CAJ | 학술논문

为解决长周期压力容器设备安全评估的低效率、低可靠性和不能人机交互等问题，利用开源的 R语言设计复杂的神经网络识别算法，并通过C＃．NET设计出一套界面友好的压力容器评估系统．实验结果表明：创建的压力容器评估系统嵌入BP神经网络算法，能精确刻画压力容器参数与状态之间的复杂非线性关系，评估准确率高；同时，软件系统实现了评估过程的交互性和自动化，具有良好的用户体验和很强的实践性．
위해결장주기압력용기설비안전평고적저효솔、저가고성화불능인궤교호등문제，이용개원적 R어언설계복잡적신경망락식별산법，병통과C＃．NET설계출일투계면우호적압력용기평고계통．실험결과표명：창건적압력용기평고계통감입BP신경망락산법，능정학각화압력용기삼수여상태지간적복잡비선성관계，평고준학솔고；동시，연건계통실현료평고과정적교호성화자동화，구유량호적용호체험화흔강적실천성．
To address the problems of low efficiency,low reliability,without human?computer interaction in the safety e-valuation of the long?periodic pressure vessel,an intelligent evaluation system based on artificial neural network algorithm is established.In which the open source R language is used to design the complex neural network intelligent algorithm and a user?friendly operating system is developed through C#.NET technology.The experimental results show that the pres-sure vessel evaluation system embedded BP neural network algorithm can precisely figure out the complex nonlinear rela-tionship between the parameters and the state of pressure vessel by significantly high accuracy.Meanwhile,the software system promotes the interactivity and automation of the evaluation process,which gives good user experience and strong practicality.