Processing math: 94%
Research article

Symmetry of positive solutions of a p-Laplace equation with convex nonlinearites

  • Received: 22 September 2022 Revised: 18 January 2023 Accepted: 25 January 2023 Published: 06 April 2023
  • MSC : 35A21, 35B06

  • In this paper, we consider the symmetry properties of the positive solutions of a p-Laplacian problem of the form

    {Δpu=f(x,u),       in     Ω,       u=g(x),          on    Ω,

    where Ω is an open smooth bounded domain in RN,N2, and symmetric w.r.t. the hyperplane Tν0(ν is a direction vector in RN,|ν|=1), f: Ω×R+R+ is a continuous function of class C1 w.r.t. the second variable, g0 is continuous, and both f and g are symmetric w.r.t. Tν0, respectively. Introducing some assumptions on nonlinearities, we get that the positive solutions of the problem above are symmetric w.r.t. the direction ν by a new simple idea even if Ω is not convex in the direction ν.

    Citation: Keqiang Li, Shangjiu Wang. Symmetry of positive solutions of a p-Laplace equation with convex nonlinearites[J]. AIMS Mathematics, 2023, 8(6): 13425-13431. doi: 10.3934/math.2023680

    Related Papers:

    [1] Takao Komatsu, Ram Krishna Pandey . On hypergeometric Cauchy numbers of higher grade. AIMS Mathematics, 2021, 6(7): 6630-6646. doi: 10.3934/math.2021390
    [2] Takao Komatsu, Wenpeng Zhang . Several expressions of truncated Bernoulli-Carlitz and truncated Cauchy-Carlitz numbers. AIMS Mathematics, 2020, 5(6): 5939-5954. doi: 10.3934/math.2020380
    [3] Taekyun Kim, Hye Kyung Kim, Dae San Kim . Some identities on degenerate hyperbolic functions arising from p-adic integrals on Zp. AIMS Mathematics, 2023, 8(11): 25443-25453. doi: 10.3934/math.20231298
    [4] Dojin Kim, Patcharee Wongsason, Jongkyum Kwon . Type 2 degenerate modified poly-Bernoulli polynomials arising from the degenerate poly-exponential functions. AIMS Mathematics, 2022, 7(6): 9716-9730. doi: 10.3934/math.2022541
    [5] Tabinda Nahid, Mohd Saif, Serkan Araci . A new class of Appell-type Changhee-Euler polynomials and related properties. AIMS Mathematics, 2021, 6(12): 13566-13579. doi: 10.3934/math.2021788
    [6] Jizhen Yang, Yunpeng Wang . Congruences involving generalized Catalan numbers and Bernoulli numbers. AIMS Mathematics, 2023, 8(10): 24331-24344. doi: 10.3934/math.20231240
    [7] Ling Zhu . Asymptotic expansion of a finite sum involving harmonic numbers. AIMS Mathematics, 2021, 6(3): 2756-2763. doi: 10.3934/math.2021168
    [8] Waseem Ahmad Khan, Kottakkaran Sooppy Nisar, Dumitru Baleanu . A note on (p, q)-analogue type of Fubini numbers and polynomials. AIMS Mathematics, 2020, 5(3): 2743-2757. doi: 10.3934/math.2020177
    [9] Letelier Castilla, William Ramírez, Clemente Cesarano, Shahid Ahmad Wani, Maria-Fernanda Heredia-Moyano . A new class of generalized Apostol–type Frobenius–Euler polynomials. AIMS Mathematics, 2025, 10(2): 3623-3641. doi: 10.3934/math.2025167
    [10] Nadia N. Li, Wenchang Chu . Explicit formulae for Bernoulli numbers. AIMS Mathematics, 2024, 9(10): 28170-28194. doi: 10.3934/math.20241366
  • In this paper, we consider the symmetry properties of the positive solutions of a p-Laplacian problem of the form

    {Δpu=f(x,u),       in     Ω,       u=g(x),          on    Ω,

    where Ω is an open smooth bounded domain in RN,N2, and symmetric w.r.t. the hyperplane Tν0(ν is a direction vector in RN,|ν|=1), f: Ω×R+R+ is a continuous function of class C1 w.r.t. the second variable, g0 is continuous, and both f and g are symmetric w.r.t. Tν0, respectively. Introducing some assumptions on nonlinearities, we get that the positive solutions of the problem above are symmetric w.r.t. the direction ν by a new simple idea even if Ω is not convex in the direction ν.



    Hermite-Hadamard inequality is a double inequality for convex functions that has a lot of literary value (please see [16]).

    Let ζ:IR, IR, ς,τI with ς<τ, be a convex function. Then

    ζ(ς+τ2)1τςτςζ(w)dwζ(ς)+ζ(τ)2, (1.1)

    the inequality holds in reversed direction if ζ is concave.

    Fejér [15] established the following double inequality as a weighted generalization of (1.1):

    ζ(ς+τ2)τςϑ(w)dwτςζ(w)ϑ(w)dwζ(ς)+ζ(τ)2τςϑ(w)dw, (1.2)

    where ζ:IR, IR, ς,τI with ς<τ is any convex function and ϑ:[ς,τ]R is non-negative integrable and symmetric with respect to w=ς+τ2.

    These inequalities have many extensions and generalizations, see [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18] and [19,20,21,22,23,24,25,26,27,28,29,30,31].

    Consider the following mappings on [0,1]:

    ˘G(ι)=12τς[ζ(ις+(1ι)ς+τ2)+ζ(ιτ+(1ι)ς+τ2)]dw,
    H(ι)=1τςτςζ(ιw+(1ι)ς+τ2)dw,
    Hϑ(ι)=1τςτςζ(ιw+(1ι)ς+τ2)ϑ(w)dw,
    L(ι)=12(τς)τς[ζ(ις+(1ι)w)+ζ(ιτ+(1ι)w)]dw

    and

    Lϑ(ι)=12τς[ζ(ις+(1ι)w)+ζ(ιτ+(1ι)w)]ϑ(w)dw,

    where ζ:[ς,τ]R is a convex function and ϑ:[ς,τ]R is non-negative integrable and symmetric with respect to w=ς+τ2.

    The important results that characterize the properties of the above mappings and inequalities are discussed by a number of mathematicians.

    Dragomir [2] established the theorem which refines the first inequality of (1.1).

    Theorem 1. [2] Let ζ:[ς,τ]R be a convex function on [ς,τ]. Then H is monotonically increasing and convex on [0,1]. Moreover, one has thefollowing inequalities

    ζ(ς+τ2)=H(0)H(ι)H(1)=1τςτςζ(w)dw.

    Dragomir et al. [7] obtained the refinements of (1.1).

    Theorem 2. [2] Let ζ, H be defined asabove. Then

    (i) The following inequality holds

    ζ(ς+τ2)2τςς+3τ43ς+τ4ζ(w)dw10H(ι)dι12[ζ(ς+τ2)+1τςτςζ(w)dw].

    (ii) If ζ is differentiable on [ς,τ], then for all ι[0,1], one has

    01τςτςζ(w)dwH(ι)(1ι)[ζ(ς)+ζ(τ)21τςτςζ(w)dw]

    and

    0ζ(ς)+ζ(τ)2H(ι)(ζ(τ)ζ(ς))(τς)4.

    Theorem 3. [7] Let ζ, H, ˘G be defined as above. We have

    (i) ˘G is convex and increasing on [0,1].

    (ii) The following hold

    infι[0,1]˘G(ι)=˘G(0)=ζ(ς+τ2)

    and

    supι[0,1]˘G(ι)=˘G(1)=ζ(ς)+ζ(τ)2.

    (iii) The inequality

    H(ι)˘G(ι)

    holds for all ι[0,1].

    (iv) Then

    2τςς+3τ43ς+τ4ζ(w)dw12[ζ(3ς+τ4)+ζ(ς+3τ4)]10˘G(ι)dι12[ζ(ς+τ2)+ζ(ς)+ζ(τ)2].

