Code Components Research Articles

Generalisation and discrimination emerge from a self-organising componential network: a speech example

It is demonstrated that a componential code emerges when a self-organising neural network is exposed to continuous speech. The code's components correspond to substructures that occur relatively independently of one another: words and phones. A capability for generalisation and discrimination develops without having been optimised explicitly. The componential structure is revealed by optimising a necessarily complicated nonlinear moment of the data's distribution, equal to the mean-squared output response of a multi-layered network of simple threshold neurons. Earlier analytical work had predicted that componential codes, generalisation and discrimination should emerge from the self-organisation of threshold neurons of this form, assuming certain properties of the pattern-space distribution of the data.

A computational, modular approach to the simulation of powerplants

The paper describes the development, the implementation and the practical application of a new modular procedure for the numerical simulation of thermal powerplants. The procedure has resulted in a FORTRAN code, named “CAMEL” (Italian acronym for “Modular Elemental Plant Calculation), which is also described in detail. The plant -any plant, but application of the procedure is presently limited to thermal powerplant- is described in terms of three matrices, named the Interconnection matrix IM, the plant matrix PM and the Topographic matrix TM, which contain in an orderly fashion all of the design data and the6“topological” description of the plant configuration. The Interconnection matrix is equivalent to a structured list of the flows (of matter and energy) which are fed to or extracted from any of the “n” plant components; the plant matrix is a table of the design values of all the fluxes in the plant, and the Topographic matrix is a reference table which allows the logical location of any component to be found both in PM and in IM. Starting from a set of design input data, CAMEL calculates, component bby component, all the values of the thermodynamic variables of interest which do not already appear in PM; more precisely, CAMEL can calculate the numerical value of all the parameters needed to uniquely determine the thermodynamic and/or energetic state of whatsoever flux in the plant by calling for each component the corresponding routine, which returns the numerical value of all the variables it can calculate at the “present” stage of the simulation. It is important to underscore that these routines can calculate a component not as whole machine but equation by equation, i.e., the routine must not necessarily calculate all of the operating parameters of the component at the same stage of the simulation, but only those which can be calculated with the data known at the current stage of the simulation and that are involved in the same physical event described by one of the equations included in the set which pertains to that component. CAMEL can perform, at the user's request, either a single plant simulation or a sensitivity analysis of the behaviour of the plant. At present, CAMEL is only capable of performing sensitivity analysis of plants at design conditions and in a limited number of (steady state) off-design conditions (i.e., the analysis is aimed at the optimization of the design operating conditions of the plant): in principle it is possible to implement a transient computation, but this has not yet been done. CAMEL has been implemented for powerplant simulation, but the core has been structured to allow the code to handle with any kind of processor-plant, provided the code's component library is large enough. The paper reports the philosophy of the approach, describes the details of the algorithm, discusses the flow chart of the numerical code and presents detailed results for an industrial case-study: a combined-cycle power station for electrical generation.

The Interface Between Beliefs and Behavior

This paper examines the interface between Henry Kissinger's operational code and his bargaining behavior during the Vietnam conflict. Kissinger's position at the pinnacle of the American foreign affairs hierarchy, amidst the existence of competing policy recommendations, may be regarded as necessary conditions to test the impact of his operational code upon American foreign policy. A comparison of his academic writings and his conduct of the Vietnam negotiations reveals a congruent relationship between his operational code and his bargaining behavior. Taken collectively, the code's components approximate game theory's “prisoner's dilemma” description of politics and specify a “general preference relation” that prescribes a consistent, predictable, metagame repertoire of responses.