Dry coal preparation control system based on data analysis

Abstract

The invention relates to the technical field of coal mining, in particular to a dry coal preparation control system based on data analysis, which comprises a real-time data acquisition module, a data analysis module, a self-adaptive control module and a feedback optimization module. The real-time data acquisition module acquires information such as granularity, density, images and moisture of coal particles through a granularity sensor, a density sensor, an image sensor and a humidity sensor, and transmits data to the central processing unit. And the data analysis module analyzes the collected data based on a random forest algorithm to generate a sorting precision prediction value. The self-adaptive control module dynamically adjusts key parameters of the coal dressing equipment according to the predicted value by utilizing a PID control algorithm, and the key parameters comprise the wind speed, the screen inclination angle, the material flow speed and the like. The system not only improves the coal dressing precision and stability, but also improves the self-adaptive capacity of the equipment through intelligent control and optimization, and the system is suitable for popularization and application. The method has remarkable economic benefits and wide application prospects.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of coal mining, and particularly relates to a dry coal separation control system based on data analysis. BACKGROUND

[0002] In the process of coal mining and utilization, coal separation technology is a key link to improve coal quality, reduce pollution, and increase resource utilization. Traditional coal separation methods mainly include wet coal separation and dry coal separation. Among them, dry coal separation gradually attracts attention due to its simple operation, water saving and environmental protection, especially in water resource scarce areas. However, the existing dry coal separation technology still has some problems to be solved, mainly including the following aspects: Insufficient system stability: Traditional dry coal separation systems mostly rely on manually set operating parameters, lack real-time monitoring and automatic adjustment capabilities, leading to a delayed response when coal quality changes. This passive control method easily leads to fluctuations in separation efficiency, affecting the final quality of coal, causing resource waste and economic loss.

[0003] Limited data analysis capability: Currently, many dry coal separation systems rely on simple statistical analysis methods to evaluate the physical properties of coal, such as particle size, density, moisture content, etc. However, these methods often fail to effectively handle complex and changing coal qualities, leading to judgment errors and decreased coal separation accuracy during the separation process.

[0004] Lack of intelligent control: The control systems in existing technologies generally lack intelligent level, and cannot dynamically optimize and adjust based on real-time collected data. This makes the coal separation process unable to adapt to the changing operating environment, reducing the flexibility and adaptability of the system.

[0005] Environmental and resource impact: In the coal separation process, failure to effectively control parameters such as wind speed and flow rate can lead to low coal separation rates and increased waste production. In addition, traditional methods have a greater impact on the environment, especially when dealing with water-containing coal, which can cause water waste and environmental pollution.

[0006] Slow technology update: Although some new technologies such as artificial intelligence and machine learning have been applied in many industries, their application in the field of dry coal separation is still lagging behind. The lack of targeted and practical technical solutions has slowed down the pace of technological updates in the entire industry, affecting the sustainable development of the coal industry.

[0007] In summary, the current dry coal preparation technology faces many challenges in system stability, data analysis capability, intelligent control, etc. In order to solve these problems, an urgent need for a dry coal preparation control system based on data analysis is needed, which can improve the accuracy and efficiency of the coal preparation process through real-time data acquisition, intelligent analysis and adaptive control, so as to promote the efficient use of coal resources and environmental protection. SUMMARY

[0008] The present application proposes a dry coal preparation control system based on data analysis, aiming to solve the problems of poor system stability, limited data analysis capability, insufficient intelligent control, etc. in the existing dry coal preparation technology. Through real-time data acquisition, intelligent data analysis, adaptive control and feedback optimization modules, the system realizes accurate control and dynamic adjustment of the coal preparation process, significantly improving the coal preparation effect and system stability.

[0009] The dry coal preparation control system of the present application consists of four main modules: real-time data acquisition module, data analysis module, adaptive control module and feedback optimization module. The real-time data acquisition module collects real-time data such as particle size, density, image and moisture content of coal particles through particle size sensor, density sensor, image sensor and humidity sensor, and stores them in the central processing unit. The data analysis module uses random forest algorithm to extract and analyze the collected data, and generates a prediction value of separation accuracy by comparing the historical data and real-time data. The adaptive control module uses PID control algorithm to automatically adjust the key parameters of the coal preparation equipment, such as wind speed, screen inclination and material flow rate, according to the prediction value provided by the data analysis module, to ensure that the system is always in the optimal working state. The feedback optimization module compares the actual separation effect with the prediction value, and performs feedback control according to the error size, and triggers the retraining of the machine learning model when necessary, to continuously optimize the separation accuracy and system performance.

[0010] The operation process of the system is as follows: first, the real-time data acquisition module obtains the physical property data of coal particles; then, the data analysis module analyzes and processes these data to predict the current separation accuracy; next, the adaptive control module dynamically adjusts the parameters of the coal preparation equipment according to the predicted separation accuracy to achieve the optimal separation effect; finally, the feedback optimization module monitors the actual separation effect in real time, and performs closed-loop feedback adjustment according to the difference between the actual effect and the prediction value, to ensure that the system can self-optimize during operation, improving the accuracy and stability of the coal preparation process.

