The article touches upon the possibility of using the method of gradientless n-parametric minimization of Powell'64 in tasks of monitoring the working process of marine diesel engines. There is given an example of finding a global minimum of the Rosenbrock test function. Using the Powell'64 method, the Least-squares functionals in the synchronization and modelling tasks of compression-expansion curves in the working cylinder are minimized. The calculation results of data synchronization for low-speed two-stroke and medium-speed four-stroke marine diesel engines are shown. The synchronization problem can be solved in terms of equation P' = 0 derived for the sector from compression starting to combustion starting in the cylinder. The selection of the boundary conditions for simulation is shown. The advantage of Powell’64 method is its high efficiency for quadratic functionals. As opposed to gradient methods, the Powell'64 method does not require calculating derivatives and is universal for minimizing complex nonlinear general functionals. The original author's algorithm of data synchronization by analyzing the indicator diagrams using the Powell'64 method has been applied in the latest versions of monitoring systems of D4.0H marine diesel engine.
Рассмотрена возможность применения метода безградиентной n-параметрической мини-мизации Powell'64 в задачах мониторинга рабочего процесса судовых дизелей. Приведен пример поиска глобального минимума тестовой функции Розенброка. С помощью метода Powell'64 минимизируются функционалы МНК в задачах синхронизации и моделирования кривых сжатия/расширения в рабочем цилиндре. Показаны случаи расчета синхронизации данных для малооборотных двухтактных и среднеоборотных четырехтактных судовых дизе-лей. Задача синхронизации решается на базе уравнения P' = 0, составленного для участка от начала сжатия до начала сгорания в цилиндре. Показан выбор граничных условий модели-рования. Преимуществом использования метода Powell'64 является его высокая эффектив-ность для квадратичных функционалов. В отличие от градиентных методов, метод Powell'64 не требует расчета производных и является универсальным для минимизации сложных нелинейных функционалов общего вида. Оригинальный авторский алгоритм синхронизации данных путем анализа индикаторных диаграмм, в котором используется метод Powell'64, применен в последних версиях систем мониторинга судовых дизелей D4.0H.Ключевые слова: безградиентная минимизация, функция Розенброка, рабочий процесс ди-зеля, политропа сжатия, функционал.
The study of influence of the method of locking the needle on the starting qualities of the engine and the consumption of the starting air, as well as on the indicators of dynamic tension, was carried out. The widely used diesel engine 4Ч 17.5/24 having increased turnover and jet mixing was taken as an example. There have been discussed the peculiarities of starting a "hot" and "cold" engine equipped with hydraulic locking fuel injection valves, when the pressure of the hydraulic locking fluid is equal to atmospheric pressure. The influence of the method of locking the needle of fuel injection valve on the efficiency of the engine has been analyzed under permanent parameters of atmosphere, cooling water temperature and circulating oil. The influence of the pressure of the hydraulic locking fluid on the economy of the diesel engine 4Ч 17.5/24 has been examined. The effect of the change in the effective cross section of sprayers within the limits encountered in the operation of diesel engines on the economy and the characteristics of the working process has been studied. It has been inferred that pressure of hydraulic locking fluid at which minimum fuel consumption is provided may change when sprayers are replaced. It should be noted that for each type of engines it is necessary to check the constancy of the pressure value of the hydraulic locking fluid, which ensures minimum fuel consumption when the effective cross section of the sprayer varies within ±20%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.