
In this paper, a decoupled finite element method for a modified Cahn-Hilliard-Hele-Shaw system with double well potential is constructed. The proposed scheme is based on the convex splitting of the energy functional in time and uses the mixed finite element discretzation in space. The computation of the velocity u is separated from the computation of the pressure p by using an operator-splitting strategy. A Possion equation is solved to update the pressure at each time step. Unconditional stability and error estimates are analyzed in detail theoretically. Furthermore, the theoretical part is verified by several numerical examples, whose results show that the numerical examples are consistent with the results of the theoretical part.
Citation: Haifeng Zhang, Danxia Wang, Zhili Wang, Hongen Jia. A decoupled finite element method for a modified Cahn-Hilliard-Hele-Shaw system[J]. AIMS Mathematics, 2021, 6(8): 8681-8704. doi: 10.3934/math.2021505
[1] | Imran Shahzad Khan, Choonkil Park, Abdullah Shoaib, Nasir Shah . A study of fixed point sets based on Z-soft rough covering models. AIMS Mathematics, 2022, 7(7): 13278-13291. doi: 10.3934/math.2022733 |
[2] | R. Mareay, Radwan Abu-Gdairi, M. Badr . Soft rough fuzzy sets based on covering. AIMS Mathematics, 2024, 9(5): 11180-11193. doi: 10.3934/math.2024548 |
[3] | Shoubin Sun, Lingqiang Li, Kai Hu, A. A. Ramadan . L-fuzzy upper approximation operators associated with L-generalized fuzzy remote neighborhood systems of L-fuzzy points. AIMS Mathematics, 2020, 5(6): 5639-5653. doi: 10.3934/math.2020360 |
[4] | Rehab Alharbi, S. E. Abbas, E. El-Sanowsy, H. M. Khiamy, K. A. Aldwoah, Ismail Ibedou . New soft rough approximations via ideals and its applications. AIMS Mathematics, 2024, 9(4): 9884-9910. doi: 10.3934/math.2024484 |
[5] | Saqib Mazher Qurashi, Ferdous Tawfiq, Qin Xin, Rani Sumaira Kanwal, Khushboo Zahra Gilani . Different characterization of soft substructures in quantale modules dependent on soft relations and their approximations. AIMS Mathematics, 2023, 8(5): 11684-11708. doi: 10.3934/math.2023592 |
[6] | Jamalud Din, Muhammad Shabir, Samir Brahim Belhaouari . A novel pessimistic multigranulation roughness by soft relations over dual universe. AIMS Mathematics, 2023, 8(4): 7881-7898. doi: 10.3934/math.2023397 |
[7] | Mostafa K. El-Bably, Radwan Abu-Gdairi, Mostafa A. El-Gayar . Medical diagnosis for the problem of Chikungunya disease using soft rough sets. AIMS Mathematics, 2023, 8(4): 9082-9105. doi: 10.3934/math.2023455 |
[8] | Jamalud Din, Muhammad Shabir, Nasser Aedh Alreshidi, Elsayed Tag-eldin . Optimistic multigranulation roughness of a fuzzy set based on soft binary relations over dual universes and its application. AIMS Mathematics, 2023, 8(5): 10303-10328. doi: 10.3934/math.2023522 |
[9] | José Sanabria, Katherine Rojo, Fernando Abad . A new approach of soft rough sets and a medical application for the diagnosis of Coronavirus disease. AIMS Mathematics, 2023, 8(2): 2686-2707. doi: 10.3934/math.2023141 |
[10] | Mohammed Shehu Shagari, Akbar Azam . Integral type contractions of soft set-valued maps with application to neutral differential equations. AIMS Mathematics, 2020, 5(1): 342-358. doi: 10.3934/math.2020023 |
In this paper, a decoupled finite element method for a modified Cahn-Hilliard-Hele-Shaw system with double well potential is constructed. The proposed scheme is based on the convex splitting of the energy functional in time and uses the mixed finite element discretzation in space. The computation of the velocity u is separated from the computation of the pressure p by using an operator-splitting strategy. A Possion equation is solved to update the pressure at each time step. Unconditional stability and error estimates are analyzed in detail theoretically. Furthermore, the theoretical part is verified by several numerical examples, whose results show that the numerical examples are consistent with the results of the theoretical part.
Traditional mathematical methods are frequently unable to address a wide range of complicated problems that arise in a variety of fields, including economics, engineering, sociology, medicine, environmental science, and many others. This is because many problems are inherently vague and uncertain. Uncertainty related issues and situations are dealt with using an antiquated and powerful instrument of probability theory, which is only appropriate when the occurrence of events is solely determined by chance. Along with probability theory, additional well-known theories have been created to address uncertainty, including fuzzy sets, intuitionistic fuzzy sets, rough sets, soft sets, and combinations of these theories.
The idea of fuzzy sets, which Lotfi. A. Zadeh [1] introduced in 1965, has shown to be a highly effective theoretical solution to ambiguity and vagueness. This idea is founded on fuzzy membership functions, which depict an element's membership in a set. By focusing on the fuzzy information granularity, fuzzy set theory can be utilized to obtain, simulate, and even explain the fuzziness in many practical materials of information. Almost all areas of mathematics, medicine, engineering, and other fields have had concepts reinterpreted using fuzzy sets as a result of their extensive uses.
Pawlak [2] proposed the theory of rough sets in the early 1980's which is another nice mathematical tool dealing the set's approximation for dealing with the uncertainty of imprecise data and vagueness. Given the facts at hand, this idea is the best replacement for fuzzy set theory and tolerance theory. Rough sets and fuzzy set theory are two distinct approaches used to address the ambiguity, imprecision, and haziness of real-world situations. Each of these theories has its inherent limitations. Many applications together with machine learning, data mining, knowledge discovery and pattern recognition can be found in [3,4,5,6,7,8]. The fundamental idea of traditional rough set theory is the lower/upper approximation concept, which is typically based on equivalence relations and partitions. Sometimes dealing with a real-world issue while constrained by an equivalence relation is challenging. In order to tackle such kind of situations, the concept of rough set has been extended to the notion of covering rough sets introduced in [9,10,11,12,13,14,15] which are an important generalizations of classical and traditional rough sets by relaxing partition of universe to covering. A more all-encompassing notion used to handle the attribute subset is covering, which is a method to expand any partition. Covering-based rough sets are more rational and logical than traditional rough sets for addressing uncertainty related issues, and this theory has attracted significant attention and produced numerous useful research outcomes.
In 1999, Molodtsov [16] introduced the concept of soft sets which is a very new and effective mathematical technique for handling uncertainties. It is a set connected with parameters and has been used in many different contexts. Several methods have been developed for addressing the imprecision, uncertainty, and ambiguity of real-world situations including fuzzy set theory, the rough, soft sets and blend of these theories. Each of these theories comes with certain built in drawbacks. [14,15,16,17,18,19,20,21] has several uses, including machine learning, data mining, pattern detection, and knowledge discovery.
In order to establish an applicable mathematical systems for covering-based rough set and promote its applications in various fields of life, it has been linked with some other theories like fuzzy set theory, soft set theory, neutrosophic set theory, graph theory and blend of theories [7,14,21,22,23,24]. The notion of a family of fuzzy complementary β-neighborhood and thus four types of covering-basedoptimistic (pessimistic) multigranulation fuzzy rough sets models are presented in [25]. Also, four new kinds of covering-based M-optimistic (pessimistic) multigranulation fuzzy rough sets models are constructed. Some characterizations of these models and its relation with Zhan's model are studied.
