{"title":"关于Hilbert第19问题中所有极值的Hölder正则性","authors":"F. Tomi, A. Tromba","doi":"10.1515/acv-2021-0089","DOIUrl":null,"url":null,"abstract":"Abstract Let Ω ⊂ ℝ n {\\Omega\\subset\\mathbb{R}^{n}} be a C 1 {C^{1}} smooth compact domain. Furthermore, let F : Ω × ℝ n N → ℝ {F:\\Omega\\times\\mathbb{R}^{nN}\\to\\mathbb{R}} , F ( x , p ) {F(x,p)} , be C 0 {C^{0}} , differentiable with respect to p, and with F p := D p F {F_{p}:=D_{p}F} continuous on Ω × ℝ n N {\\Omega\\times\\mathbb{R}^{nN}} and F strictly convex in p. Consider an n N × n N {nN\\times nN} matrix A = ( A α β i j ) ∈ C 0 ( Ω ) {A=(A^{{ij}}_{\\alpha\\beta})\\in C^{0}(\\Omega)} satisfying (0.1) A α β i j ( x ) ξ α i ξ β j = A β α j i ( x ) ξ α i ξ β j ≥ λ | ξ | 2 , λ > 0 . A^{ij}_{\\alpha\\beta}(x)\\xi^{i}_{\\alpha}\\xi^{j}_{\\beta}=A^{ji}_{\\beta\\alpha}(x)% \\xi^{i}_{\\alpha}\\xi^{j}_{\\beta}\\geq\\lambda\\lvert\\xi\\rvert^{2},\\quad\\lambda>0. Suppose that (0.2) lim | p | → ∞ 1 | p | ( D p F ( x , p ) - A ( x ) p ) = 0 , \\displaystyle\\lim_{\\lvert p\\rvert\\to\\infty}\\frac{1}{\\lvert p\\rvert}(D_{p}F(x,p% )-A(x)p)=0, (0.3) - C 0 + c 0 | p | 2 ≤ F ( x , p ) ≤ C 0 ( 1 + | p | 2 ) , \\displaystyle{-}C_{0}+c_{0}\\lvert p\\rvert^{2}\\leq F(x,p)\\leq C_{0}(1+\\lvert p% \\rvert^{2}), (0.4) | F p ( x , p ) - F p ( x , q ) | ≤ C 0 | p - q | , \\displaystyle\\lvert F_{p}(x,p)-F_{p}(x,q)\\rvert\\leq C_{0}\\lvert p-q\\rvert, (0.5) 〈 F p ( x , p ) - F p ( x , q ) , p - q 〉 ≥ c 0 | p - q | 2 \\displaystyle\\langle F_{p}(x,p)-F_{p}(x,q),p-q\\rangle\\geq c_{0}\\lvert p-q% \\rvert^{2} uniformly in x and with positive constants c 0 {c_{0}} and C 0 {C_{0}} . Consider the functional (0.6) J ( u ) := ∫ Ω F ( x , D u ( x ) ) 𝑑 x + ∫ Ω G ( x , u ) 𝑑 x , J(u):=\\int_{\\Omega}F(x,Du(x))\\,dx+\\int_{\\Omega}G(x,u)\\,dx, where G ( x , ⋅ ) ∈ C 1 ( ℝ N ) {G(x,\\cdot\\,)\\in C^{1}(\\mathbb{R}^{N})} for each x ∈ Ω {x\\in\\Omega} , G ( ⋅ , u ) {G(\\,\\cdot\\,,u)} is measurable for each u ∈ ℝ N {u\\in\\mathbb{R}^{N}} , and (0.7) | G u ( x , u ) | ≤ C 0 ( 1 + | u | s ) \\lvert G_{u}(x,u)\\rvert\\leq C_{0}(1+\\lvert u\\rvert^{s}) with s < n + 2 n - 2 {s<\\frac{n+2}{n-2}} . Under these conditions, we shall show that if n > 2 {n>2} , then any weak solution u ∈ W 1 , 2 ( Ω , ℝ N ) {u\\in W^{1,2}(\\Omega,\\mathbb{R}^{N})} of the Euler equations of J, i.e. ∑ α ∂ ∂ x α F p α i ( x , D u ) = G u i ( x , u ) , i = 1 , … , N , \\sum_{\\alpha}\\frac{\\partial}{\\partial x^{\\alpha}}F_{p^{i}_{\\alpha}}(x,Du)=G_{u% ^{i}}(x,u),\\quad i=1,\\ldots,N, is Hölder continuous in the interior of Ω and under appropriate boundary conditions also Hölder continuous up to the boundary.","PeriodicalId":49276,"journal":{"name":"Advances in Calculus of Variations","volume":" ","pages":""},"PeriodicalIF":1.3000,"publicationDate":"2022-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"On the Hölder regularity