| Relative abundance of coal resources determines the occurrence condition of its dominant position in the industry in our country, however, coal-dominated energy consumption structure leads to low efficiency of energy use and more emissions, which caused great threat to the ecological environment. Now the fossil fuels are gradually drying up, and exploiting new energy is high cost and difficult. With the increasing prevalence of international carbon tax, its economy will be hinder by new trade barriers due to no taking energy conservation and emissions reduction measures. Therefore, energy conservation and emissions reduction as an effective way to achieve a win-win economic development and environmental protection, not only is our own inherent requirement of sustainable development, but also an important contribution to mitigate global climate change.The coal industry as one of nine key energy-consuming industries, plays an important role in energy conservation and emissions reduction work. With the implementation of energy conservation and emissions reduction strategy of the "11th Five-Year Plan", the coal industry both have great progress to improve energy efficiency and reduce emissions. In order to achieve the conservation and emissions reduction goal of the "12th Five-Year Plan", we must continue to rely on the power of advanced science and technology, change the past mode of production of high input, high energy consumption and high emission, to improve energy efficiency, enhance recycling emissions, and develop circular economy.Coal mines, complex resource allocation system, need to consider a number of factors such as economic, social, environmental and technology in the strategy of energy conservation and emissions reduction. From the perspective of sustainable development, this paper simultaneously takes into account the economic benefits, energy efficiency and environmental benefits of the three goals, trying to use energy-saving equipments, invest in certain projects to control or utilization emissions, and optimize the production plan for coal mines by the approach of multi-objective optimization decisions on key processes. The main contents of this paper are as follows:(1) Potential of energy conservation and emissions reduction in coal mines. Before the development of energy saving strategies, you need to grasp the current situation of the energy consumption structure, energy efficiency, recovery and emissions of pollutants, using the data of energy consumption and pollution emissions at the end of the "11th Five-Year Plan", determine the input-output indexes, then calculate the energy efficiency and evaluate the potential of energy conservation and emissions reduction based on the CCR-DEA model investment.(2) Static multi-objective optimization model. The coal production, investments on the key process energy-saving equipment and governance or utilization projects in the "12th Five-Year Plan", are selected as the decision variables needed to be optimized, then espress the economic benefits, energy efficiency and environmental benefits, describe constraints included coal reserves, investment funds and processes relations, etc.. Finally, establish the static multi-objective optimization model for energy conservation and emissions reduction of coal mines.(3) Algorithm sovling static multi-objective model. Different from the single-objective optimization problem, multi-objective optimization problem is needed to not only distinguish feasible solutions and infeasible solutions, but also identify strengths and weaknesses among the similar solutions with multiple objectives. Solving these problems depends on the handling constraints correctly. Due to that the optimal solutions of multi-objective problem is a pareto set composed by a number of non-dominated solutions, so as much as possible non-dominated solutions searched in the solution space guarantee high-quality solution. The paper combines with NSGA-II and PSO, full play global search abilities of NSGA-II and local optimization features of PSO, and verify convergence, coverage, uniformity of the PSO-NSGA-II algorithm.(4) Dynamic multi-objective optimization model. According to the time effect of the investment decision-making, take each year of the period of "12th Five-Year Plan" as the optimization target, considerng the stage effect of energy conservation and emissions reduction, development of energy conservation and emissions reduction strategies for the next year based on the previous year’s coal production plans and investment decisions. Finally, establish the dynamic multi-objective optimization model for energy conservation and emissions reduction for coal mines.(5) Algorithm sovling dynamic multi-objective model. Objectives and constraints of the dynamic multi-objective model dynamically changes with time, leading to the decision-making environment changes continuously, which requires that the algorithm can search the pareto solution as much as possible in a fixed evolution environment, but also can detect any small changes in the evolution environment, and make correct response to environmental changes to determine the evolutionary parameters of the new environment. Based on these considerations, there needs to design the hybrid evolutionary algorithm DNSGA-II-PSO for solving dynamic multi-objective model which can detect environmental changes, maintaining population diversity, and forecasting changes in the three environments. (6) Selection of satisfied solutions. Rapid and effective decision-making is based on a small and limited number of candidate solutions. The multi-objective optimization model established is ultimately obtained a non-inferior pareto set, so candidate solutions also need to further selecte the pareto solution set, which requires the help of multi-attribute decision-making methods. Therefore, the use of hybrid clustering method SC-MTD-GAM, minimize the size of pareto set on the base of ensuring the distribution characteristics of pareto front, choose a representative pareto solution as a candidate solution with consideration of the decision maker’s preferences.Apply multi-objective optimization model and satisfied solution selection methods into the coal production of Chao Hua mine. And explore how to arrange coal production plan, which produce equipment for energy saving and how many investment projects to control or utilize pollutants, in order to achieve targets of energy conservation and emissions reduction in the "12th Five-Year Plan". Some conclusions are as follows:(1) After the analysis of energy consumption structure and pollution emissions for coal mines, some results are got that coal mines mainly consume four kinds of energy included coal, gasoline, diesel, electricity and discharge three pollutants of SO2, mine water, coal gangue to the environment. Using CCR-DEA model to assess energy efficiency and emission reduction potential indicate that the coal mines are currently at the optimal front of production, that means energy efficiency has been maximized. Further implementation of energy conservation requires the use of advanced technology to improve production efficiency.