    (v) If ζ is differentiable on [ς,τ], then for all ι[0,1], one has

    0H(ι)ζ(ς+τ2)˘G(ι)H(ι).

    Theorem 4. [7] Let ζ, H, ˘G, L be defined as above. Then

    (i) L is convex on [0,1].

    (ii) The inequalities

    ˘G(ι)L(ι)1ιτςτςζ(w)dw+ιζ(ς)+ζ(τ)2ζ(ς)+ζ(τ)2

    hold for all ι[0,1] and

    supι[0,1]L(ι)=L(1)=ζ(ς)+ζ(τ)2.

    (iii) The inequalities

    H(1ι)L(ι)andH(ι)+H(1ι)2L(ι)

    hold for all ι[0,1].

    Teseng et al. [24] proved the following result.

    Lemma 1. [24] Let ζ: [ς,τ]R be a convex function and let ς˘ϰ1w1w2˘ϰ2τ with w1+w2=˘ϰ1+˘ϰ2. Then

    ζ(w1)+ζ(w2)ζ(˘ϰ1)+ζ(˘ϰ2).

    Yang and Tseng [28] proven the theorem by using Lemma 1 which refines the first inequality of (1.2) and generalizes Theorem 1.

    Theorem 5. [28] Let ζ:[ς,τ]R be a convex function and ϑ:[ς,τ]R is non-negative integrable and symmetric with respect to w=ς+τ2. Then Hϑ is convex, increasing on [0,1], and for all ι[0,1], we have

    ζ(ς+τ2)τςϑ(w)dw=Hϑ(0)Hϑ(ι)Hϑ(1)=τςζ(w)ϑ(w)dw. (1.3)

    One of the generalizations of the convex functions is harmonic functions:

    Definition 1. [17] Define IR{0} as an interval of real numbers. A function ζ from I to the real numbers is considered to be harmonically convex, if

    ζ(w˘ϰιw+(1ι)˘ϰ)ιζ(˘ϰ)+(1ι)ζ(w) (1.4)

    for all w,˘ϰI and ι[0,1]. Harmonically concave ζ is defined as the inequality in (1.4) reversed.

    İşcan [17] used harmonic-convexity to develop the inequalities of Hermite-Hadamard type.

    Theorem 6. [17] Let ζ:IR{0}R be a harmonically convex function and ς,τI with ς<τ. If ζL([ς,τ]) then theinequalities

    ζ(2ςτς+τ)ςττςςτζ(w)w2dwζ(ς)+ζ(τ)2 (1.5)

    hold.

    Let ζ:[ς,τ](0,)R be a harmonic convex mapping and let S,V:[0,1]R be defined by

    S(ι)=ςττςτς1w2ζ(2ςτw2ςτι+(1ι)w(ς+τ))dw (1.6)

    and

    V(ι)=ςτ2(τς)τς1w2[ζ(2τw(1+ι)w+(1ι)τ)+ζ(2ςw(1+ι)w+(1ι)ς)]dw. (1.7)

    The author obtained the refinement inequalities for (1.5) related to the above mappings:

    Theorem 7. [21] Let ζ:[ς,τ](0,)R be a harmonic convex function on [ς,τ]. Then

    (i) S is harmonic convex (0,1] andincreases monotonically on [0,1].

    (ii) The following hold:

    ζ(2ςτς+τ)=S(0)S(ι)S(1)=ςττςτςζ(w)w2dw.

    Theorem 8. [21] Let ζ:[ς,τ](0,)R be a harmonic convex function on [ς,τ]. Then

    (i) V is harmonic convex (0,1] andincreases monotonically on [0,1].

    (ii) The following hold:

    ςττςτςζ(w)w2dw=V(0)V(ι)V(1)=ζ(ς)+ζ(τ)2.

    Harmonic symmetricity of a function is given in the definition below.

    Definition 2. [22] A function ϑ:[ς,τ]R{0}R is harmonically symmetric with respect to 2ςτς+τ if

    ϑ(w)=ϑ(11ς+1τ1w)

    holds for all w[ς,τ].

    Fejér type inequalities using harmonic convexity and the notion of harmonic symmetricity were presented in Chan and Wu [1].

    Theorem 9. [1] Let ζ:IR{0}R be a harmonically convex function and ς,τI with ς<τ. If ζL([ς,τ]) and ϑ:[ς,τ]R{0}R is nonnegative, integrable and harmonically symmetric with respect to 2ςτς+τ, then

    ζ(2ςτς+τ)ςτϑ(w)w2dwςτζ(w)ϑ(w)w2dwζ(ς)+ζ(τ)2ςτϑ(w)w2dw. (1.8)

    Chan and Wu [1] also defined some mappings related to (1.8) and discussed important properties of these mappings.

    Motivated by the studies conducted in [2,21,24,27], we define some new mappings in connection to (1.8) and to prove new Féjer type inequalities which indeed provide refinement inequalities as well.

    To prove the major findings of this work, we employ the given important facts about harmonic convex and convex functions.

    Theorem 10. [8,9] If [ς,τ]I(0,) and if we consider thefunction ˘h:[1τ,1ς]R defined by ˘h(ι)=ζ(1ι), then ζ is harmonicallyconvex on [ς,τ] if and only if ˘h is convex in the usual sense on [1τ,1ς].

    Theorem 11. [8,9] If I(0,) and ζ is convex and nondecreasing function then ζ isharmonic convex and if ζ is harmonic convex and nonincreasing function then ζ is convex.

    Let us now define some mappings on [0,1] related to (1.8) and prove some refinement inequalities.

    ˘G1(ι)=12[ζ(2ςτ2ςι+(1ι)(ς+τ))+ζ(2ςτ2τι+(1ι)(ς+τ))]dw,
    S(ι)=ςττςτς1w2ζ(2ςτw2ςτι+(1ι)(ς+τ)w)dw,
    Sϑ(ι)=τςζ(2ςτw2ςτι+(1ι)(ς+τ)w)ϑ(w)w2dw,
    T(ι)=ςτ2(τς)τς1w2[ζ(τwιw+(1ι)τ)+ζ(ςwιw+(1ι)ς)]dw

    and

    Tϑ(ι)=12τς[ζ(τwιw+(1ι)τ)+ζ(ςwιw+(1ι)ς)]ϑ(w)w2dw,

    where ζ:[ς,τ]R is a harmonic convex function and ϑ:[ς,τ]R is non-negative integrable and symmetric with respect to w=2ςτς+τ.

    Lemma 2. Let ζ: [ς,τ](0,)R be a harmonic convex function and let ς˘ϰ1w1w2˘ϰ2τ with w1w2w1+w2=˘ϰ1˘ϰ2˘ϰ1+˘ϰ2. Then

    ζ(w1)+ζ(w2)ζ(˘ϰ1)+ζ(˘ϰ2).

    Proof. For ˘ϰ1=˘ϰ2, the result is obvious. We observe that

    w1=˘ϰ1˘ϰ2(w1˘ϰ1˘ϰ1˘ϰ2w1˘ϰ1w1˘ϰ2)˘ϰ2+(˘ϰ1˘ϰ2w1˘ϰ2w1˘ϰ1w1˘ϰ2)˘ϰ1andw2=˘ϰ1˘ϰ2(w2˘ϰ1˘ϰ1˘ϰ2w2˘ϰ1w2˘ϰ2)˘ϰ2+(˘ϰ1˘ϰ2w2˘ϰ2w2˘ϰ1w2˘ϰ2)˘ϰ1.

