在煤矿生产中，矿工是作业主体，矿工在井下作业时，煤矿的作业环境、重复单调的作业任务和姿势容易使其产生身体疲劳和心理疲劳。矿工疲劳会降低其工作效率，使其产生麻痹、侥幸等心理，从而出现不安全行为，导致安全事故的发生。因此，解决矿工的疲劳问题，对于提高矿工的工作效率和有效预防事故发生具有重大作用。为了识别煤矿工人的疲劳状态，选择煤矿中三个班次的矿工进行研究，对不同班次矿工的主观嗜睡量表、脑电数据和行为指标进行统计分析，建立BP神经网络模型来识别矿工的疲劳程度。本文的主要研究内容如下： （1）设计并完成了煤矿工人在班前班后状态下的脑电实验，并根据行为和量表数据分析矿工的疲劳状态，实现矿工疲劳程度分级。本研究以早、中、夜三个班次各15名矿工作为被试，采用怪球实验范式收集他们班前班后的行为数据，并同步采集其脑电信号，使用卡罗林斯卡嗜睡量表（KSS）记录了被试的得分。对于三个班次矿工上班前和下班后的KSS得分和行为数据，使用SPSS22.0进行配对t检验，结果表明，早班、中班和夜班的矿工在下班后，KSS得分增加，平均反应时间和遗漏次数显著增大，正确率显著降低。采用K-means聚类分析法对矿工的KSS得分和行为数据进行聚类，得到了三级疲劳程度，第一级的疲劳状态为正常，第二级为轻微疲劳，第三级为重度疲劳。 （2）分析脑电数据，探讨矿工的疲劳状态，并筛选疲劳指标。针对脑电数据，使用EEGLAB进行预处理，通过FFT提取脑电波δ、θ、α、β的功率谱密度和功率，并计算脑电指标θ/β、α/β、（α+θ）/β和（α+θ）/（α+β），根据变化较为显著的脑电通道分析三个班次矿工的疲劳状态。研究表明，早班和夜班矿工下班后的δ、θ、α功率谱密度在脑电通道P1、P3、PO7显著降低，中班矿工无显著变化。在颞叶和侧额区域的脑电通道FT7、FT8、C5、TP7、P7和CP6，早班和夜班矿工的脑电指标在下班后均升高；中班矿工下班后的脑电指标α/β和θ/β在侧额区域增大。 （3）基于BP神经网络构建矿工疲劳识别模型。将选取的疲劳指标设置为输入层，将矿工疲劳程度的三个等级设置为输出层，以实现矿工的疲劳识别。对矿工疲劳识别模型进行训练和测试，模型训练集的识别率为96.359%，测试集识别率为74.819%，总体识别率为91.734%，平均误差为0.091，表明该模型的准确率较高。 本文通过分析矿工的实验数据，阐明了矿工在上班前和下班后的疲劳状态呈现显著差异，完成了矿工的疲劳程度分级，建立的BP神经网络模型实现了矿工的疲劳识别。本研究为矿工疲劳的识别和检测提供了依据和参考。