of all extrema in Hilbert’s 19th Problem\",\"authors\":\"F. Tomi, A. Tromba\",\"doi\":\"10.1515/acv-2021-0089\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Let Ω ⊂ ℝ n {\\\\Omega\\\\subset\\\\mathbb{R}^{n}} be a C 1 {C^{1}} smooth compact domain. Furthermore, let F : Ω × ℝ n N → ℝ {F:\\\\Omega\\\\times\\\\mathbb{R}^{nN}\\\\to\\\\mathbb{R}} , F ( x , p ) {F(x,p)} , be C 0 {C^{0}} , differentiable with respect to p, and with F p := D p F {F_{p}:=D_{p}F} continuous on Ω × ℝ n N {\\\\Omega\\\\times\\\\mathbb{R}^{nN}} and F strictly convex in p. Consider an n N × n N {nN\\\\times nN} matrix A = ( A α β i j ) ∈ C 0 ( Ω ) {A=(A^{{ij}}_{\\\\alpha\\\\beta})\\\\in C^{0}(\\\\Omega)} satisfying (0.1) A α β i j ( x ) ξ α i ξ β j = A β α j i ( x ) ξ α i ξ β j ≥ λ | ξ | 2 , λ > 0 . A^{ij}_{\\\\alpha\\\\beta}(x)\\\\xi^{i}_{\\\\alpha}\\\\xi^{j}_{\\\\beta}=A^{ji}_{\\\\beta\\\\alpha}(x)% \\\\xi^{i}_{\\\\alpha}\\\\xi^{j}_{\\\\beta}\\\\geq\\\\lambda\\\\lvert\\\\xi\\\\rvert^{2},\\\\quad\\\\lambda>0. Suppose that (0.2) lim | p | → ∞ 1 | p | ( D p F ( x , p ) - A ( x ) p ) = 0 , \\\\displaystyle\\\\lim_{\\\\lvert p\\\\rvert\\\\to\\\\infty}\\\\frac{1}{\\\\lvert p\\\\rvert}(D_{p}F(x,p% )-A(x)p)=0, (0.3) - C 0 + c 0 | p | 2 ≤ F ( x , p ) ≤ C 0 ( 1 + | p | 2 ) , \\\\displaystyle{-}C_{0}+c_{0}\\\\lvert p\\\\rvert^{2}\\\\leq F(x,p)\\\\leq C_{0}(1+\\\\lvert p% \\\\rvert^{2}), (0.4) | F p ( x , p ) - F p ( x , q ) | ≤ C 0 | p - q | , \\\\displaystyle\\\\lvert F_{p}(x,p)-F_{p}(x,q)\\\\rvert\\\\leq C_{0}\\\\lvert p-q\\\\rvert, (0.5) 〈 F p ( x , p ) - F p ( x , q ) , p - q 〉 ≥ c 0 | p - q | 2 \\\\displaystyle\\\\langle F_{p}(x,p)-F_{p}(x,q),p-q\\\\rangle\\\\geq c_{0}\\\\lvert p-q% \\\\rvert^{2} uniformly in x and with positive constants c 0 {c_{0}} and C 0 {C_{0}} . Consider the functional (0.6) J ( u ) := ∫ Ω F ( x , D u ( x ) ) 𝑑 x + ∫ Ω G ( x , u ) 𝑑 x , J(u):=\\\\int_{\\\\Omega}F(x,Du(x))\\\\,dx+\\\\int_{\\\\Omega}G(x,u)\\\\,dx, where G ( x , ⋅ ) ∈ C 1 ( ℝ N ) {G(x,\\\\cdot\\\\,)\\\\in C^{1}(\\\\mathbb{R}^{N})} for each x ∈ Ω {x\\\\in\\\\Omega} , G ( ⋅ , u ) {G(\\\\,\\\\cdot\\\\,,u)} is measurable for each u ∈ ℝ N {u\\\\in\\\\mathbb{R}^{N}} , and (0.7) | G u ( x , u ) | ≤ C 0 ( 1 + | u | s ) \\\\lvert G_{u}(x,u)\\\\rvert\\\\leq C_{0}(1+\\\\lvert u\\\\rvert^{s}) with s < n + 2 n - 2 {s<\\\\frac{n+2}{n-2}} . Under these conditions, we shall show that if n > 2 {n>2} , then any weak solution u ∈ W 1 , 2 ( Ω , ℝ N ) {u\\\\in W^{1,2}(\\\\Omega,\\\\mathbb{R}^{N})} of the Euler equations of J, i.e. ∑ α ∂ ∂ x α F p α i ( x , D u ) = G u i ( x , u ) , i = 1 , … , N , \\\\sum_{\\\\alpha}\\\\frac{\\\\partial}{\\\\partial x^{\\\\alpha}}F_{p^{i}_{\\\\alpha}}(x,Du)=G_{u% ^{i}}(x,u),\\\\quad i=1,\\\\ldots,N, is Hölder continuous in the interior of Ω and under appropriate boundary conditions also Hölder continuous up to the boundary.