Hard rock non-explosion continuous mining method and intelligent equipment

Abstract

The invention provides a hard rock non-explosion continuous mining method and intelligent equipment, and is suitable for the field of mining mechanical equipment. The method comprises the steps of sensor deployment, data processing, working parameter automatic adjustment and self-learning, a vibration sensor, a torque sensor, a rotating speed sensor and a temperature sensor are installed on mining equipment, and time sequence data are collected; the rock stratum hardness recognition is performed by utilizing feature extraction based on wavelet packet singular value decomposition, feature dimension reduction of a Relief algorithm and an improved XGBoost model, so that high-accuracy hardness detection is realized; the control center automatically controls the working parameters according to the recognition result, ensures that the working parameters are highly matched with the rock stratum hardness, realizes high efficiency and stability of the mining process, and reduces abrasion caused by vibration while reducing energy consumption; by recording and analyzing data in actual operation, the XGBoost model is continuously optimized, and the adaptability and accuracy of the system are improved; by introducing an intelligent and data-driven technology, the self-adaptive capability of a mining environment with complex and changeable rock stratum hardness is realized.

Description

Technical Field

[0001] The present invention relates to the field of mining machinery and equipment, and particularly to a method and intelligent equipment for continuous hard rock non-explosive excavation. Background Art

[0002] Hard rock excavation is a key link in mining engineering and is widely used in fields such as coal mines, metal mines, and building stones. With the increasing depletion of mineral resources and the continuous increase in mining depth, hard rock excavation technology faces higher requirements for safety, efficiency, and environmental protection. Traditional hard rock excavation methods mainly include blasting methods and non-blasting mechanical excavation methods. The blasting excavation method has the characteristics of high efficiency, strong adaptability, and relatively low cost, and can quickly break large rocks. However, it also has disadvantages such as threats to operators and equipment during the blasting process, a large amount of dust, noise, vibration, and damage to the surrounding ecological environment.

[0003] Compared with the blasting method, the non-blasting mechanical excavation method has advantages in terms of safety and environmental protection, but it still has significant deficiencies. Compared with the blasting method, the crushing efficiency of mechanical excavation is lower, and the excavation speed is slower. In rock formations with different hardnesses, it is difficult for the performance of mechanical equipment to be adaptively adjusted, resulting in unstable excavation efficiency. Moreover, continuous mechanical operation consumes a large amount of energy, and mechanical equipment wears severely under high-intensity operation, with high maintenance and replacement costs.

[0004] In view of this, it is particularly important to develop a method and equipment that can enhance excavation efficiency, reduce energy consumption, and reduce equipment wear for the demand of continuous hard rock non-explosive excavation operations. Summary of the Invention

[0005] The purpose of the present invention is to solve the problems of low, unstable excavation efficiency and large energy consumption of traditional non-blasting mechanical excavation methods. A method and intelligent equipment for continuous hard rock non-explosive excavation are proposed, which overcome the deficiencies of the prior art in terms of automation, monitoring, adaptability, and cost control, provide an efficient, safe, and environmentally friendly excavation solution, reduce unplanned downtime, improve overall operation efficiency, and can be widely applied to the field of mining machinery and equipment.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for continuous hard rock non-explosive excavation specifically includes the following steps:

[0008] S101, Sensor deployment;

[0009] The sensor deployment includes setting a vibration sensor inside the pick drum of the cutting part, setting a torque sensor on the rotating shaft driving the pick drum, setting a first rotational speed sensor on the input shaft of the transfer case, setting a second rotational speed sensor on the output shaft of the cutting part of the transfer case, and setting a temperature sensor on the housing of the hydraulic torque converter; the vibration sensor, torque sensor, first rotational speed sensor, second rotational speed sensor, and temperature sensor respectively collect the time-series data of the vibration variable V1, torque variable V2, rotational speed variable V3, rotational speed variable V4, and temperature variable V5 according to their respective set frequencies and transmit them to the control center;

[0010] S102. Data processing;

