SU739558A1

SU739558A1 - Functional converter with piecewise-nonlinear approximation

Info

Publication number: SU739558A1
Application number: SU772490179A
Authority: SU
Inventors: Александр Михайлович Косолапов; Владимир Семенович Баскаков
Original assignee: Куйбышевский политехнический институт им. В.В.Куйбышева
Priority date: 1977-05-25
Filing date: 1977-05-25
Publication date: 1980-06-05

Description

- 1 Изобретение относитс к вычислительной технике и автоматике и может быть использовано в аналоговых, цифро-анадотовых , цифровых машинах и в автоматики при воспроизведении сложных функциональных зависимостей. Известен функциональный преобразователь с .кусочно-нелинейной аппроксимацией содержащий анализаторы сигналов, первый из которых соединен со входом преобразовател , блоки формировани опорных сигналов, квадратичный преобразователь, блок управлени и сумматорУ |jl. Недостатком такого преобразователи вл етс низка точность работы и ограниченный диапазон входного сигнайа. Наиболее близким к предложенному вл етс функциональный преобразователь с кусочно-нелинейной аппроксимацией, содержащий сумматоры, первый вход пер- вого из которых соединён со входом преобразовател , второй вход - с первым вы ходом первого анализатора сигналов, а выход св зан через первый масштабный блок с первым входом второго сумматора , входы которого со второго по шестой св заны с выходами соответственно вто- « рого, третьего, четвертого, днтого и шее-, того масштабных блоков, седьмой вход второго сумматора соединен со вторым выходом первого анализатора сигн)алов, подключенного третьим выходом к управгл ющим входам первого, второго, треть-, его, четвертого, п того и шестого масштабных блоков, которые, кроме первого масштабного блока, другими входами- св 4 заны с выходами соответственно первого, второго, третьего, четвертого и п того запоминающих блоков, выход пербого сумматора соединен с сигнальными входами первого и второго. ключей, причем выход первого ключа через квадратор соединен с сигнальньш входом третьего ключа, первый выход которого св зан с первым входом первого запоминающего блока, выход второго ключа через кубический преобразователь св зан с сигнальным выходом четвертого ключа, первый - 1 The invention relates to computing and automation and can be used in analog, digital-analog, digital machines and in automatics when playing complex functional dependencies. A functional converter with a piecewise non-linear approximation is known, which contains signal analyzers, the first of which is connected to the converter input, the reference signal generation units, the quadratic converter, the control unit and the adder | jl. The disadvantage of such converters is low accuracy and limited input signal range. The closest to the proposed is a functional converter with piecewise nonlinear approximation, containing adders, the first input of the first of which is connected to the input of the converter, the second input - with the first output of the first signal analyzer, and the output through the first large-scale block with the first the input of the second adder, whose inputs from the second to the sixth are connected to the outputs of the second, third, fourth, day, and neck, respectively, of the scale blocks, the seventh input of the second adder is connected to the second output the first signal analyzer connected by the third output to the control inputs of the first, second, third, its fourth, fifth and sixth scale blocks, which, besides the first scale block, other inputs are connected to the first, second , the third, fourth and fifth memory blocks, the output of the perb adder is connected to the signal inputs of the first and second. keys, and the output of the first key through the quad is connected to the signal input of the third key, the first output of which is connected to the first input of the first storage unit, the output of the second key through the cubic converter is connected to the signal output of the fourth key, the first

