The Initial Value Problem y' = f(x, y); y(c) = d 51
boundary of the rectangle twice instead of two different boundaries. T he
direction field for the differential equation y' = -x/y and the solution of the
IVP (10) are displayed in Figure 2.9.
2 /////______ ,__ ____________ """'
/////_______ -----------"""'"
/ / / / / /----------------"-.. "-..-....._,""
//////---/ / / / / .....-::~--------,..._---~---"""'"""' """"::--.....-....._,"" '\ \
I////./"/__..,... ________ "-'..::::~'\\\\
I I I / ~ / //-------........._ "-°" ~ \ \ \ \
II/Iii///__..,----"-"-'\~\\
I I I I I I / / --"-. \ \ \ \ \ \ \
y(x) 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ 2; ~ J ~ } } } ~ t
\ \ \ \ \ \ "--"----~---/ / / I I I I I
\\\\'\""-"----~---////////
- 1 \ \ \ '\ \ '\ "-....., '\ °""-"-"-"-.. ......... ____----- ---//////I // / / / / /
'\ ""-....., "-.. ......... ___ /// / / /
" "-....., "-.. "-.. -------------------/ / / / /
"-....., "-.. "-.. -----------------------/ / / / - 2 -.....,-....., "-.. "-.. ---------- ,___---------/ / / /
- 2 -1 0 2
x
- 2 -1 0 2
Figure 2.9 Direction Field for the DE y' = -x/y and the
Solution of the IVP y' = -x/y; y(-l) = 1
When the function f(x, y) in the initial value problem (1) y' = f(x, y);
y(c) = d has the form f(x, y) = a(x)y + b(x), the initial value problem is
call ed a linear first-order initia l value proble m; otherwise, the init ial
value problem is said to be nonlinear. The initial value problems (2), (5), and
(9) which we examined in examples 1, 2, and 3, respectively, are all nonlinear
first-order initial value problems. The intervals on which the solutions to these
problems existed and were unique exhibited no discernable pattern, mainly
because the initial value problems were nonlinear. However, when we apply
the fundamental existence and uniqueness theorem to the linear first-order
initial value problem
(11) y' = a(x)y + b(x) = f(x, y); y(c) = d
we obtain an existence and uniqueness theorem in which the interval of ex-
istence and uniqueness is completely determined by the functions a(x) and
b(x). Differentiating f(x, y) = a(x)y+b(x) partially with respect toy, we find
fy(x, y) = a(x). Hence, when the functions a(x) and b(x) are both defined
and continuous functions of x on the interval (a , (3), the functions f(x, y) and
fy(x, y) are both defined and continuous functions of x and y in any finite
rectangle R = {(x, y)I a< x < (3 and / < y < b}. So the linear IVP ( 11 )
has a unique solution and this solution can be extended uniquely until the
boundary of the largest rectangle on which both a(x) and b(x) are defined