In coal mine production, the miner is the main part of operation. When the miner is working underground, the working environment of coal mine, repeated monotonous tasks and postures are likely to cause physical fatigue and mental fatigue. Fatigue of miners will reduce their work efficiency, cause them to become paralyzed, fluke and other psychology, result in unsafe behaviors, and lead to safety accidents. Therefore, solving the fatigue problems of miners plays a major role in improving the work efficiency of miners and effectively preventing accidents. In order to identify the fatigue states of coal miners, three shifts of miners in the coal mine were selected for research, the subjective sleepiness scale, EEG data and behavior indicators of miners in different shifts were statistically analyzed, and a BP neural network model was established to identify the degree of fatigue of miners. The main research contents of this paper are as follows: (1) The EEG experiment of coal miners before and after shifts was designed and completed, the fatigue status of miners was analyzed based on behavior and scale data, and the classification of miners' fatigue levels was realized. In this study, 15 mine workers in each of the three shifts of morning, middle, and night were selected as subjects. The oddball experiment paradigm was used to collect their behavior data before and after work, and their EEG signals were collected simultaneously. Karolinska scale (KSS) was used for recording the scores of the subjects. SPSS22.0 was used to perform a paired t-test to the KSS scores and behavior data of the three shift miners before and after work. The results showed that the KSS scores of miners in the morning, middle, and night shifts increased after work, and the average reaction time was increased. The number of omissions was significantly increased, and the accuracy rate was markedly reduced. The K-means cluster analysis method was used to cluster the KSS scores and behavior data of the miners, and the three levels of fatigue were obtained. The first level of fatigue is normal, the second level is mild fatigue, and the third level is severe fatigue. (2) EEG data was analyzed to explore the fatigue state of miners and screen fatigue indicators. For EEG data, EEGLAB was used for preprocessing to extract the power spectral density and power of brain waves δ, θ, α, β through FFT, and EEG indicators θ/β, α/β, (α+θ)/β and (α+θ)/(α+β) were calculated to analyze the fatigue status of the miners in three shifts according to the EEG channels that have changed significantly. Studies had shown that the δ, θ, and α power spectral densities of the early and night shift miners after work were significantly reduced in the EEG channels P1, P3, and PO7, while the middle shift miners had no significant changes. In the EEG channels FT7, FT8, C5, TP7, P7, and CP6 in the temporal lobe and lateral frontal areas, the EEG indicators of the early and night shift miners all increased after work; the EEG indicators α/ β and θ/β of the middle shift miners after work were increased in the lateral frontal area. (3) A miners’ fatigue recognition model was constructed based on BP neural network. The selected fatigue indexes were set as the input layer, and the three levels of miners’ fatigue were set as the output layer to realize the fatigue identification of the miners. The miners’ fatigue recognition model was trained and tested. The recognition rate of the model training set was 96.359%, the recognition rate of the test set was 74.819%, the overall recognition rate was 91.734%, and the average error was 0.091, indicating that the model has a high accuracy rate. By analyzing the experimental data of miners, this paper clarified the significant difference between the fatigue state of miners before and after work, and completed the classification of the fatigue degree of miners. The established BP neural network model realized the fatigue identification of miners. This research provides a basis and reference for the identification and detection of miners’ fatigue.

[1] Legault G, Clement A L, Kenny G P, et al. Cognitive Consequences of Sleep Deprivation, Shiftwork, and Heat Exposure for Underground Miners[J]. Applied Ergonomics, 2017, 58: 144-150.

[2] 田水承, 乌力吉, 寇猛 等. 基于生理实验的矿工不安全行为与疲劳关系研究[J]. 西安科技大学学报, 2016, 36(03): 324-330.

[3] Butlewski M, Dahlke G, Drzewiecka M, et al. Fatigue of Miners as a Key Factor in the Work Safety System[J]. Procedia Manufacturing, 2015, 3: 4732-4739.

[4] 乌力吉. 基于生理实验的疲劳与矿工不安全行为关系研究[D]. 西安科技大学, 2016.

[5] 刘青林. 综掘工作面作业疲劳及合理轮班方式研究[D]. 中国矿业大学, 2019.

[6] 王丽丽. 工作时间合理是小康生活的一部分[N]. 中国煤炭报, 2015-01-16(003).

[7] Chanchlani N. Health Consequences of Shift Work and Insufficient Sleep[J]. BMJ (Clinical research ed), 2017, 356: i6599.

[8] 徐景贤. 浅谈影响矿工作业可靠性的疲劳因素[J]. 中国电子商务, 2010, (004): 215-216.

[9] Bauerle T, Dugdale Z, Poplin G. Mineworker Fatigue: A Review of What We Know and Future Decisions[J]. Mining Engineering, 2018, 70(3): 33-40.

[10] 邢娟娟, 刘卫东, 王秀兰. 煤矿井下工人体力劳动负荷及疲劳调查研究[J]. 劳动保护科学技术, 1998, (04): 55-58.

[11] 倪蓉. 煤矿职工疲劳度相关因素分析与控制研究[J]. 煤矿安全, 2013, 44(10): 230-233.

[12] 董建梁, 冯宏, 李鹏, 等. 煤矿矿工井下作业疲劳检测研究[J]. 中国科技信息, 2014, (06): 171-172.

[13] 董建梁, 李鹏, 马晨光. 基于脑电检测的煤矿矿工井下疲劳作业监测研究[J]. 中国科技信息, 2014, (16): 82-83.

[14] 李红霞, 薛建文, 杨妍. 疲劳对矿工不安全行为影响的ERP实验研究[J]. 西安科技大学学报, 2015, 35(03): 376-380.