    By applying the harmonic convexity, we obtain

    ζ(w1)+ζ(w2)(w1˘ϰ1˘ϰ1˘ϰ2w1˘ϰ1w1˘ϰ2)ζ(˘ϰ1)+(˘ϰ1˘ϰ2w1˘ϰ2w1˘ϰ1w1˘ϰ2)ζ(˘ϰ2)+(w2˘ϰ1˘ϰ1˘ϰ2w2˘ϰ1w2˘ϰ2)ζ(˘ϰ1)+(˘ϰ1˘ϰ2w2˘ϰ2w2˘ϰ1w2˘ϰ2)ζ(˘ϰ2)=(w1˘ϰ1˘ϰ1˘ϰ2w1˘ϰ1w1˘ϰ2+w2˘ϰ1˘ϰ1˘ϰ2w2˘ϰ1w2˘ϰ2)ζ(˘ϰ1)+(˘ϰ1˘ϰ2w1˘ϰ2w1˘ϰ1w1˘ϰ2+˘ϰ1˘ϰ2w2˘ϰ2w2˘ϰ1w2˘ϰ2)ζ(˘ϰ2)=˘ϰ1˘ϰ2˘ϰ1((w1+w2w1w2)˘ϰ22)ζ(˘ϰ1)+˘ϰ2˘ϰ2˘ϰ1(2(w1+w2w1w2)˘ϰ1)ζ(˘ϰ2)=ζ(˘ϰ1)+ζ(˘ϰ2).

    We first prove a result similar to (1.3) for harmonically convex functions which provide refinement inequalities for (1.8).

    Theorem 12. Let ζ: [ς,τ](0,)R be a harmonic convex function, 0<ρ<1, 0<θ<1, λ=ςτρς+(1ρ)τ, τ0=(ςττς)min{ρ1θ,1ρθ} andlet ϑ:[ς,τ]R be nonnegative and integrable and ϑ(λ1θιλ)=ϑ(λ1+(1θ)ιλ), ι[0,τ0]. Then

    ζ(ςτρς+(1ρ)τ)λ1θιλλ1+(1θ)ιλϑ(w)w2dw1θθλλ1θιλζ(w)ϑ(w)w2dw+θ1θλ1+(1θ)ιλλζ(w)ϑ(w)w2dw[ρζ(τ)+(1ρ)ζ(ς)]λ1θιλλ1+(1θ)ιλϑ(w)w2dw. (2.1)

    Proof. For every τ[0,τ0], we have the identity

    λ1θιλλ1+(1θ)ιλϑ(w)w2dw=λλ1+(1θ)ιλϑ(w)w2dw+λ1θιλλϑ(w)w2dw=θι0ϑ(λ1θwλ)w2dw+(1θ)ι0ϑ(λ1θwλ)w2dw=ι0ϑ(λ1θwλ)w2dw. (2.2)

    We now prove that the mapping W:[0,τ0]R defined by

    W(ι)=(1θ)ζ(λ1θιλ)+θζ(λ1+(1θ)ιλ)

    is harmonic convex (0,τ0] and monotonically increasing on [0,τ0].

    Since the sum of two harmonic convex functions is a harmonic convex, hence W is a harmonic convex on (0,τ0]. Let ι(0,τ0], it follows from the harmonic convexity of ζ that

    W(ι)=(1θ)ζ(λ1θιλ)+θζ(λ1+(1θ)ιλ)ζ((λ1θιλ)(λ1+(1θ)ιλ)θ(λ1θιλ)+(1θ)(λ1+(1θ)ιλ))=ζ(λ)=ζ(ςτρς+(1ρ)τ). (2.3)

    We observed that 0<ρρ(τς)+ςτθιτς1, 0(1ρ)(τς)θιςττς1ρ<1, 0ρρ(τς)(1θ)ιςττςρ1 and 0<1ρ(1ρ)(τς)+(1θ)ιςττς1. Thus, by using the harmonic convexity, we obtain

    W(ι)=(1θ)ζ(λ1θιλ)+θζ(λ1+(1θ)ιλ)=(1θ)ζ(ςτ(ρ(τς)+ςτιθτς)ς+((1ρ)(τς)θιςττς)τ)+θζ(ςτ(ρ(τς)(1θ)ιςτ(τς))ς+((1ρ)(τς)+(1θ)ιςττς)τ)(1θ)((1ρ)(τς)θιςττς)ζ(ς)+(1θ)×(ρ(τς)+ιςτθτς)ζ(τ)+θ((1ρ)(τς)+(1θ)ιςτ(τς))ζ(ς)+θ(ρ(τς)(1θ)ιςττς)ζ(τ)=(1ρ)ζ(ς)+ρζ(τ). (2.4)

    From (2.3) and (2.4), we obtain

    ζ(ςτρς+(1ρ)τ)W(ι)(1ρ)ζ(ς)+ρζ(τ). (2.5)

    Finally, for ι1, ι2, such that 0<ι1<ι2ς0, since W(ι) is harmonic convex, it follows from (2.3) that

    W(ι2)W(ι1)ι2ι10.

    This shows that W is increasing on [0,τ0].

    Since ϑ is nonnegative, multiplying (2.5) by ϑ(λ1θwλ)w2, integrating the resulting inequalities over [0,ι] and using ϑ(λ1θwλ)=ϑ(λ1+(1θ)wλ), we have

    ζ(ςτρς+(1ρ)τ)ι0ϑ(λ1θwλ)w2dw(1θ)ι0ζ(λ1θwλ)ϑ(λ1θwλ)w2dw+θι0ζ(λ1+(1θ)wλ)ϑ(λ1+(1θ)wλ)w2dw[(1ρ)ζ(ς)+ρζ(τ)]ι0ϑ(λ1θwλ)w2dw. (2.6)

    By using the identity (2.2) in (2.6), we obtain (2.1).

    Remark 1. If we choose ρ=ϑϑ+q, θ=12, ι=2˘ϰ in Theorem 12, then

    ζ(ςτ(ϑ+q)ϑς+qτ)λ1λτ˘ϰλ1+λτ˘ϰϑ(w)w2dwλ1λτ˘ϰλ1+λτ˘ϰζ(w)ϑ(w)w2dw[ϑζ(τ)+qζ(ς)ϑ+q]λ1λτ˘ϰλ1+λτ˘ϰϑ(w)w2dw. (2.7)

    Remark 2. If we choose ρ= θ=12, ι=τ0=ςττς in Theorem 12, then we get (1.8).

    Remark 3. If we choose ρ= θ=12, ι=τ0=ςττς and ϑ(w)=1, w[ς,τ] in Theorem 12, then we get (1.5).

    Theorem 13. Let ζ, λ and τ0 be defined as in Theorem 12, 0<ρ<1, 0<θ<1, ρ+θ1 and let X be defined on [0,1] as

    X(ι)=ρςτ(1θ)(τς)×ρςτ(1θ)(τς)0[(1θ)ζ(λ1θιwλ)+θζ(λ1+(1θ)ιwλ)]dw. (2.8)

    Then, X is harmonically convex on (0,1] andmonotonically increasing on [0,1], and

    ζ(ςτρς+(1ρ)τ)X(ι)X(1)=ρςτ(1θ)(τς)×ρςτ(1θ)(τς)0[(1θ)ζ(λ1θwλ)+θζ(λ1+(1θ)wλ)]dw(1ρ)ζ(ς)+ρζ(τ).

    Proof. Since ζ is harmonically convex on [ς,τ] this prove the harmonic convexity of X on (0,1]. By using the condition ρ+θ1 implies that τ0=ρςτ(1θ)(τς). Since the mapping W:[0,τ0]R defined by

    W(ι)=(1θ)ζ(λ1θιλ)+θζ(λ1+(1θ)ιλ) (2.9)

    has been proved to be monotonically increasing on [0,τ0], thus the mapping X is also monotonically increasing on [0,1].

    Because X is monotonically increasing on [0,1], it follows that the inequalities (2.8) can be deduced from these inequalities (2.5). The proof of the theorem was completed as a result of this.

