论文部分内容阅读
基于Internet/Intranet松耦合环境设计分布式选矿专家系统 ,采用面向agent(智能体 )的系统开发方法和multi agent分布协作求解模式 ,采用异构异质agent,即agent可建构于广域网、局域网和单处理器紧耦合模式 ,且求解a gent可以是基于符号系统、神经网络、遗传算法等agent的协作 .此外 ,建立了统一的知识库标准 ,基于关系数据模式表征了传统的规则、框架、语义网络及谓词逻辑等知识表示方法 .其中 ,基于数据库引导及搜索技术实现了正向推理机制及反向推理机制 ,易于agent之间知识的传输 ,大大提高了通信效率 ;基于DCOM技术 ,以C + +构建框架类加速了agent的实现 .以矿石可选性预测求解为例 ,实现了多agent协作求解的仿真系统 ,较好地验证了所提出的理论模型及框架
Based on the loosely coupled environment of Internet / Intranet, a distributed mineral processing expert system is designed. The agent-oriented system development method and the multi-agent distributed cooperative solution mode are adopted. Heterogeneous heterogeneous agents, ie, agents, can be constructed on the WAN, Processor tightly coupled mode, and solving a gent can be based on the cooperation of symbology, neural network, genetic algorithm, etc. In addition, a unified knowledge base standard is established and the traditional rules, frameworks and semantic networks are characterized based on the relational data model And predicate logic etc. Among them, the forward reasoning mechanism and the reverse reasoning mechanism are implemented based on the database guidance and search technology, which facilitates the transmission of knowledge between agents and greatly improves the communication efficiency. Based on the DCOM technology, The construction of framework class accelerates the realization of agent.With the ore optional predictive solution as an example, the simulation system of multi-agent collaborative solution is realized, and the proposed theoretical model and framework are well verified