Intelligent electro-hydraulic mining equipment control method and system

Abstract

The application belongs to the technical field of equipment control, and discloses an intelligent electro-hydraulic mining equipment control method and system, which comprises the following steps: obtaining historical ore body mining data and performing screening to construct a learning data set; training an equipment parameter optimization model using the data set, the model being capable of predicting an optimal control parameter sequence according to ore body characteristics and equipment state data; in the actual mining process, ore body characteristics and equipment state data are collected in real time and input into the optimization model to obtain a control parameter sequence; and the hydraulic pump rotating speed, the main circuit pressure set value and the unloading valve opening degree are taken as target control sequences to control the electro-hydraulic mining equipment to work. The method realizes dynamic optimization of the control parameters through the data-driven intelligent control strategy, enables the equipment to timely adjust key parameters according to the ore body characteristic changes and state fluctuations, significantly improves the mining efficiency and the equipment adaptability, prolongs the service life of the equipment and reduces the maintenance cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device control, more particularly, the present application relates to an intelligent electro-hydraulic mining equipment control method and system. BACKGROUND

[0002] Electro-hydraulic mining equipment is a kind of mining equipment that combines electrical control system and hydraulic actuator, which is widely used in mining, tunnel, quarry and other fields. Through the coordinated work of electro-hydraulic system, the efficient operation of mining, crushing, conveying and other processes is realized. As a modern mining method, electro-hydraulic mining has become an important technical means for mineral resource development.

[0003] Although electro-hydraulic mining equipment has many advantages, the existing electro-hydraulic mining equipment operation control still faces some challenges and limitations. The traditional electro-hydraulic mining equipment control system is relatively extensive in parameter adjustment, often relying on the experience of operators for setting, which is difficult to accurately adapt to the changes of different ore body characteristics and mining environment. This kind of control mode dominated by human factors is easy to lead to low energy utilization efficiency, aggravate equipment wear and tear, and even cause safety hazards. At the same time, in the actual mining process, there are often great differences between ore bodies, and the existing control system mostly uses fixed parameter setting, which cannot realize real-time response and adjustment to these differences, leading to unstable mining efficiency, increased equipment failure rate and rising maintenance cost. SUMMARY

[0004] In order to overcome the problems of unstable mining efficiency, increased equipment failure rate and rising maintenance cost caused by the prior art, the present application provides an intelligent electro-hydraulic mining equipment control method and system to solve the above problems.

[0005] The present application provides the following technical solutions:

[0006] An intelligent electro-hydraulic mining equipment control method, comprising:

[0007] The present application also provides an intelligent electro-hydraulic mining equipment control system for realizing an intelligent electro-hydraulic mining equipment control method, comprising:

[0008] Obtain historical ore body mining data and filter to obtain ore body mining learning data set, the ore body mining data includes ore body characteristics, equipment state data and control parameter sequence, the control parameter sequence is composed of control parameters collected at a preset frequency in the mining process, the control parameters include hydraulic pump speed, main circuit pressure set value and unloading valve opening degree;

[0009] Train the equipment parameter optimization model using the ore body mining learning data set, the equipment parameter optimization model is used to obtain the control parameter sequence according to the ore body characteristics and the equipment state data;

[0010] obtain real-time device state data and ore body characteristics, and input the real-time device state data and ore body characteristics into the device parameter optimization model to obtain a control parameter sequence;

[0011] obtain a target control sequence by taking the hydraulic pump rotating speed in the obtained control parameter sequence as a target rotating speed, taking the main circuit pressure set value as a target pressure, and taking the unloading valve opening degree as a target opening degree; and control the electro-hydraulic mining equipment to perform mining operations according to the target control sequence.

[0012] Preferably, the obtaining of historical ore body mining data and the screening to obtain the ore body mining learning data set comprise:

[0013] For any historical ore body mining data, obtaining device state data in the ore body mining data;

[0014] determining whether the ore body mining data can be used as reference data according to the device state data;

[0015] screening out the ore body mining data with a determination result of yes to form the ore body mining learning data set.

[0016] Preferably, the ore body characteristics include target ore hardness and content, main impurity hardness and content, and hardest impurity hardness and content, and the device state data include hydraulic oil temperature change rate, maximum hydraulic oil temperature, system pressure fluctuation rate, and device vibration intensity.