    The next theorem can be proved similarly:

    Theorem 14. Let ζ, λ, τ0, ρ and θ be defined as in Theorem 13. Let X1 be defined on [0,1] as

    X1(ι)=ρςτ(1θ)(τς)×ρςτ(1θ)(τς)0[(1θ)ζ(λ(1θ)ςτ(1θ)ςτθ(ρ(τς)w(1ι)(1θ)ςτ)λ)+θζ(λ(1θ)ςτ(1θ)ςτ+(1θ)(ρ(τς)w(1ι)(1θ)ςτ)λ)]dw. (2.10)

    Then, X1 is harmonically convex monotonically increasing on [0,1], and

    (1θ)2ςτρθ(τς)(1θ)ςτ(1θ)ςρ(τς)λζ(w)w2dw+θςτρ(τς)λςζ(w)w2dwX1(ι)X1(1)=(1θ)ζ((1θ)ςτ(ςτ)ρ+(1θ)τ)+θζ(ς)(1ρ)ζ(ς)+ρζ(τ). (2.11)

    Remark 4. Taking ρ=θ=12 in the inequality (2.8) reduces to

    S(ι)=ςττςτςζ(2ςτw2ςτι+(1ι)w(ς+τ))dww2.

    Remark 5. Taking ρ=θ=12 in the inequality (2.10) reduces to

    V(ι)=ςτ2(τς)τς1w2[ζ(2τw(1+ι)w+(1ι)τ)+ζ(2ςw(1+ι)w+(1ι)ς)]dw. (2.12)

    Theorem 15. Let ζ, ρ, θ, λ, τ0 be defined as in Theorem 13 and let ϑ be defined as in Theorem 12. Let Y be a function defined on [0,1] by

    Y(ι)=s0[(1θ)ζ(λ1θλιw)ϑ(λ1θλw)+θζ(λ1+(1θ)λιw)ϑ(λ1+(1θ)λw)]dw (2.13)

    for some s[0,τ0]. Then Y isharmonic convex and monotonically increasing on (0,1] and

    ζ(ςτρς+(1ρ)τ)λ1θλsλ1+(1θ)λsϑ(w)dwY(ι)Y(1)=1θθλλ1θλsζ(w)ϑ(w)w2dw+θ1θλ1+(1θ)λsλζ(w)ϑ(w)w2dw. (2.14)

    Proof. Since ζ is harmonic convex and ϑ is nonnegative, we see that Y is harmonic convex on (0,1]. Let w[0,s], where s[0,τ0], from Theorem 12 we get ˘h(ιw)=(1θ)ζ(λ1θλιw)+θζ(λ1+(1θ)λιw) is increasing for ι[0,1]. Therefore the inequalities (2.14) are achieved immediately.

    Theorem 16. Let ζ, ρ, θ, λ, τ0 be defined as in Theorem 15 and let ϑ be defined as in Theorem 12. Let Y1 be a function defined on [0,1] by

    Y1(ι)=s0[(1θ)ζ(λ1θs+θw(1ι)λ)ϑ(λ1θ(sw)λ)+θζ(λ1+(1θ)s(1θ)w(1ι)λ)ϑ(λ1+(1θ)(sw)λ)]dw (2.15)

    for some s[0,τ0]. Then Y1 isharmonic convex (0,1] and monotonically increasing on [0,1], and

    1θθλλ1θλsζ(w)ϑ(w)w2dw+θ1θλ1+(1θ)λsλζ(w)ϑ(w)w2dwY(ι)Y(1)=[(1θ)ζ(λ1θsλ)+(1θ)ζ(λ1+(1θ)sλ)]×λ1+(1θ)λsλ1θλsϑ(w)w2dw[(1ρ)ζ(ς)+ρζ(τ)]λ1+(1θ)λsλ1θλsϑ(w)w2dw. (2.16)

    Proof. Since ζ is harmonic convex and ϑ is nonnegative, we see that Y1 is harmonic convex on (0,1]. Next, for each w[0,ι], where ι[0,τ0], it follows from Theorem 12 that ˘h(ι)=(1θ)ζ(λ1θλι)+θζ(λ1+(1θ)λι) and k(ι)=s(1ι)w are increasing on [0,τ0] and [0,1] respectively. Hence

    ˘h(k(ι))=(1θ)ζ(λ1θs+θw(1ι)λ)ϑ(λ1θ(sw)λ)+θζ(λ1+(1θ)s(1θ)w(1ι)λ)ϑ(λ1+(1θ)(sw)λ)

    is increasing on [0,1]. Using the identity ϑ(λ1θλι)=ϑ(λ1+(1θ)λι) we see that Y(ι) is increasing on [0,1]. Therefore the inequalities (2.16) follows from

    ζ(ςτρς+(1ρ)τ)W(k(ι))(1ρ)ζ(ς)+ρζ(τ)

    and (2.16).

    Remark 6. Choose ρ=θ=12, s=τ0=ςττς in Theorems 15 and 16. Then the inequalities (2.14) and (2.16) reduce to

    ζ(2ςτς+τ)τςϑ(w)w2dwY(ι)Y(1)=τςζ(w)ϑ(w)w2dwY1(ι)Y1(1)=ζ(ς)+ζ(τ)2τςϑ(w)w2dw, (2.17)

    where

    Y(ι)=ςττςτςζ(2ςτw2ςτι+(1ι)w(ς+τ))ϑ(w)w2dw

    and

    Y1(ι)=12τς1w2[ζ(2τw(1+ι)w+(1ι)τ)ϑ(2ςwς+w)+ζ(2ςw(1+ι)w+(1ι)ς)ϑ(2wτw+τ)]dw. (2.18)

    Remark 7. The inequalities (2.17) provide weighted generalizations of Theorems 9 and 15.

    In the coming results we provide weighted generalizations of Theorems 2–4 for harmonic convex functions by using Lemma 2.

    Theorem 17. Let ζ, ϑ, Sϑ be defined as above. Then

    (i) The inequality

    ζ(2ςτς+τ)τςϑ(w)w2dw24ςτ3ς+τ4ςτς+3τζ(w)ϑ(2ςτw4ςτ(ς+τ)w)dww210Sϑ(ι)dι12[ζ(2ςτς+τ)τςϑ(w)w2dw+τςζ(w)ϑ(w)w2dw] (2.19)

    holds.

    (ii) If ζ is differentiable on [ς,τ] and ϑ isbounded on [ς,τ], then theinequalities

    0τςζ(w)ϑ(w)w2dwSϑ(ι)(1ι)[(τςςτ)[ζ(ς)+ζ(τ)2]τςζ(w)w2dw]ϑ, (2.20)

    hold for all ι[0,1], where ϑ=supw[ς,τ]ϑ(w).

    (iii) If ζ is differentiable on [ς,τ], then, for all ι[0,1], then

    0ζ(ς)+ζ(τ)2τςϑ(w)w2dwSϑ(ι)(τς)(τ2ζ(τ)ς2ζ(ς))4ςττςϑ(w)w2dw. (2.21)

    Proof. (i) Using techniques of integration and the hypothesis of ϑ, we have the following identities:

    ζ(2ςτς+τ)τςϑ(w)w2dw=42ςτς+τς120ζ(2ςτς+τ)ϑ(w)w2dιdw, (2.22)
    24ςτ3ς+τ4ςτς+3τζ(w)w2ϑ(2ςτw4ςτ(ς+τ)w)dw=22ςτς+τς120[ζ(4ςτw2ςτ+(ς+τ)w)+ζ(4ςτw3(ς+τ)w2ςτ)]ϑ(w)w2dιdw, (2.23)
    10Sϑ(ι)dι=2ςτς+τς120[ζ(2ςτwι(ς+τ)w+2(1ι)ςτ)+ζ(2ςτw2ςτι+(1ι)(ς+τ)w)]×ϑ(w)w2dιdw+2ςτς+τς120[ζ(2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w)+ζ(2ςτw2(1ι)(ςw+τwςτ)+ι(ς+τ)w)]ϑ(w)w2dιdw (2.24)

    and

    12[ζ(2ςτς+τ)τςϑ(w)w2dw+τςζ(w)ϑ(w)w2dw]=2ςτς+τς120[ζ(w)+ζ(2ςτς+τ)]ϑ(w)w2dιdw+2ςτς+τς120[ζ(2ςτς+τ)+ζ(ςτwς+τw)]ϑ(w)w2dιdw. (2.25)

