An intelligent planning model for the development and utilization of urban underground space with an application to the Luohu District in Shenzhen

Highlights

• An intelligent planning model for UUS development and utilization was established.

• An improved AIGA with an additional correlation operator was developed.

• A three-dimensional UUS planning scheme of Luohu District in Shenzhen was derived.

Abstract

Development and utilization of Urban underground space (UUS) is one of the key ways for developing sustainable cities. However, the development and utilization of UUS is currently occurring in a passive manner, lacking proactive and quantitative planning. In this paper, an intelligent planning model was established to guide the future development and utilization of UUS. The proposed model, in which more than twenty factors were considered, is capable of analysing the suitability of UUS development in different regions and at different depths. In addition, an improved artificial intervention genetic algorithm (AIGA) with an additional correlation operator was integrated into the proposed model to achieve the optimal two-dimensional (2D) planning scheme by quantitively analysing the matching of the ground surface and underground space, and the compactness and suitability of underground space. A three-dimensional (3D) planning scheme was obtained by analysing the stratification of the underground space and hierarchical development. The proposed intelligent planning model was, subsequently, applied to the development and utilization of UUS in Luohu District in Shenzhen. An analysis of the geological conditions and the current construction status of the ground surface showed that 69.4% of the underground space is suitable for shallow layer developments, and this proportion further increases to 76% for the sub-shallow, sub-deep and deep layers, indicating that the Luohu District is highly suitable for developing underground space. Furthermore, the derived optimal 2D planning scheme indicated that the suitability component plays a greater important role in the objective values than the matching and compactness components, and more attentions should be paid to the development of municipal facilities in the shallow layer in Luohu District. The derived 3D planning scheme of Luohu District showed the characteristics of an “inverted cone”, where the utilization rate significantly decreases with increasing depth. The findings in this paper provide new insights for the future development and utilization of UUS.

Introduction

Due to the intense urbanization, existing cultivated land is reaching the limit to support the population, particularly in some large cities. A series of “urban syndromes”, e.g., urban green space reduction, traffic congestion, environmental pollution and resource shortages, have formed and seriously restricted the sustainability of city developments (Inmaculada, 2008, Kalnay and Cai, 2003, Xie, 2011). The development and utilization of urban underground space (UUS) is regarded as an effective way to solve these problems (Bobylev, 2006, Carmody and Sterling, 1983, Monnikhof et al., 1998, Sterling, 1997, Zhao and Künzli, 2016). In addition, the development of UUS has demonstrated prominent advantages than that of ground space in terms of urban space expansion and environmental improvement (Rönkä et al. 1998).

Underground space is a non-renewable resource because it is difficult to transform once assigned to a specific function (Sterling 1983). However, a great deal of underground infrastructure has been passively designed and constructed, causing numerous problems. For example, the random and reckless exploitation of underground coal leads to many goafs and other potential crises (Gong et al. 2008). Due to the lack of overall planning, the use rate of underground commercial space is low, and a large portion of underground space has been abandoned, leading to the waste of underground space resources (Pan 2012). Moreover, the layout of previous underground civil defence structures has influenced the current planning of underground infrastructure such as underground utility tunnels (Zhang et al. 2012). Therefore, the scientific and systematic planning of the development and utilization of UUS is essential.

A suitability evaluation of UUS development is crucial to the intelligent planning of sustainable cities. A UUS suitability evaluation provides key information such as geological conditions and land usage of ground surface and underground space for the intelligent planning of cities (Malone, 1996, Peng et al., 2009, Liu et al., 2018, Zhao et al., 2016). Sterling and Nelson (1982) identified different geological factors, i.e., rock characteristics, soil layer distribution, hydrogeological distribution and topographic gradient, and classified the effects of different ground and underground infrastructure on the development and utilization of UUS. In addition, some other factors, e.g., society, economy, psychology and actualities, are also taken into consideration in a suitability evaluation (Bobylev, 2005, Monnikhof et al., 1998). Many analytical methods have therefore been developed to systematically analyse and characterize the effect of the abovementioned factors on the development and utilization of UUS. For example, the analytic hierarchy process (AHP) is widely used to comprehensive evaluate multi-factors and guided the design of urban underground development (Bobylev, 2011, Feng and Lin, 1999, Liu et al., 2015). One multilayer framework based on the fuzzy analytic hierarchy process (AHP) was proposed to evaluate the geological engineering effect on the suitability of UUS exploitation and was further applied to study a railway station area (Lu et al. 2016). Compared to the fuzzy AHP, the classical AHP method is more suitable to evaluate the suitability of the underground space for specific decision makers. Peng and Peng (2018a) synthetically used the most unfavourable grading method (MUGM) and the exclusive method (EM) on the analysis platform of Geographic Information System (GIS) to determine suitability classifications. However, the collection of influencing factors was not standardized and varied form different situations and scholars (Ron and Jyrki, 2001, Wu et al., 2013). Moreover, collecting information on the abovementioned factors is also difficult (Yu et al. 2006).