[0017] The hydraulic oil temperature change rate is calculated by using a linear regression method based on all the hydraulic oil temperature data collected at a preset frequency during the mining process; the maximum hydraulic oil temperature is the maximum value of all the collected hydraulic oil temperature data; the system pressure fluctuation rate is calculated by taking the ratio of the standard deviation to the average value of the system pressure data collected at a preset frequency during the mining process; and the device vibration intensity is obtained by calculating the root mean square value of the vibration frequency data collected at a preset frequency during the mining process.

[0018] Preferably, the determining whether the ore body mining data can be used as reference data according to the device state data comprises:

[0019] determining whether the hydraulic oil temperature change rate is less than a preset hydraulic oil temperature change rate threshold value;

[0020] determining whether the maximum hydraulic oil temperature is less than a preset maximum hydraulic oil temperature threshold value;

[0021] determining whether the system pressure fluctuation rate is less than a preset system pressure fluctuation rate threshold value;

[0022] determining whether the device vibration intensity is less than a preset device vibration intensity threshold value;

[0023] If the hydraulic oil temperature change rate, the maximum hydraulic oil temperature, the system pressure fluctuation rate, and the equipment vibration intensity are all less than the corresponding threshold values, it is determined that the ore body mining data can be used as reference data; otherwise, it is determined that the ore body mining data cannot be used as reference data.

[0024] Preferably, the training of the equipment parameter optimization model using the ore body mining learning data set comprises:

[0025] An ore body mining learning data set is obtained, and the data is preprocessed, including anomaly value detection and processing, feature normalization, and standardization.

[0026] The data in the data set is randomly divided into a training set and a test set.

[0027] A hardness comprehensive index is calculated according to the ore body characteristics, and a stability index is calculated according to the equipment state data.

[0028] The hardness comprehensive index and the stability index are used as main features, combined with the ore body characteristics and the equipment state data features, to construct a feature matrix.

[0029] A gradient boosting decision tree model is built, and learning rate, maximum depth of tree, and minimum number of split samples, etc. are set as hyperparameters.

[0030] The feature matrix is used as an input variable, and the control parameter sequence is used as an output variable; the model is trained using the training set data, with root mean square error as the loss function, and the hyperparameters are optimized using the cross-validation method.

[0031] The test set is used to evaluate the model performance, and when the model accuracy reaches a preset threshold, the model is saved as an equipment parameter optimization model; if the accuracy does not reach the threshold, the hyperparameters are adjusted and retrained until the model accuracy reaches the preset threshold.

[0032] Preferably, the acquisition of real-time equipment state data and ore body characteristics comprises:

[0033] Before the mining operation, samples of the ore body to be mined are collected, and the hardness and content of the target ore are obtained through mineral analysis technology; the main impurities in the samples are identified through mineral composition analysis technology, and their hardness and content are measured; the impurities with the highest hardness in the samples are identified through mineral hardness testing technology, and their hardness and content are measured; and the ore body characteristics are obtained by combination.

[0034] The equipment is started for low-load trial operation, and hydraulic oil temperature data, system pressure data, and equipment vibration data are collected at a preset frequency; and the current equipment state data is calculated according to the collected data.

[0035] Preferably, the input of real-time equipment state data and ore body characteristics into the equipment parameter optimization model comprises:

[0036] preprocessing the real-time orebody property and equipment state data;

[0037] calculating an extraction hardness comprehensive index according to the orebody property data and a stability index according to the equipment state data;

[0038] combining the hardness comprehensive index and the stability index as main features with orebody property and equipment state data features to construct a feature matrix;

[0039] inputting the feature matrix into the equipment parameter optimization model.

[0040] Preferably, the mining operation controlled by the target control sequence includes:

[0041] dividing the target control sequence into multiple control periods in chronological order, each control period corresponding to a group of control parameter values;

[0042] setting a hydraulic pump speed safety threshold, a main circuit pressure safety threshold, and an unloading valve opening safety threshold;

[0043] for each control period, comparing the corresponding target speed with the hydraulic pump speed safety threshold, if the target speed exceeds the safety threshold, using the safety threshold as the actual control value, transmitting it to the hydraulic pump motor through the frequency converter to adjust the actual speed of the hydraulic pump;

[0044] comparing the corresponding target pressure with the main circuit pressure safety threshold, if the target pressure exceeds the safety threshold, using the safety threshold as the actual control value, setting the main circuit pressure value through the electro-hydraulic proportional valve controller;

[0045] comparing the corresponding target opening with the unloading valve opening safety threshold, if the target opening exceeds the safety threshold, using the safety threshold as the actual control value, adjusting the opening of the unloading valve through the electro-hydraulic proportional control module.