[15] 张巨峰, 杨峰峰, 郑超, 等. 不同轮班制度对煤矿井下一线矿工疲劳的影响[J]. 煤矿安全, 2017, 48(01): 235-237.

[16] 沈剑, 李红霞. 矿工作业疲劳对煤矿险兆事件的影响机理——基于情感耗竭中介变量的分析[J]. 安全与环境学报, 2019, 19(02): 527-534.

[17] 田一丹, 张同全, 胡琬雪. 煤矿一线员工过度劳动评价与影响因素——基于四个省份十二个煤矿调查数据的分析[J]. 中国劳动关系学院学报, 2019, 33(05): 104-115.

[18] 任大伟, 徐小华, 梁亚敏, 等. 煤矿井下作业疲劳影响因素及对策研究[J]. 煤矿安全, 2019, 50(03): 231-234.

[19] 王新平, 畅涛涛, 孙林辉. 体脑疲劳交互影响下煤矿工人作业改善研究[J]. 工业工程, 2020, 23(04): 148-153+182.

[20] THERON, W. J, Heerden V, et al. Fatigue Knowledge - a New Lever in Safety Management[J]. J S Afr Inst Min Metall, 2011, 111: 1-10.

[21] Drews F A, Rogers W P, Talebi E, et al. The Experience and Management of Fatigue: A Study of Mine Haulage Operators[J]. Mining, Metallurgy & Exploration, 2020, 37(6): 1837-1846.

[22] Butlewski M, Misztal A. The Concept of Process Changes in the Worker Fatigue Management System[J]. Zeszyty Naukowe Małopolskiej Wyższej Szkoły Ekonomicznej w Tarnowie / The Malopolska School of Economics in Tarnow Research Papers Collection, 2016, 31.

[23] Butlewski M, Dahlke G, Drzewiecka-Dahlke M, et al. Use of the Methodology of Network Thinking for a Fatigue Criteria Investigation Based on the Example of Mining Companies[J]. Teh Vjesn, 2020, 27(4): 1037-1043.

[24] Brake D J, Bates G P. Fatigue in Industrial Workers under Thermal Stress on Extended Shift Lengths[J]. Occupational Medicine, 2001, (7): 456-463.

[25] Yu H, Chen H, Long R. Mental Fatigue, Cognitive Bias and Safety Paradox in Chinese Coal Mines[J]. Resources Policy, 2017, 52: 165-172.

[26] Ren D W, Li Z Z, Xu X H, et al. Relations between Human Energy Consumption and Physical State in the Simulation Environment of Coal Mine[J]. Fresenius Environ Bull, 2017, 26(11): 6632-6636.

[27] Kowal M, Kubal S, Piotrowski P, et al. Wireless Platform for the Fatigue Measurement of Workers Employed in Underground Mines[J]. Elektron Elektrotech, 2017, 23(4): 24-28.

[28] Chen S K, Xu K L, Zheng X, et al. Linear and Nonlinear Analyses of Normal and Fatigue Heart Rate Variability Signals for Miners in High-Altitude and Cold Areas[J]. Comput Meth Programs Biomed, 2020, 196(4): 13.

[29] 孙昕, 顾延中. 基于面部特征识别的管制员疲劳检测算法研究[J]. 网络安全技术与应用, 2021, (01): 48-51.

[30] 王艳玲. 基于神经网络的疲劳驾驶识别系统[J]. 汽车实用技术, 2021, 46(01): 89-91.

[31] 罗映雪, 贾博, 裘旭益, 等. 基于Gamma深度信念网络的飞行员脑疲劳状态识别[J]. 电子学报, 2020, 48(06): 1062-1070.

[32] 李军, 幸坚炬, 黄超生, 等. 人脸识别在疲劳驾驶检测中的应用研究[J]. 广东技术师范学院学报, 2019, 40(03): 22-27.

[33] 楚文慧, 吴超仲, 张晖, 等. 基于个性化行为模型的驾驶疲劳识别方法[J]. 中国安全科学学报, 2018, 28(06): 43-48.

[34] 杨昌其, 谭娟, 仇争平. 基于BP神经网络的管制员疲劳状态识别研究[J]. 航空计算技术, 2018, 48(01): 17-20+25.