Lattice theory and partial order play an important role in many fields of engineering and computer science and they have many applications in distributed computing, that is, vector clocks and global predicate detection, concurrency theory, occurrence nets and pomsets, programming language semantics (fixed-point semantics), and data mining[26]. They are also useful in other disciplines of mathematics such as combinatorics, group theory and number theory. Many authors have combined the rough set theory and lattice theory, and some useful results have been obtained. Based on the existing works about the connection of rough sets and lattice theory, Chen et al. [4] used the notion of covering to define the approximation operators on a completely distributive lattice and set up a unified framework for generalizations of rough sets. Shah et al. [27] discussed another approach to roughness of soft graphs with applications in decision making, see also [19,24,28,29,30,31]. Rough approximation models via graphs based on neighborhood systems can be seen in [32]. Some applications of soft graphs linked with rough sets and soft sets can be seen in [27,33,34,35,36]. Further, the concept of dual hesitant fuzzy graphs (DHFGs) proposed by [37], where a two-stage MADM approach is constructed by means of DHFGs for addressing complicated MADM situations with correlations and prioritization relationships. He et al. [38] revised the "tight" bounds for path-factor critical avoidable graphs. Since the avoidable graph is a special case of deleted graphs, a link is established between the path-factor critical graph and the path-factor deleted graph. In [34], Praba defined a novel rough set called minimal soft rough set by using minimal soft description of the objects. They also analyzed the relation between modified soft rough set and minimal soft rough set. They proposed a lattice structure on minimal soft rough sets. Uncertainty measures associated with neighborhood based soft covering rough graphs such as roughness measure, entropy measure and granularity are proposed in [36]. Atef and Nada [39] introduced the concept of the complementary fuzzy soft neighborhood as a generalization of Zhan's method, which increases the lower approximation and decreases the upper approximation. As a result, three new types of soft fuzzy rough covering models are constructed. These constructions' properties are discussed. They define three categories of fuzzy soft measure degrees in light of these results. A decision-making algorithm is then described based on the suggested operations, and its performance is illustrated with a numerical example. Further the relationships among these three models and Zhan's model are presented [39,40]. Li and Zhu [11] introduced the lattice structures of fixed points of the lower approximations of two types of covering-based rough sets in which they discussed that under what conditions two partially ordered sets are some lattice structures. They defined two types of sets called the fixed point set of neighborhoods and the fixed point set of covering, respectively. Fixed points of covering upper and lower approximation operators are introduced in [41] in which by using some results about the Feynman paths, they have shown that the family of all fixed points of covering upper and lower approximation operators is an atomic frame and a complete lattice, respectively.
Z-soft rough covering models introduced by Zhan et al. [42] are important generalizations of classical rough set theory to deal with data structure and more complex problems of the real world. It can be seen in [43] that the CSR approach uses property soft neighborhoods to establish models that successfully grow the lower approximation and lower the upper approximation and study the relationships between these models and some of the topological properties based on the CSR approach. In order to solve MGDM problems, they finally developed an algorithm for the presented model. Different kinds of uncertainty measures related to Li-soft rough covering sets and their limitations are presented in [44]. The concept of fixed point sets by using Z-soft rough covering models is introduced by Imran et al. [45], where they have discussed different algebraic structures along with their limitations connecting both lattices and Z-soft rough covering models. The purpose of introducing soft graphs is to discretize these fundamental mathematical ideas, which are inherently continuous, and to provide new tools for applying mathematical analysis technology to real-world applications including imperfect and inexact data or uncertainty. Li et al. presented the idea of soft rough covering models (briefly, SRC-Models), a novel theory that addresses uncertainty. Two new notions have been introduced in the current paper. Li-soft rough covering graphs (Li-SRCGs) and the notion of fixed point sets, also known as Li-SRCFP sets of such graphs. Several types of approximation operators and their related properties are discussed using Li-SRCGs. We also investigated a few algebras that dealt with the fixed points of Li-SRCGs. Applications of the algebraic structures available in covering soft sets to soft graphs may open up new areas of graph theory. We go over the prerequisites for the family of Li-SRCFP sets acquiring lattice structure, distributive lattice, complete lattice, and some algebra pertaining to soft graphs. This paper is organized as follows.
The basics of rough sets, covering soft sets, Li-soft rough covering models, lattices and soft graphs are reviewed in Section 2 of this article. In Section 3, we define the concept of Li-SRCGs, Li-soft reduct of covering soft sets, and their attributes based on Li-SRCGs. Further, we investigate the conditions in which the soft neighborhood Li-SRCFP sets transform into particular lattice structures. In Section 4, we explore the idea of soft graphs'Li-soft rough covering fixed point sets (Li-SRCFP sets). We have talked about the idea of bounds for any two Li-SRCGs elements. Several algebraic structures related to Li-SRCGs are also addressed. In Section 5, we finally put our paper to conclusion.
This section provides a succinct review of certain essential concepts, results, and core ideas that will be useful in comprehending the remaining chapters of this thesis. The universe V is assumed to be a non-void finite set throughout this article, along with the void (empty) set ∅ and R, the parameter's set.
Definition 2.1. A family C of non-void subsets of U is said to be a covering of U if ⋃C=U. Also, for a subset Y of U, the sets
(i) given below, denoted by LC(Y) and HC(Y), are called respectively, the second type covering lower and upper approximation of Y,
FL(Y)=∪{K∈C:K⊆Y},FH(Y)=∪{K∈C:K∩Y≠∅}, |
(ii) given below, denoted by FL(Y) and FH(Y), are called respectively, covering lower and upper approximations of sixth type of Y
LC(Y)={x∈U:N(x)⊆Y},HC(Y)={x∈U:N(x)∩Y≠∅}, |
where N(x)=⋂{Q∈C:x∈Q} is called neighborhood of x with respect to C.
Proposition 2.1. For any subset Y of U, the following laws always hold true:
(i) FL(∅)=∅ and LC(∅)=∅
(ii) FL(U)=U and LC(U)=U
(iii) FL(Y)⊆Y and LC(Y)⊆Y (iv) FL(FL(Y))=FL(Y) and LC(LC(Y))=LC(Y) (v) Y⊆X implies FL(Y)⊆FL(X) and LC(Y)⊆LC(X) (vi) for all K∈C, FL(K)=(K) and LC(K)=LC(K).
Definition 2.2. Let τ:R→P(U) is a set valued mapping, then the ordered pair T=(π,R) is called a soft set over U. In this case, the pair D=(U,T), is called soft approximation space (briefly,SAS).
Definition 2.3. A soft set (τ,R) is called covering soft set (briefly,CSS)if
(i) it is full, that is, if ⋃σ∈Rτ(σ)=U and,
(ii) for every σ∈R, τ(σ)≠∅.
In such case, the ordered pair T=(U, CV) is called SCAS (soft covering approximation space). Then, for W⊆U, the following two sets:
F_CV(W)=∪{τ(σ)∈CV:τ(σ)⊆W} and¯FCV(W)=∪{τ(σ)∈CV:τ(σ)∩W≠∅} |
In the above sets the operators F_CV(W) and ¯FT(W) are called respectively, Li−SCLA operator and Li−SCUA operator.
Definition 2.4. A partial ordered set (or briefly, a poset) is an ordered pair (L,⪯), consisting of a non-void L and a partial order ⪯ on L.
Definition 2.5. A lattice is a poset (L,⪯) in which a∧b=inf(a,b) and a∨b=sup(a,b) exist for any pair of elements a and b of L. Also, L is said to have a lower bound 0 if and only for any t in L we have, 0⪯t. Analogously, L is said to have an upper bound 1 if and only for any t in L we have, t⪯1. A lattice (L,⪯) is bounded if and only if has both 0 and 1.
Definition 2.6. A lattice (L,⪯) is said to be distributed if and only if for any a, b, t in L we have, a∧(b∨t)=(a∧b)∨(a∧t) and a∨(b∧t)=(a∨b)∧(a∨t). Otherwise, (L,⪯) is said to be non distributive.
Definition 2.7. Let 0 be a lower bound and 1 be an upper bound in a lattice (L,⪯). An element p in L is said to be
(i) join irreducible if p=a∨b implies p=a or p=b.
(ii) complement of t if p∨t=1 and p∧t=0.
(iii) pseudocomplement of t ∈L, if p∧t=0 and for all r ∈L, t∧r=0 implies r⪯p.
(iv) dual pseudocomplement of t ∈L, if p∨t=1 and for all r ∈L, t∨r=1 implies p⪯r.