[0011] The data processing includes that after the control center receives the time-series data, it respectively performs feature extraction based on wavelet packet singular value decomposition and feature data dimensionality reduction based on the Relief algorithm on the time-series data of the vibration variable V1, torque variable V2, rotational speed variable V3, rotational speed variable V4, and temperature variable V5, and then generates high-order feature cross terms according to the feature vectors of the vibration variable V1, torque variable V2, rotational speed variable V3, rotational speed variable V4, and temperature variable V5, uses the high-order feature cross terms as the feature vectors of the new variable V6, and normalizes the feature vectors of the vibration variable V1, torque variable V2, rotational speed variable V3, rotational speed variable V4, temperature variable V5, and new variable V6 and uses them as the overall input parameters, inputs them into the improved XGBoost model, and then performs rock formation hardness identification based on the improved XGBoost;

[0012] The feature extraction based on wavelet packet singular value decomposition includes using the wavelet packet algorithm to decompose the time-series data layer by layer, decomposing the time-series data to n nodes according to the preset decomposition layer number, and obtaining the wavelet packet coefficients of each node; reconstructing the n nodes to obtain the reconstructed time-series data of each node, and combining the reconstructed time-series data of the n nodes to obtain the time-frequency matrix;

[0013] The feature data dimensionality reduction based on the Relief algorithm includes performing singular value decomposition on the time-frequency matrix, using the Relief algorithm to measure the classification ability of each order of singular values through the relevant statistics corresponding to each order of singular values, selecting the first k order of singular values with the largest to smallest relevant statistics, and using the k order of singular values as the feature vectors of the time-series data;

[0014] The rock hardness identification based on the improved XGBoost includes using the Bayesian optimization algorithm, combined with the Gaussian process surrogate model, to optimize the key hyperparameters of XGBoost, and then, based on the approximately optimal parameters found by Bayesian optimization, performing grid search to fine-tune the key hyperparameters; the improved XGBoost model obtains the probability distribution of each hardness level according to the overall input parameters, and selects the hardness level with the highest probability as the rock hardness identification result;

[0015] S103. Automatic adjustment of working parameters;

[0016] The automatic adjustment of the working parameters includes the improved XGBoost model completed by local deployment training of the intelligent hard rock non-explosive continuous mining equipment. The parameter adjustment decision module built in the control center automatically adjusts the engine power, transmission gear, and main boom downward pressure that are most matched with the current rock hardness level according to the hardness identification result;

[0017] S104. Self-learning;

[0018] The self-learning includes recording the time series data collected each time and the hardness identification result. When the mining stops due to the inconsistency between the hardness identification result and the actual situation, manually adjust the engine power, transmission gear, and main boom downward pressure. When the mining continues after adjustment, record the adjusted engine power, transmission gear, and main boom downward pressure, evaluate the correct hardness identification result according to the adjusted engine power, transmission gear, and main boom downward pressure, and then use the sliding window technique to regularly introduce new data and eliminate old data to fine-tune the improved XGBoost model.

[0019] As a preferred technical solution of the present invention, in step S102, the high-order feature cross-term is obtained by weighted summation of the feature vectors of any variable and the feature vectors of another variable.

[0020]

[0021] In the formula, - High-order feature cross-term, - Feature vector of any variable, - Feature vector of another variable, t - Weight.

[0022] As a preferred technical solution of the present invention, in step S103, the engine power, transmission gear, and main boom downward pressure that are most matched with the current rock hardness level are determined through on-site tests. The evaluation reference indicators for the best match are the mining speed, energy consumption, and equipment vibration.

[0023] As a preferred technical solution of the present invention, in step S102, the standardization process uses Z-score standardization.

[0024]

[0025] wherein, f′ i,j - the i-th value in the eigenvector of the j-th variable V j , μ j - the mean value of the eigenvector of the j-th variable V j , σ j - the standard deviation of the eigenvector of the j-th variable V j .

[0026] As a preferred technical solution of the present invention, in step S102, the hardness grades adopt fifteen grades of I, II, III, IIIa, IV, IVa, V, Va, VI, VIa, VII, VIIa, VIII, IX, and X classified according to the Protodyakonov coefficient.

[0027] As a preferred technical solution of the present invention, in step S104, the determination method of the mining stagnation is that the output power of the engine remains basically unchanged, the oil temperature of the hydraulic torque converter suddenly rises abnormally, and the numerical value of the rotational speed sensor of the output shaft of the cutting part of the transfer case rapidly decreases or even decreases to 0.

