مدل برنامه ریزی دو مرحله ای نادرست-تصادفی برای برنامه ریزی انتشار تجارت دی اکسید کربن تحت عدم قطعیت

In this study, a two-stage inexact-stochastic programming (TISP) method is developed for planning carbon dioxide (CO2) emission trading under uncertainty. The developed TISP incorporates techniques of interval-parameter programming (IPP) and two-stage stochastic programming (TSP) within a general optimization framework. The TISP can not only tackle uncertainties expressed as probabilistic distributions and discrete intervals, but also provide an effective linkage between the pre-regulated greenhouse gas (GHG) management policies and the associated economic implications. The developed method is applied to a case study of energy systems and CO2 emission trading planning under uncertainty. The results indicate that reasonable solutions have been generated. They can be used for generating decision alternatives and thus help decision makers identify desired GHG abatement policies under various economic and system-reliability constraints.

Currently, a large amount of electricity relies primarily on nonrenewable energy supplies, such as coal, natural gas and petroleum [1]. Greenhouse gas (GHG) is primary gas emitted from these fossil fuels combustion, and increasing concentration of GHG [e.g., carbon dioxide (CO2)] is likely to accelerate the rate of global warming [2], [3], [4], [5], [6], [7] and [8]. The present measured concentration of CO2 in the atmosphere is approximately 30% higher than Pre-Industrial Revolution (1850s) levels [5]. Many scientists concern about the increase of global CO2 and other GHG emissions, which lead to the increase in surface temperature, the change in the global climate, and the rise in sea level. Some of them question that whether energy supplies can meet GHG mitigation standards with increasing electricity demands. Moreover, a number of researchers are in a puzzle about how to balance increasing electricity demands (due to the population growth and the economic development), less fossil fuel consumption, and mandated requirement for reducing GHG emission [1]. A large number of research works were undertaken for the planning of GHG mitigation in integrated energy and environmental management systems. For example, economic incentive (typically a carbon tax) was proposed to promote less carbon-intensive fuels and to develop alternatives [9]. Renewable energy sources or less GHG intensive fuels were used, such as nuclear power and natural gas [10], [11] and [12]. Sequestration facilities were built up and used to capture GHG emitted from power plants during electricity generation process [12] and [13]. Besides, GHG emission trading was envisaged within the Kyoto protocol as one of the so-called flexible mechanisms, it was introduced to help attain reduction of GHG emission in a cost-effective way [14], [15] and [16]. Previously, deterministic methods were extensively used for managing GHG emission in energy systems [17], [18], [19], [20], [21], [22] and [23]. However, an integrated energy and environmental management system often contains various uncertainties that may exist in electricity demand and supply, electricity generation processes, related economic parameters, GHG emission inventories, and errors in the measurement instruments. For example, GHG emissions from the electricity generation sector can be influenced by stochastic events such as electricity demand, which may fluctuate from time to time. Meanwhile, the quality of information on generated energy and cost/benefit coefficients are not sufficient, which may vacillate within a certain interval. As a result, a number of research efforts were conducted for dealing with various uncertainties in the integrated energy and environmental management systems, such as interval mathematical programming (IMP) and stochastic mathematical programming (SMP) [24], [25], [26], [27], [28] and [29]. IMP allows uncertainties to be directly communicated into the optimization process and resulting solutions, it does not lead to more complicated intermediate models and does not require distribution information for model parameters [30]. Nevertheless, IMP has difficulties when the right-hand sides of a model are highly uncertain, especially with uncertainties expressed as possibilistic and/or probabilistic distributions, which may lead to the loss of valuable information in many real-world decision-making problems [31] and [32]. In comparison, SMP is effective for decision problems whose coefficients (input data) are uncertain but could be represented as chances or probabilities, which has been extensively applied to energy systems planning [33], [34], [35] and [36]. Two-stage stochastic programming (TSP) is a typical SMP method, which is an effective alternative for tackling problems where an analysis of policy scenarios is desired and the right-hand-side coefficients are random with known probability density functions (PDFs) [3], [37], [38] and [39]. In TSP, the first-stage decision is to be made before uncertain information is revealed, whereas the second-stage one (recourse) is to adapt to the previous decision based on the further information; the second-stage decision is used to minimize ‘penalties’ that may appear due to any infeasibility [38], [40], [41], [42], [43], [44] and [45]. However, the major problem of stochastic programming method is that there are increased data requirements for the specification of the probability distribution of the coefficients which may affect the practical applicability [46]. For example, in an integrated energy and environmental system, a planner may know that the daily pollutant and/or GHG emission rate fluctuates within a certain interval, but he may find it is difficult to state a meaningful probability distribution for this variation [31] and [47]. Therefore, one potential approach for better accounting for the uncertainties and economic penalties is to incorporate the interval-parameter programming (IPP) and TSP techniques within a general optimization framework. This will lead to a two-stage inexact-stochastic linear programming method. For example, Li et al. [29] developed an inexact fuzzy-robust two-stage programming model for managing sulfur dioxide abatement in an energy system under uncertainty, where fuzzy programming was introduced into a TSP framework to deal with uncertainties presented in terms of fuzzy sets and random variables. Huang and Loucks [39] proposed an inexact two-stage stochastic programming (ITSP) model to address the uncertainties. In their study, the concept of inexact optimization was incorporated within a two-stage stochastic programming framework. The model was applied to a case study of water resources management. Moreover, few research works focused on the TSP method for GHG emission trading planning within an integrated energy and environmental management system. Therefore, the objective of this study aims to develop a two-stage inexact-stochastic programming (TISP) method for CO2 emission trading planning within an integrated energy and environmental management system. The developed TISP will integrate techniques of IPP and TSP into a general optimization framework. Uncertainties expressed as probabilistic distributions and interval values will be reflected. A case study will then be provided for demonstrating applicability of the developed method. A number of policy scenarios that are associated with different mitigation levels of CO2 emission permits will be analyzed. The results can help decision makers not only discern optimal energy-allocation patterns, but also gain deep insights into the tradeoffs between CO2 emission trading and economic objective. The paper is organized as follows: Section 2 describes the statement of energy and environmental management problem, and formulates the CO2 emission trading and non-trading models; Section 3 provides the results analysis of the case study; Section 4 discusses the potential limitations and extensions of the proposed TISP method; Section 5 presents conclusions of the work; Appendix A depicts the detailed methodology of the proposed model.