Definition 2.8. A lattice (L,⪯) is said to be
(i) complemented if it is (a) bounded and (b) every element of L has a complement ;
(ii) pseudocomplemented (briefly, pseudCd) if every member of L has pseudocomplement;
(iii) a Stone algebra if it is a lattice which is (a) distributive (b) pseudCd and (c) satisfying the identity p∗∨p∗∗=1, for all p ∈L,(where p∗is pseudocomplemented of p);
(iv) a dual pseudocomplemented (briefly, D-pseudCd) if each of its member has dual pseudocomplement;
(v) a dual Stone algebra, if the lattice meets the conditions of being (a) distributive (b) D-pseudCd, and (c) satisfying the identity (dual Stone identity) p∗∧p∗∗=1, for all p ∈L;
(vi) a double p-algebra if it is simultaneously (a) pseudCd and (b) D-pseudCd;
(vii) a double Stone algebra, if it is (a) Stone algebra and (b) dual Stone algebra.
Definition 2.9. Let Ω=(V, E) be a graph. Then, a quadruple Θ=(Ω,δ,γ,R) is called a soft graph, provided
(i) δ:R→P(V) is a soft set over vertex V;
(ii) γ:R→P(E) is a soft set over E and for every σ∈R, the pair H=(δ(σ),γ(σ)) represents a subgraph of Ω.
Further, if Θ is a soft graph such that
(a) ⋃σ∈Rδ(σ)=V, then Θ is called full soft vertex graph;
(b) ⋃σ∈Rγ(σ)=E, then Θ is called full soft vertex graph;
(c) (⋃σ∈Rδ(σ),⋃σ∈Rγ(σ))=(V,E),then Θ is called full soft graph.
Li-SRC sets are significant mathematical tools for addressing challenges in real world that involve uncertainty. Another useful tool for displaying information through matrices, relations, and diagrams is graph theory, which has obvious applications. This section focuses on the description of a novel blend of Li-soft rough covering sets and graphs, called Li-soft rough covering graphs (Li-SRCGs). This approach will assist us discuss the idea of uncertainty in these concepts as well as improve the application of Li-SRC sets and soft graphs. It is shown that the family of Li-SRCFP sets is a lattice. We have also proposed some basic properties and related examples in details.
Definition 3.1. Let Θ be a CSV -Graph and Q=(V,CV) be a SVCAS. Then, for any W⊆V, SVCL and SVCU approximation operators are respectively, defined as:
F_CV(W)=∪{τ(σ)∈CV:τ(σ)⊆W} and¯FCV(W)=∪{τ(σ)∈CV:τ(σ)∩W≠∅}. |
In the above sets, the operators F_CV(W) and ¯FT(W) are called respectively, Li-SVCL operator and Li-SVCU operator. In case, F_CV(W)=¯FCV(W), then W is called Li -soft vertex covering definable, where the graph GQ:=(V,E) is called Li-SVC definable. But on the other hand, if F_CV(W)≠¯FCV(W), then the set W is called Li-SRVC set and the graph GQ is called Li-SRVC
Example 3.1. Suppose Ω=(V,E) is a graph given in Figure 1 below, with V={x1,x2,...,x7}, E={e1,e2,...,e10},
Let Q=(V,CV) be a SVCAS, where CV=(δ,R), R={σ1,σ2,...,σ6} is parameters set and δ:R→P(V) is a set valued mapping, presented in Table 1 such that δ(σ1)={x3,x4,x5}, δ(δ2)={x1,x2}, δ(σ3)={x3,x5,x6}, δ(σ4)={x3,x4}, δ(σ5)={x1,x2,x3},δ(σ6)={x1,x2,x5},
R╲V | x1 | x2 | x3 | x4 | x5 | x6 | x7 |
σ1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 1 | 0 | 1 | 1 | 1 |
σ4 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
σ5 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ6 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
Then, clearly, the pair Q=(V,CV) is a SVCAS. Suppose we have a subset W={x1,x2,x4} of vertex set V. Then, F_CV(W) and ¯FCV(W) can be calculated in the following manners:
F_CV(W)=∪{τ(σ)∈CV:τ(σ)⊆W}={x1,x2} and¯FCV(W)=∪{τ(σ)∈CV:τ(σ)∩W≠∅}={x1,x2,x3,x4,x5}. |
Since F_CV(W) ≠ ¯FCV(W), so W is a SVCR and ΔQ:=(V,E) is Li-SRVC -Graph with
Δ_Q=(F_CV(W),E)=({x1,x2},{e1,e2,...,e10}), |
¯ΔQ=(¯FCV(W),E)=({x1,x2,x3,x4},{e1,...,e10}) and |
Note if W={x1,x2}⊆V, then F_CV(W) = ¯FCV(W)={x1,x2} showing that W is a Li-SVC definable setF_CV(W) = ¯FCV(W)={x1,x2}. Also,
Δ_Q=(F_CV(W),E)=({x1,x2},{e1,e2,...,e10})=¯ΔQ=(¯FCV(W),E) |
Definition 3.2. A full soft edge graph Θ, such that γ(σ)≠∅ for all σ∈R, is called CSE Graph. In this case, D=(E,CE) is called SECAS.
Definition 3.3. Let D=(E,CE) be a SECAS then, the sets
F_CE(N)=∪{γ(σ)∈CE:γ(σ)⊆N} and¯FCE(N)=∪{γ(σ)∈CE:γ(σ)∩N≠∅}, N⊆E, |
are called the Li-SECL and Li-SECU approximations of N, respectively.
Also, if ¯FCE(N)=F_CE(N), where N is a subset of E. Then, N is called Li-SEC definable set and the graph ΔQ:=(V,E) is called Li-SEC definable. The graph ΔQ is called Li-SECAS Graph only if ¯FCE(N)≠F_CE(N). In this case, the subset N of E is called Li-SECR set.
Example 3.2. Continued from Example 3.2, if Q=(E,CE) represents a SECAS with γ(σ1)={e1,e3,e5,e10}, γ(σ2)={e4}, γ(σ3)={e4,e5,e6}, γ(σ4)={e1,e2,e5,e6,e10}, γ(σ5)={e1,e3} and γ(σ6)={e6,e7,e9}, see Table 2 below.
R╲E | e1 | e2 | e3 | e4 | e5 | e6 | e7 | e8 | e9 | e10 |
σ1 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
σ2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ4 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 |
σ5 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
σ6 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 0 |
Let N={e1,e3,e4,e10}⊆E. Then
F_CE(N)=∪{γ(σ)∈CE:γ(σ)⊆N}={e1,e3,e4} and¯FCE(N)=∪{γ(σ)∈CE:γ(σ)∩N≠∅}={e1,e2,e3,e4,e5,e6,e10}. |
Here, N is a Li-SREC set and ΔD:=(V,E) is Li-SRECG because F_CE(N) ≠¯FCE(N) such that
G_D=(V,F_CE(N))=({x1,x2,...,x7},{e1,e3,e4}) and |
¯GD=(V,¯FCE(N))=({x1,x2,...,x7},{e1,..,e6,e10}). |
Definition 3.4. Suppose Θ is a soft graph. Then, Θ is called
(i) Li-soft covering definable if F_CV(W)=¯FCV(W) and ¯FCE(N)=F_CE(N),
(ii) Li-soft rough covering graph (briefly,Li-SCR Graph) if F_CV(W)≠¯FCV(W) and ¯FCE(N)≠F_CE(N), where W⊆V and N⊆E.
Definition 3.5. Let Θ be a CSE-Graph such that Q=(V,CV) is a SVCAS and δ(σ)∈Cv. Then, δ(σ) is called Li-soft union reducible element (briefly, Li-SUred element) if δ(σi) is the union of some δ(σj)∈CV−{δ(σi)} for i≠j. Any other element which is not Li-SUred element, is called Li-soft union irreducible element (Li-SUirred element). If every δ(σi)∈CV is Li-SUirred element, then CV is called Li-soft irreducible, otherwise CV is called Li-soft reducible.
It can be seen that if δ(σi) is a Li-SUred element of CV, then CV−{δ(σi)} is still a covering soft set over the universe set V.