    By using Lemma 2, we observe that the following inequalities hold for all ι[0,12] and w[ς,2ςτς+τ]:

    4ζ(2ςτς+τ)2[ζ(4ςτw2ςτ+(ς+τ)w)+ζ(4ςτw3(ς+τ)w2ςτ)], (2.26)
    2ζ(4ςτw2ςτ+(ς+τ)w)ζ(2ςτwι(ς+τ)w+2(1ι)ςτ)+ζ(2ςτw2ςτι+(1ι)(ς+τ)w), (2.27)
    2ζ(4ςτw3(ς+τ)w2ςτ)ζ(2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w)+ζ(2ςτw2(1ι)(ςw+τwςτ)+ι(ς+τ)w), (2.28)
    ζ(2ςτwι(ς+τ)w+2(1ι)ςτ)+ζ(2ςτw2ςτι+(1ι)(ς+τ)w)ζ(w)+ζ(2ςτς+τ) (2.29)

    and

    ζ(2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w)+ζ(2ςτw2(1ι)(ςw+τwςτ)+ι(ς+τ)w)ζ(2ςτς+τ)+ζ(ςτwς+τw). (2.30)

    Multiplying the inequalities (2.26)–(2.30) by ϑ(w)w2 and integrating them over ι on [0,12], over w on [ς,2ςτς+τ] and using identities (2.22)–(2.25), we derive (2.19).

    (ii) Since ζ:[ς,τ]R is harmonic convex on [ς,τ], hence ˘h:[1τ,1ς]R defined by ˘h(w)=ζ(1w) is convex on [1τ,1ς]. Thus, by integration by parts, we get that following identity holds:

    ς+τ2ςτ1τ(ς+τ2ςτw)[˘h(1ς+1τw)˘h(w)]dw=(τς2ςτ)[˘h(1ς)+˘h(1τ)]ς+τ2ςτ1τ[˘h(1ς+1τw)+˘h(w)]dw. (2.31)

    The equality (2.31) is equivalent to the equality:

    2ςτς+τς1w2(1wς+τ2ςτ)[ζ(11ς+1τ1w)(1ς+1τ1w)2w2ζ(w)]dw=(τςςτ)[ζ(ς)+ζ(τ)2]τςζ(w)w2dw. (2.32)

    Using substitution rules for integration and the hypothesis of ϑ, we have the following identities:

    τςζ(w)ϑ(w)w2dw=2ςτς+τς[ζ(w)+ζ(11ς+1τ1w)]ϑ(w)w2dw (2.33)

    and

    Sϑ(ι)=2ςτς+τς[ζ(2ςτw2ςτι+(1ι)(ς+τ)w)+ζ(2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w)]ϑ(w)w2dw. (2.34)

    Now, using the convexity of ˘h(w)=ζ(1w) on [1τ,1ς] and the hypothesis of ϑ, the following inequality holds for all ι[0,1] and w[1τ,ς+τ2ςτ]:

    [˘h(w)˘h(ιw+(1ι)(ς+τ2ςτ))]ϑ(1w)+[˘h(1ς+1τw)˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))]ϑ(1w)(1ι)(wς+τ2ςτ)˘h(w)ϑ(1w)+(1ι)(ς+τ2ςτw)˘h(1ς+1τw)ϑ(1w)=(1ι)(ς+τ2ςτw)[˘h(1ς+1τw)˘h(w)]ϑ(1w) (2.35)

    which is equivalent to

    [ζ(1w)ζ(1ιw+(1ι)(ς+τ2ςτ))]ϑ(1w)+[ζ(11ς+1τw)ζ(1ι(1ς+1τw)+(1ι)(ς+τ2ςτ))]ϑ(1w)(1ι)(ς+τ2ςτw)[ζ(1w)w2ζ(11ς+1τw)(1ς+1τw)2]ϑ(1w). (2.36)

    Integrating the above inequalities over w on [1τ,ς+τ2ςτ], we get

    ς+τ2ςτ1τ[ζ(1w)ζ(1ιw+(1ι)(ς+τ2ςτ))]ϑ(1w)dw+ς+τ2ςτ1τ[ζ(11ς+1τw)ζ(1ι(1ς+1τw)+(1ι)(ς+τ2ςτ))]ϑ(1w)dw(1ι)ς+τ2ςτ1τ(ς+τ2ςτw)[ζ(1w)w2ζ(11ς+1τw)(1ς+1τw)2]ϑ(1w)dw. (2.37)

    After making use of suitable substitution, the inequality (2.37) takes the form:

    2ςτς+τς1w2[ζ(1ι1w+(1ι)(ς+τ2ςτ))ζ(w)]ϑ(w)dw+2ςτς+τς1w2[ζ(1ι(1ς+1τ1w)+(1ι)(ς+τ2ςτ))ζ(11ς+1τ1w)]ϑ(w)dwϑ(1ι)2ςτς+τς1w2(1wς+τ2ςτ)[ζ(11ς+1τ1w)(1ς+1τ1w)2w2ζ(w)]dw. (2.38)

    Inequality (2.20) follows from (2.31)–(2.34) and (2.38).

    (iii) We use the fact that ζ:[ς,τ]R is harmonic convex on [ς,τ], hence ˘h:[1τ,1ς]R defined by ˘h(w)=ζ(1w) is convex on [1τ,1ς]. Thus

    ˘h(1τ)˘h(ς+τ2ςτ)2ςτ4ςτ˘h(1τ)

    and

    ˘h(1ς)˘h(ς+τ2ςτ)2τς4ςτ˘h(1ς).

    Adding the above inequalities

    ˘h(1ς)+˘h(1τ)2˘h(ς+τ2ςτ)(τς)(˘h(1ς)˘h(1τ))4ςτ. (2.39)

    The inequality (2.39) becomes

    ζ(ς)+ζ(τ)2ζ(2ςτς+τ)(τς)(τ2ζ(τ)ς2ζ(ς))4ςτ. (2.40)

    Multiplying (2.40) both sides by ϑ(w)w2 and integrating over [ς,τ], we get

    ζ(ς)+ζ(τ)2τςϑ(w)w2dwζ(2ςτς+τ)τςϑ(w)w2dw(τς)(τ2ζ(τ)ς2ζ(ς))4ςττςϑ(w)w2dw. (2.41)

    From (2.17) and (2.41) we get (2.21).

    Corollary 1. Suppose that the assumption of Theorem 17 are satisfiedand ϑ(w)=ςττς, w[ς,τ], then

    (i) The inequalities

    ζ(2ςτς+τ)24ςτ3ς+τ4ςτς+3τζ(w)w210S(ι)dι12[ζ(2ςτς+τ)+ςττςτςζ(w)w2dw] (2.42)

    holds.

    (ii) The inequalities

    0ςττςτςζ(w)w2dwS(ι)(1ι)[ζ(ς)+ζ(τ)2ςττςτςζ(w)w2dw] (2.43)

    hold for all ι[0,1].

    (iii) The inequalities

    0ζ(ς)+ζ(τ)2S(ι)(τς)(τ2ζ(τ)ς2ζ(ς))4ςτ (2.44)

    are valid for all ι[0,1].

    In the following theorems, we discuss inequalities for the functions S, Sϑ, ˘G1, T and Tϑ as considered above:

    Theorem 18. Let ζ, ϑ, ˘G1, Sϑ be defined as above. Then

    (i) The inequality

    Sϑ(ι)˘G1(ι)τςϑ(w)w2dw (2.45)

    holds for all ι[0,1].