[0028] A non-explosive continuous hard rock mining intelligent equipment, which is used in the non-explosive continuous hard rock mining method, includes a frame arranged on a walking track assembly, a main arm arranged at the front end of the frame, a cutting part located at the front end of the main arm, an engine, a gearbox, an electric control system arranged at the rear end of the frame, and a transmission system arranged on the whole frame. A hydraulic torque converter is arranged between the engine and the gearbox, a transfer case is arranged between the cutting part and the gearbox, power is transmitted between the transfer case and the gearbox through the input shaft of the transfer case, and power is transmitted between the transfer case and the cutting part through the output shaft of the cutting part of the transfer case and the driven shaft of the cutting part. The cutting part includes two pick drums, a rotating shaft arranged at the central axis of the two pick drums, and a transmission cutting chain arranged between the two pick drums. One end of the transmission cutting chain is connected to the rotating shaft, and the other end is connected to the driven shaft of the cutting part. Picks are welded on both the pick drums and the transmission cutting chain. Driven by the engine, after passing through the transmission system, the pick drums are driven to rotate through the transmission cutting chain. It is characterized in that: a torque sensor is arranged on the rotating shaft driving the pick drums, rotational speed sensors are arranged on the input shaft of the transfer case and the output shaft of the cutting part of the transfer case, and a temperature sensor is arranged on the outer shell of the hydraulic torque converter. An electric control board integrating a control center and a server locally deploying an improved XG Boost model are added to the electric control system.

[0029] As a preferred technical solution of the present invention, after the control center receives the timing data collected by the sensors, performs feature extraction and dimensionality reduction processing on the feature data, it inputs the data into the improved XGBoost model for rock hardness identification, and then sends instructions to adjust the engine power, transmission gear, and main boom downward pressure according to the hardness identification result.

[0030] As a preferred technical solution of the present invention, the control center records all the collected timing data, hardness identification results, and the adjusted engine power, transmission gear, and main boom downward pressure in the storage device. When the storage device capacity is insufficient, it automatically issues a capacity warning to remind the maintenance personnel to replace the storage device in time.

[0031] The beneficial effects of the present invention are as follows: By installing vibration sensors, torque sensors, rotational speed sensors, and temperature sensors on the mining equipment, collecting timing data, and using feature extraction based on wavelet packet singular value decomposition, feature dimensionality reduction by the Relief algorithm, and an improved XGBoost model for rock hardness identification, high-accuracy hardness detection is achieved; According to the identification result, the control center automatically adjusts the engine power, transmission gear, and main boom downward pressure to ensure a high degree of matching between the mining parameters and the rock hardness, realizing the efficiency and stability of the mining process, reducing energy consumption while reducing wear caused by vibration; In addition, through the self-learning mechanism, the data in actual operation is recorded and analyzed to continuously optimize the XGBoost model, improving the adaptability and accuracy of the system; By introducing intelligent and data-driven technical means, the adaptive ability in the complex and changeable mining environment of rock hardness is realized. Description of the Drawings

[0032] Figure 1 is the flow chart of the hard rock non-explosive continuous mining method of the present invention;

[0033] Figure 2 is the structural schematic diagram of the hard rock non-explosive continuous mining intelligent equipment of the present invention;

[0034] Figure 3 is the schematic diagram of the engine and transmission system of the hard rock non-explosive continuous mining intelligent equipment of the present invention;

[0035] In the figure, reference numerals: 1 - walking track assembly, 2 - frame, 4 - main boom, 5 - cutting unit, 51 - cutting tooth drum, 52 - rotating shaft, 53 - driving cutting chain, 54 - driven shaft, 6 - power take-off box, 61 - input shaft, 6 - cutting unit output shaft, 7 - transmission, 8 - hydraulic torque converter, 9 - engine. Detailed Embodiments

[0036] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments given here are only used to illustrate and explain the present invention and cannot be used to limit the present invention. It should be noted that many specific details are set forth in the following description in order to fully understand the present invention. However, the present invention may have other embodiments and variations, and therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0037] Embodiment 1. The hard-rock non-explosive continuous excavation method is as follows:

[0038] A hard-rock non-explosive continuous excavation method specifically includes the following steps:

[0039] S101. Sensor deployment;