\",\"PeriodicalId\":49276,\"journal\":{\"name\":\"Advances in Calculus of Variations\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2022-06-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Advances in Calculus of Variations\",\"FirstCategoryId\":\"100\",\"ListUrlMain\":\"https://doi.org/10.1515/acv-2021-0089\",\"RegionNum\":3,\"RegionCategory\":\"数学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MATHEMATICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advances in Calculus of Variations","FirstCategoryId":"100","ListUrlMain":"https://doi.org/10.1515/acv-2021-0089","RegionNum":3,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICS","Score":null,"Total":0}
引用次数: 0
摘要
设Ω∧∈n {\Omega\subset\mathbb{R}^{n}} 是C {c ^{1}} 光滑紧致域。更进一步,设F: Ω x, n, n,→ {f:\Omega\times\mathbb{R}^{nN}\to\mathbb{R}} , F∑(x, p) {F(x,p)} ,是C 0 {c ^{0}} ,对p可导,对F可导,p = dp∑F {f_{p}:= d_{p}f} 在Ω上连续的 {\Omega\times\mathbb{R}^{nN}} F在p中是严格凸的,考虑n n × n n {nN\times nN} 矩阵A = (A α _ β i _ j)∈c0 _ (Ω) {a =(a ^{{ij}}_{\alpha\beta})\in c ^{0}(\Omega)} 满足(0.1)A α减去β i减去j减去(x)减去ξ α i减去ξ β j = A β减去α j减去i减去(x)减去ξ α i减去ξ β j≥λ减去| ξ | 2, λ > 0。a ^{ij}_{\alpha\beta}(x)\xi^{I}_{\alpha}\xi^{j}_{\beta}= a ^{ji}_{\beta\alpha}(x)% \xi^{i}_{\alpha}\xi^{j}_{\beta}\geq\lambda\lvert\xi\rvert^{2},\quad\lambda>0. Suppose that (0.2) lim | p | → ∞ 1 | p | ( D p F ( x , p ) - A ( x ) p ) = 0 , \displaystyle\lim_{\lvert p\rvert\to\infty}\frac{1}{\lvert p\rvert}(D_{p}F(x,p% )-A(x)p)=0, (0.3) - C 0 + c 0 | p | 2 ≤ F ( x , p ) ≤ C 0 ( 1 + | p | 2 ) , \displaystyle{-}C_{0}+c_{0}\lvert p\rvert^{2}\leq F(x,p)\leq C_{0}(1+\lvert p% \rvert^{2}), (0.4) | F p ( x , p ) - F p ( x , q ) | ≤ C 0 | p - q | , \displaystyle\lvert F_{p}(x,p)-F_{p}(x,q)\rvert\leq C_{0}\lvert p-q\rvert, (0.5) 〈 F p ( x , p ) - F p ( x , q ) , p - q 〉 ≥ c 0 | p - q | 2 \displaystyle\langle F_{p}(x,p)-F_{p}(x,q),p-q\rangle\geq c_{0}\lvert p-q% \rvert^{2} uniformly in x and with positive constants c 0 {c_{0}} and C 0 {C_{0}} . Consider the functional (0.6) J ( u ) := ∫ Ω F ( x , D u ( x ) ) 𝑑 x + ∫ Ω G ( x , u ) 𝑑 x , J(u):=\int_{\Omega}F(x,Du(x))\,dx+\int_{\Omega}G(x,u)\,dx, where G ( x , ⋅ ) ∈ C 1 ( ℝ N ) {G(x,\cdot\,)\in C^{1}(\mathbb{R}^{N})} for each x ∈ Ω {x\in\Omega} , G ( ⋅ , u ) {G(\,\cdot\,,u)} is measurable for each u ∈ ℝ N {u\in\mathbb{R}^{N}} , and (0.7) | G u ( x , u ) | ≤ C 0 ( 1 + | u | s ) \lvert G_{u}(x,u)\rvert\leq C_{0}(1+\lvert u\rvert^{s}) with s < n + 2 n - 2 {s 2 {n>2} , then any weak solution u ∈ W 1 , 2 ( Ω , ℝ N ) {u\in W^{1,2}(\Omega,\mathbb{R}^{N})} of the Euler equations of J, i.e. ∑ α ∂ ∂ x α F p α i ( x , D u ) = G u i ( x , u ) , i = 1 , … , N , \sum_{\alpha}\frac{\partial}{\partial x^{\alpha}}F_{p^{i}_{\alpha}}(x,Du)=G_{u% ^{i}}(x,u),\quad i=1,\ldots,N, is Hölder continuous in the interior of Ω and under appropriate boundary conditions also Hölder continuous up to the boundary.
