基于神经网络的单元自动机CA及真实和忧化的城市模拟

提出了一种基于神经网络的单元自动机（ＣＡ）。ＣＡ已被越来越多地应用在城市及其它地理现象的模拟中。ＣＡ模拟所碰到的最大问题是如何确定模型的结构和参数。模拟真实的城市涉及到使用许多空间变量和参数。当模型较复杂时，很难确定模型的参数值。本模型的结构较简单，模型的参数能通过对神经网络的训练来自动获取。分析表明，所提出的方法能获得更高的模拟精度，并能大大缩短寻找参数所需要的时间。通过筛选训练数据，本模型还可以进行优化的城市模拟，为城市规划提供参考依据。     

基于多智能体的土地利用模拟与规划模型

利用多智能体和元胞自动机对城市土地资源的可持续利用进行了规划。根据环境经济学资源分配原理和可持续发展理论,提出结合多智能体及元胞自动机的微观规划模型,在时间和空间上合理分配及规划城市土地资源的利用,以避免浪费不可再生的土地资源。该模型由相互作用的多智能体层、元胞自动机层和环境因素层组成,可方便地探索不同土地利用政策下城市土地利用发展情景,能够为城市规划提供有用的决策依据。以广州市海珠区为实验区,在可持续发展为前提的规划下,模拟了1995-2010年的城市扩展的动态变化,并讨论了在不同规划情景下城市土地资源的利用效率及合理性。     

顾及空间尺度效应的城市土地利用变化精细化模拟

城市土地利用变化模拟是优化土地资源配置的科学依据,提高其精细化程度和可靠性有助于准确把握城市用地发展趋势,对城市土地资源精准调控具有重要意义。基于宏观遥感分类的土地利用变化模拟,难以在街区尺度上揭示城市用地社会功能变化及精细化模拟中空间尺度效应来源和作用机理。本文联合遥感影像和POI数据识别出城市土地利用精细化特征,运用响应面法率定土地利用精细化模拟的最优空间尺度组合,在此基础上,利用CA-Markov模型开展了未来土地利用变化的精细化模拟。以武汉市中心城区为应用案例,研究结果表明：基于POI 的城市土地利用精细化识别方法,可以深度解析城市建设用地的社会功能,极大改善了传统基于遥感的土地覆被宏观解译效果;研究区土地利用变化元胞自动机精细化模拟的最优空间尺度组合是30 m元胞、7×7邻域以及冯诺依曼邻域类型,采用最优空间尺度组合能够提高土地利用变化精细化模拟的可靠性。响应面试验设计结果可有效识别精细化模拟过程中空间尺度效应的主要来源,并区分其对模拟精度的影响程度与正负效应;预计到2025年,研究区建设用地范围将继续向周边扩张,各类型用地之间互为交织,土地利用空间格局将呈更加破碎化趋势。     

Simulating the 1999 Capbreton canyon turbidity current with a Cellular Automata model

A numerical model has been developed for the simulation of turbidity currents driven by nonuniform, non cohesive sediment and flowing over a complex three dimensional submarine topography. The model is based on an alternative approach known as Cellular Automata paradigm. The model is validated by comparing a simulation with a reported field-scale event. The chosen case is a turbidity current which occurred in Capbreton Canyon and was initiated by a storm in December 1999. Using data from recent oceanographic cruises, the deposit of the event has been precisely described, which constrain values of model parameters. The model simulates the 1999 turbidity current over the actual canyon topography and related turbidite using three different types of particle. The model successfully simulates areas of erosion and deposition in the canyon. It predicts the vertical and longitudinal grain size evolution, and shows that the fining-up sequence can be deposited by several phases of deposition and erosion related to the current energetic variation during its evolution. This result could explain the presence of intrabed contacts or the frequent lack of facies in Bouma sequences.     

Cellular automata as analysis and synthesis engines at the geomorphology–ecology interface

The linkages between ecology and geomorphology can be difficult to identify because of physical complexity and the limitations of the current theoretical representations in these two fields of study. Deep divisions between these disciplines are manifest in the methods used to simulate process, such as rigidly physical-deterministic methods for many aspects of geomorphology compared with purely stochastic simulations in many models of change in landcover. Practical and theoretical research into ecology–geomorphology linkages cannot wait for a single simulation schema which may never come; as a result, studies of these linkages often appear disjointed and inconsistent.The grid-based simulation framework for cellular automata (CA) allows simultaneous use of competing schemas. CA use in general geographic studies has been primarily limited to urban simulations models of change for land cover, both highly stochastic and/or expert rule-based. In the last decade, however, methods for describing physically deterministic systems in the CA framework have become much more accurate. The possibility now exists to merge separate CA simulations of different environmental systems into unified “multiautomata” models. Because CAs allow transition rules that are deterministic, probabilistic, or expert rule-based, they can immediately incorporate the existing knowledge rules in ecology and geomorphology. The explicitly spatial nature of CA provides a map-like framework that should allow a simple and deeply rooted connection with the mapping traditions of the geosciences and ecological sciences.     

