基于PSO-BP的应急通信感知装备效能评价方法

doi:10.12305/j.issn.1001-506X.2022.11.20

摘要/Abstract

摘要：

应急通信感知装备效能评价可支撑相关装备的发展规划, 而现有评价方法主观性强, 且自适应能力有待提升。因此, 提出一种基于粒子群优化(particle swarm optimization, PSO)算法的改进反向传播(back propagation, BP)神经网络的应急通信感知装备效能评价方法, 旨在建立客观精准的效能评价。首先面向实战效能构建了三级效能评价指标体系, 然后将样本数据进行主成分分析法降维, 建立BP神经网络回归模型, 并结合PSO算法对模型的连接权值与阈值进行优化, 形成PSO-BP模型以避免局部极小值问题, 获得可评价具体装备效能时的神经网络模型。实例分析表明, PSO-BP相较于BP神经网络模型评价的均方误差减少了28.18%, 表明PSO-BP模型具有更高的准确性。

关键词: 应急通信感知, 效能评价, 反向传播神经网络, 粒子群优化, 主成分分析

Abstract:

The effectiveness evaluation of emergency communication and sensing equipment can support the development plan of corresponding equipment, while the existing evaluation methods are highly subjective and the adaptive capabilities need to be improved. Therefore, an improved back propagation (BP) neural network based on particle swarm optimization (PSO) algorithm for the evaluation of the effectiveness of emergency communication and sensing equipment is proposed. The method aims to establish objective and accurate effectiveness evaluation. Firstly, a three-level effectiveness evaluation index system is established which oriented to actual combat effectiveness. Secondly, the sample data is reduced in dimensionality through principal component analysis (PCA). Next, a BP neural network regression model is built, the weights and thresholds of BP neural network are optimized by the particle swarm algorithm and a PSO-BP neural network model is formed to avoid the local minimum problem. Finally, a neural network model that can evaluate the effectiveness of specific equipment is obtained. The case analysis shows that compared with the BP neural network model, the mean square error of PSO-BP is reduced by 28.18%. The results show that the PSO-BP model has higher accuracy.

Key words: emergency communication and sensing, effectiveness evaluation, back propagation (BP) neural network, particle swarm optimization (PSO), principal component analysis (PCA)

中图分类号:

TN91

石宸睿, 田露, 徐湛, 职如昕, 陈晋辉. 基于PSO-BP的应急通信感知装备效能评价方法[J]. 系统工程与电子技术, 2022, 44(11): 3455-3462.

Chenrui SHI, Lu TIAN, Zhan XU, Ruxin ZHI, Jinhui CHEN. Effectiveness evaluation method of emergency communication and sensing equipment based on PSO-BP[J]. Systems Engineering and Electronics, 2022, 44(11): 3455-3462.