[0040] A vibration sensor is arranged inside the pick drum 51 of the cutting part 5, a torque sensor is arranged on the rotating shaft 52 driving the pick drum 51, speed sensors are arranged on the input shaft 61 of the transfer case 6 and the cutting part output shaft 62 of the transfer case 6, and a temperature sensor is arranged on the housing of the hydraulic torque converter 8; the vibration sensor, torque sensor, first speed sensor, second speed sensor, and temperature sensor respectively collect the time series data of the vibration variable V1, torque variable V2, speed variable V3, speed variable V4, and temperature variable V5 according to their respective set frequencies and transmit them to the control center;

[0041] S102. Data processing;

[0042] After receiving the time series data, the control center respectively performs feature extraction based on wavelet packet singular value decomposition and feature data dimensionality reduction based on the Relief algorithm on the time series data of the vibration variable V1, torque variable V2, speed variable V3, speed variable V4, and temperature variable V5;

[0043] The feature extraction based on wavelet packet singular value decomposition includes using the wavelet packet algorithm to decompose the time series data layer by layer. According to the preset decomposition layer number, the time series data is decomposed to n nodes and the wavelet packet coefficients of each node are obtained; the n nodes are reconstructed to obtain the reconstructed time series data of each node, and the reconstructed time series data of the n nodes are combined to obtain a time-frequency matrix;

[0044] The feature data dimensionality reduction based on the Relief algorithm includes performing singular value decomposition on the time-frequency matrix, using the Relief algorithm to measure the classification ability of each order of singular value through the relevant statistics corresponding to each order of singular value, selecting the first k orders of singular values with the largest to smallest relevant statistics, and taking the k orders of singular values as the feature vectors of the time series data; taking the time-frequency matrix of the vibration variable V1 as an example, given m samples with determined rock layer hardness grades, denoted as {(x1,y1),(x2,y2),…,(x i, y i ), …, (x m , y m )}, where y i is the rock formation hardness grade corresponding to the sample x i . For each sample x i , its r-th singular value is denoted as Find the sample with the same rock formation hardness grade as this sample and the closest distance as the negative near neighbor x i,- , and find the sample with a different rock formation hardness grade from this sample and the closest distance as the positive near neighbor x i,+ . Denote the relevant statistic as δ, and the relevant statistic component δ j of the j-th singular value is

[0045]

[0046] In the formula,[[]] - The distance between sample x i and x i,+ at the j-th singular value, - The distance between sample x i and x i,- at the j-th singular value; if the distance between sample x i and its positive near neighbor x i,+ at the j-th singular value is greater than the distance between sample x i and its negative near neighbor x i,- at the j-th singular value, it indicates that the j-th singular value is beneficial for distinguishing different rock formation hardness grades. The relevant statistic component δ j of the j-th singular value will increase as the distance difference increases, and the classification ability of the j-th singular value to distinguish different rock formation hardness grades is stronger;

[0047] Then, according to the eigenvectors of the vibration variable V1, the eigenvectors of the torque variable V2, the eigenvectors of the rotational speed variable V3, the eigenvectors of the rotational speed variable V4, and the eigenvectors of the temperature variable V5, generate high-order feature cross terms, and use the high-order feature cross terms as the eigenvectors of the new variable V6 to further enhance the model's ability to capture complex feature relationships. The high-order feature cross terms are obtained by weighted summation of the eigenvectors of any variable and the eigenvectors of another variable.[[]]

[0048]

[0049] In the formula,[[]] - High-order feature cross terms,[[]] - The eigenvector of any variable,[[]] - The eigenvector of another variable, t - weight;[[]]