    (ii) The inequalities

    24ςτ3ς+τ4ςτς+3τζ(w)ϑ(2ςτw4ςτ(ς+τ)w)ϑ(w)w2dw12[ζ(4ςτ3ς+τ)+ζ(4ςτς+3τ)]τςϑ(w)w2dwτςςτ10λ1(ι)ϑ(ςτ(1ι)ς+ιτ)dι12[ζ(2ςτς+τ)+ζ(ς)+ζ(τ)2]τςϑ(w)w2dw (2.46)

    hold.

    (iii) If ζ is differentiable on [ς,τ] and ϑ isbounded on [ς,τ], then, forall ι[0,1], then

    0Sϑ(ι)ζ(2ςτς+τ)τςϑ(w)w2dw(τςςτ)[˘G1(ι)S(ι)]ϑ, (2.47)

    where ϑ=supw[ς,τ]ϑ(w).

    Proof. (i) Using integration by substitution and the assumptions on ϑ, we have that the following identity holds on [0,1]:

    ˘G1(ι)τςϑ(w)w2dw=2ςτς+τς[ζ(2ςτ2ςι+(1ι)(ς+τ))+ζ(2ςτ2τι+(1ι)(ς+τ))]ϑ(w)w2dw. (2.48)

    By Lemma 2, the following inequality holds for all w[ς,2ςτς+τ] with

    w1=2ςτw2ςτι+(1ι)(ς+τ)w,w2=2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w,˘ϰ1=2ςτ2τι+(1ι)(ς+τ)and˘ϰ2=2ςτ2ςι+(1ι)(ς+τ):
    ζ(2ςτw2ςτι+(1ι)(ς+τ)w)+ζ(2ςτw2ι(ςw+τwςτ)+(1ι)(ς+τ)w)ζ(2ςτ2τι+(1ι)(ς+τ))+ζ(2ςτ2ςι+(1ι)(ς+τ)). (2.49)

    Multiplying both sides of (2.49) with ϑ(w)w2, integrating over [ς,2ςτς+τ] and using (2.34) and (2.49), we obtain (2.45).

    (ii) We can observe that

    12[ζ(4ςτ3ς+τ)+ζ(4ςτς+3τ)]τςϑ(w)w2dw=[ζ(4ςτ3ς+τ)+ζ(4ςτς+3τ)]2ςτς+τςϑ(w)w2dw. (2.50)

    By using harmonic symmetric assumption on ϑ, we get

    24ςτ3ς+τ4ςτς+3τζ(w)ϑ(2ςτw4ςτ(ς+τ)w)ϑ(w)w2dw=2ςτς+τς[ζ(4ςτw2ςτ+(ς+τ)w)+ζ(4ςτw3(ς+τ)w2ςτ)]ϑ(w)w2dw. (2.51)

    We can also see that the following identity holds:

    τςςτ10˘G1(ι)ϑ(ςτ(1ι)ς+ιτ)dι=τςςτ×[112ζ(12ςι+(1ι)(ς+τ))ϑ(ςτ(1ι)ς+ιτ)dι+120ζ(12ςι+(1ι)(ς+τ))ϑ(ςτ(1ι)τ+ις)dι+120ζ(12τι+(1ι)(ς+τ))ϑ(ςτ(1ι)τ+ις)dι+112ζ(12τι+(1ι)(ς+τ))ϑ(ςτ(1ι)ς+ιτ)dι]=122ςτς+τς[ζ(2wςς+w)+ζ(2ςτw2ςw+τwςτ)+ζ(2τwτ+w)+ζ(2ςτwςw+2τwςτ)]ϑ(w)w2dw. (2.52)

    Finally we also have

    12[ζ(2ςτς+τ)+ζ(ς)+ζ(τ)2]τςϑ(w)w2dw=[ζ(2ςτς+τ)+ζ(ς)+ζ(τ)2]2ςτς+τςϑ(w)w2dw. (2.53)

    By Lemma 2, the following inequalities hold for all w[ς,2ςτς+τ]:

    The inequality

    ζ(4ςτw2ςτ+(ς+τ)w)+ζ(4ςτw3(ς+τ)w2ςτ)ζ(4ςτ3ς+τ)+ζ(4ςτς+3τ) (2.54)

    holds with the choices of w1=4ςτw2ςτ+(ς+τ)w, w2=4ςτw3(ς+τ)w2ςτ, ˘ϰ1=4ςτ3ς+τ and ˘ϰ2=4ςτς+3τ.

    The inequality

    ζ(4ςτ3ς+τ)12[ζ(2ςτw2ςw+τwςτ)+ζ(2ςwς+w)] (2.55)

    holds with the choices of w1=w2=4ςτ3ς+τ, ˘ϰ1=2ςwς+w, ˘ϰ2=2ςτw2ςw+τwςτ.

    The inequality

    ζ(4ςτς+3τ)12[ζ(2ςτwςw+2τwςτ)+ζ(2τwτ+w)] (2.56)

    holds with the choices of w1=w2=4ςτς+3τ, ˘ϰ1=2τwτ+w, ˘ϰ2=2ςτwςw+2τwςτ.

    The inequality

    ζ(2ςτw2ςw+τwςτ)+ζ(2ςwς+w)ζ(ς)+ζ(2τςς+τ) (2.57)

    holds with the choices of w1=2ςwς+w, w2=2ςτw2ςw+τwςτ, ˘ϰ1=ς, ˘ϰ2=2τςς+τ.

    The inequality

    ζ(2ςτwςw+2τwςτ)+ζ(2τwτ+w)ζ(2τςς+τ)+ζ(τ) (2.58)

    holds with the choices of w1=2τwτ+w, w2=2ςτwςw+2τwςτ, ˘ϰ1=2τςς+τ, ˘ϰ2=τ.

    Multiplying (2.54)–(2.58) by ϑ(w), integrating them over [ς,2ςτς+τ] and using (2.50)–(2.53), we get (2.46).

    (iii) By integration by parts, we get

    ις+τ2ςτ1τ[(wς+τ2ςτ)˘h(ιw+(1ι)(ς+τ2ςτ))+(ς+τ2ςτw)˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))]dw=ι1ς1τ(wς+τ2ςτ)˘h(ιw+(1ι)(ς+τ2ςτ))dw=τς2ςτ[ζ(2ςτ2τι+(1ι)(ς+τ))+ζ(2ςτ2ςι+(1ι)(ς+τ))]τς1w2ζ(2ςτw2ςτι+(1ι)(ς+τ))dw=(τςςτ)[˘G1(ι)S(ι)]. (2.59)

    Using the convexity of ˘h and the hypothesis of ϑ, the inequality holds for all ι[0,1] and w[1τ,ς+τ2ςτ]:

    [˘h(ιw+(1ι)(ς+τ2ςτ))˘h(ς+τ2ςτ)]ϑ(1w)+[˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))˘h(ς+τ2ςτ)]ϑ(1w)ι(wς+τ2ςτ)˘h(ιw+(1ι)(ς+τ2ςτ))ϑ(1w)+ι(ς+τ2ςτw)˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))ϑ(1w)=ι(ς+τ2ςτw)[˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))˘h(ιw+(1ι)(ς+τ2ςτ))]ϑ(1w)ι(ς+τ2ςτw)[˘h(ι(1ς+1τw)+(1ι)(ς+τ2ςτ))˘h(ιw+(1ι)(ς+τ2ςτ))]ϑ. (2.60)

    Integrating (2.60), using (2.59) and (2.17), we get (2.47).

    Corollary 2. According to the assumptions of Theorem 18 with ϑ(w)=ςττς, w[ς,τ], then

    (i) The inequality

    S(ι)˘G1(ι)

    holds for all ι[0,1].

