相关性是可拓学[1-2]中进行主动变换及引起传导变换的基础。复杂的相关关系网中,只有在两个信息元之间存在相关性的前提下,当对其中一个信息元特征的量值实施主动变换时,才会引起另一个信息元发生传导变换。然而,现实情况中,当对一个信息元实施一次主动变换时,并不能引起与之相关的信息元发生传导变换。而只有积累实施N次主动变换后,才能引起与之相关的信息元发生传导变换。N非固定取值,也会随着时间或场景的变化而不同。例如,某酒店在把饭菜价格提高后,其就餐人数并未发生变化,营业额和利润增加了。某段时间后,该酒店再次将饭菜价格进行了提高。本次提价后,酒店的就餐人数、营业额和利润都减少了。目前,关于类似累计的主动变换及传导效应的研究文献还未见。
对信息元某特征的量值而言,如何从累积多次的主动变换中挖掘出有用的传导知识,将会对分析所实施的主动变换有着至关重要的作用。为此,本文在主动变换、相关性、传导变换、传导效应的基础上,结合灵敏度的概念,重点研究某特征的量值累积多次主动变换后, 引起的传导变换的传导知识挖掘理论,丰富了可拓学的传导知识数据挖掘[3-4]理论。为传导知识的数据挖掘提供了一个新方法和新思路,使传导知识数据挖掘理论更加全面。
1 积累多次主动变换的传导特征为了方便,仅研究两个相关信息元之间的主动变换[5-8]及其传导变换。记信息元
$\begin{array}{l} {{{I}}_1}(t) = ({{{O}}_1}(t),{{{C}}_1}(t),{{{U}}_1}(t)) = \left( {\begin{array}{*{20}{c}} {{{{O}}_1}(t),}&{{c_{11}},}&{{u_{11}}(t)}\\ {}&{{c_{12}},}&{{u_{12}}(t)}\\ {}& \vdots & \vdots \\ {}&{{c_{1m}},}&{{u_{1m}}(t)} \end{array}} \right) \end{array}$ | (1) |
$\begin{array}{l} {{{I}}_2}(t) = ({{{O}}_2}(t),{{{C}}_2}(t),{{{U}}_2}(t)) = \left( {\begin{array}{*{20}{c}} {{{{O}}_2}(t),}&{{c_{21}},}&{{u_{21}}(t)}\\ {}&{{c_{22}},}&{{u_{22}}(t)}\\ {}& \vdots & \vdots \\ {}&{{c_{2n}},}&{{u_{2n}}(t)} \end{array}} \right) \end{array}$ | (2) |
在
当在
定义1 称特征
定义2 称特征
定义3 若集合
一般情况下,若信息元
${{{I}}_1}({t_0}) = \left( {\begin{array}{*{20}{c}} {{{{O}}_1}({t_0}),}&{{c_{11}},}&{{u_{11}}({t_0})} \\ {}&{{c_{12}},}&{{u_{12}}({t_0})} \\ {}& \vdots & \vdots \\ {}&{{c_{1m}},}&{{u_{1m}}({t_0})} \end{array}} \right)$ | (1) |
信息元
${{{I}}_2}({t_0}) = \left( {\begin{array}{*{20}{c}} {{{{O}}_2}({t_0}),}&{{c_{21}},}&{{u_{21}}({t_0})} \\ {}&{{c_{22}},}&{{u_{22}}({t_0})} \\ {}& \vdots & \vdots \\ {}&{{c_{2n}},}&{{u_{2n}}({t_0})} \end{array}} \right)$ | (1) |
在
在解决实际矛盾问题中,若信息元
$\begin{split} {{{I}}_{2k}}({t_p}) = & \left( {\begin{array}{*{20}{c}} {{{{O}}_2}({t_p}),}&{{c_{2k}},}&{{v_{2k}}({t_p})} \end{array}} \right),\\ & p \in \left\{ {1,2, \cdots ,q} \right\} \end{split}$ | (3) |
$\begin{split} & {{{I}}_1}({t_p}) = \left( {\begin{array}{*{20}{c}} {{{{O}}_1}({t_p}),}&{{c_{1i}},}&{{v_{1i}}({t_p})} \end{array}} \right),i \in \{ 1,2, \cdots ,m\} \\ & \qquad\qquad\qquad\qquad p \in \left\{ {1,2, \cdots ,q} \right\} \end{split}$ | (4) |
记
${{{E}}_{_{11}}} = \frac{{\sum\limits_{x = 1}^e {({v_{11}}({t_p}) - {v_{11}}({t_0}))} }}{q},$ |
${{{E}}_{12}} = \frac{{\sum\limits_{x = 1}^e {({v_{12}}({t_p}) - {v_{12}}({t_0}))} }}{q}, \cdots ,$ |
${{{E}}_{1n}} = \frac{{\sum\limits_{x = 1}^e {({v_{1n}}({t_p}) - {v_{1n}}({t_0}))} }}{q}$ |
为
归一化得
$\begin{split} & {{{E}}_{1l}}^\prime = \frac{{{{{E}}_{1l}}}}{{\mathop {\max }\limits_{p \in \{ 1,2, \cdots ,q\} } \left\{ {{v_{1l}}({t_p}) - {v_{1l}}({t_0})} \right\}}}\\ & \qquad\;\; l = 1,2, \cdots ,r,r \leqslant m \end{split}$ | (5) |
记
${D_r} = \frac{{\sum\limits_{x = 1}^e {({v_{kr}}({t_p}) - {v_{kr}}({t_0}))} }}{e},l = 1,2, \cdots ,r,r \leqslant m$ |
为由
归一化得
$\begin{split} & {D_l}^\prime = \frac{{{D_l}}}{{\mathop {\max }\limits_{p \in \{ 1,2, \cdots ,e\} } \left\{ {{v_{2kl}}({t_p}) - {v_{2kl}}({t_0})} \right\}}},\\ & \quad \;\;\;\;\;\;\; l = 1,2, \cdots ,r,r \leqslant m \end{split}$ | (6) |