On the Hölder regularity of all extrema in Hilbert’s 19th Problem
Abstract Let Ω ⊂ ℝ n {\Omega\subset\mathbb{R}^{n}} be a C 1 {C^{1}} smooth compact domain. Furthermore, let F : Ω × ℝ n N → ℝ {F:\Omega\times\mathbb{R}^{nN}\to\mathbb{R}} , F ( x , p ) {F(x,p)} , be C 0 {C^{0}} , differentiable with respect to p, and with F p := D p F {F_{p}:=D_{p}F} continuous on Ω × ℝ n N {\Omega\times\mathbb{R}^{nN}} and F strictly convex in p. Consider an n N × n N {nN\times nN} matrix A = ( A α β i j ) ∈ C 0 ( Ω ) {A=(A^{{ij}}_{\alpha\beta})\in C^{0}(\Omega)} satisfying (0.1) A α β i j ( x ) ξ α i ξ β j = A β α j i ( x ) ξ α i ξ β j ≥ λ | ξ | 2 , λ > 0 . A^{ij}_{\alpha\beta}(x)\xi^{i}_{\alpha}\xi^{j}_{\beta}=A^{ji}_{\beta\alpha}(x)% \xi^{i}_{\alpha}\xi^{j}_{\beta}\geq\lambda\lvert\xi\rvert^{2},\quad\lambda>0. Suppose that (0.2) lim | p | → ∞ 1 | p | ( D p F ( x , p ) - A ( x ) p ) = 0 , \displaystyle\lim_{\lvert p\rvert\to\infty}\frac{1}{\lvert p\rvert}(D_{p}F(x,p% )-A(x)p)=0, (0.3) - C 0 + c 0 | p | 2 ≤ F ( x , p ) ≤ C 0 ( 1 + | p | 2 ) , \displaystyle{-}C_{0}+c_{0}\lvert p\rvert^{2}\leq F(x,p)\leq C_{0}(1+\lvert p% \rvert^{2}), (0.4) | F p ( x , p ) - F p ( x , q ) | ≤ C 0 | p - q | , \displaystyle\lvert F_{p}(x,p)-F_{p}(x,q)\rvert\leq C_{0}\lvert p-q\rvert, (0.5) 〈 F p ( x , p ) - F p ( x , q ) , p - q 〉 ≥ c 0 | p - q | 2 \displaystyle\langle F_{p}(x,p)-F_{p}(x,q),p-q\rangle\geq c_{0}\lvert p-q% \rvert^{2} uniformly in x and with positive constants c 0 {c_{0}} and C 0 {C_{0}} . Consider the functional (0.6) J ( u ) := ∫ Ω F ( x , D u ( x ) ) 𝑑 x + ∫ Ω G ( x , u ) 𝑑 x , J(u):=\int_{\Omega}F(x,Du(x))\,dx+\int_{\Omega}G(x,u)\,dx, where G ( x , ⋅ ) ∈ C 1 ( ℝ N ) {G(x,\cdot\,)\in C^{1}(\mathbb{R}^{N})} for each x ∈ Ω {x\in\Omega} , G ( ⋅ , u ) {G(\,\cdot\,,u)} is measurable for each u ∈ ℝ N {u\in\mathbb{R}^{N}} , and (0.7) | G u ( x , u ) | ≤ C 0 ( 1 + | u | s ) \lvert G_{u}(x,u)\rvert\leq C_{0}(1+\lvert u\rvert^{s}) with s < n + 2 n - 2 {s<\frac{n+2}{n-2}} . Under these conditions, we shall show that if n > 2 {n>2} , then any weak solution u ∈ W 1 , 2 ( Ω , ℝ N ) {u\in W^{1,2}(\Omega,\mathbb{R}^{N})} of the Euler equations of J, i.e. ∑ α ∂ ∂ x α F p α i ( x , D u ) = G u i ( x , u ) , i = 1 , … , N , \sum_{\alpha}\frac{\partial}{\partial x^{\alpha}}F_{p^{i}_{\alpha}}(x,Du)=G_{u% ^{i}}(x,u),\quad i=1,\ldots,N, is Hölder continuous in the interior of Ω and under appropriate boundary conditions also Hölder continuous up to the boundary.
期刊介绍:
Advances in Calculus of Variations publishes high quality original research focusing on that part of calculus of variation and related applications which combines tools and methods from partial differential equations with geometrical techniques.