Phil Lucht Math & Physics Archive
Home / Math and Physics Files / Math / Lagrange Multipliers / All Support

four inequalities

DOCX · 19.9 KB
Open DOCX file

Short worked note, dated 6.6.15 in its heading, from Phil's Lagrange multipliers files. Assuming energies ε3 > ε2 > ε1, it rewrites the linear relations for N1 and N2 in terms of N3, then examines the four inequalities 0 ≤ N1, N2 ≤ M one at a time. These combine into a single range for N3, which is evaluated for an example with u = 1.5. Many fractions are lost in extraction, so some bounds are unclear.

AI-written summary; may contain errors. This description is approximate.

Extracted text (machine-read; may contain errors)
The four inequalities problem PhL 6.6.15 Assume, ε3 > ε2 > ε1 . From (6.3.1), N1 = + (ε2- ε1)-1[ ε2M - U + (ε3- ε2)N3] = N1(N3) N2 = - (ε2- ε1)-1[ ε1M - U + (ε3 - ε1)N3] = N2(N3) . (6.3.1) Rewrite these as N1 = + (ε2- ε1)-1[(ε2-u)M + (ε3- ε2)N3] = N1(N3) N2 = - (ε2- ε1)-1[ (ε1-u)M + (ε3 - ε1)N3] = N2(N3) . (6.3.1) Rewrite again as N1 = + (ε2- ε1)-1[(ε2-u)M + (ε3- ε2)N3] = N1(N3) N2 = +(ε2- ε1)-1[ (u-ε1)M - (ε3 - ε1)N3] = N2(N3) . (6.3.1) There are four inequalities 0 ≤ N1 ≤ M 0 ≤ N2 ≤ M Examine the four inequalities one at a time. Inequality #1: N1 ≥ 0 (ε2-u)M + (ε3- ε2)N3 ≥ 0 N3 ≥ M This is only a constraint if u > ε2, Inequality #2: N2 ≥ 0 (u-ε1)M - (ε3 - ε1)N3 ≥ 0 N3 ≤ M Since ε1 ≤ u ≤ ε3, we always have u ≥ ε1 so in general this is always a constraint. Inequality #3: N1 ≤ M (ε2-u)M + (ε3- ε2)N3 ≤ (ε2-ε1)M (ε3- ε2)N3 ≤ (ε2-ε1)M + (u-ε2)M = (u-ε1)M N3 ≤ M Since ε1 ≤ u ≤ ε3, we always have u ≥ ε1 so in general this is always a constraint. Inequality #4: N2 ≤ M (u-ε1)M - (ε3 - ε1)N3 ≤ (ε2-ε1)M (ε3 - ε1)N3 ≥ (u-ε1)M + (ε1-ε2)M = (u-ε2)M N3 ≥ M This is only a constraint if u > ε2. Summary N3 ≥ M N3 ≥ M N3 ≤ M N3 ≤ M Combine these to get N3 ≥ max ( M, M) only a constraint if u > ε2 N3 ≤ min( M, M) In the first line, the second denominator is larger, so this second term is smaller, and we get N3 ≥ M as the summary of both constraints In the second line, the second denominator is larger, so this second term is smaller, so N3 ≤ M So this entire mess can be summarized in this simple way M ≤ N3 ≤ M In my application this says (u-2) M ≤ N3 < (1/2) (u-1) M If I choose u = 1.5 for the example, we get -(1/2) M ≤ N3 < (1/2) (1/2) M and this boils down to 0 ≤ N3 ≤ M/4