Example 3.3. Consider a CSV-Graph Θ=(Ω,δ,γ,R) where, V= a finite universe (vertex set)={r1,r2,r3,r4} and R={σ1,σ2,σ3,σ4} such that (δ,R) is CSS over V, see below in Table 3, so that δ(σ1)={r1}, δ(σ2)={r2,r3}, δ(σ3)={r1,r2,r3}, δ(σ4)={r2,r4} and CV={δ(σ1),δ(σ2), δ(σ3),δ(σ4)}={{r1}, {r2,r3}, {r1,r2,r3},{r2,r4}}.
R∖V | r1 | r2 | r3 | r4 |
σ1 | 1 | 0 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 1 | 1 | 0 |
σ4 | 0 | 1 | 0 | 1 |
Clearly, δ(σ3)={r1,r2,r3}=δ(σ1)∪δ(σ2), where δ(σ1), δ(σ2)∈CV−{δ(σ3)} showing that δ(σ3) is a Li-SUred element in CV. The elements δ(σ1),δ(σ2) and δ(σ4) are Li-SUirred elements in CV.
Definition 3.6. Let CSV-Graph be Θ=(Ω,δ,γ,R) and Q=(V,CV), be a SVCAS. Then, family Ffix(CV) of subsets of V, defined by Ffix(CV)={W∈P(V):F_CV(W)= W}, is called Li-soft rough vertex covering fixed point set (briefly Li-SRVCFP set) induced by CV.
Example 3.4. Consider a CSV-Graph Θ=(Ω,δ,γ,R) as shown in Figures 2 and 3, where Ω=(V, E) with vertex set V={x1,x2,x3,x4}, E={e1,e2,,,,e8} and parameters set R={σ1,σ2,σ3,σ4}, (δ,R) is a CSS over V,below in Table 4, so that δ(σ1)={x1,x2}, δ(σ2)={x3}, δ(σ3)={x1,x2,x3}, δ(σ4)=V.
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 1 | 0 |
σ3 | 0 | 0 | 1 | 0 |
σ4 | 1 | 1 | 1 | 1 |
Here, CV={{x1,x2},{x3}, {x1,x2,x3},,{x1,x2,x3,x4}}. Let W={x1,x2}⊆V, then F_CV(W)=∪{δ(σ)∈CV:δ(σ)⊆W}={r1,r2}= W. That is, F_CV(W)=W In this case, W is a member of Li-SRCFP set or W∈ Ffix(CV).
Similarly, when W={r1,r2,r3} then F_CV(W)= W, showing that W∈ Ffix(CV).
Proposition 3.1. Suppose Θ=(Ω,δ,γ,R) is a CSV-Graph and δ(σ) is a Li-SUirred element of CV, then Ffix(CV)=Ffix((CV)−{δ(σ)}).
Proof. By definition, Ffix(CV)={ W∈P(V):F_CV(W)=W} and Ffix((CV)−{δ(σ)})={W∈P(V):F_CV−{δ(σ)}(W)=W}. But F_CV-{δ(σ)}(W)=F_CV(W) for any W∈P(V). Therefore, Ffix(CV)=Ffix((CV)−{δ(σ)}).
Now, the following proposition shows that Li-SRVCFP set with respect to soft covering CV is the same as that of one, induced by the reduction of the CSS CV.
Definition 3.7. If every δ(σi)∈CV is Li-SUirred element, then CV is called Li-soft irreducible, otherwise CV is called Li-soft reducible. It can be seen that if δ(σi) is a Li-SUred element of CV, then CV−{δ(σi)} is still a CSS over the universe set V. For CV, the newly Li-soft union irreducible covering soft set with respect to the above reduction, is called Li-soft reduct of CV, and is denoted by Li-Sredct(CV).
Proposition 3.2. Let Θ=(Ω,δ,γ,R) be a CSV-Graph and let CV is a covering soft set over V, then Ffix(CV)=Ffix(Li-Sredct(CV)).
Proof. By definition,
Ffix(CV)={W∈P(V):F_CV(W)=W} andFfix(Li-Sredct(CV))={W∈P(V):F_Li-Sredct(CV)(W)=W}. |
Since, F_Li-Sredct(CV)(W)=F_CV(W), for any W∈P(V). Thus, Ffix(CV)=Ffix(Li-Sredct(CV)).
The Li-SRCFP- set induced by CV together with the set inclusion, (Ffix(CV),⊆), is a POS. Actually, we see that whether this POS is a lattice or not. In the theorem given below, we show that Ffix(CV) (Li-SRCFP- set) is a lattice and for any two members of this lattice. And for such lattice, we have find its lub and glb.
Proposition 3.3. Let Q=(V,CV), be a SVCAS for CSV-Graph Θ=(Ω,δ,γ,R). Suppose, W, Y∈Ffix(CV). Then, (Ffix(CV),⊆) is a lattice, where W∨Y= W∪Y and W∧Y=F_CV(W∩Y).
Proof. We have to prove only that W∪Y∈Ffix(CV) as well as F_CV(W∩Y)∈Ffix(CV) for any W, Y∈Ffix(CV). If possible, suppose W∪Y ∉ Ffix(CV), then there exists r∈ W∪Y such that for all δ(σ)∈CV and x∈δ(σ) implies δ(σ)⊈ W∪Y. Since r∈ W∪Y, so r∈ W or r∈Y. Hence δ(σ)⊈ W or δ(σ)⊈ Y, that is, r∉F_CV(W) or r∉F_CV(Y), which gives a contradictory with the fact W, Y∈Ffix(CV). Therefore, W∪Y∈Ffix(CV). Now, for any W, Y∈Ffix(CV), if W∩Y⊆V then by using the fact F_CV(F_CV(M))=F_CV(M), we get F_CV(F_CV(W∩Y))=F_CV(W∩Y). This implies that F_CV(W∩Y)∈Ffix(CV) for any W, Y∈Ffix(CV). Thus, (Ffix(CV),⊆) is lattice.
The above theorem shows that the Li-SRCFP- set having the relation of set inclusion is a POS and for any two members of the Li-SRCFP- set, the lub is the join of these two members, while the glb is the lower approximation of the intersection of these two members. Actually, (Ffix(CV),∧,∨) is defined from the view point of algebra and (Ffix(CV),⊆) is defined from the viewpoint of partially ordered set. Further, (Ffix(CV),∧,∨) and (Ffix(CV),⊆) are both lattices.
Remark 3.1. In above Proposition, we have seen that (Ffix(CV),⊆) is a lattice such that the sets U and ∅ are respectively, the greatest and least of (Ffix(CV),⊆). But we know a lattice having least and greatest elements is bounded. Therefore, (Ffix(CV),⊆) is a lattice which is bounded.
Therefore, in the assertion that follows, we shall demonstrate that any Li-SUirred element of CV is also a Join Li-SUirred element of CV, and any Li-SUred element of CV is a Join Li-SUred element of Ffix(CV).
Proposition 3.4. Let Θ=(Ω,δ,γ,R) be a CSV-Graph and Q=(V,CV), be a SVCAS such that δ(σi)∈CV. Then,
(i) the set δ(σi) is a Join Li-SUirred element of Ffix(CV), if δ(σi) is a Li-SUirred element of CV.
(ii) the set δ(σi) is Join Li-SUred element of Ffix(CV), if δ(σi) is a Join Li-SUred element of CV for parameters σi.
Proof. Since for any parameter σi∈Q and δ(σi)∈CV, F_CV(δ(σi))=δ(σi). This shows δ(σi)∈Ffix(CV) as Ffix(CV)={ W∈P(V):F_CV(W)=W}.
(i) Suppose there are elements Y, Z in Ffix(CV) such that δ(σi)=Y∪Z, then there exists some elements δ(σj),(j∈I) and δ(σk),(k∈K) in CV such that δ(σj)⊆Y, δ(σk)⊆Z and
Y=⋃{δ(σj)∈CV:δ(σj)⊆Y (j∈I)},Z=⋃{δ(σk)∈CV:δ(σk)⊆Z (k∈K)},where I,K⊆{1,2,...,|CV|}. |
Therefore, δ(σi)=(⋃δ(σj)⊆Y (j∈I)δ(σj))∪(⋃δ(σk)⊆Z (k∈K)δ(σk)). Since δ(σi) is a Li-SUirred element of CV, so there exists t∈I∪K so that δ(σi)=δ(σt). But δ(σi)=Y∪Z implies Y⊆δ(σi) and Z⊆δ(σi). If δ(σt)⊆Y, then δ(σi)=Y. If δ(σt)⊆Z δ(σi)=Z. This indicates that δ(σi)=Y or δ(σi)=Z. Thus, δ(σi) is a Join Li-SUirred element of Ffix(CV). Hence, the set δ(σi) is a Join Li-SUirred element of Ffix(CV), if δ(σi) is a Li-SUirred element of CV.
(ii) Suppose δ(σi) is a Li-SUred element of CV for parameters σi∈Q. Then there are some elements in CV−{δ(σi)} in such away that δ(σi) is the union of those elements. In other words, there are some elements δ(σj), (j∈I) such that δ(σi)=⋃j∈Iδ(σj), where I⊆S={1,2,...,|CV|}.
Since, for any H⊆S, we have δ(σk)(k∈H)⊆⋃j∈Hδ(σj). Then,
F_CV(⋃j∈Hδ(σj))=⋃{δ(σk)(k∈H):δ(σk)⊆⋃j∈Hδ(σj)}=⋃j∈Hδ(σj), |
Therefore, for any K, T ⊆ I,
F_CV(⋃k∈Kδ(σk))=⋃k∈Kδ(σk), |
and
F_CV(⋃t∈Tδ(σt))=⋃t∈Tδ(σt). |
Hence, we have ⋃k∈Kδ(σk)∈Ffix(CV) and ⋃t∈Tδ(σt)∈Ffix(CV). As a result, K1 and T1⊆I exists such that ⋃j∈Iδ(σj)=δ(σi)=(⋃k∈K1δ(σk))∪=(⋃t∈T1δ(σt)). This shows that δ(σi) is a Join Li-SUred element for Ffix(CV). Thus, the set δ(σi) is a Join Li-SUred element of Ffix(CV), if δ(σi) is a Li-SUred element of CV for parameters σi.
Proposition 3.5. Let Q=(V,CV), be a SVCAS of CSV-Graph Θ=(Ω,δ,γ,R). Then, for any r∈V, NB(r) is a Join Li-SUirred element of the lattice (Ffix(CV),⊆).
Proof. Suppose (Ffix(CV),⊆) is a lattice and there exist W,Y∈ Ffix(CV) such that NB(r)= W∪Y. Since r∈N(r), r∈ W∪Y. Therefore, r∈ W or r∈Y. Furthermore, as W,Y∈ Ffix(CV), then NB(r)⊆ W⊆ W∪Y=NB(r) or NB(r)⊆Y⊆ W∪Y=NB(r) Therefore, NB(r) = W or NB(r)=Y. Thus NB(r) is a Join Li-SUirred element of the lattice Ffix(CV), for every r∈V.
We have already shown that the set (Ffix(CV),⊆) is a lattice if and only if W∨Y= W∪Y and W∧Y=F_CV(W∩Y). Now in the following we show that (Ffix(CV),⊆), is a complete lattice.
Proposition 3.6. Let Θ=(Ω,δ,γ,R) be a CSV-Graph and Q=(V,CV), be a SVCAS. Then, the lattice (Ffix(CV),⊆) is complete.
Proof. For any G⊆Ffix(CV), we need to prove that ∧G ∈ Ffix(CV) and ∨G ∈ Ffix(CV). Equivalently, we have to prove that F_CV(∩G)∈ Ffix(CV) and ∪G ∈ Ffix(CV). Since G⊆Ffix(CV) then ∩G⊆V. According to the fact that F_CV(F_CV(∩G))=F_CV(∩G), that is, F_CV(∩G)∈Ffix(CV). If ∪G∉Ffix(CV) then there exists t∈∪G such that δ(σi)⊊∪G, for any δ(σi)∈CV and t∈δ(σi). Hence, there exists an element W in G so that t∈ W and δ(σi)⊊ W, for any δ(σi)∈CV and t∈δ(σi). So, t∉F_CV(W) that is, W∉Ffix(CV). Which is contradictory with the fact that W∈Ffix(CV). Hence ∪G ∈ Ffix(CV). Therefore, (Ffix(CV),⊆) is a complete lattice.
Note that the Li-SRCFP-set induced by CV need not be always a distributive lattice. The example that follows will support our assertion.
Example 3.5. Suppose we have a CSV-Graph Θ=(Ω,δ,γ,R) such that Q=(V,CV), is a SVCAS where V={x1,x2,x3,x4} is a vertex set and R={σ1,σ2,σ3}, set representing all parameters σ1,σ2,σ3,σ4 as shown in Table 5, with δ(σ1)={x1,x2}, δ(σ2)={x2,x3}, δ(σ3)={x1,x3,x4}. Also, CV={δ(σ1),δ(σ2),δ(σ3)}={{x1,x2},{x2,x3},{x1,x3,x4}}.
R╲V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 0 | 1 | 1 |
Then, by using the expression Ffix(CV)={ W∈P(V):F_CV(W)=W}, we obtain Ffix(CV)={∅, {x1,x2},{x1,x2,x3},{x2,x3},{x1,x3,x4},V}. Also, {x1,x2,x3}∧({x1,x3,x4}∨{x1,x2})={x1,x2,x3}, but ({x1,x3,x4}∧{x1,x2,x3})∨({x1,x2,x3}∧{x1,x2})={x1,x2}. Which shows that Ffix(CV)={∅, {x1,x2},{x2,x3},{x1,x3,x4},{x1,x2,x3},V} is not a distributive lattice. In other words, Li-SRCFP- set, that is not distributive.
In the statement that follows, we investigate the requirement for the Li-SRCFP set to transform into a distributive lattice.
Proposition 3.7. Let Θ be a CSV-Graph and Q=(V,CV), be a SVCAS. If CV is soft unary, then (Ffix(CV),⊆) is a distributive lattice.
Proof. Suppose W, Y and Z∈Ffix(CV) where W, Y, Z⊆V. then for parameters σi, σj, σk there are some elements δ(σi), (i∈I), δ(σj),(j∈I) and δ(σk),(k∈K) in CV such that δ(σi)⊆W, δ(σj)⊆Y, δ(σk)⊆Z with
W=⋃{δ(σi)∈(CV):δ(σi)⊆W, (i∈I)},Y=⋃{δ(σJ)∈(CV):δ(σJ)⊆Y, (j∈J)},Z=⋃{δ(σk)∈(CV):δ(σk)⊆Z, (k∈K)}, |
where I, J, K ⊆{1,2,...,|CV|}.
It can easily be seen that
W∧(Y∨Z)=F_CV(W∩(Y∪Z))=F_CV((W∩Y)∪(W∩Z))=F_CV((⋃δ(σi)⊆W(i∈I)δ(σi)∩⋃δ(σj)⊆Y(j∈J)δ(σj))∪(⋃δ(σi)⊆W(i∈I)δ(σi)∩⋃δ(σk)⊆Z(k∈K)δ(σk)))=F_CV((⋃δ(σi)⊆W(i∈I)δ(σj)⊆Y(j∈J){δ(σi)∩δ(σj)})∪(⋃δ(σi)⊆W(i∈I)δ(σk)⊆Z(k∈K){δ(σi)∩δ(σk)})). |
Moreover,
F_CV(W∩Y)∪F_CV(W∩Z)=(W∧Y)∨(W∧Z)=F_CV(⋃δ(σi)⊆W(i∈I)δ(σj)⊆Y(j∈J){δ(σi)∩δ(σj)})∪F_CV(⋃δ(σi)⊆W(i∈I)δ(σk)⊆Z(k∈K){δ(σi)∩δ(σk)}) |
Since, CV is unary, then for every x∈V, |MdesT(x)|=1. We suppose, MdesT(x)={δ(σx)} for any x∈V. Then, δ(σi)∩δ(σj) is the union of finite elements in CV. Hence,
δ(σi)∩δ(σj)=⋃x∈δ(σi)∩δ(σj)δ(σx). Therefore, W∧(Y∨Z)=F_CV(W∩(Y∪Z)))=⋃x∈(⋃δ(σi)⊆W(i∈I)δ(σj)⊆Y(j∈J){δ(σi)∩δ(σj)})∪(⋃δ(σi)⊆W(i∈I)δ(σk)⊆Z(k∈K){δ(σi)∩δ(σk)})δ(σx)=(⋃y∈(⋃δ(σi)⊆W(i∈I)δ(σj)⊆Y(j∈J){δ(σi)∩δ(σj)})δ(σy))∪(⋃z∈(⋃δ(σi)⊆W(i∈I)δ(σk)⊆Z(k∈K){δ(σi)∩δ(σk)})δ(σz))=(W∧Y)∨(W∧Z). |
Hence, Ffix(CV) is a distributive lattice.