ыход которого соединен со входом второго запоминающего блока, вторые входь первого и второго ключей соединены с первым входом седьмого масштабного блока и выходом третьего сумматора, первый вход которого подключен ко второму выходу первого запоминающего блока , первый выход первого анализатора сигналов соединен со вторым входом третьего сумматора, второй выход третьего ключа соединен с первым входом осьмого масштабного блока и первым входом четвертого сумматора, второй вход которого подключен ко второму выходу второго анализатора, сигналов, соединенного третьим вьрсодом с первым входом п того сумматора, второй выход ветвертого ключа св дан с первым входом дев того масштабного блока и с первым входом шестого сумматора, второй вход которого соединен с четвертым ны %одом второго анализатора сигналов, св занного п тым выходом со вторыми входами седьмого, восьмого и дев того масштабных блоков, выход дев того Мйсштабного блока подключен ко второму вхо- , йу п того сумматора, третий н четвертый входи которого соединены соответственНО с первьши выходами седьмого и восьмого масштабных блоков, подключенных . вторыми выходами соответственно к третьему и четвертому входам шестого сумматори , причем третий выход -седьмого масштабного блока соединен с третьим, выходом четвертого сумматора, подключенного выходом к выходу третьего запоминающего блока, выход п того сумматора подключен ко входу четвертого запомийающего блока, выход шестого сумматора - ко входу п того запоминающего блока, выход блока -управлени - к управл ющим входам ключей, а выход второго сумматора -к выходу преобразовател 2/.the output of which is connected to the input of the second storage unit, the second input of the first and second keys are connected to the first input of the seventh scale unit and the output of the third adder, the first input of which is connected to the second output of the first storage unit, the first output of the first signal analyzer is connected to the second input of the third adder, The second output of the third key is connected to the first input of the seventh scale unit and the first input of the fourth adder, the second input of which is connected to the second output of the second analyzer, signals connected by the third signal to the first input of the fifth adder, the second output of the wind key is connected to the first input of the ninth scale unit and to the first input of the sixth adder, the second input of which is connected to the fourth input of the second output signal analyzer with the second inputs of the seventh, eighth and ninth scale blocks, the output of the ninth Myscale block is connected to the second inlet, y of the nth adder, the third and fourth inputs of which are connected respectively to the first outputs of the seventh and seventh scale blocks connected. the second outputs, respectively, to the third and fourth inputs of the sixth totalizer, the third output of the seventh scale unit is connected to the third, the output of the fourth adder connected to the output of the third storage unit, the output of the fifth adder is connected to the input of the fourth storing unit, the output of the sixth adder is co the input of the fifth storage unit, the output of the control unit to the control inputs of the keys, and the output of the second adder to the output of the converter 2 /.

Недостатком такого преобразовател вл етс низка точность воспроизведени функций, ТИК к.ак в нем не полностью . используетс динамический диапазон эле- ментов схемы. . .The disadvantage of such a converter is the low accuracy of reproduction of the functions, the TEC is not completely complete. The dynamic range of the circuit elements is used. . .

Цель изобретени - повышение точности воспроизведени заданной зависимости за счет полного использовани динамического диапазона элементов функциональной схемы на всех участках аппроксимации.The purpose of the invention is to improve the accuracy of reproduction of a given dependence due to the full use of the dynamic range of the elements of the functional diagram in all parts of the approximation.

Поставленна цель достигаетс тем, что в преобразователе третий вход первого сумматора соединен с третьим выхо- ao первого анализатора сигналов, а выход третьего сумматорп соединен со входом второго анализатора сигналов.The goal is achieved by the fact that in the converter the third input of the first adder is connected to the third output of the first signal analyzer, and the output of the third adder is connected to the input of the second signal analyzer.

На чертеже представлена схема преоб разовател .The drawing shows a schematic of the converter.

Функциональный преобразователь с.кусочно-нелинейной аппроксимацией содержит сумматор 1 с переменным коэффициентом передачи, соединенный со входом преобразовател 2, мас1утабные блоки 30 8, пороговые элементы 9-12, блоки i3i 14 формировани опорных сигналов и уп. равлени масштабными блоками, ключи 15 18, сумматоры 19-23, масштабные блоки 24-26, запоминающие блоки .27-31, , 5 квадратор 32, кубический преобразователь 33 и блок 34 управлени . Цифрой 35 обозначен выход устройства, 36 и 37 - соответственно первый и второй анализаторы сигналов.The functional transducer of a non-linear non-linear approximation contains an adder 1 with a variable transmission coefficient connected to the input of the transducer 2, 30–8 maslint block units, threshold elements 9–12, reference signal generating units i3i 14. Scale blocks, keys 15-18, adders 19-23, scale blocks 24-26, storage blocks .27-31,, 5 quad 32, cubic converter 33 and control block 34. The number 35 denotes the device output, 36 and 37, respectively, the first and second signal analyzers.

Преобразователь работает следующим образом.The Converter operates as follows.