[0046] a historical data screening module for obtaining historical orebody mining data and screening to obtain an orebody mining learning data set, the orebody mining data including orebody properties, equipment state data, and control parameter sequences, the control parameter sequences composed of control parameters collected at a preset frequency during the mining process, the control parameters including hydraulic pump speed, main circuit pressure set value, and unloading valve opening;

[0047] a model training module for training the equipment parameter optimization model using the orebody mining learning data set, the equipment parameter optimization model for obtaining a control parameter sequence according to orebody properties and equipment state data;

[0048] The control parameter acquisition module is configured to acquire real-time device state data and ore body characteristics, and input the real-time device state data and ore body characteristics into the device parameter optimization model to obtain a control parameter sequence.

[0049] The control execution module is configured to obtain a target control sequence by taking the rotational speed of the hydraulic pump in the obtained control parameter sequence as a target rotational speed, taking the main circuit pressure set value as a target pressure, and taking the unloading valve opening degree as a target opening degree, and control the electro-hydraulic mining equipment to perform mining operations according to the target control sequence.

[0050] The present application provides an intelligent electro-hydraulic mining equipment control method and system, which has the following beneficial effects:

[0051] By systematically screening historical ore body mining data, based on key indicators such as hydraulic oil temperature change rate, maximum hydraulic oil temperature, system pressure fluctuation rate and equipment vibration intensity, unstable, abnormal or potential risk mining data can be effectively eliminated. This screening mechanism ensures the high quality of the learning data set, providing a reliable foundation for subsequent model training, thereby significantly improving the accuracy and practicality of the model. Compared with traditional control methods that rely on human experience, this learning method based on high-quality data can more objectively and comprehensively capture the relationship between ore body characteristics and control parameters.

[0052] By real-time acquisition of ore body characteristics and device state data, combined with the trained model for parameter prediction, dynamic optimization of control parameters is realized. This real-time adaptive adjustment mechanism enables the electro-hydraulic mining equipment to adjust key parameters such as hydraulic pump rotational speed, main circuit pressure set value and unloading valve opening degree in a timely manner according to changes in ore body characteristics and fluctuations in device state, thereby maintaining optimal mining efficiency. Compared with traditional fixed parameter setting control systems, this intelligent control method significantly improves the adaptability of the equipment to different ore bodies, thereby improving mining efficiency and resource utilization, prolonging the service life of the equipment and reducing maintenance costs. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 The flowchart of the intelligent electro-hydraulic mining equipment control method of the present application;

[0054] Figure 2 The module schematic diagram of the intelligent electro-hydraulic mining equipment control system of the present application. DETAILED DESCRIPTION

[0055] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0056] Embodiment 1

[0057] Please refer to Figure 1 In the embodiment, a control method of an intelligent electro-hydraulic mining equipment comprises the following steps.

[0058] S1, obtaining historical ore body mining data and screening to obtain an ore body mining learning data set, the ore body mining data comprising ore body characteristics, equipment state data and a control parameter sequence, the control parameter sequence being composed of control parameters collected at a preset frequency during a mining process, the control parameters comprising a hydraulic pump rotating speed, a main circuit pressure set value and an unloading valve opening degree;

[0059] The obtaining of the historical ore body mining data and the screening to obtain the ore body mining learning data set comprises the following steps.

[0060] For any ore body mining data in history, the equipment state data in the ore body mining data is obtained.

[0061] It is judged according to the equipment state data whether the ore body mining data can be used as reference data.

[0062] The ore body mining data with a yes judgment result is screened out to form the ore body mining learning data set.

[0063] The ore body characteristics comprise target ore hardness and content, main impurity hardness and content, and hardest impurity hardness and content, and the equipment state data comprises a hydraulic oil temperature change rate, a maximum hydraulic oil temperature, a system pressure fluctuation rate and an equipment vibration intensity.

[0064] The hydraulic oil temperature change rate is calculated by using a linear regression method based on all the hydraulic oil temperature data collected at the preset frequency during the mining process; the maximum hydraulic oil temperature is the maximum value of all the collected hydraulic oil temperature data; the system pressure fluctuation rate is calculated by using a ratio of a standard deviation to an average value of the system pressure data collected at the preset frequency during the mining process; and the equipment vibration intensity is obtained by calculating a root mean square value of the vibration frequency data collected at the preset frequency during the mining process.

