Generalized circuits. Elements of a mathematical theory (Q796864)

The author develops a mathematical theory of generalized electrical circuits. He defines such a circuit as a system \(Ct=\{\Lambda;\quad G;\quad C(X),\quad X\in V(G)\},\) where \(\Lambda\) is an abelian group or vector space with an involutive automorphism, G is an oriented graph, and, for each vertex X of G, C(X) is a subset of \(\Lambda^ p\) (where p is the valency of X). The vertices of the graph correspond to the circuit elements, its edges to the metallic connections; the transport in a terminal is given by an element of \(\Lambda\) ; for each X, C(X) is the set of all its possible states. A large part of the paper deals with the structure of and the main operations with these generalized circuits; in particular, for \(\Lambda ={\mathbb{R}}^ m\), those circuits are discussed, for which the sets C(X) are convex subsets or linear subspaces. Then an important class of generalized circuits, called N-circuits, are introduced; these admit a partition \(\{ {\mathcal M},{\mathcal N}\}\) of the set of vertices, such that all edges of the graph have the source in \({\mathcal N}\) and the aim in \({\mathcal M}\). For these circuits, the author constructs a lattice of subgraphs and a sheaf of states over it; in order to describe the set of states, he applies some methods of algebraic topology, including Leray's sheaf cohomology theory and his own algebraic theory of stacks [C. R. Acad. Sci., Paris, Sér. A 281, 613-616, 691-694 (1975; Zbl 0321.55015, Zbl 0321.55016); Ann. Sc. Norm. Super. Pisa, Cl. Sci., IV. Ser. 3, 37-46 (1976; Zbl 0322.06015); Stacks and Cohomology, Dept. Math., Univ. California, Berkeley, CA (May, 1976)]. Unlike in the classical theory, here the problem arises to determine the lower dimensional cohomology groups, provided the higher dimensional were known. To this aim, the construction of lower dimensional cochains from a given cochain is approached and some obstructions to this construction are described. The author shows that his theory, applied to the special case of the classical electrical circuits, provides all the known facts about them. However, the interest of the general theory consists in its possible applications to the more complicated biological systems, such as the central nervous system. Different possible approaches as well as some objections to the applicability of this theory are discussed.