    (ii) The inequalities

    2ςττς4ςτ3ς+τ4ςτς+3τζ(w)ϑ(2ςτw4ςτ(ς+τ)w)dww212[ζ(4ςτ3ς+τ)+ζ(4ςτς+3τ)]τςςτ10˘G1(ι)×ϑ(ςτ(1ι)ς+ιτ)dι12[ζ(2ςτς+τ)+ζ(ς)+ζ(τ)2] (2.61)

    hold.

    (iii) The inequality

    0S(ι)ζ(2ςτς+τ)dw(τςςτ)[˘G1(ι)S(ι)] (2.62)

    holds for all ι[0,1].

    Theorem 19. Let ζ, ϑ, ˘G1, Sϑ, Tϑ bedefined as above. Then

    (i) Tϑ is harmonic convex on (0,1].

    (ii) The inequalities

    ˘G1(ι)τςϑ(w)w2dwTϑ(ι)(1ι)τςζ(w)ϑ(w)w2dw+ιζ(ς)+ζ(τ)2τςϑ(w)w2dwζ(ς)+ζ(τ)2τςϑ(w)w2dw, (2.63)
    Sϑ(1ι)Tϑ(ι) (2.64)

    and

    Sϑ(ι)+Sϑ(1ι)2Tϑ(ι) (2.65)

    hold for all \mathfrak{\iota}\in\left[ 0, 1\right] .

    (iii) The following bound is true:

    \begin{equation} \underset{\mathfrak{\iota}\in\left[ 0,1\right] }{\sup}\mathbb{T} _{\mathbb{\vartheta}}\left( \mathfrak{\iota}\right) = \frac{\mathbb{\zeta }\left( \mathfrak{\varsigma}\right) +\mathbb{\zeta}\left( \mathbb{\tau }\right) }{2}\int_{\mathfrak{\varsigma}}^{\mathbb{\tau}}\frac {\mathbb{\vartheta}\left( \mathfrak{w}\right) }{\mathfrak{w}^{2} }d\mathfrak{w}. \end{equation} (2.66)

    Proof. (i) Since \mathbb{\zeta} is harmonic convex and \mathbb{\vartheta} is nonnegative, we see that \mathbb{T}_{\mathbb{\vartheta}} is harmonic convex on \left(0, 1\right] .

    (ii) We observe that the following identity holds on \left[ 0, 1\right] :

    \begin{array}{l} \mathbb{T}_{\mathbb{\vartheta}}\left( \mathfrak{\iota}\right) = \frac{1} {2}\int_{\mathfrak{\varsigma}}^{\frac{2\mathfrak{\varsigma}\mathbb{\tau} }{\mathfrak{\varsigma}+\mathbb{\tau}}}\left[ \mathbb{\zeta}\left( \frac{\mathbb{\tau}\mathfrak{w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota }\right) \mathbb{\tau}}\right) +\mathbb{\zeta}\left( \frac {\mathfrak{\varsigma}\mathbb{\tau}\mathfrak{w}}{\mathfrak{\varsigma w\iota }+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w}-\mathfrak{\varsigma}\mathbb{\tau}\right) }\right) \right. \\ \left. +\mathbb{\zeta}\left( \frac{\mathfrak{\varsigma w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}}\right) +\mathbb{\zeta}\left( \frac{\mathfrak{\varsigma}\mathbb{\tau}\mathfrak{w} }{\mathbb{\tau}\mathfrak{w\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w}-\mathfrak{\varsigma }\mathbb{\tau}\right) }\right) \right] \mathbb{\vartheta}\left( \mathfrak{w}\right) d\mathfrak{w}. \end{array} (2.67)

    By Lemma 2, the following inequalities hold for all \mathfrak{w} \in\left[ \mathfrak{\varsigma}, \frac{2\mathfrak{\varsigma}\mathbb{\tau} }{\mathfrak{\varsigma}+\mathbb{\tau}}\right] :

    \begin{array}{l} 2\mathbb{\zeta}\left( \frac{2\mathfrak{\varsigma}\mathbb{\tau}} {2\mathfrak{\varsigma\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma}+\mathbb{\tau}\right) }\right) \\ \leq\mathbb{\zeta}\left( \frac{\mathbb{\tau}\mathfrak{w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathbb{\tau}}\right) +\mathbb{\zeta }\left( \frac{\mathfrak{\varsigma}\mathbb{\tau}\mathfrak{w}} {\mathfrak{\varsigma w\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w}-\mathfrak{\varsigma }\mathbb{\tau}\right) }\right) \end{array} (2.68)

    with

    \begin{align*} \mathfrak{w}_{1} & = \mathfrak{w}_{2} = \frac{2\mathfrak{\varsigma} \mathbb{\tau}}{2\mathfrak{\varsigma\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma}+\mathbb{\tau}\right) }{, }\breve{\varkappa }_{1} = \frac{\mathbb{\tau}\mathfrak{w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathbb{\tau}}\\ {\rm{and }}\;\breve{\varkappa}_{2} & = \frac{\mathfrak{\varsigma}\mathbb{\tau }\mathfrak{w}}{\mathfrak{\varsigma w\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w}-\mathfrak{\varsigma }\mathbb{\tau}\right) }. \end{align*}
    \begin{array}{l} 2\mathbb{\zeta}\left( \frac{2\mathfrak{\varsigma}\mathbb{\tau}} {2\mathbb{\tau}\mathfrak{\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma}+\mathbb{\tau}\right) }\right) \\ \leq\mathbb{\zeta}\left( \frac{\mathfrak{\varsigma w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}}\right) +\mathbb{\zeta}\left( \frac{\mathfrak{\varsigma}\mathbb{\tau}\mathfrak{w} }{\mathbb{\tau}\mathfrak{w\iota}+\left( 1-\mathfrak{\iota}\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w}-\mathfrak{\varsigma }\mathbb{\tau}\right) }\right) \end{array} (2.69)

    with

    \begin{align*} \mathfrak{w}_{1} & = \mathfrak{w}_{2} = \frac{2\mathfrak{\varsigma} \mathbb{\tau}}{2\mathbb{\tau}\mathfrak{\iota}+\left( 1-\mathfrak{\iota }\right) \left( \mathfrak{\varsigma}+\mathbb{\tau}\right) }{, } \breve{\varkappa}_{1} = \frac{\mathfrak{\varsigma w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}}\\ {\rm{and }}\;\breve{\varkappa}_{2} & = \frac{\mathfrak{\varsigma}\mathbb{\tau }\mathfrak{w}}{\mathbb{\tau}\mathfrak{w\iota}+\left( 1-\mathfrak{\iota }\right) \left( \mathfrak{\varsigma w}+\mathbb{\tau}\mathfrak{w} -\mathfrak{\varsigma}\mathbb{\tau}\right) }. \end{align*}