[0065] The judging of whether the ore body mining data can be used as reference data according to the equipment state data comprises the following steps.

[0066] It is judged whether the hydraulic oil temperature change rate is less than a preset hydraulic oil temperature change rate threshold value.

[0067] determining whether the maximum hydraulic oil temperature is less than a preset maximum hydraulic oil temperature threshold value;

[0068] determining whether the system pressure fluctuation rate is less than a preset system pressure fluctuation rate threshold value;

[0069] determining whether the equipment vibration intensity is less than a preset equipment vibration intensity threshold value;

[0070] If the hydraulic oil temperature change rate, the maximum hydraulic oil temperature, the system pressure fluctuation rate and the equipment vibration intensity are all less than the respective threshold values, it is determined that the ore body mining data can be used as reference data; otherwise, it is determined that the ore body mining data cannot be used as reference data.

[0071] In the embodiment, the process of obtaining historical ore body mining data and screening to obtain the ore body mining learning data set can be as follows: First, export the historical ore body mining data records from the enterprise mine data management system. Each set of data contains three categories of information: ore body characteristics, equipment state data and control parameter sequence. The ore body characteristics data includes target ore hardness and content, main impurity hardness and content, and the hardest impurity hardness and content. These data can be obtained by a mineral hardness tester and a composition analysis device. The control parameter sequence is a control parameter group collected at a preset frequency (such as every 5 seconds) during the mining process, including hydraulic pump speed, main circuit pressure set value and unloading valve opening degree.

[0072] For each set of historical ore body mining data, four key equipment state indicators need to be calculated, which can be obtained by analyzing the sequence data composed of various parameters collected at a preset frequency (such as every 5 seconds). The calculation of the hydraulic oil temperature change rate is from the hydraulic oil temperature time sequence data, and the temperature change rate is calculated by using the linear regression method. The maximum hydraulic oil temperature is the maximum value found from the hydraulic oil temperature sequence data. The system pressure fluctuation rate is calculated by the ratio of the standard deviation to the average value of the system pressure time sequence data. This indicator reflects the stability of the system pressure, and a lower fluctuation rate indicates that the equipment runs smoothly, which is beneficial to prolong the service life of the equipment and improve the mining efficiency. The equipment vibration intensity is obtained by calculating the root mean square value of the vibration frequency sequence data. This indicator can comprehensively reflect the energy size of the equipment vibration at each frequency, and is a key parameter for evaluating the equipment running state. Excessive vibration intensity often means that the equipment has hidden faults or poor working condition.

[0073] In the reference data screening process, reasonable screening thresholds can be preset according to equipment manufacturer recommendations or industry safety standards, including hydraulic oil temperature change rate threshold, maximum hydraulic oil temperature threshold, system pressure fluctuation rate threshold, and equipment vibration intensity threshold. For each set of historical ore body mining data, determine whether each indicator is less than the corresponding preset threshold. If the above four indicators are all less than the respective threshold, mark the data set as qualified and can be used as reference data; otherwise, mark it as unqualified.

[0074] This screening method can effectively eliminate unstable, abnormal or potentially risky mining data, ensuring the quality of the learning data set. By excluding those working condition data that may cause equipment damage or low efficiency, the accuracy and practicality of the subsequent model can be significantly improved. After screening, all the data marked as qualified are combined to form the ore body mining learning data set. These data have good equipment operating state characteristics and can provide high-quality samples for subsequent model training. Through this scientific data screening and learning data set construction method, the intelligent electro-hydraulic mining equipment control system can learn based on high-quality historical experience data, thereby improving the adaptive control ability and mining efficiency of the equipment, and reducing the equipment failure rate and maintenance cost.