Входной сигнал с входа преобразовател подаетс на вход сумматора 1 с переменным коэффициентом передачи, работающего в режиме вычитани . Порог срабатывани первого порогового элемента 9 настроен на верхнее (максимальное) значение линейной зоны выходного сигнала сумматора 1. Порог срабатывани второго порогового элемента 1О настроен на нижнее , (минимальное) значение линейной зоны выходного сигнала сумматора 1. Пороговые элементы 9 и 1О управл ют работой первого блока 13 формировани , который формирует сигналы дл управлени коэффициентом передачи сумматора 1 таким, образом, что при любом i 1, 2, 3, . , . выполн етс условиеThe input signal from the converter input is fed to the input of the adder 1 with a variable transmission coefficient operating in the subtraction mode. The trigger threshold of the first threshold element 9 is set to the upper (maximum) value of the linear zone of the output signal of the adder 1. The trigger threshold of the second threshold element 1O is set to the lower (minimum) value of the linear zone of the output signal of the adder 1. Threshold elements 9 and 1O control the operation of the first a generation unit 13 which generates signals for controlling the transmission coefficient of the adder 1 in such a way that for any i 1, 2, 3,. , condition is met

0 ,. moA- TT выходной сигнал сумматора 1. Кроме0,. moA- TT output of the adder 1. Except

того, блок 13 формирует -опорный сигнал Х , сигнал К пропорциональный значению реализуемой функции в опорной точке Xgv 1 а также сиг.налы дл управлени масштабными блоками 3-8. 1 Сигнал К Ч-Р noCTyn et на вход сумма-, тора 19. Сигнал Х подаетс на вход сумматора 1, где вычитаетс из выходного . На выходе сумл атора 1 будет деист вовать сигналK -K..(g)), ГПР К -коэффициент передач; сумматора 1, соответствующий юпорной точке; , который выбираетс из сформированного выше услови .. In addition, block 13 generates an X reference signal, a K signal proportional to the value of the function being implemented at the reference point Xgv 1, as well as signals to control the scale blocks 3-8. 1 A KCh-R signal of noCTyn et to the input of the sum-, torus 19. A signal X is fed to the input of the adder 1, where it is subtracted from the output. At the output of the sum of ator 1, the signal K –K .. (g)) will act, the GPR K will have a gear ratio; adder 1, corresponding to the junction point; which is selected from the conditions formed above.