图/表 14

图1

图2

图3

图4

图5

表1

表2

表3

表4

表5

图6

图7

表6

图8

参考文献 31

1	周玉臣, 林圣琳, 马萍, 等. 武器装备效能评估研究进展[J]. 系统仿真学报, 2020, 32 (8): 1413- 1424. doi: 10.16182/j.issn1004731x.joss.20-0442
	ZHOU Y C , LING S L , MA P , et al. Research progress of weapon equipment effectiveness evaluation[J]. Journal of System Simulation, 2020, 32 (8): 1413- 1424. doi: 10.16182/j.issn1004731x.joss.20-0442
2	CUI X Y , LU F , HE Y F , et al. Two-dimensional evaluation model of electrical equipment based on combined weighting and rating algorithm[J]. Energy Reports, 2021, 7, 443- 448. doi: 10.1016/j.egyr.2021.01.048
3	CHEN B , LI X H , LIU H W , et al. Hybrid subjective and objective evaluation method of the equipment for first class distribution network[J]. Energy Procedia, 2019, 158, 3452- 3457. doi: 10.1016/j.egypro.2019.01.928
4	郝亮亮, 张金生, 李婷, 等. 导弹武器模拟训练考评系统设计[J]. 系统工程与电子技术, 2020, 42 (4): 843- 850.
	HAO L L , ZHANG J S , LI T , et al. Design of evaluation system for missile simulation training[J]. Systems Engineering and Electronics, 2020, 42 (4): 843- 850.
5	HUANG Z A , ZHAO W , ZHANG Y H , et al. Comprehensive safety evaluation of emergency training for building ruins scenario based on analytic hierarchy process-grey fuzzy evaluation[J]. IEEE Access, 2020, 8, 147776- 147789. doi: 10.1109/ACCESS.2020.3015829
6	GU Y L, XIE J J, LIU H J, et al. Evaluation and analysis of comprehensive performance of a brake pedal based on an improved analytic hierarchy process[J]. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering, 2021, 235(5): 095440702098773.
7	NGUYEN T T M , NGUYEN H T T , DOAN T , et al. Application analytic hierarchical process (AHP) in setting up local community urban environmental quality of life index in a developed metropolitan area in Ho Chi Minh City, Vietnam[J]. Current Urban Studies, 2021, 9 (3): 376- 391. doi: 10.4236/cus.2021.93023
8	WANG Y , LI Y , LIU W , et al. Assessing operational ocean observing equipment (OOOE) based on the fuzzy comprehensive evaluation method[J]. Ocean Engineering, 2015, 107 (10): 54- 59.
9	胡利平, 刘锦帆, 王洪叶, 等. 基于模糊综合评判的车辆目标SAR仿真图像评估方法[J]. 系统工程与电子技术, 2019, 41 (3): 534- 540.
	HU L P , LIU J F , WANG H Y , et al. Vehicle SAR simulation images validation method based on fuzzy comprehensive evaluation[J]. Systems Engineering and Electronics, 2019, 41 (3): 534- 540.
10	YOU G D , XU B , SU H L , et al. Evaluation of aquaculture water quality based on improved fuzzy comprehensive evaluation method[J]. Water, 2021, 13 (8): 1019. doi: 10.3390/w13081019
11	ZHOU Z J , CHEN L Y , HAN X X , et al. An interval evidential reasoning-based dynamic performance evaluation method for complex systems[J]. Computers & Industrial Engineering, 2021, 162, 107735.
12	WANG R , FENG Y . Evaluation research on green degree of equipment manufacturing industry based on improved particle swarm optimization algorithm[J]. Chaos, Solitons & Fractals, 2020, 131, 109502.
13	XU R , QIN Q Q , SUN T , et al. Comprehensive evaluation model for resettlement site selection in karst areas using multiobjective particle swarm optimization[J]. IEEE Access, 2021, 9, 113759- 113769. doi: 10.1109/ACCESS.2021.3101636
14	KAMRUZZAMAN M , BHUSAL N , BENIDRIS M . A convolutional neural network-based approach to composite power system reliability evaluation[J]. International Journal of Electrical Power & Energy Systems, 2021, 135 (4): 107468.
15	ZHENG X Y, HU P, GAO J N. Application of BP neural network in fire risk assessment of comprehensive shopping mall[C]//Proc. of the International Conference on Intelligent Computing, Automation and Systems, 2019.
16	程磊, 赵昊, 耿剑统, 等. 基于BP神经网络的矿井热环境评价体系研究[J]. 中国矿业, 2020, 29 (4): 152- 156.
	CHENG L , ZHAO H , GENG J T . Research on mine thermal environment evaluation system based on BP neural network[J]. China Mining Magazine, 2020, 29 (4): 152- 156.
17	HOU H, MENG H. Evaluation of material suppliers based on BP neural network under the background of big data[C]//Proc. of the Management Science Informatization and Economic Innovation Development Conference, 2020: 12-16.
18	WANG L , BI X H . Risk assessment of knowledge fusion in an innovation ecosystem based on a GA-BP neural network[J]. Cognitive Systems Research, 2021, 66, 201- 210.
19	LIM J Y, KIM T W, WANG X Y, et al. Evaluation of compressive strength of sustainable concrete using genetic algorithm assisted artificial neural networks[C]//Proc. of the Materials Science Forum, 2021.
20	LUO Y Y , REN D . Influence of the enterprise's intelligent performance evaluation model using neural network and genetic algorithm on the performance compensation of the merger and acquisition parties in the commitment period[J]. Plos One, 2021, 16 (3): e0248727.
21	何华锋, 何耀民, 徐永壮. 基于改进型BP神经网络的导引头测高性能评估[J]. 系统工程与电子技术, 2019, 41 (7): 1544- 1550.
	HE H F , HE Y M , XU Y Z . High performance evaluation of seeker measurement based on improved BP neural network[J]. Systems Engineering and Electronics, 2019, 41 (7): 1544- 1550.
22	FENG J Y , YUAN B Y , LI X , et al. Evaluation on risks of sustainable supply chain based on optimized BP neural networks in fresh grape industry[J]. Computers and Electronics in Agriculture, 2021, 183, 105988.
23	CHENG X L. Research on fuzzy comprehensive evaluation model of clothing brand based on neural network and data envelopment analysis[C]//Proc. of the 12th International Conference on Measuring Technology and Mechatronics Automation, 2020.
24	ZHANG M C . Prediction of rockburst hazard based on particle swarm algorithm and neural network[J]. Neural Computing and Applications, 2022, 34 (4): 2649- 2659.
25	VIET D T , PHUONG V V , DUONG M Q , et al. Models for short-term wind power forecasting based on improved artificial neural network using particle swarm optimization and genetic algorithms[J]. Energies, 2020, 13 (11): 2873.
26	ALCANTARA G M R N , DRESCH D , MELCHERT W R , et al. Use of non-volatile compounds for the classification of specialty and traditional Brazilian coffees using principal component analysis[J]. Food Chemistry, 2021, 360, 130088.
27	冯建鑫, 王雅雷, 王强, 等. 基于改进粒子群算法的快速反射镜自抗扰控制[J]. 系统工程与电子技术, 2021, 43 (12): 3675- 3682.
	FENG J X , WANG Y L , WANG Q , et al. Fast steering mirror ADRC based on improved particle swarm optimizer[J]. Systems Engineering and Electronics, 2021, 43 (12): 3675- 3682.
28	魏科宇. 基于改进粒子群算法的城市供水系统优化研究[D]. 重庆: 重庆大学, 2020.
	WEI K Y. A research on the optimization of municipal water distribution systems based on improve particle swarm method[D]. Chongqing: Chongqing University, 2020.
29	EBERHART R C, SHI Y H. Particle swarm optimization: development, applications and resources[C]//Proc. of the Congress on Evolutionary Computation, 2001.
30	王生亮, 刘根友. 一种非线性动态自适应惯性权重PSO算法[J]. 计算机仿真, 2021, 38 (4): 249- 253.249-253, 451
	WANG S L , LIU G Y . A nonlinear dynamic adaptive inertial weight PSO algorithm[J]. Computer Simulation, 2021, 38 (4): 249- 253.249-253, 451
31	LU H J , ZOU N , JACOBS R , et al. Error assessment and optimal cross-validation approaches in machine learning applied to impurity diffusion[J]. Computational Materials Science, 2019, 169, 109075.