[0075] S2, training an equipment parameter optimization model using the ore body mining learning data set, the equipment parameter optimization model being used to obtain a control parameter sequence according to the ore body characteristics and the equipment state data;

[0076] Training the equipment parameter optimization model using the ore body mining learning data set includes:

[0077] Obtaining the ore body mining learning data set, and preprocessing the data, including outlier detection and processing, feature normalization and standardization;

[0078] Randomly dividing the data in the data set into a training set and a test set;

[0079] Calculating and extracting a hardness comprehensive index according to the ore body characteristics, and calculating a stability index according to the equipment state data;

[0080] Combining the hardness comprehensive index and the stability index as main features with the ore body characteristics and the equipment state data features to construct a feature matrix;

[0081] Building a gradient boosting decision tree model, and setting learning rate, maximum depth of tree, and minimum number of split samples, etc. as hyperparameters;

[0082] Taking the feature matrix as an input variable and the control parameter sequence as an output variable; training the model using the training set data, taking the root mean square error as the loss function, and using the cross-validation method to optimize the hyperparameters;

[0083] The model performance is evaluated using the test set, and when the model accuracy reaches the preset threshold, the model is saved as a device parameter optimization model; if the accuracy does not reach the threshold, the hyperparameters are adjusted and retrained until the model accuracy reaches the preset threshold.

[0084] In this embodiment, the process of training the device parameter optimization model using the ore body mining learning data set can be as follows: first, the screened ore body mining learning data set is obtained, and the data is preprocessed to improve the effect and efficiency of subsequent model training. The preprocessing steps include outlier detection and processing, feature normalization and standardization. Outlier detection can use statistical methods such as box plots or Z-score to identify data points that deviate from the normal distribution range. For detected outliers, you can choose to delete or replace them with median values. Feature normalization and standardization are to convert features of different dimensions to the same scale, which can use Min-Max normalization or Z-score standardization methods, which helps to improve the convergence speed and stability of the model.

[0085] After preprocessing, the data in the data set is randomly divided into a training set and a test set according to an 8:2 ratio. The training set is used for the learning process of the model, while the test set is used to evaluate the generalization ability of the model. In order to make full use of the information of the ore body characteristics and device state data, it is necessary to construct more representative features. According to the ore body characteristics, the hardness comprehensive index is calculated and extracted, which can use the weighted average method to calculate the target ore hardness, the main impurity hardness and the hardest impurity hardness according to their content proportion, to obtain an index that can comprehensively reflect the overall hardness characteristics of the ore body. Similarly, according to the device state data, the stability index is calculated, which can be weighted and synthesized after normalization processing of the hydraulic oil temperature change rate, the highest hydraulic oil temperature, the system pressure fluctuation rate and the device vibration intensity, to form an index that can reflect the overall stability of the device.

[0086] Next, the hardness comprehensive index and the stability index are used as the main features, combined with the original ore body characteristics and device state data features to construct a feature matrix. This feature engineering method can preserve the original data information while introducing higher-level comprehensive features, which helps the model better capture the complex relationship between ore body characteristics and device state and control parameters.

[0087] In terms of model selection, this embodiment adopts gradient boosting decision tree models such as XGBoost or LightGBM, which perform well in handling complex nonlinear relationships and mixed-type features. During model building, key hyperparameters need to be set, including learning rate (usually set in the range of 0.01-0.1), maximum tree depth (can be set to 3-10), and minimum number of split samples (can be set to 5-20). These parameters directly affect the complexity and generalization ability of the model, and need to be determined through subsequent optimization process to find the best value.

[0088] The feature matrix is used as the input variable, and the control parameter sequence is used as the output variable. The model is trained using the training set data. During training, the root mean square error (RMSE) is used as the loss function, which can effectively measure the deviation between the predicted value and the actual value. To determine the hyperparameter combination, the K-fold cross-validation method (such as K=5) can be used, dividing the training set into K subsets, using K-1 subsets for training each time, and the remaining 1 subset for validation. After K cycles, the average performance is taken as the evaluation result of this group of hyperparameters. Through grid search or Bayesian optimization, the hyperparameter space can be systematically explored to find the best hyperparameter combination.

[0089] After training, the model performance is evaluated using the test set. When the accuracy of the model on the test set (measured by the coefficient of determination) reaches the preset standard, the model is saved as the final device parameter optimization model. If the model performance does not meet expectations, the hyperparameters need to be adjusted and retrained, or additional features, additional training samples, or other model structures can be considered until the model accuracy reaches the preset threshold. Through this iterative optimization method, the final device parameter optimization model can have good prediction ability, accurately predicting the optimal control parameter sequence based on the input ore body characteristics and device state data, providing reliable decision support for intelligent electro-hydraulic mining equipment.