The intelligent planning of UUS aims to rationally guide the development and utilization of underground space according to actual economical, social, ecological and other needs and to be included in the master planning and detailed planning (Bobylev, 2009, Brian, 2015, Li et al., 2013a, Zhao et al., 2016). Master planning plays a critical role in coordinating the different needs of UUS in various regions and different urban systems, and has a significant influence on the overall development of urban areas and guides the detailed planning process (Qiao and Peng 2016). Detailed planning involves the use of a specific design layout that is based on master planning (Huang and Yu 2014). Numerous studies have been conducted to guide and optimize UUS planning. Rönkä et al. (1998) proposed the principles of underground space planning and suggested the optimal depths for different activities. Li et al. (2013b) proposed using integrated planning in Suzhou, China and conducted a comprehensive assessment of underground resources, i.e., underground space, groundwater, geomaterials and geothermal energy. Zhao et al. (2016) proposed that government at all levels should develop UUS planning based on local economic, social and ecological development needs and urban master planning.

However, UUS planning is still facing many problems when it is applied to practice. First, the development of UUS planning is still in a preliminary stage, lagging behind actual demand and lacking relevant laws, standards, systems, etc. (Lu and Ji, 2012, Sterling et al., 2012). Second, current UUS planning is basically qualitative in nature and is not based on mature theories and methods (Volchko et al. 2020). UUS planning is currently in a rough and preliminary planning state due to the difficulty to quantitatively consider actual economic, social and ecological needs. Note that, such quantitative manner is to obtain an optimal solution of the underground space planning by analyzing the effects of economic, social, ecological and geological factors using a certain mathematical algorithm. Notably, although the quantitative intelligent planning models with mature algorithms are seldom used in the real practices of UUS planning, they have been adopted to guide the development of urban ground surface (Balling et al., 1999, Day et al., 1999, Liu et al., 2015, Stewart et al., 2004). It is suggested that quantitative ground surface planning maximizes the economic, social and ecological benefits by arranging different types of ground surface into spatial units, which is challenging because it not only addresses the tradeoffs of multiple objectives but also conciliates the competition of different land uses for the same spatial unit (Chen et al., 2018, Aerts et al., 2003, Eastman et al., 1995).

Three main challenges may hinder the development and utilization of UUS in a quantitative manner. First, it is difficult to establish the qualitative correlation between underground space and ground surface, which inhibits the ability of the current ground development status to drive intelligent planning for underground space (Liu and Liao, 2017, Zhang et al., 2018). Second, the objective function of underground space is difficult to solve using traditional algorithms because traditional algorithms, e.g., the genetic algorithm, are difficult to address constraints that widely exist in the UUS planning, e.g., excluding the unusable space. (Anastasios and Dimitrios, 2016, Liu et al., 2014). Third, it is difficult to combine the three-dimensional (3D) planning of underground space with the principle of hierarchical development (Admiraal 2006). Therefore, further quantitative research on underground space planning is needed.

To proactively and quantitatively design UUS planning, an intelligent planning model for the development and utilization of UUS, which is capable of analysing the correlation between ground surface and underground space and further giving the optimal two-dimensional (2D) and 3D planning schemes, is proposed in this paper. First, the suitability evaluation method of the development and utilization of UUS is introduced. Second, based on the correlation between ground surface and underground space, the 2D planning of UUS was generated by using the artificial intervention genetic algorithm (AIGA). Third, the 3D planning of UUS is proposed according to the principles of depth stratification and hierarchical development. Finally, the proposed intelligent planning model is applied to the development and utilization of UUS in Luohu District, Shenzhen. The findings in this paper could provide innovative insights for the future development and utilization of UUS.