(2) Multi-objective investment decision-making models of energy conservation and emissions reduction for coal mines established in this paper meet the need of investment decisions of the energy conservation and emissions reduction. This model takes the coal production largest, energy consumption and pollution emissions minimum as objectives, with consideration of multiple constraints including resources, processes, capital and environmental protection, which better describes investment decision needs of energy conservation and emissions reduction of the typical coal mines in China at the present stage.(3) Hybrid multi-objective algorithm PSO-NSGA-Ⅱ proposed has better convergence, coverage, and uniformity than NSGA-II algorithm. For the multi-objective optimization model established characterized by decision variables of0-1and real numbers, this paper proposes a hybrid algorithm combined real-coded PSO and binary-coded NSGA-Ⅱ. Compared with NSGA-Ⅱ through the three performance indicators of the center distance, coverage and spcing, the results indicate that PSO-NSGA-II has the advantages of two algorithms with a combination of PSO and NSGA-Ⅱ,which has better convergence, coverage, uniformity.(4) Hybrid multi-objective algorithm DNSGA-Ⅱ-PSO proposed is more able to detect any small changes in the evolution environment, maintain population diversity, avoid the algorithm precocity trapping in local optimum, and can predict changes in the environment. According to the characteristics of dynamic decisions, this paper takes each year as a decision-making stage, then establishes a dynamic multi-objective model with a combination of real and0-1variables, and proposes the hybrid algorithm DNSGA-Ⅱ-PSO which fully uses DNSGA-Ⅱ to guide the global search direction, PSO to optimize quickly in the local area. This hybrid algorithm uses a environment detection operator to detect any changes of the investment environment, forecasts the new environment with a combination of the three methods of the inertial prediction, Gaussian distribution, and randomly generation to produce new individuals. Compared with NSGA-II through three performance indicators of the center distance, coverage and spcing, the results show that DNSGA-II-PSO has better performance of detection and predictive.(5) Hybrid clustering method based on SC-MTD-GAM with introduction of manager preferences is uesd to selecte satisfied solutions from pareto sets, and ultimately get three types of energy conservation and emissions reduction programs included the economy preferred, energy-saving and harmonious development. The situation of coal production before and after energy conservation and emissions reduction is compared:the economic-preferred investment program maximizes coal production by means of ensuring maximum production capacity and making full use of investment funds under the premise of targets of energy conservation and emissions reduction in the "12th Five-Year Plan"; the energy-saving investment program prefers to reduce energy consumption and pollutant emissions, to a large extent which is dependent on redution of coal production; and the coordinated developmental programs achieve targets of energy conservation and emissions reduction in three ways with reducing coal production, invest energy-saving engineering and emission reduction projects. In either program, although there is a temporary reduction in coal production, but it is eventually in trend to increase and finally exceed, with the implementation of of measures energy conservation and emissions reduction and the continuous accumulation the effect of energy conservation and emissions reduction. This shows that the work of energy conservation and emissions reduction is a very necessary and long but valuable one, because of not only completing emission reduction targets, but also optimizing the allocation of resources for the coal production fundamentally, which promotes the transformation of green production model called as "low-power, high efficiency, zero emission".Innovation of this paper is mainly reflected in the following three aspects:(1) A multi-objective optimization model of energy conservation and emissions reduction for coal mines is established. The paper establishes two multi-objective optimization models for static and dynamic perspectives respectively, taking the coal production largest, energy consumption and pollution emissions minimum during the planning period as three objectives with a consideration of multiple constraints of resources, processes, capital and environmental protection. The modesl better describe investment decision needs of energy conservation and emissions reduction of the typical coal mines in China at the present stage:in the planning period, the coal mines select which energy-saving equipment and comprehensive control and utilization projects to invest? When to invest? How much capital for each project in order to achieve targets of energy conservation and emissions reduction?(2) Two hybrid evolutionary algorithms of PSO-NSGA-Ⅱ and DNSGA-Ⅱ-PSO are proposed. Due to the multi-objective optimization model with the characteristics of both types of decision variables0-1and real numbers, this paper proposes a hybrid algorithm PSO-NSGA-Ⅱ combined real-coded PSO and binary-coded NSGA-Ⅱ; for the stage characteristics of the dynamic decision-making, DNSGA-Ⅱ-PSO uses DNSGA-Ⅱto guide the global evolution directions of solutions and accelerate the convergence of the algorithm by PSO. This two hybrid evolutionary algorithm effectively inherite the advantages of both NSGA and PSO, which can search more non-dominated solutions in the solution space within the scope as large as possible, while constantly approach the true pareto front.(3) A hybrid clustering method SC-MTD-GAM is proposed. It chooses the solution as a representative candidate with consideration of decision-makers’preferences on different objectives based on ensurance to in the shape of the pareto front distribution, and finally get three types of decision-making programs included the economy preferred, energy-saving and harmonious development, which provide scientific references for the coal mines when it develops investment plans of energy conservation and emissions reduction. |
PDF Full Text Request