    Multiplying the inequalities (2.68) and (2.69) by \mathbb{\vartheta }\left(\mathfrak{w}\right) , integrating them over \mathfrak{w} on \left[ \mathfrak{\varsigma}, \frac{2\mathfrak{\varsigma}\mathbb{\tau} }{\mathfrak{\varsigma}+\mathbb{\tau}}\right] and using identities (2.48) and (2.67), we derive the first inequality of (2.63). Using the harmonic convexity of \mathbb{\zeta} and the inequality (2.17), the last part of (2.63) holds. Using again the harmonic convexity of \mathbb{\zeta} , we get

    \begin{array}{l} \mathbb{S}_{\mathbb{\vartheta}}\left( 1-\mathfrak{\iota}\right) = { \int\limits_{\mathfrak{\varsigma}}^{\mathbb{\tau}}} \mathbb{\zeta}\left( \frac{2\mathfrak{\varsigma}\mathbb{\tau}\mathfrak{w} }{2\mathfrak{\varsigma}\mathbb{\tau}\left( 1-\mathfrak{\iota}\right) +\mathfrak{\iota}\left( \mathfrak{\varsigma}+\mathbb{\tau}\right) \mathfrak{w}}\right) \frac{\mathbb{\vartheta}\left( \mathfrak{w}\right) }{\mathfrak{w}^{2}}d\mathfrak{w} \\ = { \int\limits_{\mathfrak{\varsigma}}^{\mathbb{\tau}}} \mathbb{\zeta}\left( \frac{1}{\frac{1}{2}\left( \frac{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}} {\mathfrak{\varsigma w}}\right) +\frac{1}{2}\left( \frac{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathbb{\tau}}{\mathbb{\tau} \mathfrak{w}}\right) }\right) \frac{\mathbb{\vartheta}\left( \mathfrak{w} \right) }{\mathfrak{w}^{2}}d\mathfrak{w}\\ \leq\frac{1}{2} { \int\limits_{\mathfrak{\varsigma}}^{\mathbb{\tau}}} \left[ \mathbb{\zeta}\left( \frac{\mathfrak{\varsigma w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}}\right) +\mathbb{\zeta}\left( \frac{\mathfrak{\varsigma w}}{\mathfrak{\iota w}+\left( 1-\mathfrak{\iota}\right) \mathfrak{\varsigma}}\right) \right] \frac{\mathbb{\vartheta}\left( \mathfrak{w}\right) }{\mathfrak{w}^{2} }d\mathfrak{w} = \mathbb{T}_{\mathbb{\vartheta}}\left( \mathfrak{\iota}\right) . \end{array} (2.70)

    From (2.45), (2.63) and 2.70), we get (2.65).

    (iii) (2.66) holds due to the inequality (2.63).

    The subject of mathematical inequalities using convex functions has been seen to be an emerging topic during the past more than three decades. The researchers are trying to find new generalizations of convex functions and as a result new results are being adding to the theory of inequalities. In the current research we have used harmonic convex functions to generalize a number of results that hold for convex functions. In order to get the novel results in this study, we defined some new mappings over the interval [0, 1] . We have discussed some interesting properties of these mappings and obtained new refinements of the Hermite-hadamard and Fejér type inequalities already proven for harmonic convex functions. We believe that the results of this paper could be a source of inspiration for mathematicians working in this field and young researchers thinking to start their career in this fascinating field of mathematics.

    The author is very thankful to all the anonymous referees for their very useful and constructive comments in order to present the paper in the present form. This work is supported by the Deanship of Scientific Research, King Faisal University under the Nasher Track 2021 (Research Project Number NA000177) which has been converted to Ambitious Researcher Track (Research Project Number GRANT931).

    The authors declare that there are no conflicts of interest regarding the publication of this article.



    [1] J. Serrin, A symmetry problem in potential theory, Arch. Ration. Mech. Anal., 43 (1971), 304–318. https://doi.org/10.1007/BF00250468 doi: 10.1007/BF00250468
    [2] B. Gidas, Symmetry of positive solutions of nonlinear elliptic equations in R^{n}, Adv. Math. Suppl. Stud., 7 (1981), 369–402.
    [3] B. Gidas, W. M. Ni, L. Nirenberg, Symmetry and related properties via the maximum principle, Commun. Math. Phys., 68 (1979), 209–243. https://doi.org/10.1007/BF01221125 doi: 10.1007/BF01221125
    [4] H. Berestycki, L. Nirenberg, On the method of moving planes and the sliding method, Bol. Soc. Bras. Mat., 22 (1991), 1–37. https://doi.org/10.1007/BF01244896 doi: 10.1007/BF01244896
    [5] L. Damascelli, F. Pacella, Monotonicity and symmetry of solutions of p-Laplace equations, 1 < p < 2, via the moving plane method, Ann. Sc. Norm. Super. Pisa Cl. Sci., 26 (1998), 689–707.
    [6] L. Damascelli, F. Pacella, Monotonicity and symmetry results for p-Laplace equations and applications, Adv. Differ. Equations, 5 (2000), 1179–1200. https://doi.org/10.57262/ade/1356651297 doi: 10.57262/ade/1356651297
    [7] L. Damascelli, F. Pacella, M. Ramaswamy, Symmetry of ground states of p-Laplace equations via the moving plane method, Arch. Ration. Mech. Anal., 148 (1999), 291–308. https://doi.org/10.1007/s002050050163 doi: 10.1007/s002050050163
    [8] J. Serrin, H. Zou, Symmetry of ground states of quasilinear elliptic equations, Arch. Ration. Mech. Anal., 148 (1999), 265–290. https://doi.org/10.1007/s002050050162 doi: 10.1007/s002050050162
    [9] J. M. do Ó, R. da Costa, Symmetry properties for nonnegative solutions of non-uniformally elliptic equations in hyperbolic spaces, J. Math. Anal. Appl., 435 (2016), 1753–1711. https://doi.org/10.1016/j.jmaa.2015.11.031 doi: 10.1016/j.jmaa.2015.11.031
    [10] F. Pacella, Symmetry resluts for solutions of semilinear elliptic equations with convex nonlinearities, J. Funct. Anal., 1992 (2002), 271–282. https://doi.org/10.1006/jfan.2001.3901 doi: 10.1006/jfan.2001.3901
    [11] O. A. Ladyzhenskaya, Linear and quasilinear elliptic equations, New York: Academic Press, 1968.
    [12] J. Simon, Régularité de la solution d'un problème aux limites non linéaires, Ann. Fac. Sci. Toulouse, 3 (1981), 247–274. https://doi.org/10.5802/AFST.569 doi: 10.5802/AFST.569
    [13] L. Damascelli, Comparison theorems for some quasilinear degenerate elliptic operators and applications to symmetry and monotonicity results, Ann. Inst. H. Poincaré Anal. Non Linéaire, 15 (1998), 493–516. https://doi.org/10.1016/S0294-1449(98)80032-2 doi: 10.1016/S0294-1449(98)80032-2
    [14] M. Badiale, E. Nabana, A note on radiality of solutions of p-Laplacian, Appl. Anal., 52 (1994), 35–43. https://doi.org/10.1080/00036819408840222 doi: 10.1080/00036819408840222
  • This article has been cited by:

    1. Takao Komatsu, Two types of hypergeometric degenerate Cauchy numbers, 2020, 18, 2391-5455, 417, 10.1515/math-2020-0030
    2. Takao Komatsu, Continued fraction expansions of the generating functions of Bernoulli and related numbers, 2020, 31, 00193577, 695, 10.1016/j.indag.2020.06.006
    3. Takao Komatsu, Wenpeng Zhang, Several expressions of truncated Bernoulli-Carlitz and truncated Cauchy-Carlitz numbers, 2020, 5, 2473-6988, 5939, 10.3934/math.2020380
    4. Takao Komatsu, Ram Krishna Pandey, On hypergeometric Cauchy numbers of higher grade, 2021, 6, 2473-6988, 6630, 10.3934/math.2021390
    5. Narakorn Rompurk Kanasri, Takao Komatsu, Vichian Laohakosol, Cameron’s operator in terms of determinants and hypergeometric numbers, 2022, 28, 1405-213X, 10.1007/s40590-021-00401-8
    6. Beáta Bényi, Toshiki Matsusaka, Combinatorial aspects of poly-Bernoulli polynomials and poly-Euler numbers, 2023, 34, 2118-8572, 917, 10.5802/jtnb.1234
    7. James C. Fu, Wan-Chen Lee, Hsing-Ming Chang, On Distribution of the Number of Peaks and the Euler Numbers of Permutations, 2023, 25, 1387-5841, 10.1007/s11009-023-09987-0
    8. Takao Komatsu, Guo-Dong Liu, Congruence properties of Lehmer-Euler numbers, 2025, 0001-9054, 10.1007/s00010-024-01150-5
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1610) PDF downloads(75) Cited by(0)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog