Modelling And Research on Water Level Control of Great Lakes Based on Neural Network PID Algorithm

Lingfei Zhang; Jiayu Zhou; Gaofeng Zhang

doi:10.54097/a0q10t93

Authors

Lingfei Zhang
Jiayu Zhou
Gaofeng Zhang

DOI:

https://doi.org/10.54097/a0q10t93

Keywords:

Pearson correlation coefficient, network, PID control system, SHAP algotithm.

Abstract

The Great Lakes are situated in the border region between the United States and Canada, which has a significant impact on the climate and the lives of those living in the surrounding areas. The objective of this paper is to establish a network of the Great Lakes through Pearson's correlation coefficient analysis and to construct a two-tank water level model based on a PID control system in order to effectively manage the dynamics of the Great Lakes. Firstly, the strength and direction of the linear relationship between two variables is quantified through Pearson's correlation coefficient analysis, which involves the collection of observational data and the calculation of mean values. This analysis serves as a fundamental basis for predictive modelling and hypothesis testing. Secondly, based on the flow balance principle, mathematical expressions are constructed to simulate the water flow, and a PID control system is constructed to achieve optimal water level maintenance. By analysing the Pearson's correlation coefficient, the interrelationships among the variables in the Great Lakes network can be understood, thereby providing guidance for scientific research and decision-making. The results demonstrate that the constructed two-tank water level model combined with the PID control system and SHAP algotithm can effectively manage the water level of the Great Lakes and achieve optimal water level regulation.

Downloads

Download data is not yet available.

References

[1] U.S. Geological Survey. Manning's Roughness Coefficients for South Carolina Streams[R]. Washington, DC: U.S. Geological Survey, 2022.

[2] Benesty, J., Chen, J., Huang, Y. On the Importance of the Pearson Correlation Coefficient in Noise Reduction[J]. IEEE Transactions on Audio, Speech, and Language Processing, 2020, 28(4): 757-765.

[3] Gabryel, M., Scherer, R., Damaševičius, R. Pearson Correlation-Based Feature Selection for Document Classification Using Balanced Training[J]. Sensors, 2020, 20(23): 6793.

[4] Adrian Gambier. MPC and PID Control Based on Multi-Objective Optimization[C]// Proceedings of the 2008 American Control Conference. IEEE, 2008: 4727-4732.

[5] Chen, M., Liu, W. Robust Self-Learning PID Control of an Aircraft Anti-Skid Braking System[J]. Mathematics, 2022, 10(8): 1290. DOI: 10.3390/math10081290.

[6] Hossein Bonakdari, Isa Ebtehaj, Pijush Samui, Bahram Gharabaghi. Lake Water-Level Fluctuations Forecasting Using Minimax Probability Machine Regression, Relevance Vector Machine, Gaussian Process Regression, and Extreme Learning Machine[J]. Water Resources Management, 2019, 33: 3965-3984.

[7] Bahrudin Hrnjica, Ognjen Bonacci. Lake level prediction using feed forward and recurrent neural networks[J]. Water Resources Management, 2019, 33(7): 2471-2484.

[8] Ahmet Sancak Sanlı, Erdal Kesgin, R. Ilayda Tan, Hayrullah Agaccioglu, Ibrahim Demir, Ilke Sinan. Effect of lake-water budget management preferences on optimum operating conditions and neighboring basins interacting: case of lake Beysehir (Turkey)[J]. Sustainable Water Resources Management, 2022, 8: 1-18.

[9] Pengfei Xue, Aditya Wagh, Gangfeng Ma, Yilin Wang, Yongchao Yang, Tao Liu, Chenfu Huang. Integrating deep learning and hydrodynamic modeling to improve the Great Lakes forecast[J]. Remote Sensing, 2022, 14(11): 2640.

[10] E. Maria, E. Budiman, M. Taruk, et al. Measure distance locating nearest public facilities using haversine and euclidean methods[C]// Journal of Physics: Conference Series, 2020, 1450: 012080. IOP Publishing.