[0011] The beneficial effects of the present application are: 1.The present application improves system stability: the present application introduces an adaptive control module and a feedback optimization module to overcome the problem of insufficient system stability in the prior art. By real-time monitoring and dynamic adjustment of the working parameters of the coal preparation equipment, the system can automatically adapt to the change of coal quality, maintain high stability and accuracy of the coal preparation process, and avoid the problem of reduced separation accuracy caused by fluctuation of coal quality. The system of the present application uses advanced machine learning techniques such as random forest algorithm to analyze multi-dimensional data such as particle size, density, moisture and image of coal particles, and generates a separation accuracy prediction value based on these data. Compared with traditional simple statistical analysis method, machine learning algorithm significantly improves the accuracy and reliability of data analysis, which can effectively process complex and variable coal quality data, thereby improving the separation accuracy. The adaptive control module combines with the PID control algorithm to realize the automatic adjustment of the coal preparation equipment. According to the real-time data and prediction value, the system can automatically adjust the wind speed, screen inclination and material flow rate and other key parameters to ensure that the equipment always works in the best state. This intelligent control reduces manual intervention, improves operation efficiency, and ensures high precision and consistency of the separation process. The feedback optimization module can automatically adjust according to the error between the actual separation rate and the predicted accuracy, and start the retraining mechanism of the machine learning model when necessary. This closed-loop feedback system ensures that the equipment can continuously optimize the separation process, so that the system can maintain high efficiency and stable working state after a long time of operation. The present system reduces resource waste and maximizes the utilization rate of coal through accurate parameter control and separation optimization. At the same time, since the system can realize accurate dry coal preparation, it does not need additional water resource consumption, especially suitable for areas with water resource shortage, and has good environmental protection effect. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only illustrate the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0013] Figure 1 The present application is a system architecture schematic diagram of the embodiment. Figure 2 The present application is a flowchart of the data analysis module of the embodiment. Figure 3 The present application is a PID control flowchart of the adaptive control module of the embodiment. Figure 4 The present application is a closed-loop control schematic diagram of the feedback optimization module of the embodiment. DETAILED DESCRIPTION

[0014] To make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in combination with specific embodiments.

[0015] 1. Real-time data acquisition module 1.1 Sensor configuration Particle size sensor: A laser particle size analyzer is used to measure the particle size distribution of coal particles in real time.

[0016] Measurement range: 1 µm to 50 mm, accuracy ±0.1 µm.

[0017] Sampling frequency: 100 samples per second.

[0018] Density sensor: An ultrasonic densimeter is used to measure the density of coal by measuring the reflection of sound waves.

[0019] Measurement range: 0.5 g / cm³ to 2.5 g / cm³, accuracy ±0.01 g / cm³.

[0020] Image sensor: A high-resolution industrial camera is configured to capture images of coal blocks.

[0021] Image processing software (such as OpenCV) is used to analyze the shape and color of coal.

[0022] Humidity sensor: A capacitive humidity sensor is used to monitor the moisture content in coal.

[0023] Measurement range: 0% to 100%, accuracy ±1%.

[0024] 1.2 Data acquisition process The data collected by the sensors is transmitted to the central processing unit (CPU) in the form of digital signals, and the data format is JSON.

[0025] The data is stored in a local database (such as MySQL) for subsequent processing and querying.

[0026] The data acquisition frequency is set to once per second to ensure real-time monitoring.

[0027] 2. Data analysis module 2.1 Data preprocessing Noise filtering: The median filter algorithm is used to suppress noise in the collected data.

[0028] Outlier detection: The Z-score method is used to identify and eliminate outliers, with a standard deviation threshold of 3.

[0029] 2.2 Feature extraction Features extracted from the collected coal particle data include: Mean particle size Particle size standard deviation Particle shape factor (based on image processing results) Moisture content Density 2.3 Machine Learning Model Model Selection: Random Forest algorithm is used for training, suitable for handling non-linear features and high-dimensional data.

[0030] Training Process: 70% of historical data is used for model training, and 30% of data is used for verification.

[0031] Model output is the prediction value of sorting accuracy.

[0032] Prediction Model Formula: Where, is the feature, is the corresponding weight, is the bias term.

[0033] 3. Adaptive Control Module 3.1 Control Objectives Ensure that the key parameters of the coal preparation equipment are always in an optimized state, including air speed, screen inclination, material flow rate.

[0034] 3.2 Control Algorithm Use PID control algorithm for dynamic adjustment.

[0035] Parameter Settings: Proportional Gain , Integral Gain , Differential Gain Preliminary settings are made according to historical operation data, and adjustments are made through tests Optimization.

[0036] PID Control Formula: : Error between target sorting accuracy and current predicted accuracy.

[0037] 3.3 Adjustment Execution The control system updates the equipment parameters every 5 seconds, automatically adjusts the air speed of the air separator, the screen angle of the vibrating screen, and the material flow rate according to the calculation results.

[0038] 4. Feedback Optimization Module 4.1 Real-time Feedback Collection Actual sorting effect is obtained through sorting rate sensors, and real-time coal quality data is recorded.

