1

Known conditions: both angles a and b are obtuse angles

AND

\[Pi]/2 < a < \[Pi], \[Pi]/2 < b < \[Pi], 
Sin[a/2]^2 + Cos[a + \[Pi]/3] == (5 - Sqrt[15])/10, 
Sin[b] == Sqrt[10]/10

Required values for angles a+b

Reduce[{\[Pi]/2 < a < \[Pi], \[Pi]/2 < b < \[Pi], 
   Sin[a/2]^2 + Cos[a + \[Pi]/3] == (5 - Sqrt[15])/10, 
   Sin[b] == Sqrt[10]/10, t == a + b}, t, {a, b}, 
  Reals] // FullSimplify

The above methods did not obtain accurate values

To get the value of the question is: (7 π)/4

csn899
  • 3,953
  • 6
  • 13
  • "The above methods did not obtain accurate values" They're accurate values, but expressed as Root[…]s. Please read this post for more info: https://mathematica.stackexchange.com/a/126156/1871 Since you can read Chinese, read this also: https://tieba.baidu.com/p/7231221217 – xzczd Jul 26 '23 at 13:11
  • Please clarify the following: 1. Is my understanding in my previous comment correct? 2. Is your actual goal obtaining (7 π)/4 as the answer? – xzczd Jul 26 '23 at 23:59
  • 以防你没看懂我的话,我现在用汉语重复一遍:1. 我之前的评论对不对?你是不是误解了Reduce的输出? 2. 你真正的目标,是不是想让Mathematica输出(7 π)/4作为答案? – xzczd Jul 27 '23 at 02:22
  • 如果是的话,你把你的问题改一改写清楚,这个问题是完全可以重开的。当然,如果你觉得你已经拿到你想要的东西了,懒得费时间改帖子,那也是你的自由。 – xzczd Jul 27 '23 at 02:23
  • 我前面已经说过了,你这问题陈述是错的,Reduce输出的已经是准确答案了,只不过它用了Root来表示它。请注意,跟“准确答案”相对的是“近似答案”! – xzczd Jul 27 '23 at 02:41
  • 没改干净,这样是不可能重开的。 – xzczd Jul 28 '23 at 01:58

3 Answers3

6
a + b /. 
  First@Solve[{π/2 < a < π, π/2 < b < π, 
     Sin[a/2]^2 + Cos[a + π/3] == (5 - Sqrt[15])/10, 
     Sin[b] == Sqrt[10]/10}, {a, b}, Reals] // FullSimplify

(7 π)/4

cvgmt
  • 72,231
  • 4
  • 75
  • 133
4

Probably OP won't bother to understand this, but here's a direct fix for OP's approach:

Reduce[{π/2 < a < π, π/2 < b < π, 
  Sin[a/2]^2 + Cos[a + π/3] == (5 - Sqrt[15])/10, Sin[b] == Sqrt[10]/10, 
  t == a + b}, t, {a, b}, Reals] // FullSimplify

% /. a_Root :> Select[x /. Solve[a[[1, 1]][x] == 0, x], # == a[[1, 2]] &][[1]] // Map@FullSimplify (* t == (7 π)/4 *)

xzczd
  • 65,995
  • 9
  • 163
  • 468
3
$Version

(* "13.3.0 for Mac OS X ARM (64-bit) (June 3, 2023)" *)

Clear["Global`*"]

sys = {π/2 < a < π, π/2 < b < π, 
   Sin[a/2]^2 + Cos[a + π/3] == (5 - Sqrt[15])/10, Sin[b] == Sqrt[10]/10, 
   t == a + b};

sol = Reduce[sys, t, {a, b}, Reals] // FullSimplify // ToRules

enter image description here

Pi*RootApproximant[(t /. sol // N)/Pi]

(* (7 π)/4 *)

sol2 = Map[FullSimplify, Reduce[sys, {t, a, b}, Reals] // ToRules, {2}]

(* {t -> (7 π)/4, a -> 4 ArcTan[2 - Sqrt[5] + Sqrt[10 - 4 Sqrt[5]]], b -> -a + t} *)

Verification is quite slow

sys //. sol2 // FullSimplify // AbsoluteTiming

(* {30.8832, {True, True, True, True, True}} *)

EDIT: Using Backsubstitution

sol2r = Map[FullSimplify, 
  Reduce[sys, {t, a, b}, Reals, Backsubstitution -> True] // ToRules, {2}]

(* {t -> (7 π)/4, a -> 4 ArcTan[2 - Sqrt[5] + Sqrt[10 - 4 Sqrt[5]]], b -> π - ArcCot[3]} *)

sys //. sol2r // FullSimplify // AbsoluteTiming

(* {20.2925, {True, True, True, True, True}} *)

Bob Hanlon
  • 157,611
  • 7
  • 77
  • 198