Claims

The signal X. is fed to the first 15 and third 17 keys, as well as through the first scale unit 3 of the first-degree polynomial to the input of the adder 19. At the output 573 of the time scale | b) of the first block 3, the CNP X., Kj, - 1, X WU, GDR k, is the coefficient 4 | of the fi e ment nepeaa iH of the first scale unit 3, corresponding to the reference point A, -. In the following, the value of the implemented function is calculated in two clocks. First beat The first switch 15 connects the output of the adder 1 to the input of the quadrant 32, the output of which through the second switch 16 is connected to the first memory block 27. The third switch 17 connects the output of the adder 1 to the input of the cubic converter 33, the output of which through the fourth switch 18 connects to the second to the storage unit 28, the signals at the output of the quadrtor 32 and the cubic converter 33 are formed: S2-SzNV 5x-, where K 2 and b are the transfer coefficients of the quadrant 32 and the cubic converter 33, respectively. At the outputs of the second and third scale blocks 4 and 5 signals will be formed. . Хххv where К - and К с are the transfer coefficients of the second and third scale blocks 4 and 5, respectively, corresponding to the reference point of the second cycle. The first key 15 connects the quad input 32 to the output of the adder 20 .. The output of the quad 32 via the second key 16 is connected to the inputs of the adder 21 and the eighth scale unit 25. The third key 17 connects the output of the adder 20 to the input of the cubic converter 33, the output of which through the fourth key 18 is connected to the codes of the adder 23 and the ninth scale unit 26. From the output of the first storage unit 27, the signal X2-j - 32 is fed to the input of the adder 2O, the output of which includes the third and fourth threshold elements 11 and 12. Threshold third time Horn element 11 is set to the upper (maximum) value of the linear zone of the output signal of the adder 2O. The trigger threshold of the fourth threshold element 12 is set to the lower (minimum) value of the linear zone output signal 5. adder 20. Threshold elements 11 and 12 control the operation of the second generation unit 14, which generates signals, to control the scale inputs 24i 25, 26, the signals arriving at the inputs of adders 21, 22 and 23, as well as the second reference signal, which is fed to the input of the adder 2O. At the output of the adder 20, a signal Xdo-KaoSZ- :: -) is generated. The transfer coefficient of the adder 2 O. The signal Xjo is fed through the first key 15 to the input of the quad 32, through the third key 17 to the input of the cubic converter 33, through the seventh scale unit 24 to the inputs of the adders 21-23. Thus, the output of the quad. 32, a signal x -K k is formed. Through the second key 16, this signal is sent to the inputs of the adder 21 and the eighth scale unit 25. At the output of the cubic converter 33, a signal is formed (, which is fed through the fourth key 18 to the outputs of the adder 23 and the scale unit 26, at the output of which the signal will act where K ,,, is the transfer coefficient of the ninth scale block 26, corresponding to the value of the second reference point j. By similar considerations, equating the product of the coefficients as a unit or functions according to U are obtained for the signal at the output of the adder 21: Xj - K 2o "for the signal at the output of the adder 22: -v, for the signal at the output of the adder 2b- (2t 2o. The signals X, 2 and 23 are memorized respectively in the third, fourth, fifth storage blocks 29-31 and a pair of corresponding scale blocks 6-8 are fed to the inputs of the adder 19. As a result, at the output of the converter 35, we get b5 bx, x i 4xi (-x) + -; xxxx-c D ...- 1p (-, i .. rue (- the value of the function being implemented and its derivatives at a point. Thus, the range of variation of the signal is divided into M sections; n approximation function iruets on kaShbM of them corresponds to a polynomial of degree 6 with respect prelozhenie preobrazov131Tol make it possible ZHSH TyrShey be anY yShtor from: ygaalbv, SCHSH1 zhaZhyoDva yrbnzvblnom threshold elements including the supporting points that euSchestvenno uyrschaot their skhom Gyu realy- zatsyk pvyshaet and accuracy of pre-forming. The variable transfer coefficient of the sum of the matrix Kj. allows you to achieve a constant dynamic range of the output of the output signal & summation of the Hai each approximation segment. At the same time, the dynamic range of all elements of the fuctional one is used if the Aeruniform split by the reference points of the range of variation of the input signal is adopted, which will increase the total tokenness. --- Invention Functional transducer with piecewise nonlinear approximation, adjoining adders, the first input of the first one is connected to the input of the converter, the second input is connected to the first B1 input of the first signal analyzer, and the output is connected through the first large-scale block the first input of the second sum of 136 and 6 of the sixth is connected to the outputs of the second, third, fourth, fifth and sixth scale blocks, the seventh input of the second adder is connected to the second B1 of the first 6m analyzer of the second, on the C The third output to the first, second, fourth, fifth, and sixth scaled control units, which, besides the first mass block, are connected by other inputs to the outputs of the first, second, third, third | TrGB and 1 st of the main terminal. blocks, the output of the first adder is connected to the signal inputs of the first and second keys, and the output of the first key is connected via a quad. With the signal input of the third key, ШШ КШ1) | Р by the departure of the first key, the output of the second key via the cubic converter is connected to the signal output of the fourth key, the first pinch key is connected to the second recording unit, the second inputs of the first and second keys are connected to the first input of the seventh scale unit and the output of the third adder, the first input. which is connected to the second output of the first storage unit, the first output of the first signal analyzer is connected to the second input of the third adder, the second output of the third key is connected to the first input of the eighth scale unit and the first input of the fourth adder, the second input of which is connected to the second output of the second signal analyzer connected the third output with the first input of the fifth adder, the second output of the fourth key connected with i the first input of the ninth scale unit and the first input of the sixth adder, W The input of which is connected to the fourth output of the second signal analyzer, connected by the fifth output to the second inputs of the seventh, eighth and ninth scale blocks, the output of the ninth Macuy-block block is connected to the second input of the fifth adder; the third and fourth inputs of which are connected respectively to the first outputs of the seventh and eighth large blocks, connected by the second outputs, respectively, to the third and fourth entrances of the sixth adder, the third output being. the third scale output connected to the input of the third storage block, the output of the fifth adder connected to the input of the fourth memory block, the output of the sixth adder to the input of the fifth memory block, the output of the control unit to the control outputs of keys, and the output of the second adder is connected to the output of the converter, characterized in that, in order to increase the accuracy of operation, the third input of the first adder is connected to the third output of the first signal analyzer, and the output of the transducer is Another adder is connected to the input of the second signal analyzer. Sources of information taken into account in the examination. 1. Avtora certificate of the USSR N 374622, cl. G 06 G 7/28, 1970..

2. The copyright certificate in application M 237315О / 24, cl. a 06 Q 7/26, priority 14.06.76 (prototype).