[0050] The hardness grades adopt fifteen grades of I, II, III, IIIa, IV, IVa, V, Va, VI, VIa, VII, VIIa, VIII, IX, and X classified according to the Protodyakonov coefficient. Only the intelligent continuous mining in the relatively hard rock strata of grades II - Va is considered. Through on-site tests, the time-series data of the vibration variable V1 of the pick drum 51, the time-series data of the torque variable V2 of the rotating shaft 52 of the pick drum, the time-series data of the rotational speed variable V3 of the input shaft 61 of the transfer case 6, the time-series data of the rotational speed variable V4 of the output shaft 62 of the cutting part of the transfer case 6, the time-series data of the temperature variable V5 of the housing of the hydraulic torque converter 8, the power of the engine 9, the gear position of the gearbox 7, the downward pressure of the main boom 4, the mining speed, the energy consumption, and the equipment vibration data are obtained; the equipment vibration data is obtained through vibration sensors installed in the cab of the vehicle frame 2. Calculate the Pearson correlation coefficients between the vibration data of the pick drum 51, the torque data of the rotating shaft 52 of the pick drum, the rotational speed data of the input shaft 61 of the transfer case 6, the rotational speed data of the output shaft 62 of the cutting part of the transfer case 6, the temperature data of the housing of the hydraulic torque converter 8, and the hardness grades, which are denoted as ρ1, ρ2, ρ3, ρ4, ρ5 respectively. Then, the importance ratio of the time-series data of each variable to the hardness recognition result is,

[0051]

[0052] In the formula, I i - The importance of the time-series data of the i-th variable V i to the hardness recognition result; further determine the weight t according to the importance,

[0053]

[0054] In the formula, I b - The importance of the time-series data of any variable to the hardness recognition result, I c - The importance of the time-series data of another variable to the hardness recognition result;

[0055] After normalizing the eigenvectors of the vibration variable V1, the torque variable V2, the rotational speed variable V3, the rotational speed variable V4, the temperature variable V5, and the eigenvector of the new variable V6, use them as the overall input parameter X = [f′ i,j , input it into the improved XGBoost model, and then perform the rock stratum hardness recognition based on the improved XGBoost; the normalization process uses Z-score normalization,

[0056]

[0057] In the formula, f′ i,j - The j-th variable V jThe i-th value in the eigenvector, μ j - The j-th variable V j The mean of the eigenvector, σ j - The j-th variable V j The standard deviation of the eigenvector;

[0058] The rock stratum hardness identification based on the improved XGBoost includes using the Bayesian optimization algorithm and combining with the Gaussian process surrogate model to optimize the key hyperparameters of XGBoost. XGBoost has multiple hyperparameters, and the common key hyperparameters include the learning rate η, the number of trees n es , the maximum depth d max , the minimum subsample weight sum cw min , the column sampling ratio cs b , the regularization parameters λ, α. Define the hyperparameter space according to the maximum and minimum values of each key hyperparameter,

[0059] Θ = {η ∈ (0.01, 0.3), n es ∈ [100, 1000], d max ∈ [3, 10],...} (6)

[0060] Select several initial hyperparameter combinations {θ1, θ2, …, θ m} and train the XGBoost model, record the cross-validation score g(θ i ), use the current hyperparameter combination and its corresponding score g(θ i ) to construct the Gaussian process surrogate model, and select the next hyperparameter combination θ m+1 by optimizing the EI acquisition function. In the framework of the Gaussian process surrogate model, EI can be specifically expressed as,

[0061] EI(θ) = (μ(θ) - g(θ + ))Φ(Z) + σ(θ)φ(Z) (7)

[0062]

[0063] where, μ(θ) - the predicted mean of the Gaussian process at θ, σ(θ) - the predicted standard deviation of the Gaussian process at θ, g(θ + ) - the currently observed best cross-validation score, Φ(Z) - the cumulative distribution function of the standard normal distribution, φ(Z) - the probability density function of the standard normal distribution; select the hyperparameter combination θ m+1 that maximizes EI,

[0064]

[0065] At θ m+1Train the model, record the cross - validation score, update the Gaussian process surrogate model, and repeat the above process until the approximate optimal hyperparameter combination θ is found. * ;

[0066] Then, based on the approximate optimal parameters found by Bayesian optimization, perform a grid search to fine - tune the key hyperparameters. Based on the results of Bayesian optimization, set the fine - tuning range for each hyperparameter to obtain the fine - tuned hyperparameter space Θ. fine , set the step size for each hyperparameter according to requirements, evaluate each hyperparameter combination within the fine - tuning range, and select the hyperparameter combination θ that gives the best cross - validation score. fine ;

[0067]

[0068] Use the hyperparameter combination θ after Bayesian optimization and grid - search fine - tuning. fine , train the XGBoost model.

[0069] F(X)=XGBoost(X;θ fine ) (11)

[0070] In the formula, F(X) - prediction score function. Apply the softmax function to the prediction score to obtain the probability distribution of each hardness level, and select the hardness level with the highest probability as the hardness identification result of the rock formation.