性能指标	测量量程	精度	稳定性能	功耗	权重
测量量程	1	4	2	3	0.46
精度	1/4	1	1/3	1/2	0.10
稳定性能	1/2	3	1	2	0.28
功耗	1/3	2	1/2	1	0.16

结构指标	重量	材质	防护等级	权重
重量	1	1/3	1/5	0.11
材质	3	1	1/3	0.26
防护等级	5	3	1	0.63

指标	组合权重	指标	组合权重
测量量程	0.30	重量	0.04
精度	0.07	材质	0.09
稳定性能	0.18	防护等级	0.22
功耗	0.10

序号	测量量程	精度	稳定性能	功耗	重量	材质	防护等级	评分
1	4	3	3	8	4	4	4	4.15
2	4	4	4	7	5	2	4	4.16
3	4	4	4	7	5	3	4	4.25
4	5	4	4	6	6	2	4	4.40
5	4	4	4	8	4	2	5	4.44
6	6	4	3	5	7	2	4	4.46
7	5	4	4	7	6	2	4	4.50
8	6	4	4	6	5	2	4	4.66
9	4	4	4	9	6	3	5	4.71
10	5	4	3	10	8	4	4	4.88
11	5	4	6	2	5	2	8	5.20
12	6	5	5	4	3	4	7	5.47
13	6	5	6	4	4	4	7	5.69
14	6	6	5	4	4	6	7	5.76
15	6	6	6	4	4	6	7	5.94
16	8	6	5	5	2	4	6	5.98
17	7	6	6	4	3	3	8	6.15
18	6	5	7	3	5	5	9	6.34
19	7	7	5	5	5	7	8	6.58
20	8	7	6	4	3	7	7	6.66
21	8	7	7	4	4	6	8	7.01
22	8	8	8	3	3	6	8	7.12
23	7	9	9	4	4	8	7	7.17
24	9	7	7	3	3	4	9	7.21
25	8	8	8	4	4	7	8	7.35
26	9	7	7	4	4	7	9	7.62
27	9	8	8	3	3	6	9	7.64
28	8	9	9	4	4	9	8	7.78
29	9	8	8	4	5	6	9	7.82
30	9	8	8	5	4	7	9	7.97

隐藏层节点数	MSE
2	0.094 2
3	0.092 3
4	0.090 5
5	0.087 2
6	0.094 8
7	0.088 6
8	0.088 1
9	0.094 6
10	0.089 7
11	0.102 2
12	0.111 9