Improving the uncertainty of measured and calculated data in engine problems using the reconciliation algorithm
In experimental testing of internal combustion engines, there are often situations where the number of calculated quantities of unknowns is less than the number of independent equations linking these quantities. Such cases are often observed when carrying out measurements to determine unknown quantities that are difficult or impossible to measure directly. The calculation algorithm then uses equations in which there are measured quantities and these calculated unknowns. The unknowns, determined directly from these equations, can be calculated in a number of ways, depending on the choice of equation set. The remaining equations will not be satisfied exactly due to inevitable measurement errors, and as a consequence the same calculated quantities will take different values. In order to avoid these differences (obtain a single parameter value) and to get consistency of all equations, it is necessary to carry out the equation reconciliation procedure using the methods of equalization calculus. The essence of this calculus is the correction of measurement results after which the calculation results of the unknown quantity, determined from different sets of equations, will be the same. The paper presents a selected example of the application of equalization calculus concerning the problem of selecting the parameters of the theoretical Seiliger-Sabathe cycle according to the experimentally determined real engine cycle (the so-called inverse problem). The reconciliation covered the engine load parameters and the heat distribution number. The calculations carried out and the results obtained confirmed the need to use equalizing calculus to reduce deviations of the calculation results.
Tematyka artykułu: Badania silników i modelowanie procesów zachodzących w silnikach
Autor: Zbigniew Żmudka
Współautor(zy):