[35] 马发民, 吴红乐, 张林, 等. 基于BP神经网络的疲劳判定[J]. 计算机与数字工程, 2017, 45(06): 1187-1193.

[36] 何明山, 罗卫东, 陈启新, 等. 基于DSP虹膜识别防疲劳驾驶报警系统的研究[J]. 中国安全生产科学技术, 2016, 12(01): 127-131.

[37] 郭孜政, 谭永刚, 马国忠, 等. 基于BP神经网络的驾驶精神疲劳识别方法[J]. 哈尔滨工业大学学报, 2014, 46(08): 118-121.

[38] 王斐, 王少楠, 王惜慧, 等. 基于脑电图识别结合操纵特征的驾驶疲劳检测[J]. 仪器仪表学报, 2014, 35(02): 398-404.

[39] Wang R, Guo L, Tong B, et al. Monitoring Mouth Movement for Driver Fatigue or Distraction with One Camera[C]// proceedings of the 7^ International IEEE Conference on Intelligent Transportation Systems, Itsc. IEEE, 2004.

[40] Karvekar S, Abdollahi M, Rashedi E. A Data-Driven Model to Identify Fatigue Level Based on the Motion Data from a Smartphone[C]// proceedings of the 2019 IEEE Western New York Image and Signal Processing Workshop (WNYISPW). IEEE, 2019.

[41] Li J, Li H, Umer W, et al. Identification and Classification of Construction Equipment Operators' Mental Fatigue Using Wearable Eye-Tracking Technology[J]. Automation in Construction, 2020, 109: 103000.

[42] Ulinskas M, Damaševičius R, Maskeliūnas R, et al. Recognition of Human Daytime Fatigue Using Keystroke Data[J]. Procedia Computer Science, 2018, 130: 947-952.

[43] Lal S, Craig A, Boord P, et al. Development of an Algorithm for an EEG-Based Driver Fatigue Countermeasure[J]. Journal of Safety Research, 2003, 34(3): p.321-328.

[44] Yeo M, Li X, Shen K, et al. Can Svm Be Used for Automatic EEG Detection of Drowsiness During Car Driving?[J]. Safety Science, 2009, 47(1): 115-124.

[45] Sant’Ana M, Li G, Zhang H. A Decentralized Sensor Fusion Approach to Human Fatigue Monitoring in Maritime Operations[C]// proceedings of the 2019 IEEE 15th International Conference on Control and Automation (ICCA). 16-19 July 2019.

[46] Shangguan P, Qiu T, Liu T, et al. Feature Extraction of EEG Signals Based on Functional Data Analysis and Its Application to Recognition of Driver Fatigue State[J]. Physiological Measurement, 2020, 41(12): 125004 (125012pp).

[47] Dinges D F, Mallis M M, Maislin G, et al. Evaluation of Techniques for Ocular Measurement as an Index of Fatigue and as the Basis for Alertness Management[J]. 1998.

[48] Kubitz K A, Mott A A. EEG Power Spectral Densities During and after Cycle Ergometer Exercise[J]. Research Quarterly for Exercise & Sport, 1996, 67(1): 91-96.

[49] Ng S C, Raveendran P. EEG Peak Alpha Frequency as an Indicator for Physical Fatigue[D]. Springer Berlin Heidelberg, 2007.

[50] Jap B T, Lal S, Fischer P, et al. Using EEG Spectral Components to Assess Algorithms for Detecting Fatigue[J]. Expert Systems with Applications, 2009, 36(2, Part 1): 2352-2359.

[51] Zadry H R, Dawal S Z, Taha Z. Effect of Load on Upper Limb Muscle and Brain Activity in Light Assembly Task[J]. Lecture Notes in Engineering & Computer ence, 2010, 2182(1).

[52] Trejo L J, Kubitz K, Rosipal R, et al. EEG-Based Estimation and Classification of Mental Fatigue[J]. Psychology, 2015, 6(6): 572-589.

[53] 王永祥. 基于EEG的疲劳检测技术的研究与应用[D]. 南京邮电大学, 2019.

[54] 王莉莉, 杨勇. 不同流量等级下管制员特情反应时的相关性分析[J]. 安全与环境学报, 2020, 20(03): 999-1006.

[55] 徐明伟, 金龙哲, 张安琪. 基于EEG的有限空间检修工脑疲劳特性研究[J]. 中国安全生产科学技术, 2020, 16(06): 171-176.