[0071] S103. Automatic adjustment of working parameters;

[0072] The parameter adjustment decision module built in the control center automatically adjusts the power of the engine 8, the gear of the transmission 7, and the downward pressure of the main boom 4 to the most suitable ones for the current rock formation hardness level according to the hardness identification result. Obtain the power of the engine 8, the gear of the transmission 7, the downward pressure of the main boom 4, the excavation speed, the energy consumption, and the equipment vibration data under different hardness levels of II - Va through on - site tests. Use the particle swarm optimization algorithm to find the most suitable parameter combination for optimizing the performance index under specific hardness level conditions. Under the same hardness level conditions, take the power of the engine 8, the gear of the transmission 7, and the downward pressure of the main boom 4 corresponding to fast excavation speed, low energy consumption, and small equipment vibration as the most suitable evaluation criteria. Then the established objective function is

[0073] obj = β(1 - S)+γE+ξA (12)

[0074] In the formula, obj - objective function. The smaller this value is, the more suitable the parameter combination is. β, γ, ξ - weight coefficients, reflecting the importance of each index, and the sum of the three is 1. In this embodiment, take β = 0.5, γ = 0.3, ξ = 0.2 in the order of importance of excavation speed, energy consumption, and equipment vibration.

[0075] S104, Self-learning;

[0076] Record the timing data and hardness recognition results collected each time. When the hardness recognition result does not match the actual situation and leads to the suspension of mining, the judgment method for the suspension of mining is that the output power of the engine 9 remains basically unchanged, the oil temperature of the hydraulic torque converter 8 suddenly rises abnormally, and causes the temperature of the outer shell of the hydraulic torque converter 8 to rise suddenly. The value of the output shaft speed sensor of the cutting part of the transfer case 6 decreases rapidly or even decreases to 0. Then manually adjust the power of the engine 8, the gear of the gearbox 7, and the downward pressure of the main boom 4. When the mining continues after adjustment, record the power of the engine 8, the gear of the gearbox 7, and the downward pressure of the main boom 4 after adjustment. According to the power of the engine 8, the gear of the gearbox 7, and the downward pressure of the main boom 4 after adjustment, evaluate the correct hardness recognition result, and then adopt the sliding window technology to regularly introduce new data and eliminate old data to fine-tune and improve the XGBoost model.

[0077] Embodiment 2, A non-explosive continuous hard rock mining intelligent equipment, including a frame 2 provided on a walking track assembly 1, a main boom 4 provided at the front end of the frame 2, a cutting part 5 located at the front end of the main boom 4, an engine 9, a gearbox 7, an electronic control system provided at the rear end of the frame 2, and a transmission system provided on the entire frame 2. A hydraulic torque converter 8 is provided between the engine 9 and the gearbox 7. A transfer case 6 is provided between the cutting part 5 and the gearbox 7. Power is transmitted between the transfer case 6 and the gearbox 7 through the input shaft 61 of the transfer case. Power is transmitted between the transfer case 6 and the cutting part 5 through the cutting part output shaft 62 of the transfer case 6 and the driven shaft 54 of the cutting part 5. The cutting part 5 includes two pick drums 51, a rotating shaft 52 provided at the central axis of the two pick drums 51, and a transmission cutting chain 53 provided between the two pick drums 51. One end of the transmission cutting chain 53 is connected to the rotating shaft 52, and the other end is connected to the driven shaft 54 of the cutting part 5. Picks are welded on both the pick drum 51 and the transmission cutting chain 53. Driven by the engine 9, the pick drum 51 is driven to rotate through the transmission system by the transmission cutting chain 53. A torque sensor is provided on the rotating shaft 52 of the pick drum 51 for driving, speed sensors are provided on the input shaft 61 of the transfer case 6 and the cutting part output shaft 62 of the transfer case 6, and a temperature sensor is provided on the outer shell of the hydraulic torque converter 8. An electronic control board integrating a control center and a server for locally deploying and improving the XGBoost model are added to the electronic control system.