Simulation of urban expansion via integrating artificial neural network with Markov chain – cellular automata

Accurate simulations and predictions of urban expansion are critical to manage urbanization and explicitly address the spatiotemporal trends and distributions of urban expansion. Cellular Automata integrated Markov Chain (CA-MC) is one of the most frequently used models for this purpose. However, the urban suitability index (USI) map produced from the conventional CA-MC is either affected by human bias or cannot accurately reflect the possible nonlinear relations between driving factors and urban expansion. To overcome these limitations, a machine learning model (Artificial Neural Network, ANN) was integrated with CA-MC instead of the commonly used Analytical Hierarchy Process (AHP) and Logistic Regression (LR) CA-MC models. The ANN was optimized to create the USI map and then integrated with CA-MC to spatially allocate urban expansion cells. The validated results of kappa and fuzzy kappa simulation indicate that ANN-CA-MC outperformed other variously coupled CA-MC modelling approaches. Based on the ANN-CA-MC model, the urban area in South Auckland is predicted to expand to 1340.55 ha in 2026 at the expense of non-urban areas, mostly grassland and open-bare land. Most of the future expansion will take place within the planned new urban growth zone.     

Modelling sustainable urban growth in a rapidly urbanising region using a fuzzy-constrained cellular automata approach

This article presents an application of a fuzzy-constrained cellular automata model to simulate the spatio-temporal processes of urban growth in the rapidly growing Gold Coast City in Southeast Queensland, Australia. Urban growth has been captured in the model as a continuous process in space and over time, which has been affected by a set of primary and secondary transition rules. The primary transition rules deal with the propensity of a local area for development and the impact of its neighbouring cells on such development, while the secondary transition rules reflect the influences of environmental and institutional factors on urban growth. Application of the model demonstrates its re-applicability to different regions and the effectiveness of the cellular automata technique in studying urban dynamics. It also provides tools to explore sustainable urban growth options under different socio-environmental and planning control factors. A sustainable urban future of the region is achievable if development is managed to maintain a balance amongst ecological conservation, economic growth and the contemporary Australian lifestyle.     

The inclusion of differentially assessed lands in urban growth model calibration: a comparison of two approaches using SLEUTH

This research sought to understand the role that differentially assessed lands (lands in the United States given tax breaks in return for their guarantee to remain in agriculture) play in influencing urban growth. Our method was to calibrate the SLEUTH urban growth model under two different conditions. The first used an excluded layer that ignored such lands, effectively rendering them available for development. The second treated those lands as totally excluded from development. Our hypothesis was that excluding those lands would yield better metrics of fit with past data. Our results validate our hypothesis since two different metrics that evaluate goodness of fit both yielded higher values when differentially assessed lands are treated as excluded. This suggests that, at least in our study area, differential assessment, which protects farm and ranch lands for tenuous periods of time, has indeed allowed farmland to resist urban development. Including differentially assessed lands also yielded very different calibrated coefficients of growth as the model tried to account for the same growth patterns over two very different excluded areas. Excluded layer design can greatly affect model behavior. Since differentially assessed lands are quite common through the United States and are often ignored in urban growth modeling, the findings of this research can assist other urban growth modelers in designing excluded layers that result in more accurate model calibration and thus forecasting.     

A heuristic cellular automata approach for modelling urban land-use change based on simulated annealing

This article presents a novel cellular automata (CA) approach to simulate the spatio-temporal process of urban land-use change based on the simulated annealing (SA) algorithm. The SA algorithm enables dynamic optimisation of the CA's transition rules that would otherwise be difficult to configure using conventional mathematical methods. In this heuristic approach, an objective function is constructed based on a theoretical accumulative disagreement between the simulated land-use pattern and the actual land-use pattern derived from remotely sensed imagery. The function value that measures the mismatch between the actual and the simulated land-use patterns would be minimised randomly through the SA process. Hence, a set of attribution parameters that can be used in the CA model is achieved. An SA optimisation tool was developed using Matlab and incorporated into the cellular simulation in GIS to form an integrated SACA model. An application of the SACA model to simulate the spatio-temporal process of land-use change in Jinshan District of Shanghai Municipality, PR China, from 1992 to 2008 shows that this modelling approach is efficient and robust and can be used to reconstruct historical urban land-use patterns to assist with urban planning policy-making and actions. Comparison of the SACA model with a typical CA model based on a logistic regression method without the SA optimisation (also known as LogCA) shows that the SACA model generates better simulation results than the LogCA model, and the improvement of the SACA over the LogCA model is largely attributed to higher locational accuracy, a feature desirable in most spatially explicit simulations of geographical processes.     

基于耗费场的最优路径算法研究

在GIS中,有一类应用是基于连续分布耗费场的最优路径计算问题,如公路选线等。该文首先采用规则格网对耗费场进行建模,进而基于规则格网表现,分别建立网络实现模型和元胞自动机实现模型,并探讨了基于元胞自动机模型的最优路径算法,通过实例研究,说明了算法的正确性。