Proposition 3.8. Let Q=(V,CV), be a SVCAS such that CV is soft unary. Then, P∩ Q ∈Ffix(CV), for any P, Q ∈Ffix(CV).
Proof. Let p ∈P∩ Q then p∈P and p∈ Q. Since, CV is soft unary so, for every x∈V, |MdesT(x)|=1. We suppose, MdesT(x)={δ(σx)} for any x∈V. As, P, Q ∈Ffix(CV), then δ(σy)⊆P and δ(σy)⊆Q, that is, δ(σy)⊆P∩Q. therefore,
F_CV(P∩Q)=⋃{δ(σi)∈(CV):δ(σi)⊆P∩Q}=⋃{δ(σy):y∈P∩Q}=P∩Q |
showing that F_CV(P∩Q)=P∩Q.
Thus, P∩Q∈Ffix(CV), that is, an intersection of any two members of (Ffix(CV)) Li-SRCFP set induced by CV, a soft unary covering, is closed.
In this section, we will prove Li-SRCFP set, with respect to a soft unary covering CV over the vertex set V is a pseudCd lattice and a D- pseudCd lattice. It means any element of Li-SRCFP set has both a pseudocomplement and a D- pseudocomplement. Also, we will see that for any member of Li-SRCFP set, its pseudCmt(pseudocomplement) represents the SLA of its complement and D- pseudCmt represents the union of all Join Li-SUirred elements containing the element in its complement. We also discuss some algebras connected with Li-SRCFP sets.
Proposition 4.1. Let T=(V, CV) be a SCAS such that CV is soft unary. Then,
(i) Ffix(CV) is a pseudCd lattice such that W∗=F_CV(∼W), where W∈ Ffix(CV);
(ii) Ffix(CV) is a D- pseudCd lattice, and W+=⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi), for any W∈ Ffix(CV), parameter (σi)∈Q, where ∼ W is the complement of W in V and J(Ffix(CV)) denotes all Join Li-SUirred elements in Ffix(CV).
Proof. (i) Suppose W∈ Ffix(CV), then we have F_CV(F_CV(∼W))=F_CV(∼W), showing that F_CV(∼W)∈Ffix(CV). But (F_CV(∼W))⊆(∼W), so W∩(F_CV(∼W))=∅. Therefore, F_CV(W∩(F_CV(∼W)))=∅. Now it is needed to prove that Y⊆(F_CV(∼W)) if F_CV(W∩Y)=∅ for any Y∈ Ffix(CV), W∩Y∈Ffix(CV) for W, Y∈Ffix(CV). Hence, F_CV(W∩Y)=(W∩Y). Further, if F_CV(W∩Y)=∅ then W∩Y=∅. Therefore, for any Y∈ Ffix(CV) if F_CV(W∩Y)=(W∩Y) then (W∩Y)=∅. As (W∩Y)=∅ so Y⊆(∼W). Which shows F_CV(Y)⊆(F_CV(∼W)). But Y∈Ffix(CV) gives Y=F_CV(Y) and so Y=F_CV(Y)⊆(F_CV(∼W)). Thus, W∗= F_CV(∼W) for any W∈ Ffix(CV). That is, Ffix(CV) is a pseudCd lattice.
(ii) For any W∈Ffix(CV),
F_CV(⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi))=⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi). |
So,
⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)∈Ffix(CV), for any W∈Ffix(CV). |
Further it is easy to show that W∪(⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi))=V. now we need only to show that for any
Y∈Ffix(CV),if W∪Y=V,then⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)⊆Y. |
The following two cases serve as evidence for it.
Case 1: If ⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)=∼ W, then ⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi) ⊆Y.
Case 2: If ∼ W⊂⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi), then ∼ W⊂Y. If ∼ W=Y, then F_CV(∼W)=F_CV(Y)=Y=∼ W. Since,
F_CV(∼W)=∪{δ(σi)∈CV:δ(σi)⊆∼W}=∪{δ(σx)∈ MdesT(x):x∈MdesT(x)}=∪{δ(σi)∈J(Ffix(CV)):x∈∼W∧x∈δ(σi)}=⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi),⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)=∼W, |
which is contradictory with ∼W⊂⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi).
Suppose Y⊂⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi), then there will exist y∈⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi) such that y∉Y. So, y∉∼ W showing that there exists z∈∼ W such that y∈δ(σi) for any δ(σi)∈J(Ffix(CV) and z∈δ(σi). Since, ∼ W is properly contained in Y and z∈Y, so, δ(σi)⊊Y for any δ(σi)∈J(Ffix(CV)) and z∈δ(σi). That is, z∉F_CV(Y). Equivalently, we can say that F_CV(Y)≠Y, which is a contradiction to the fact that Y∈Ffix(CV). Hence, ⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)⊆Y. As a Consequence, W+=⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi) for any W∈Ffix(CV), that is, Ffix(CV) is a D- pseudCd lattice. Thus, we have seen that Li-SRCFP- set induced by any soft unary covering CV over V represents a pseudCd lattice and also a D- pseudCd lattice. This shows that Ffix(CV) is a double p-algebra.
Remark 4.1. In general, no soft unary covering by the Li-SRCFP set will generate a Stone algebra or a dual Stone algebra.
Example 4.1. Let Θ=(Ω,δ,γ,R) represents a CSV-Graph and Q=(V,CV), be a SVCAS, where V={x1,x2,x3,x4} is a vertex set and R={σ1,σ2,σ3,σ4}, set representing all parameters σ1,σ2,σ3,σ4 as shown in Table 6, with δ(σ1)={x3}, δ(σ2)={x1}, δ(σ3)={x1,x3,x4}, δ(σ4)={x2,x3}.
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 1 | 0 |
σ2 | 1 | 0 | 0 | 0 |
σ3 | 1 | 0 | 1 | 1 |
σ4 | 0 | 1 | 1 | 0 |
Let
CV={{x3},{x1},{x1,x3,x4},{x2,x3}}. |
Then,
Ffix(CV)={∅,{x1},{x3},{x1,x3},{x2,x3},{x1,x3,x4},{x1,x2,x3},V}. |
Let
M={x3} and N={x2,x3}. |
Then,
M∗=F_CV(∼M)={x1}, |
since,
M∗=F_CV(∼M)={x1}, so M∗∗=F_CV(∼M∗)={x2,x3}. |
That is, M∗∪M∗∗≠V. Therefore, Ffix(CV) is not a Stone algebra.
Also,
N+=⋃x∈∼N(x∈δ(σi)∈J(Ffix(CV)))δ(σi)={x1,x3,x4},N++=⋃x∈∼N+(x∈δ(σi)∈J(Ffix(CV)))δ(σi)={x2,x3}, |
showing that
N+∩N++≠∅. |
Thus, Ffix(CV) is not a dual Stone algebra.
The purpose of the following claim is to investigate the circumstances and conditions under which the Li-SRCFP-set, which is caused by any soft covering, yields a double Stone algebra and a boolean lattice, respectively.
Proposition 4.2. Suppose a CSV-Graph is represented by Θ having a SVCAS, denoted by Q=(V,CV). If Li-Sredct(CV) is a partition of V, then Ffix(CV) represents a boolean lattice.