[0078] In this embodiment, the control center receives the timing data collected by the sensor, performs feature extraction and dimensionality reduction processing on the feature data, then inputs it into the improved XGBoost model for rock hardness recognition, and then sends instructions to adjust the power of the engine 9, the gear of the gearbox 7, and the downward pressure of the main boom 4 according to the hardness recognition result.

[0079] In this embodiment, the control center records all the collected timing data, hardness identification results, and the adjusted power of the engine 9, the gear of the transmission 7, and the downward pressure of the main boom 4 in the storage device. When the capacity of the storage device is insufficient, a capacity warning is automatically issued to remind the maintenance personnel to replace the storage device in a timely manner.

[0080] In summary, the hard rock non-explosive continuous mining method and intelligent equipment of the present invention have the characteristics of high efficiency and smoothness in the mining process, reducing equipment wear in the field of mining machinery and equipment, and can intelligently adapt to the mining environment with complex and variable rock formations.

[0081] It should be understood that the above embodiments are one or more embodiments of the present invention. Based on the present invention, there are many other embodiments and their variations; when ordinary technicians in this industry do not make pioneering innovations, the variations and modifications made through the present invention all fall within the protection scope of the present invention.

Claims

1. A method for continuous non-explosive mining of hard rock, characterized in that The specific steps include: S101, sensor deployment; The sensor deployment includes arranging a vibration sensor in the cutting tooth drum of the cutting part, arranging a torque sensor on the rotating shaft of the driving cutting tooth drum, arranging a first speed sensor on the input shaft of the transfer case, arranging a second speed sensor on the cutting part output shaft of the transfer case, and arranging a temperature sensor on the torque converter housing; the vibration sensor, the torque sensor, the first speed sensor, the second speed sensor, and the temperature sensor respectively collect time series data of the vibration variable V1, the torque variable V2, the speed variable V3, the speed variable V4, and the temperature variable V5 according to their respective set frequencies, and transmit them to the control center; S102, data processing; The data processing includes, after the control center receives the time series data, performing feature extraction based on wavelet packet singular value decomposition and feature data dimension reduction based on the Relief algorithm on the time series data of the vibration variable V1, the torque variable V2, the speed variable V3, the speed variable V4, and the temperature variable V5, and then generating high-order feature cross terms according to the feature vector of the vibration variable V1, the feature vector of the torque variable V2, the feature vector of the speed variable V3, the feature vector of the speed variable V4, and the feature vector of the temperature variable V5, using the high-order feature cross terms as the feature vector of the new variable V6, and standardizing the feature vector of the vibration variable V1, the feature vector of the torque variable V2, the feature vector of the speed variable V3, the feature vector of the speed variable V4, the feature vector of the temperature variable V5, and the feature vector of the new variable V6 as the overall input parameters, and inputting them into the improved XGBoost model, and then performing rock formation hardness identification based on the improved XGBoost; The feature extraction based on wavelet packet singular value decomposition includes using a wavelet packet algorithm to decompose the time series data layer by layer, decomposing the time series data into n nodes according to a preset number of decomposition layers, and obtaining the wavelet packet coefficient of each node; reconstructing the n nodes to obtain the reconstructed time series data of each node, and combining the reconstructed time series data of the n nodes to obtain a time-frequency matrix; The feature data dimensionality reduction based on the Relief algorithm includes performing singular value decomposition on the time-frequency matrix, using the Relief algorithm to measure the classification ability of each order singular value through the relevant statistics corresponding to each order singular value, selecting the first k order singular values ​​of the relevant statistics from large to small, and using the k order singular values ​​as the feature vectors of the time series data; The rock formation hardness identification based on improved XGBoost includes using the Bayesian optimization algorithm in combination with the Gaussian process proxy model to optimize the key hyperparameters of XGBoost, and then performing a grid search based on the approximate optimal parameters found by Bayesian optimization to fine-tune the key hyperparameters; the improved XGBoost model obtains the probability distribution of each hardness level according to the overall input parameters, and selects the hardness level with the highest probability as the hardness identification result of the rock formation; S103, automatic adjustment of working parameters; The automatic adjustment of the working parameters includes an improved XGBoost model trained and deployed locally in the hard rock non-explosive continuous mining intelligent equipment, and a parameter adjustment decision module built into the control center automatically adjusts the engine power, gearbox gear and main arm downforce to the most matching rock hardness level according to the hardness identification result; S104, self-learning; The self-learning includes recording each collected time series data and hardness recognition result. When the hardness recognition result does not match the actual situation and causes mining to stagnate, the engine power, gearbox gear and main arm downforce are manually adjusted. When mining continues after adjustment, the adjusted engine power, gearbox gear and main arm downforce are recorded. Based on the adjusted engine power, gearbox gear and main arm downforce, the correct hardness recognition result is evaluated, and then the sliding window technology is used to regularly introduce new data and eliminate old data to fine-tune and improve the XGBoost model.

