基于Multi-agent的排污权交易系统建模与仿真

Multi-agent-based Modeling and Simulation of Emissions Trading System

摘要:

摘要：排污权交易政策在我国试点工作中，由于交易成本过高、政府监管不力、市场机制不完善等因素导致市场交易量较少，难以在全国推广。基于复杂适应系统（CAS）和多Agent系统（MAS），构建排污权交易系统的多Agent仿真模型，模拟排污权交易系统中的主体行为，并在Netlogo平台进行仿真。结果表明：交易成本会降低交易主体的平均收益，随着交易量的上升，交易成本逐渐下降；买卖双方议价能力的悬殊会使议价能力弱的一方利益受损，并促使合谋行为的产生，不利于排污权交易市场的发展；政府监管可以有效减少合谋行为的产生，但当监管程度在达到一定水平后随着监管投入的增加，效果提升缓慢。

Abstract:

Abstract: Due to the high transaction costs, lack of government regulation and imperfect market mechanism in our country emissions trading policy pilot work, market trading volume is less and that makes the emissions trading policy difficult to promote throughout the country. Based on Complex Adaptive System (CAS) and Multi-Agent Systems (MAS), a multi-Agent simulation model is built to simulate the behavior of the emissions trading agents on the platform of Netlogo. The results showed that: (1) The transaction costs will reduce the average incomes of the trading agents. With the trading volume rising, transaction costs decrease. (2) The bargaining power of buyers and sellers will damage the interests of the weaker party and promote the collusion behaviors; (3) Government regulation can effectively reduce the collusion behaviors, however, with the increase of regulatory investment, the effect of government regulation lifts slowly at a certain level.