[0039] Actual data is compared with predicted data to calculate feedback error.

[0040] 4.2 Optimization algorithm When the error exceeds the set threshold, the model retraining mechanism is triggered.

[0041] Feedback closed-loop control formula: Online learning for actual 4.3 The system uses new data and feedback results to update the machine learning model at the end of each cycle, improving its accuracy and response speed.

[0042] 5. Overall system integration All modules are integrated through industrial control systems (such as PLC or embedded systems) to ensure real-time data transmission and processing.

[0043] The interface is user-friendly, and operators can monitor data in real time and manually intervene through the touch screen.

[0044] The error between the actual sorting precision and the predicted precision, is the feedback gain.

[0045] The present invention is intended to cover all such alternatives, modifications, and variations as fall within the broad scope of the appended claims. Accordingly, any omission, modification, substitution, improvement, made in the spirit and principle of the invention, should be included in the protection scope of the invention.

Claims

1. A dry coal preparation control system based on data analysis, characterized in that: include: Real-time data acquisition module, used to collect coal particle size, density, moisture and image data in real time through multiple sensors; A data analysis module is used to preprocess and extract features from the collected data, and generate a predicted value of sorting accuracy through a machine learning algorithm; Adaptive control module for adjusting key parameters of coal preparation equipment based on predicted values, including wind speed, screen inclination, and material flow rate; The feedback optimization module is used to monitor the sorting effect in real time and make closed-loop adjustments based on the error between the actual effect and the predicted value.

2. A dry coal preparation control system based on data analysis according to claim 1, characterized in that: The real-time data acquisition module includes a particle size sensor, a density sensor, an image sensor and a humidity sensor. The particle size sensor is used to measure the size distribution of coal particles. The density sensor is used to measure the density of coal through sound wave reflection or ultrasound. The image sensor is used to obtain images of coal particles and analyze their shape and color visual characteristics. The humidity sensor monitors the moisture content in the coal in real time. The data collected by all sensors are aggregated and stored in a database through the central processing unit (CPU), which can provide sufficient data support for subsequent data processing and analysis.

3. The dry coal preparation control system based on data analysis according to claim 1, characterized in that: The data analysis module processes the collected data using the random forest algorithm. The data first goes through preprocessing steps, noise filtering and outlier detection to clean up inaccurate data, and then extracts important feature information, such as the average particle size of coal particles, the standard deviation of the particle size distribution, the density and moisture of coal particles. Then, the random forest algorithm is used to analyze and process these features. The algorithm compares historical data with the currently collected data to predict the sorting accuracy of the current coal preparation process. The random forest algorithm is an integrated learning method that improves the accuracy of prediction by constructing multiple decision trees and generates an overall sorting accuracy prediction value. The formula is: ,in represents the eigenvector, is the weight of the feature, This process can more accurately reflect the characteristics and separation effect of coal.

4. The dry coal preparation control system based on data analysis according to claim 1, characterized in that: The adaptive control module uses a PID control algorithm to adjust the operating parameters of the coal preparation equipment according to the separation accuracy prediction value provided by the data analysis module, including adjusting the wind speed, the inclination angle of the vibrating screen, and the flow rate of the material. PID control is a common feedback control algorithm. Its core idea is to continuously adjust the equipment parameters based on the difference between the current separation accuracy and the target separation accuracy, that is, the error value. Its control formula is: ,in is the error value, 、 and They are proportional, integral and differential coefficients respectively. By adjusting these coefficients, the control effect can be optimized, so that the operating parameters of the coal preparation equipment can be kept in the optimal state, thereby improving the sorting accuracy. The working frequency of the adaptive control module is real-time, and it can dynamically adjust the equipment parameters according to the actual feedback value.

5. The dry coal preparation control system based on data analysis according to claim 1, characterized in that: The feedback optimization module monitors the coal sorting quality during the actual coal preparation process by real-time monitoring of the sorting rate, that is, by setting a sorting rate sensor. The data will be compared with the predicted sorting accuracy. The system will determine whether the equipment parameters need to be further adjusted based on this error value. If the error exceeds the preset threshold, the feedback optimization module will activate the feedback closed-loop control mechanism. The control formula of this process is: ,in Indicates the error in sorting accuracy, For feedback gain, the module adjusts the control parameters according to the difference between the actual feedback and the predicted value to ensure the stability and efficiency of the system. In addition, when the error is large, the system will trigger the retraining mechanism of the machine learning model and continuously optimize the model, making the system's adaptive control more intelligent and continuously improving the sorting accuracy and stability.

6. The dry coal preparation control system based on data analysis according to claim 1, characterized in that: The system is integrated through an industrial control system, which provides a user-friendly graphical interface. Operators can monitor the particle size, density, moisture, and image feature data of coal particles in real time through a touch screen or remote terminal, and can also manually adjust key parameters. All modules in the system cooperate with each other to ensure the accuracy and stability of the entire coal preparation process through a closed-loop operation of data collection, analysis, feedback and control. At the same time, the system also supports data recording and historical data analysis to help operators optimize the production process.

Images