[56] 王保国，王新泉等. 安全人机工程学[M]. 北京: 机械工业出版社, 2009.

[57] 郭伏, 钱省三, 李兴东 等. 21世纪工业工程专业规划教材 人因工程学[M]. 2006.

[58] 伏开放. 员工作业疲劳影响因素与测量研究综述[J]. 中小企业管理与科技(中旬刊), 2019, 572(04): 7-8.

[59] Grandjean E. Fatigue in Industry[J]. British Journal of Industrial Medicine, 1979, 36(3): 175-186.

[60] Abdelfattah H M, Abdelazeim F H, Elshennawy S. Physical and Cognitive Consequences of Fatigue: A Review[J]. Journal of Advanced Research, 2015, 6(3): 351-358.

[61] Phillips R O. A Review of Definitions of Fatigue - and a Step Towards a Whole Definition[J]. Transportation Research Part F Psychology & Behaviour, 2015, 29: 48-56.

[62] Chen M, Lu S, Mao I. Subjective Symptoms and Physiological Measures of Fatigue in Air Traffic Controllers[J]. International Journal of Industrial Ergonomics, 2019, 70: 1-8.

[63] Ahsberg E. Dimensions of Fatigue in Different Working Populations[J]. Scandinavian journal of psychology, 2000, 41(3): 231-241.

[64] Knauth P. Extended Work Periods[J]. Industrial Health, 2007, 45(1): 125-136.

[65] Fan J, Smith A P. A Preliminary Review of Fatigue among Rail Staff[J]. Frontiers in Psychology, 2018, 9: 634-634.

[66] 郭晓波, 郭海林. 影响作业疲劳的因素及对策研究[J]. 中国安全生产科学技术, 2009, 5(06): 189-192.

[67] 鞠星, 牛海波, 张贺然, 等. 基于权重分析的航空物探遥感空勤人员疲劳影响因素[J]. 科学技术与工程, 2020, 20(01): 417-422.

[68] Theron W J, Heerden G. Fatigue Knowledge - a New Lever in Safety Management[J]. Journal of the Southern African Institute of Mining & Metallurgy, 2011, 111(1): 1-10.

[69] Martell M. Mine Worker Fatigue and Circadian Rhythms: How Biological Clocks Respond to Light and Darkness[J]. Engineering & Mining Journal, 2018, (2): 38-40.

[70] Borg G. Psychophysical Scaling with Applications in Physical Work and the Perception of Exertion[J]. Scandinavian Journal of Work Environment & Health, 1990, (suppl 1): 55-58.

[71] Chen Y L, Chen C C, Hsia P Y, et al. Relationships of Borg's Rpe 6-20 Scale and Heart Rate in Dynamic and Static Exercises among a Sample of Young Taiwanese Men[J]. Perceptual & Motor Skills, 2013, 117(3): 971.

[72] 陈燕达, 屠彦, 王莉莉, 等. 视觉疲劳状态下的生理指标研究[J]. 电子器件, 2015, v.38(06): 1245-1248.

[73] 石乔莉, 王延辉, 李信政. 基于脑电信号的驾驶疲劳的研究[J]. 世界最新医学信息文摘(电子版), 2017, (55): 32-32.

[74] 徐胜, 金龙哲, 徐明伟. 受限空间钻孔工人的肌肉疲劳特性研究[J]. 中国安全科学学报, 2019, 029(003): 76-81.

[75] 王黎, 于涛, 闻邦椿. 基于脑电α波的非线性参数人体疲劳状态判定[J]. 东北大学学报, 2005, (12): 1174-1177.

[76] Fouladi Dehaghi B, Mohammadi A, Nematpour L. Mental Fatigue Assessment Using Recording Brain Signals: Electroencephalography[J]. Journal of Ergonomics, 2019, 7(2): 45-53.

[77] 姚娟娟, 路堃, 马兴宇, 等. 基于脑电和眼电的疲劳检测方法的研究[J]. 电子设计工程, 2020, 28(06): 115-120.

[78] Oldfield R C. The Assessment and Analysis of Handedness: The Edinburgh Inventory[J]. Neuropsychologia, 1971, (1): 97-113.