Proof. We need first to show that ifLi-Sredct(CV) is a partition, then CV is a soft unary covering. We assume on contrary that CV is not soft unary covering. Therefore, there will be an element x∈V such that |MdesT(x)|≩1. So, we have parameters σ1, σ2∈R with δ(σ1), δ(σ2)∈Li-Sredct(CV) so that δ(σ1), δ(σ2)∈ MdesT(x), that is, x∈δ(σ1)∩δ(σ2), which is contradictory with the fact that Li-Sredct(CV) is a partition of V. Therefore, one can see that CV is a soft unary covering. But, Ffix(CV) is distributive lattice. Moreover, Ffix(CV) represents a bounded lattice. Now, we will have to show only that Ffix(CV) is a lattice which is complemented. Equivalently, we have to prove that ∼ W∈ Ffix(CV) for any W∈ Ffix(CV). If we assume that ∼ W ∉ Ffix(CV), then there is an element y in ∼ W such that for σi∈R, δ(σi)⊊∼ W. Since, CV is soft unary then for every vertex x∈V we have MdesT(x)={δ(σx)}. So, δ(σy)⊊ ∼ W. As a result, there is x∈ W such that x∈δ(σy) exists. But it is given Li-Sredct(CV) is partitioning the universe V, so we have δ(σy)=δ(σx). Thus, δ(σx)⊊W, i.e, x∉F_CV(W). In other words, F_CV(W)≠W, which is contradictory with W∈ Ffix(CV). Therefore, ∼ W∈ Ffix(CV), showing that Ffix(CV) is a complemented lattice. Hence, as a consequence, Ffix(CV) is a boolean lattice.
Proposition 4.3. Suppose a CSV-Graph is represented by Θ having a SVCAS, denoted by Q=(V,CV). If Li-Sredct(CV) partitioning the vertex set V, then Ffix(CV) is a double Stone algebra.
Proof. For any W∈Ffix(CV), we prove W∗=∼ W= W+. We know that Ffix(CV) is a boolean lattice, therefore ∼W∈Ffix(CV) that is, F_CV(∼W)=∼W. So, W∗=F_CV(∼W)=∼ W. For any y∈∼ W, if δ(σy)∈MdesT(y), then δ(σy) is a Join Li-SUred element of CV. Using Proposition 4.1, δ(σy)∈J(Ffix(CV)). Assume that x∈ W exists such that x∈ δ(σy). Since Li-Sredct(CV), being a partition of V, leads to the fact that δ(σy)∈MdesT(x). In the light of this, we conclude that, δ(σy)⊊W, that is, x∉F_CV(W). So, F_CV(W)≠W or W∉ Ffix(CV). Which is a contradiction because W∈Ffix(CV). Therefore,
W+=⋃x∈∼W(x∈δ(σi)∈J(Ffix(CV)))δ(σi)=⋃x∈∼Wδ(σi)∈MdesT(y)δ(σi)=∼W |
Similarly, it is easy to prove that W∗∗=∼∼W=W=W++. Therefore, W∗∪ W∗∗=V, W+∩ W++=∅, demonstrating that Ffix(CV) is both a dual Stone algebra and a Stone algebra.
Soft set theory and rough set theory are two newer tools to discuss uncertainty. Soft graph theory is a nice way to depict certain information. In order to discuss uncertainty in soft graphs, a possible amalgamation of three different concepts, that is, rough sets, soft sets and graphs is discussed. In this paper, we have introduced two new concepts Li-soft rough covering graphs (Li-SRCGs) and the concept of fixed point sets, called Li-SRCFP set, induced by soft covering CV of such graphs.Li-SRCGs are used to discuss various kinds of approximation operators and the properties associated with them. We review the conditions for the family of Li-SRCFP sets to become a lattice structure, distributive lattice and complete lattice. It is shown that the Li-SRCFP set is both a double Stone algebra and a boolean lattice. Furthermore, we looked into some algebras that dealt with the fixed points of Li-SRCGs. Applications of the algebraic structures available in covering soft sets to soft graphs may reveal new facets of graph theory. This work shows a novel approach for dealing with fixed points based on soft rough covering graphs. The future work will be focused on
(i) constructing the fixed points sets based on some other kind of soft rough graphs;
(ii) constructing the fixed points sets based on multi-granular soft rough covering sets;
(iii) constructing the fixed points sets based on upper approximation operators;
(iv) comparison between the proposed study and the study to be used in (iii); $
(v) developing the family of SRCFP sets based on soft neighborhood (of elements of universe set) and studying the conditions that these sets become some lattice structure. Further, some algebra can be discussed related to SRCFP sets based on soft neighborhoods.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
The authors acknowledge the financial support provided by the Center of Excellence in Theoretical and Computational Science (TaCS-CoE), KMUTT.
This research was funded by National Science, Research and Innovation Fund (NSRF), and King Mongkut's University of Technology North Bangkok with Contract no. KMUTNB-FF-66-05.
The authors declare that they do not have any competing interests.
[1] |
H. G. Lee, J. S. Lowengrub, J. Goodman, Modelling pinchoff and reconnection in a Hele-Shaw Cell I: The models and their calibration, Physics of Fluids, 14 (2002), 492-513. doi: 10.1063/1.1425843
![]() |
[2] |
H. G. Lee, J. S. Lowengrub, J. Goodman, Modeling pinchoff and reconnection in a Hele-Shaw cell II: Analysis and simulation in the nonlinear regime, Phys. Fluids, 14 (2002), 514-545. doi: 10.1063/1.1425844
![]() |
[3] |
A. E. Diegel, C. Wang, X. M. Wang, S. M. Wise, Convergence analysis and error estimates for a second order accurate finite element method for the Cahn-Hilliard-Navier-Stokes system, Numer. Math., 137 (2017), 495-534. doi: 10.1007/s00211-017-0887-5
![]() |
[4] |
S. M. Wise, J. S. Lowengrub, H. B. Frieboes, V. Cristini, Three-dimensional multispecies nonlinear tumor growth-I Model and numerical method, J. Theor. Biol., 253 (2008), 524-543. doi: 10.1016/j.jtbi.2008.03.027
![]() |
[5] | M. Ebenbeck, H. Garcke, Analysis of Cahn-Hilliard-Brinkman models for tumour growth, PAMM, 19 (2019), e201900021. |
[6] |
H. Garcke, K. F. Lam, E. Sitka, V. Styles, A Cahn-Hilliard-Darcy model for tumour growth with chemotaxis and active transport, Math. Models Methods Appl. Sci., 26 (2016), 1095-1148. doi: 10.1142/S0218202516500263
![]() |
[7] |
M. Ebenbeck, P. Knopf, Optimal medication for tumors modeled by a Cahn-Hilliard-Brinkman equation, Calc. Var. Partial Dif., 58 (2019), 1-31. doi: 10.1007/s00526-018-1462-3
![]() |
[8] |
A. Shinozaki, Y. Oono, Spinodal decomposition in a Hele-Shaw cell, Phys. Rev. A, 45 (1992), R2161. doi: 10.1103/PhysRevA.45.R2161
![]() |
[9] |
L. Ded, H. Garcke, K. F. Lam, A Hele-Shaw-Cahn-Hilliard model for incompressible two-phase flows with different densities, J. Math. Fluid Mech., 20 (2018), 531-567. doi: 10.1007/s00021-017-0334-5
![]() |
[10] |
X. Wang, Z. Zhang, Well-posedness of the Hele-Shaw-Cahn-Hilliard system, Ann. Inst. H. Poincaré Anal. Non Linéaire, 30 (2013), 367-384. doi: 10.1016/j.anihpc.2012.06.003
![]() |
[11] |
A. E. Diegel, W. Cheng, S. M. Wise, Stability and convergence of a second-order mixed finite element method for the Cahn-Hilliard equation, IMA J. Numer. Anal., 36 (2016), 1867-1897. doi: 10.1093/imanum/drv065
![]() |
[12] |
R. Guo, Y. Xia, Y. Xu, An efficient fully-discrete local discontinuous Galerkin method for the Cahn-Hilliard-Hele-Shaw system, J. Comput. Phys., 264 (2014), 23-40. doi: 10.1016/j.jcp.2014.01.037
![]() |
[13] |