In this study, a two-stage inexact-stochastic linear program- ming (TISP) method has been developed for planning CO 2 emission mitigation with trading scheme. The TISP method can effectively deal with uncertainties presented as both probabilities and inter- vals within a multi-period, multi-demand-level, and multi-option context. Solutions of the model provide an effective linkage be- tween the pre-regulated energy and environmental policies and the associated economic implications (e.g., losses and penalties caused by improper policies). The solutions are combinations of deterministic, interval and distributional information, and can thus facilitate the reflection for different forms of uncertainties. The interval solutions can help managers obtain multiple decision alternatives, as well as provide bases for further analyses of trade- offs between system benefit and system-failure risk. The devel- oped model can also help analyze various trading policies under different CO 2 emission allowances and mitigation efficiencies. The developed method has been applied to a case study of CO 2 emission trading for a regional power system. The results obtained indicate that CO 2 emission scheme can be efficiently performed to maximize the power net benefits through trading when the level of CO 2 emission permit is low. In addition, a number of scenarios cor- responding to different CO 2 emission management policies under varied mitigation levels of total emission permits have been ana- lyzed. The results indicate that CO 2 emission trading is effective for CO 2 permit reallocation and different policies for CO 2 manage- ment are associated with different levels of CO 2 management cost and CO 2 mitigation-failure risk. Although application of TISP model to CO 2 emission trading is a new attempt and the TISP may be fur- ther enhanced or extended, the results obtained imply that the developed model is applicable and effective in CO 2 emission miti- gation through coupled mechanism between emission trading and control measure.