称
假设某型导弹武器Q的导航系统[14-17]中,部件A和部件B存在相关关系。部件A某个模块的振动值记为
将导弹武器系统Q的整体性能、部件A和部件B分别用信息元刻画为
${{{I}}_{\rm{A}}}(t) = \left( {\begin{array}{*{20}{c}} {A,}&{\text{性能}11,}&{{u_{11}}(t)}\\ {}&{\text{振动值}a,}&{a(t)}\\ {}& \vdots & \vdots \\ {}&{\text{模块}1m,}&{{u_{1m}}(t)} \end{array}} \right)$ |
${{{I}}_{\rm{B}}}(t) = \left( {\begin{array}{*{20}{c}} {B,}&{\text{性能}21,}&{{u_{21}}(t)}\\ {}&{\text{平衡性能}b,}&{b(t)}\\ {}& \vdots & \vdots \\ {}&{\text{性能}2m,}&{{u_{2m}}(t)} \end{array}} \right)$ |
和
选取导弹武器系统的整体性能
通过计算,特征
${C_{2\phi }} = b({t_2}) - b({t_0}) = 0.812 - 0.81 = 0.002,$ |
${C_{3\phi }} = b({t_3}) - b({t_0}) = 0.814 - 0.81 = 0.004,$ |
${C_{4\phi }} = b({t_4}) - b({t_0}) = 0.815 - 0.81 = 0.005,$ |
${C_{5\phi }} = b({t_5}) - b({t_0}) = 0.818 - 0.81 = 0.008,$ |
${C_{6\phi }} = b({t_6}) - b({t_0}) = 0.822 - 0.81 = 0.011,$ |
${C_{7\phi }} = b({t_7}) - b({t_0}) = 0.819 - 0.81 = 0.009,$ |
${C_{8\phi }} = b({t_8}) - b({t_0}) = 0.818 - 0.81 = 0.008,$ |
${C_{9\phi }} = b({t_9}) - b({t_0}) = 0.815 - 0.81 = 0.005,$ |
${C_{10\phi }} = b({t_{10}}) - b({t_0}) = 0.813 - 0.81 = 0.003\text{。}$ |
从表1中可见,在第6次试验(对部件A实施第6次主动变换
目标特征
${L_{a(t)c(t)}} = \frac{{0.004}}{{0.0062}} = 0.645$ |
本文仅通过对10次实验数据的分析研究,可以得到特征
现实问题中,经常遇到前N次可拓变换未引起与之相关的信息元发生传导变换,而N+1次主动变换才能引起与之相关的信息元发生传导变换的情况。本文为了挖掘N+1次积累主动变换的传导知识理论,将可拓变换、传导效应和灵敏度的概念引入到传导知识的挖掘中,充实和完善了传导知识的挖掘理论。下一步,将结合该型导弹武器系统工厂试验及设计、定型试验过程中产生的大量试验数据,利用计算机编程仿真,深入挖掘各项特征对该型战术导弹武器系统性能特征的灵敏度,为其定型、列装和改进提供理论和实践参考依据。
[1] | 杨春燕, 蔡文. 可拓学[M]. 北京: 科学出版社, 2014: 21–78. (0) |
[2] | 蔡文, 杨春燕, 陈文伟, 等. 可拓集与可拓数据挖掘[M]. 北京: 科学出版社, 2008: 14–95. (0) |
[3] |
杨春燕, 蔡文. 挖掘同对象信息元的传导知识[J]. 智能系统学报, 2008, 3(4): 305-308. YANG Chunyan, CAI Wen. Mining conductive knowledge in information elements possessed by an identical object[J]. CAAI transactions on intelligent systems, 2008, 3(4): 305-308. (0) |
[4] | WANG Feng, ZHANG Jinchun, LI Rihua. Process element on expansion of the random process’s mean function and variance function[C]//Proceedings of Extenics and Innovation Methods-Proceedings of the International Symposium on Extenics and Innovation Methods. Beijing, China, 2013: 29. (0) |
[5] |
郭强, 邹广天. 基于决策树分类的可拓建筑策划预测方法[J]. 智能系统学报, 2017, 12(1): 117-123. GUO Qiang, ZOU Guangtian. Prediction methods for extension architecture programming based on decision tree classification[J]. CAAI transactions on intelligent systems, 2017, 12(1): 117-123. (0) |
[6] | YAN Siwei, FAN Rui, CHEN Yuefeng, et al. Research on web services-based extenics aided innovation system[J]. Procedia computer science, 2017, 107(C): 103-110. (0) |
[7] | LI Xingsen, LIU Haitao, CHEN Cheng, et al. A hybrid information construction model on factor space and extenics[J]. Procedia computer science, 2017, 122: 1168-1174. DOI:10.1016/j.procs.2017.11.488 (0) |
[8] |
