基于优化神经网络算法的滚球控制系统设计  

Design of the rolling ball control system based on optimized neural network algorithm

作　　者：张宇翔 毛晓东 杜宝麒 王惠 邬可谊 Zhang Yuxiang;Mao Xiaodong;Du Baoqi;Wang Hui;Wu Keyi(Drilling Technology Research Institute,SINOPEC Shengli Petroleum Engineering Co.,Ltd.,Shandong Dongying,257029,China;SINOPEC Group Shared Services Co.,Ltd.,Shandong Dongying,257029,China)

机构地区：[1]中石化胜利石油工程有限公司钻井工艺研究院,山东东营257029 [2]中国石化集团共享服务有限公司,山东东营257029

出　　处：《机械设计与制造工程》2025年第4期64-68,共5页Machine Design and Manufacturing Engineering

摘　　要：针对传统的滚球控制系统在应对复杂任务时通常存在的效率低和误差大等问题,引入优化粒子群算法来改进滚球滑模控制中的径向基函数,构建了一种新型滚球滑模控制模型。为了验证该模型的有效性,进行了轨迹跟踪实验,并对不同的参数组合进行了性能测试。结果表明,当函数中心值为0.45、宽度参数为0.3、权值参数为-0.1时的模型轨迹跟踪率最高为80%、响应时间最短为1.20 s。采用Chaos Toy滚球系统进行室内测试,该模型能够达到约90%的真实轨迹跟踪率。由此可知,提出的模型在控制误差、超调量稳定性和能耗方面表现出显著优势。Aiming at the low efficiency and large error that usually exists in the traditional roller ball control system when coping with complex tasks,an optimized particle swarm algorithm is introduced to improve the radial basis function in the sliding mode control of the roller ball,and a new type of roller ball sliding mode control model is constructed.In order to verify the effectiveness of the model,trajectory tracking experiments are carried out,and performance tests are conducted for different parameter combinations.The results show that the model trajectory tracking rate when the function center value is 0.45,the width parameter is 0.3,and the weight parameter is-0.1 is up to 80%,and the response time is the shortest of 1.20 s.Indoor tests using the Chaos Toy rolling ball system show that the new model is able to achieve about 90%of the real trajectory tracking rate.It can be seen that the proposed model shows significant advantages in terms of control error,stability of overshooting amount and energy consumption.

关 键 词：滚球 控制系统 数学模型 径向基函数 粒子群算法 

分 类 号：TP214[自动化与计算机技术—检测技术与自动化装置]