A Ball Mill Load Detection Method Based on Genetic Algorithm Optimizing BP Neural Network
A BP neural network and genetic algorithm technology, applied in the field of ball mill load detection based on genetic algorithm optimization of BP neural network, can solve problems such as low accuracy, waste, energy waste, etc., to achieve easy implementation, improve automatic control, and adaptability strong effect
Active Publication Date: 2016-08-10
西安帝和电子科技有限公司
View PDF2 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0002] The ball mill is widely used in the material crushing industry. The determination of the load of the existing ball mill mainly depends on the long-term work experience of the on-site operators. The error is large and the accuracy is not high, resulting in certain waste or loss. The safe operation of the ball mill cannot be guaranteed. The long-term operation of the ball mill in the under-load state affects the processing capacity and product quality of the ball mill, and also causes energy waste.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0049]For load prediction of the wet ball mill in the mining plant, a medium-sized mill of Φ3.6m×4m is used, the diameter of the steel ball is about 70mm, the ore is fine-vein disseminated molybdenum ore, and the speed is 18-22r / min; 16-bit single-channel sampling is adopted. The sampling frequency is 44100Hz, and the implementation steps are as follows:
[0050] Step 1: Collect the grinding sound of starting grinding for 20 minutes and grinding sound of stopping grinding for 18 minutes;
[0051] Step 2: Pinch the beginning and end of the grinding sound and stop grinding sound, add Hamming window processing, and then perform fast Fourier transform on each windowed data segment, and the number of Fourier transform points is N FFT = 44100 × 60 = 2646000, take the square of each transformation result, and divide it by the number of fast Fourier transform points as an estimate of the power spectrum estimate; then convert this estimate into a sound pressure level; get figure 2 , ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for detecting the load of a ball mill based on a genetic algorithm to optimize a BP neural network. The steps include: step 1: collecting the starting and stopping sounds of the ball mill; step 2: processing the grinding sounds off-line; step 3: selecting Effective frequency band of grinding noise; Step 4: Calibrate mill load data; Step 5: Optimize BP neural network modeling based on genetic algorithm and predict mill load, and obtain load detection results of ball mill. The method of the present invention selects the effective frequency range of the grinding sound by analyzing the grinding sound spectrum, obtains multiple sets of experimental data, and uses the experimental data to train the load model. After the training, the model can be used to predict the load of the grinding machine; The inventive method is easy to implement and has strong adaptability, can provide relevant detection data for the optimal control of the ball mill, and improves the automatic control, energy saving and consumption reducing capabilities of the ball mill.
Description
technical field [0001] The invention belongs to the technical field of automatic detection and relates to a ball mill load detection method based on genetic algorithm optimization BP neural network. Background technique [0002] The ball mill is widely used in the material crushing industry. The determination of the load of the existing ball mill mainly depends on the long-term work experience of the on-site operators. The error is large and the accuracy is not high, resulting in certain waste or loss. The safe operation of the ball mill cannot be guaranteed. The long-term operation of the ball mill in the under-load state affects the processing capacity and product quality of the ball mill, and also causes the problem of energy waste. The energy consumption of ball mills accounts for a large proportion of the entire industry. Therefore, for ball mills, a nonlinear and complex system with large time delays, accurate detection of the mill load is of great significance for ene...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06N3/02G06N3/12
Inventor 弋英民惠瑜
Owner 西安帝和电子科技有限公司