization event. We prove results about the distribution of the time of the hybridization event. In particular we calculate a formula for all moments, and show that under various conditions, the distribution tends to an exponential with rate twice that of the birth rate for the Yule process.

q-bio.PE↗

arXiv2012

Statistical Inference of Allopolyploid Species Networks in the Presence of Incomplete Lineage Sorting

Polyploidy is an important speciation mechanism, particularly in land plants. Allopolyploid species are formed after hybridization between otherwise intersterile parental species. Recent theoretical progress has led to successful implementation of species tree models that take population genetic parameters into account. However, these models have not included allopolyploid hybridization and the special problems imposed when species trees of allopolyploids are inferred. Here, two new models for the statistical inference of the evolutionary history of allopolyploids are evaluated using simulations and demonstrated on two empirical data sets. It is assumed that there has been a single hybridization event between two diploid species resulting in a genomic allotetraploid. The evolutionary history can be represented as a network or as a multiply labeled tree, in which some pairs of tips are labeled with the same species. In one of the models (AlloppMUL), the multiply labeled tree is inferred directly. This is the simplest model and the most widely applicable, since fewer assumptions are made. The second model (AlloppNET) incorporates the hybridization event explicitly which means that fewer parameters need to be estimated. Both models are implemented in the BEAST framework. Simulations show that both models are useful and that AlloppNET is more accurate if the assumptions it is based on are valid. The models are demonstrated on previously analyzed data from the genus Pachycladon (Brassicaceae) and from the genus Silene (Caryophyllaceae).

q-bio.PE↗