Abstract: This paper proposes a method to evolving parameter sets for Risk-Based（RB） bidding strategy in CDA（Continuous Double Auction） market by using the Particle Swarm Optimization algorithm.The strategy involves the agent forming a bid or asks by assessing the degree of risk involved and making a prediction about the competitive equilibrium price that is likely to be reached in the marketplace.However，there are some parameters setting that significantly influences on the trader’s behaviors and the market efficiency.This experiment results show the evolutionary RB bidding strategy can rapidly adapt to the market circumstance contrasted to the original RB bidding strategy.

摘要： Risk-Based策略是基于风险行为的代理策略。为了改善Risk-Based代理的行为，使交易价格迅速收敛于市场均衡价格，提高市场效率，提出利用粒子群优化算法演化Risk-Based策略参数。首先分析了影响Risk-Based代理行为的关键参数；之后提出了改进的粒子群优化算法演化Risk-Based策略关键参数的模型。最后，在基于市场控制的模拟系统中采用连续双向拍卖机制对演化Risk-Based策略进行了实验评价，结果表明演化后的Risk-Based策略比演化前的策略更为优秀。