TY - GEN
T1 - Modeling and Simulation of Chemical Machinery Performance Based on GA-Bp Algorithm
AU - Zhang, Zijian
AU - Xu, Bo
AU - Chai, Junhui
AU - Shen, Jianmin
AU - Lv, Zhongjie
AU - Zhang, Xiaolong
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - As a whole energy system, chemical machinery equipment has become a research topic in today's research through the research on the thermal behavior of its working fluid, especially the effective utilization of energy directly related to its internal energy conversion, transfer and expansion. One of the important contents of chemical machinery. The purpose of this paper is to study the modeling and simulation of chemical mechanical properties based on GA-Bp algorithm. Traditional chemical-mechanical models do not have the ability to simulate, and the results obtained are often very inaccurate. To address this problem, a new chemical-mechanical performance model based on the GA-Bp algorithm is developed in this paper. GA-Bp algorithm has the advantages of Bp algorithm and genetic algorithm at the same time, and it is easy to use. Combined with the breadth of the Bp algorithm and the characteristics of the genetic algorithm full search algorithm, combined with the general ability, mapping ability and full search ability of the two algorithms, a neural network training algorithm was developed. The GA-Bp algorithm reduces the search for the best solution by increasing the size of the unit, thereby increasing the number of training sessions for neural network training. The experimental results show that the chemical model based on the GA-Bp algorithm is more accurate. Experiments have proved that the gap between the chemical mechanical performance model of this paper and the traditional pure Bp is about 3-4 times, and the prediction performance speed can be reduced to 1/10 compared with the traditional algorithm.
AB - As a whole energy system, chemical machinery equipment has become a research topic in today's research through the research on the thermal behavior of its working fluid, especially the effective utilization of energy directly related to its internal energy conversion, transfer and expansion. One of the important contents of chemical machinery. The purpose of this paper is to study the modeling and simulation of chemical mechanical properties based on GA-Bp algorithm. Traditional chemical-mechanical models do not have the ability to simulate, and the results obtained are often very inaccurate. To address this problem, a new chemical-mechanical performance model based on the GA-Bp algorithm is developed in this paper. GA-Bp algorithm has the advantages of Bp algorithm and genetic algorithm at the same time, and it is easy to use. Combined with the breadth of the Bp algorithm and the characteristics of the genetic algorithm full search algorithm, combined with the general ability, mapping ability and full search ability of the two algorithms, a neural network training algorithm was developed. The GA-Bp algorithm reduces the search for the best solution by increasing the size of the unit, thereby increasing the number of training sessions for neural network training. The experimental results show that the chemical model based on the GA-Bp algorithm is more accurate. Experiments have proved that the gap between the chemical mechanical performance model of this paper and the traditional pure Bp is about 3-4 times, and the prediction performance speed can be reduced to 1/10 compared with the traditional algorithm.
KW - artificial neural network
KW - chemical machinery
KW - performance modeling
KW - simulation research
UR - http://www.scopus.com/inward/record.url?scp=85150676513&partnerID=8YFLogxK
U2 - 10.1109/ICKECS56523.2022.10060637
DO - 10.1109/ICKECS56523.2022.10060637
M3 - Conference contribution
AN - SCOPUS:85150676513
T3 - IEEE International Conference on Knowledge Engineering and Communication Systems, ICKES 2022
BT - IEEE International Conference on Knowledge Engineering and Communication Systems, ICKES 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE International Conference on Knowledge Engineering and Communication Systems, ICKES 2022
Y2 - 28 December 2022 through 29 December 2022
ER -