Near‐optimal control of a stochastic vegetation‐water system with reaction diffusion

Abstract

In this article, we study the near-optimal control of a class of stochastic vegetation-water model. The near-optimal control is one problem in which the density of vegetation and water is higher at the lowest cost. We have provided a priori estimates of the vegetation and water densities and obtained the sufficient and necessary conditions for the system's near-optimal control problem by applying the maximum condition of the Hamiltonian function and the Ekeland principle. A numerical simulation is presented to verify our theoretical results. KEYWORDSHamiltonian function, near-optimal, sufficient and necessary conditions, vegetation-water model MSC CLASSIFICATION 60H10 INTRODUCTIONSevere desertification has led to the extreme degradation of ecosystems, decline in living standards of residents, and large economic losses. These issues have aroused extensive scholarly attention in different branches, which has achieved some research results. 1-4 However, the restoration of degraded grasslands still faces great challenges. The majority of the research results obtained by ecologists are mainly experimental results, and there is a lack of quantitative analysis of the degraded grassland restoration problem. It is meaningful for scholars to formulate mathematical models based on experimental data and provide strategies for desertification restoration.Mathematical modeling is well-known as a tool for exploring the mechanisms of vegetation patterns. [5][6][7][8][9] To date, a variety of ecosystem models have been established. Ying 10 established a grassland plant-water coupling dynamics model and simulated the dynamic changes of plants and water under different precipitation parameters, regardless of the influence of the external environment. Suna 11 analyzed the trend of plant changes in semiarid areas, used complex network technology to from an early warning index for the oasis-desertization transition, and applied the indicators to the local positive feedback vegetation model. Liu et al 12 proposed a cross-diffusion vegetation system based on the reaction-diffusion equation and revealed that cross-diffusion is an important mechanism for dynamic vegetation changes.Note that the above research results assume that the system parameters are deterministic. However, in real life, the transmission coefficient may be affected by many random environment factors, such as temperature, wind, rain, and snow. 13 Stochastic fluctuation has been included in vegetation-water models. [14][15][16] For example, Zhang et al analyzed the dynamic evolution of a vegetation-water system disturbed by Gauss noise. 14 Guttal et al studied a simple ecological model in which mutations occur when control parameters change. 16 However, optimal control with stochastic vegetation-water models has not been widely studied.Math Meth Appl Sci. 2020;43:6043-6061.wileyonlinelibrary.com/journal/mma

Analysis of a nonlinear perturbed HLIV model with viral production and multiple latent stages

“…One may propose some more realistic and interesting models, we can consider the effects of other types of random perturbations (such as discrete Markov switching or Lévy jumps et al) on the transmission dynamics of HLIV models. The motivation is that the dynamics of viral establishment may encounter human interventions and sudden environmental changes [32]. As far as we know, there is little literature studying high dimensional HLIV models with Lévy jumps because it is extremely difficult to solve the relevant Fokker-Planck equation due to its high dimension.…”

Dynamics analysis and optimal control of SIVR epidemic model with incomplete immunity

“…First, we define the objective function as follows. Let the objective function corresponding to the control system ( 4.1 ) be as follows: where , [ 29 ]. The values of other parameters are the same as in Sect.…”