蔡文, 杨春燕. 评价信息元及其原信息元的获取方法[J]. 智能系统学报, 2009, 4(3): 234-238. CAI Wen, YANG Chunyan. A method for evaluation of information-elements and acquirement of the original information element[J]. CAAI transactions on intelligent systems, 2009, 4(3): 234-238. DOI:10.3969/j.issn.1673-4785.2009.03.007 (0) |
[9] |
刘长亮, 鲍传美, 包化, 等. 可拓数据挖掘在高性能兵棋推演系统中的应用[J]. 指挥信息系统与技术, 2018, 9(1): 62–67. LIU Changliang, BAO Chuanmei, BAO Hua, et al. Application of extensible data mining in high performance wargaming system[J]. Command information system and technology, 2018, 9(1): 62–67. (0) |
[10] |
汤龙, 杨春燕. 复杂基元相关网下的传导变换[J]. 智能系统学报, 2016, 11(1): 104–110. TANG Long, YANG Chunyan. Conductive transformation under complicated basic-element correlative network[J]. CAAL transactions on intelligent systems, 2016, 11(1): 104–110. (0) |
[11] |
张园园, 张巨伟, 赵斌. 安全领域不相容问题的数学分析[J]. 辽宁石油化工大学学报, 2018, 38(1): 55-59. ZHANG Yuanyuan, ZHANG Juwei, ZHAO Bin. Mathematical analysis of incompatible problems in security domain[J]. Journal of Liaoning Shihua University, 2018, 38(1): 55-59. DOI:10.3969/j.issn.1672-6952.2018.01.010 (0) |
[12] |
温树勇, 李卫华. 本体知识拓展分析树在可拓策略生成系统的应用[J]. 智能系统学报, 2014, 9(1): 115-120. WEN Shuyong, LI Weihua. Application of ontology knowledge expansion analysis tree in the extension strategy generation system[J]. CAAI transactions on intelligent systems, 2014, 9(1): 115-120. (0) |
[13] |
韩睿, 郑竞宏, 朱守真, 等. 基于灵敏度分析的同步发电机参数分步辨识策略[J]. 电力自动化设备, 2012, 32(5): 74-80. HAN Rui, ZHENG Jinghong, ZHU Shouzhen, et al. Step identification of synchronous generator parameters based on sensitivity analysis[J]. Electric power automation equipment, 2012, 32(5): 74-80. DOI:10.3969/j.issn.1006-6047.2012.05.014 (0) |
[14] |
王丰, 张磊, 胡春万. 基于多级优度评价方法的导弹武器系统效能评估[J]. 指挥控制与仿真, 2017, 39(3): 74–77. WANG Feng, ZHANG Lei, HU Chunwan. Efficiency evaluation of missile weapon system based on multi-level priority degree evaluation method[J]. Command control & simulation, 2017, 39(3): 74–77. (0) |
[15] |
花黄伟, 杨春燕. 可拓识别方法及其在无人车识别障碍物中的应用研究[J]. 广东工业大学学报, 2016, 33(4): 1–6. HUA Huangwei, YANG Chunyan. Research on extension recognition method and its application on obstacles recognition of self-driving car[J]. Journal of Guangdong University of Technology, 2016, 33(4): 1–6. (0) |
[16] |
刘杨, 宋贵宝, 刘泽坤. 国外舰空导弹武器系统发展综述[J]. 舰船电子工程, 2017, 37(9): 10-15, 22. LIU Yang, SONG Guibao, LIU Zekun. A summary of the development of foreign ship-to-air missile weapon system[J]. Ship electronic engineering, 2017, 37(9): 10-15, 22. (0) |
[17] |
陈飞, 王鑫. 数据融合技术在导弹武器系统抗干扰性能评估中的应用[J]. 航天电子对抗, 2017, 33(3): 1–4. CHEN Fei, WANG Xin. Application of data fusion in anti-jamming performance evaluation of missile weapon equipment system[J]. Aerospace electronic warfare, 2017, 33(3): 1–4. (0) |
[18] |
王丰, 张磊. 可拓论的防空导弹武器系统效能的研究[J]. 现代防御技术, 2018, 46(3): 9-17. WANG Feng, ZHANG Lei. Effectiveness of air defense missile weapon system based on extension theory[J]. Modern Defense Technology, 2018, 46(3): 9-17. (0) |