[0090] S3, obtaining real-time device state data and ore body characteristics, and inputting the real-time device state data and ore body characteristics into the device parameter optimization model to obtain a control parameter sequence;

[0091] Obtaining real-time device state data and ore body characteristics includes:

[0092] Before mining operations, samples of the ore body to be mined are collected, and the hardness and content of the target ore are obtained through mineral analysis technology; the main impurities in the sample are identified through mineral composition analysis technology, and their hardness and content are measured; the impurities with the highest hardness in the sample are identified through mineral hardness testing technology, and their hardness and content are measured; and the ore body characteristics are obtained by combining them.

[0093] The device is started for low-load commissioning, and hydraulic oil temperature data, system pressure data and device vibration data are collected at a preset frequency; according to the collected data, the current device state data is calculated.

[0094] Inputting real-time device state data and ore body characteristics into the device parameter optimization model includes:

[0095] Pretreating real-time ore body characteristics and device state data;

[0096] Calculating a hardness comprehensive index according to the ore body characteristic data and a stability index according to the device state data;

[0097] Taking the hardness comprehensive index and the stability index as main features, combining with the ore body characteristic and device state data features to construct a feature matrix;

[0098] Inputting the feature matrix into the device parameter optimization model.

[0099] In the present embodiment, the process of obtaining real-time device state data and ore body characteristics can be as follows: before actual mining operation, first, samples of the to-be-mined ore body need to be collected for analysis. A drilling device can be used to collect multiple ore body samples in the to-be-mined area according to a grid distribution manner, to ensure the representativeness of the samples. The collected samples are sent to a field mobile laboratory or a mine analysis center for rapid analysis. The target ore hardness and content data are obtained through the foregoing mineral analysis technology, the hardness and content of the main impurities and the hardest impurities are identified, and a complete ore body characteristic description is formed.

[0100] After obtaining the ore body characteristics, device state data collection is needed. The electro-hydraulic mining device is started for low-load commissioning. The purpose of this step is to evaluate the initial running state of the device without causing excessive burden to the device. The low-load commissioning is usually set to 30% to 50% of the rated load, and the duration is 15 to 30 minutes, to ensure that the device systems are fully preheated and reach a stable working state. During the commissioning, through the sensor network installed on the device, hydraulic oil temperature data, system pressure data and device vibration data are collected in real time at a preset frequency (such as once per second). The hydraulic oil temperature can be collected by temperature sensors installed at key points of the hydraulic system; the system pressure data are obtained by pressure sensors installed on the main circuit and key branches; and the device vibration data are collected by acceleration sensors fixed on key structural parts of the device.

[0101] According to the data collected during the commissioning, the state data of the current device are calculated according to the foregoing method, including the hydraulic oil temperature variation rate, the maximum hydraulic oil temperature, the system pressure fluctuation rate and the device vibration intensity. These real-time calculated device state indicators can reflect the current working condition of the device, and provide basic data for subsequent parameter optimization.

[0102] Next, the real-time acquired ore body characteristics and equipment state data are input into the trained device parameter optimization model. First, the real-time data need to be preprocessed in the same way as in the training stage, including outlier checking, feature normalization and standardization, etc. After preprocessing, the hardness comprehensive index is calculated according to the ore body characteristics data, and the stability index is calculated according to the equipment state data, using the same calculation method as in the training stage to ensure consistency in data processing.

[0103] The calculated hardness comprehensive index and stability index are used as main features, combined with the preprocessed ore body characteristics and equipment state data features, and a feature matrix is constructed according to the feature arrangement order determined during model training. Finally, the constructed feature matrix is input into the device parameter optimization model, and the model will predict the optimal control parameter sequence according to the input feature data. This real-time data-based model prediction method can enable the electro-hydraulic mining equipment to adaptively adjust the control parameters according to the current ore body characteristics and equipment state, realize intelligent mining operation, and improve mining efficiency.

[0104] S4, the rotational speed of the hydraulic pump in the obtained control parameter sequence is used as the target rotational speed, the main circuit pressure set value is used as the target pressure, and the unloading valve opening is used as the target opening to obtain a target control sequence; the electro-hydraulic mining equipment is controlled according to the target control sequence to perform mining operation.

