Data from: An intelligent control algorithm for the gas precise drainage problem based on model predictive control
Data files
Jan 20, 2026 version files 130.73 KB
-
README.md
2.64 KB
-
xiangmushuju.csv
128.09 KB
Abstract
Intelligent extraction of coal seam gas constitutes a crucial development direction for managing underground gas disasters. Building on an established mathematical model, this study develops an intelligent control model for gas extraction. In this model, controlled variables include gas extraction concentration, gas extraction flow rate, negative pressure, and extraction pump efficiency ratio, while control variables are defined as the valve opening of extraction boreholes and the power of extraction pumps. The ideal curve of the controlled quantity with time is obtained by using the recurrent neural network (SimpleRNN), and the controlled quantity is intelligently controlled by the model predictive control (MPC) algorithm so that the actual value of controlled quantity approaches the reference value at the corresponding time of its ideal curve. Taking the simulated gas extraction data as an example, an algorithm simulation experiment is performed. The experimental results show that the ideal reference curve of the controlled quantity obtained by the cyclic neural network has a good data fitting degree. The dynamic control of the controlled quantity by the model predictive control algorithm can overcome the interference of environmental and nonlinear factors and achieve a better control effect, which provides a certain reference for the intelligent control of gas drainage.
Data and File Structure
The data serves as the basis for the algorithm simulation. The four control variables for the gas extraction system are: gas extraction concentration, gas extraction purity, gas extraction negative pressure, and gas extraction pump efficiency ratio. The raw data were cleaned, and missing and outlier values were imputed using the nearest mean method. Due to the presence of values with different dimensions in the dataset, the data was normalized and scaled to the range of 0-1. Files and Variables File: xiangmushuju.csv Description: Variables
- Date/Time
- 3356
- 3357
- 3358
- 3460
3356, 3357, 3358, 3460 correspond to Gas extraction concentration, Gas extraction purity, Gas extraction pump efficiency ratio, Gas extraction negative pressure
- Gas extraction concentration: The overall trend of gas extraction concentration shows a decrease from 40% to 5% over time, with the actual value at the current moment being infinitely close to the optimal value of the ideal reference curve. （Figure 6a in the manuscript）
- Gas extraction purity: From an overall trend perspective, gas purity gradually decreases over time (with the original range specified as 9.0-5.0 m³/min), and the actual value at the current moment is infinitely close to the optimal value of the ideal reference curve.（Figure 6b in the manuscript）
- Gas extraction pump efficiency ratio: This ratio should be maintained at approximately 1.5 m³/(kW·h) to meet the requirements of economical gas extraction.（Figure 6d in the manuscript）
- Gas extraction negative pressure: The negative pressure during the extraction process must be maintained within the safe range of 10-30 kPa.（Figure 6c in the manuscript）
Code/software
Algorithm: Model epochs, batch size, and iteration count settings and conversion relationship.p Input: Dataset dataset
- train_size=int(len(dataset)*0.70)
- train, test=split_dataset(dataset,train_size,look_back)
- Randomly divide 70% of the dataset into the training set data
- batch_size=20
- for in trange(i—iteration count)
- history=model.fit(train_x,train_y,
- batch_size=20—batch size,
- nb_epoch—number of epochs
- validation_data
- verbose,
- Shuffle
- Each epoch requires 7000 data samples for training
- Each epoch requires 7000/20+1=35 batches
- Each iteration uses 7000 data for training, with weights updated 351 times