2. The hard rock non-explosive continuous mining method according to claim 1, characterized in that: In step S102, the high-order feature cross term is obtained by weighted summing of the feature vector of any variable and the feature vector of another variable. In the formula, - high-order feature interactions, - the eigenvector of any variable, -Eigenvector of another variable, t-weight.

3. The hard rock non-explosive continuous mining method according to claim 1, characterized in that: In step S103, the engine power, gearbox position and main arm downforce that best match the current rock formation hardness grade are determined through field tests, and the best matching evaluation reference indicators are mining speed, energy consumption and equipment vibration.

4. The hard rock non-explosive continuous mining method according to claim 1, characterized in that: In step S102, the standardization process uses Z-score standardization. Where f′ i,j -jth variable V j The i-th value in the eigenvector of j -jth variable V j The mean of the eigenvectors, σ j -jth variable V j The standard deviation of the eigenvector.

5. The hard rock non-explosive continuous mining method according to claim 1, characterized in that: In step S102, the hardness grade adopts fifteen grades of I, II, III, IIIa, IV, IVa, V, Va, VI, VIa, VII, VIIa, VIII, IX, and X classified according to the Proctor coefficient.

6. The method for continuous non-explosive mining of hard rock according to claim 1, characterized in that: In step S104, the mining stagnation is judged in that the output power of the engine remains substantially unchanged, the oil temperature of the torque converter rises abnormally, and the value of the output shaft speed sensor of the cutting part of the transfer case decreases rapidly or even decreases to 0.

7. An intelligent equipment for continuous non-explosive mining of hard rock, used in the continuous non-explosive mining method of hard rock as claimed in any one of claims 1 to 6, comprising a frame arranged on a walking crawler assembly, a main arm arranged at the front end of the frame, a cutting part located at the front end of the main arm, an engine arranged at the rear end of the frame, a gearbox, an electronic control system and a transmission system arranged on the entire frame, wherein a hydraulic torque converter is arranged between the engine and the gearbox, a transfer case is arranged between the cutting part and the gearbox, and power is transmitted between the transfer case and the gearbox through an input shaft of the transfer case The transfer case and the cutting part transmit power through the cutting part output shaft of the transfer case and the driven shaft of the cutting part. The cutting part includes two pick rollers, a rotating shaft arranged at the central axis of the two pick rollers, and a transmission cutting chain arranged between the two pick rollers. One end of the transmission cutting chain is connected to the rotating shaft, and the other end is connected to the driven shaft of the cutting part. The pick roller and the transmission cutting chain are both welded with picks, which are driven by the engine and drive the pick roller to rotate through the transmission system and the transmission cutting chain. The characteristics are as follows: A torque sensor is provided on the rotating shaft of the driving cutter drum, a speed sensor is provided on the input shaft of the transfer case and the output shaft of the cutting part of the transfer case, and a temperature sensor is provided on the torque converter housing. An electronic control board integrating a control center and a server for locally deploying an improved XGBoost model are added to the electronic control system.

8. The intelligent equipment for non-explosive continuous mining of hard rock according to claim 7 is characterized by: The control center receives the time series data collected by the sensor and performs feature extraction and dimensionality reduction processing on the feature data, and then inputs it into the improved XGBoost model to identify the hardness of the rock formation. Then, based on the hardness identification result, the control center sends instructions to adjust the engine power, gearbox gear and main arm downforce.

9. The intelligent equipment for non-explosive continuous mining of hard rock according to claim 7 is characterized in that: The control center records all collected time series data, hardness identification results, and adjusted engine power, transmission gear position, and main arm downforce in a storage device. When the storage device capacity is insufficient, a capacity warning is automatically issued to remind maintenance personnel to replace the storage device in time.

Images