[79] Nishikitani M, Nakao M, Karita K, et al. Influence of Overtime Work, Sleep Duration, and Perceived Job Characteristics on the Physical and Mental Status of Software Engineers[J]. Industrial Health, 2005, 43(4): 623-629.

[80] Åhsberg E, Kecklund G, Åkerstedt T, et al. Shiftwork and Different Dimensions of Fatigue[J]. International Journal of Industrial Ergonomics, 2000, 26(4): 457-465.

[81] Akerstedt T, Gillberg M. Subjective and Objective Sleepiness in the Active Individual[J]. International Journal of Neuroscience, 1990, 52(1-2): 29-37.

[82] Riethmeister V, Matthews R W, Dawson D, et al. Time-of-Day and Days-on-Shift Predict Increased Fatigue over Two-Week Offshore Day-Shifts[J]. Applied Ergonomics, 2019, 78: 157-163.

[83] Petru R, Wittmann M, Nowak D, et al. Effects of Working Permanent Night Shifts and Two Shifts on Cognitive and Psychomotor Performance[J]. International Archives of Occupational and Environmental Health, 2005, 78(2): 109-116.

[84] Delorme A, Makeig S. Eeglab: An Open Source Toolbox for Analysis of Single-Trial EEG Dynamics Including Independent Component Analysis[J]. Journal of Neuroscience Methods, 2004, 134(1): 9-21.

[85] 杨竹青, 李勇, 胡德文. 独立成分分析方法综述[J]. 自动化学报, 2002, (05): 762-772.

[86] Halvani G H, Zare M, Mirmohammadi S J. The Relation between Shift Work, Sleepiness, Fatigue and Accidents in Iranian Industrial Mining Group Workers[J]. Industrial Health, 2009, 47(2): 134-138.

[87] Hirshkowitz M, Whiton K, Albert S M, et al. National Sleep Foundation's Updated Sleep Duration Recommendations: Final Report[J]. Sleep Health, 2015: 233-243.

[88] Maisey G, Cattani M, Devine A, et al. The Sleep of Shift Workers in a Remote Mining Operation: Methodology for a Randomized Control Trial to Determine Evidence-Based Interventions[J]. Frontiers in Neuroscience, 2021, 14(1380).

[89] Charbonnier S, Roy R N, Bonnet S, et al. EEG Index for Control Operators’ Mental Fatigue Monitoring Using Interactions between Brain Regions[J]. Expert Systems with Applications, 2016, 52: 91-98.

[90] Lal S, Craig A. Reproducibility of the Spectral Components of the Electroencephalogram During Driver Fatigue[J]. International Journal of Psychophysiology, 2005, 55(2): 137-143.

[91] Belyavin A, Wright N A. Changes in Electrical Activity of the Brain with Vigilance[J]. Electroencephalography and Clinical Neurophysiology, 1987, 66(2): 137-144.

[92] 王利, 艾玲梅, 王四万, 等. 驾驶疲劳脑电信号节律的特征分析[J]. 生物医学工程学杂志, 2012, 29(04): 629-633.

[93] 崔玉洁. 基于脑电技术的密闭空间作业脑力疲劳识别方法[D]. 天津体育学院, 2020.

[94] Ana M S, Li G, Zhang H. A Decentralized Sensor Fusion Approach to Human Fatigue Monitoring in Maritime Operations[C]// proceedings of the IEEE International Conference on Control and Automation.

[95] Rumelhart D, Hinton G E, Williams R J. Learning Representations by Back Propagating Errors[J]. Nature, 1986, 323(6088): 533-536.

[96] 蔡靖, 李玉涛, 宋雪丰, 等. 基于脑电两节律和BP神经网络的运动想象分类研究[J]. 电子测量与仪器学报, 2020, 34(06): 176-182.

[97] 郭孜政, 谭茜, 吴志敏, 等. 基于脑电信号的动车司机对突发事件反应时间的预测方法研究[J]. 铁道学报, 2018, 40(12): 1-6.

[98] Liang Y, University Y N. Design of Association Mining Model Based on Improved BP Neural Network[J]. Modern Electronics Technique, 2018.

[99] 吉珊珊, 柯钢. 基于改进BP神经网络的网络流量预测[J]. 计算机与数字工程, 2020, 48(07): 1682-1686.

[100]刘天明. 基于BP神经网络的配网线损计算[D]. 南昌大学, 2020.