S. M. Wise, Unconditionally stable finite difference, nonlinear multigrid simulation of the Cahn-Hilliard-Hele-Shaw system of equations, J. Sci. Comput., 44 (2010), 38-68. doi: 10.1007/s10915-010-9363-4
![]() |
[14] | W. Chen, Y. Liu, C. Wang, S. M. Wise, Convergence analysis of a fully discrete finite difference scheme for the Cahn-Hilliard-Hele-Shaw equation, Math. Comput., 85 (2016), 2231-2257. |
[15] |
Y. Liu, W. Chen, C. Wang, S. M. Wise, Error analysis of a mixed finite element method for a Cahn-Hilliard-Hele-Shaw systemn, Numer. Math., 135 (2017), 679-709. doi: 10.1007/s00211-016-0813-2
![]() |
[16] |
Y. Guo, H. Jia, J. Li, M. Li, Numerical analysis for the Cahn-Hilliard-Hele-Shaw system with variable mobility and logarithmic Flory-Huggins potential, Appl. Numer. Math., 150 (2020), 206-221. doi: 10.1016/j.apnum.2019.09.014
![]() |
[17] |
H. Jia, H. Hu, L. Meng, A large time-stepping mixed finite method of the modified Cahn-Hilliard equation, B. Iran. Math. Soc., 46 (2020), 1551-1569. doi: 10.1007/s41980-019-00342-z
![]() |
[18] | Y. Xian, Uncertainty quantification of modified Cahn-Hilliard equation for image inpainting, arXiv: 1906.07264. |
[19] | A. C. Aristotelous, O. Karakashian, S. M. Wise, A mixed discontinuous Galerkin, convex splitting scheme for a modified Cahn-Hilliard equation and an efficient nonlinear multigrid solver, Discrete Cont. Dyn. B, 18 (2013), 2211-2238. |
[20] |
T. T. Medjo, Unique strong and attractor of a three dimensional globally modified Cahn-Hilliard-Navier-Stokes model, Appl. Anal., 96 (2017), 2695-2716. doi: 10.1080/00036811.2016.1236924
![]() |
[21] | H. Suzuki, Existence and stability of multi-layered solutions in modified Cahn-Hilliard systems, Memoirs of the Faculty of Education Shiga University Natural Science, 56 (2006), 97-108. |
[22] |
J. W. Cahn, J. E. Hilliard, Free energy of a nonuniform system. I. Interfacial free energy, J. Chem. Phys., 28 (1958), 258-267. doi: 10.1063/1.1744102
![]() |
[23] |
J. W. Cahn, Free energy of a nonuniform system. II. Thermodynamic basis, J. Chem. Phys., 30 (1959), 1121-1124. doi: 10.1063/1.1730145
![]() |
[24] |
J. W. Cahn, J. E. Hilliard, Free energy of a nonuniform aystem, III. Nucleation in a two component incompressible fluid, J. Chem. Phys., 31 (1959), 688-699. doi: 10.1063/1.1730447
![]() |
[25] |
H. Jia, Y. Guo, J. Li, Y. Huang, Analysis of a novel finite element method for a modified Cahn-Hilliard-Hele-Shaw system, J. Comput. Appl. Math., 376 (2020), 112846. doi: 10.1016/j.cam.2020.112846
![]() |
[26] |
D. Han, A decoupled unconditionally stable numerical scheme for the Cahn-Hilliard-Hele-Shaw system, J. Sci. Comput., 66 (2016), 1102-1121. doi: 10.1007/s10915-015-0055-y
![]() |
[27] |
S. Minjeaud, An unconditionally stable uncoupled scheme for a triphasic Cahn-Hilliard/Navier-Stokes model, Numer. Meth. Part. D. E., 29 (2013), 584-618. doi: 10.1002/num.21721
![]() |
[28] |
Y. Gao, R. Li, L. Mei, Y. Lin, A second-order decoupled energy stable numerical scheme for Cahn-Hilliard-Hele-Shaw system, Appl. Numer. Math., 157 (2020), 338-355. doi: 10.1016/j.apnum.2020.06.010
![]() |
[29] |
J. Zhao, X. Yang, J. Shen, Q. Wang, A decoupled energy stable scheme for a hydrodynamic phase-field model of mixtures of nematic liquid crystals and viscous fluids, J. Comput. Phys., 305 (2016), 539-556. doi: 10.1016/j.jcp.2015.09.044
![]() |
[30] | J. Zhao, Second-order decoupled energy-stable schemes for Cahn-Hilliard-Navier-Stokes equations, arXiv: 2103.02210. |
[31] | J. Zhao, A general framework to derive linear, decoupled and energy-stable schemes for reversible-irreversible thermodynamically consistent models: Part I Incompressible Hydrodynamic Models, arXiv: 2103.02203. |
[32] |
J. Shin, H. G. Lee, J. Y. Lee, Convex splitting Runge-Kutta methods for phase-field models, Comput. Math. Appl., 73 (2017), 2388-2403. doi: 10.1016/j.camwa.2017.04.004
![]() |
[33] |
X. Wu, G. J. van Zwieten, K. G. van der Zee, Stabilized second-order convex splitting schemes for Cahn-Hilliard models with application to diffuse-interface tumor-growth models, Int. J. Numer. Meth. Bio., 30 (2014), 180-203. doi: 10.1002/cnm.2597
![]() |
[34] |
N. Alikatos, P. W. Bates, X. Chen, Convergence of the Cahn-Hilliard equation to the Hele-Shaw model, Arch. Ration. Mech. An., 128 (1994), 165-205. doi: 10.1007/BF00375025
![]() |
[35] |
J. L. Guermond, P. Minev, J. Shen, An overview of projection methods for incompressible flows, Comput. Method. Appl. M., 195 (2006), 6011-6045. doi: 10.1016/j.cma.2005.10.010
![]() |
[36] |
A. E. Diegel, X. H. Feng, S. M. Wise, Analysis of a mixed finite element method for a Cahn-Hilliard-Darcy-Stokes system, SIAM J. Numer. Anal., 53 (2015), 127-152. doi: 10.1137/130950628
![]() |
R╲V | x1 | x2 | x3 | x4 | x5 | x6 | x7 |
σ1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 1 | 0 | 1 | 1 | 1 |
σ4 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
σ5 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ6 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
R╲E | e1 | e2 | e3 | e4 | e5 | e6 | e7 | e8 | e9 | e10 |
σ1 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
σ2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ4 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 |
σ5 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
σ6 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 0 |
R∖V | r1 | r2 | r3 | r4 |
σ1 | 1 | 0 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 1 | 1 | 0 |
σ4 | 0 | 1 | 0 | 1 |
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 1 | 0 |
σ3 | 0 | 0 | 1 | 0 |
σ4 | 1 | 1 | 1 | 1 |
R╲V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 0 | 1 | 1 |
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 1 | 0 |
σ2 | 1 | 0 | 0 | 0 |
σ3 | 1 | 0 | 1 | 1 |
σ4 | 0 | 1 | 1 | 0 |
R╲V | x1 | x2 | x3 | x4 | x5 | x6 | x7 |
σ1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 1 | 0 | 1 | 1 | 1 |
σ4 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
σ5 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ6 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
R╲E | e1 | e2 | e3 | e4 | e5 | e6 | e7 | e8 | e9 | e10 |
σ1 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
σ2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 |
σ3 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 |
σ4 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 |
σ5 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
σ6 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 0 |
R∖V | r1 | r2 | r3 | r4 |
σ1 | 1 | 0 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 1 | 1 | 0 |
σ4 | 0 | 1 | 0 | 1 |
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 1 | 1 | 1 | 0 |
σ3 | 0 | 0 | 1 | 0 |
σ4 | 1 | 1 | 1 | 1 |
R╲V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 0 | 0 |
σ2 | 0 | 1 | 1 | 0 |
σ3 | 1 | 0 | 1 | 1 |
R∖V | x1 | x2 | x3 | x4 |
σ1 | 1 | 1 | 1 | 0 |
σ2 | 1 | 0 | 0 | 0 |
σ3 | 1 | 0 | 1 | 1 |
σ4 | 0 | 1 | 1 | 0 |