In order to solve the problems of low accuracy of commonly used time series prediction algorithms and difficulty in tuning parameters, a load prediction method was proposed based on a multilayer bidirectional long short-term memory (BiLSTM) neural network. The method includes network model design, adaptive parameter setting, and improved particle swarm optimization (PSO). The data was input into the BiLSTM network model for training and uses an adaptive algorithm for automatically parameter adjustment. The method uses a model evaluation method based on the benchmark model to calculate the fitness of the improved particle swarm optimization algorithm. The method uses improved particle swarm optimization to optimize the prediction results of the model. The average absolute percentage error of this method was reduced by 3.6% to 7.2% compared with several typical forecasting algorithms, and the training time was reduced by more than 10%, Through experimental comparison with a variety of typical time series prediction algorithms. Experimental results show that the method has higher accuracy and stronger applicability in time series forecasting, and provide an important scientific basis for the use of load forecast results for elastic expansion and contraction.
Fig.4Flowchart of adaptive parameter adjustment algorithm
Fig.5Application load of request volume in one-day
Fig.6Application load of request volume in one-week
Fig.7Application load of request volume in 35 days
模型
RMSE
MAE
MAPE/%
Q
Pa-BiLSTM
121.05
78.67
3.97
0.37
PSO-BiLSTM
157.13
92.53
4.63
0.45
BiLSTM
176.25
105.24
5.65
0.49
Pa-LSTM
158.39
102.84
5.38
0.46
LSTM
267.40
155.03
7.64
0.65
ARIMA（1，1，1）
366.84
167.46
5.87
0.73
HoltWinters
196.13
162.49
8.21
0.58
Prophet
234.92
198.94
11.18
0.67
基准模型
452.22
347.03
29.62
1.00
Tab.1Load prediction errors comparison
Fig.8Forecast accuracy of different models under different data sizes
Fig.9Training time of different models on different data sizes
Fig.10Comparison chart of fitness between oPaBiLSTM and Pa-BiLSTM
[1]
YAN Y Q, GUO P Predicting resource consumption in a Web server using ARIMA model[J]. Journal of Beijing Institute of Technology, 2014, 23 (4): 502- 510
[2]
CALHEIROS R N, MASOUMI E, RANJAN R, et al Workload prediction using ARIMA model and its impact on cloud applications' QoS[J]. IEEE Transactions on Cloud Computing, 2014, 3 (4): 449- 458
[3]
PRASSANNA J, VENKATARAMAN N Adaptive regressive holt–winters workload prediction and firefly optimized lottery scheduling for load balancing in cloud[J]. Wireless Networks, 2019, 1- 19
[4]
YAZDANIAN P, SHARIFIAN S. Cloud workload prediction using ConvNet and stacked LSTM [C]// 2018 4th Iranian Conference on Signal Processing and Intelligent Systems. Tehran: IEEE, 2018: 83-87.
[5]
谢晓兰, 张征征, 郑强清, 等 基于APMSSGA-LSTM 的容器云资源预测[J]. 大数据, 2019, 5 (6): 62- 72 XIE Xiao-lan, ZHANG Zheng-zheng, ZHENG Qiang-qing, et al Container cloud resource prediction based on APMSSGA-LSTM[J]. Big Data Research, 2019, 5 (6): 62- 72
[6]
LIU J, TAN X Y, WANG Y. CSSAP: software aging prediction for cloud services based on ARIMA-LSTM hybrid model [C]// 2019 IEEE International Conference on Web Services. Milan: IEEE, 2019: 283-290.
[7]
KUMAR S D, SUBHA D P. Prediction of depression from EEG signal using long short term memory [C]// 2019 3rd International Conference on Trends in Electronics and Informatics. Tirunelveli: IEEE, 2019: 1248-1253.
[8]
ZI?BA M, TOMCZAK S K, TOMCZAK J M Ensemble boosted trees with synthetic features generation in application to bankruptcy prediction[J]. Expert Systems with Applications, 2016, 58: 93- 101
doi: 10.1016/j.eswa.2016.04.001
[9]
AKYUZ A O, UYSAL M, BULBUL B A, et al. Ensemble approach for time series analysis in demand forecasting: Ensemble learning [C]// 2017 IEEE International Conference on Innovations in Intelligent Systems and Applications. Gdynia: IEEE, 2017: 7-12.
[10]
BIN Y, YANG Y, SHEN F, et al. Bidirectional long-short term memory for video description [C]// Proceedings of the 24th ACM International Conference on Multimedia. New York: ACM, 2016: 436-440.
[11]
ZENNAKI O, SEMMAR N, BESACIER L. Inducing multilingual text analysis tools using bidirectional recurrent neural networks [C]// 26th International Conference on Computational Linguistics: Technical Papers. Osaka: COLING. 2016: 450-460.
[12]
SCHUSTER M, PALIWAL K K Bidirectional recurrent neural networks[J]. IEEE Transactions on Signal Processing, 1997, 45 (11): 2673- 2681
doi: 10.1109/78.650093
[13]
SUN W, XU L, HUANG X, et al. Bidirectional LSTM for ionospheric vertical total electron content forecasting [C]// 2017 IEEE Visual Communications and Image Processing. St. Petersburg: IEEE, 2017: 1-4.
[14]
HU P, TONG J, WANG J, et al. A hybrid model based on CNN and Bi-LSTM for urban water demand prediction [C]// 2019 IEEE Congress on Evolutionary Computation. Wellington: IEEE, 2019: 1088-1094.
[15]
HAN S, ZHANG F, XI J, et al. Short-term vehicle speed prediction based on convolutional bidirectional LSTM networks [C]// 2019 IEEE Intelligent Transportation Systems Conference. Auckland: IEEE, 2019: 4055-4060.
[16]
KENNEDY J, EBERHART R. Particle swarm optimization [C]// Proceedings of ICNN'95-International Conference on Neural Networks. Perth: IEEE, 1995, 4: 1942-1948.
[17]
CHEN C, TWYCROSS J, GARIBALDI J M A new accuracy measure based on bounded relative error for time series forecasting[J]. PloS One, 2017, 12 (3): e0174202
doi: 10.1371/journal.pone.0174202
张利彪, 周春光, 马铭, 等 基于粒子群算法求解多目标优化问题[J]. 计算机研究与发展, 2004, 41 (7): 1286- 1291 ZHANG Li-biao, ZHOU Chun-guang, MA Ming, et al Solving multi-objective optimization problems based on particle swarm optimization[J]. Journal of Computer Research and Development, 2004, 41 (7): 1286- 1291
[20]
DING W, FANG W. Target tracking by sequential random draft particle swarm optimization algorithm [C]// 2018 IEEE International Smart Cities Conference. Kansas City: IEEE, 2018: 1-7.
[21]
何明慧, 徐怡, 王冉, 等 改进的粒子群算法优化神经网络及应用[J]. 计算机工程与应用, 2018, (19): 17 HE Ming-hui, XU Yi, WANG Ran, et al Improved particle swarm optimization neural network and its application[J]. Computer Engineering and Applications, 2018, (19): 17