[0105] Controlling the electro-hydraulic mining equipment to perform mining operation according to the target control sequence includes:

[0106] The target control sequence is divided into multiple control periods in time sequence, and each control period corresponds to a group of control parameter values;

[0107] Set the hydraulic pump rotational speed safety threshold, the main circuit pressure safety threshold and the unloading valve opening safety threshold;

[0108] For each control period, compare the corresponding target rotational speed with the hydraulic pump rotational speed safety threshold, if the target rotational speed exceeds the safety threshold, use the safety threshold as the actual control value, pass it to the hydraulic pump motor through the frequency converter, and adjust the actual rotational speed of the hydraulic pump;

[0109] Compare the corresponding target pressure with the main circuit pressure safety threshold, if the target pressure exceeds the safety threshold, use the safety threshold as the actual control value, set the main circuit pressure value through the electro-hydraulic proportional valve controller;

[0110] Compare the corresponding target opening with the unloading valve opening safety threshold, if the target opening exceeds the safety threshold, use the safety threshold as the actual control value, adjust the opening of the unloading valve through the electro-hydraulic proportional control module.

[0111] Embodiment 2

[0112] Referring to Figure 2 The present application provides an intelligent electro-hydraulic mining equipment control system for implementing an intelligent electro-hydraulic mining equipment control method, comprising:

[0113] A historical data screening module is configured to obtain historical ore body mining data and screen the ore body mining data to obtain an ore body mining learning data set. The ore body mining data includes ore body characteristics, equipment state data, and a control parameter sequence. The control parameter sequence is composed of control parameters collected at a preset frequency during the mining process. The control parameters include hydraulic pump speed, main circuit pressure set value, and unloading valve opening degree.

[0114] A model training module is configured to train an equipment parameter optimization model using the ore body mining learning data set. The equipment parameter optimization model is configured to obtain a control parameter sequence based on ore body characteristics and equipment state data.

[0115] A control parameter acquisition module is configured to obtain real-time ore body characteristics and equipment state data of a to-be-mined ore body, and input the real-time ore body characteristics and equipment state data into the equipment parameter optimization model to obtain a control parameter sequence.

[0116] A control execution module is configured to obtain a target control sequence by setting the hydraulic pump speed in the obtained control parameter sequence as a target speed, the main circuit pressure set value as a target pressure, and the unloading valve opening degree as a target opening degree. The control execution module is further configured to control the electro-hydraulic mining equipment to perform mining operations according to the target control sequence.

[0117] In this embodiment, the hydraulic pump speed in the obtained control parameter sequence is set as the target speed, the main circuit pressure set value is set as the target pressure, and the unloading valve opening degree is set as the target opening degree to obtain a target control sequence. The process of controlling the electro-hydraulic mining equipment to perform mining operations according to the target control sequence can be as follows: first, extract three key control parameters from the control parameter sequence output by the equipment parameter optimization model: hydraulic pump speed, main circuit pressure set value, and unloading valve opening degree. Set them as target speed, target pressure, and target opening degree respectively to form a target control sequence. This target control sequence is a time series data that reflects the control parameter combination to be used at different times during the mining process.

[0118] In order to achieve precise timing control, the target control sequence needs to be divided into multiple control periods in chronological order, and each control period corresponds to a group of control parameter values. The division of control periods can be based on a preset control frequency, such as every 30 seconds as a control period, or dynamically adjusting the period length according to the changes in ore body characteristics. This segmented control method can enable the equipment to adapt to changes in ore body characteristics during the mining process, achieving more refined adaptive control.

[0119] Before actual control, the safety threshold of each control parameter needs to be set to ensure that the equipment operates within a safe range. The safety threshold of hydraulic pump speed is usually set at 85% to 95% of the rated speed, the safety threshold of main circuit pressure is set at about 90% of the system design pressure, and the safety threshold of unloading valve opening is determined according to the system design and safety requirements. The setting of these safety thresholds needs to consider various factors such as the technical specifications of the equipment manufacturer, the service life of the equipment, and the ambient temperature.

[0120] For each control period, the system needs to perform the following control process: first, compare the target speed corresponding to the current period with the safety threshold of the hydraulic pump speed. If the target speed is lower than or equal to the safety threshold, use the target speed as the actual control value directly; if the target speed exceeds the safety threshold, use the safety threshold as the actual control value to ensure safe operation of the equipment. After determining the actual control value, transmit the control command to the frequency converter through the industrial control network, and the frequency converter adjusts the output frequency according to the command to control the actual speed of the hydraulic pump motor. The adjustment of the hydraulic pump speed directly affects the system flow and is the basis for controlling the working capacity of the mining equipment.

[0121] Similarly, compare the target pressure corresponding to the current period with the safety threshold of the main circuit pressure. If the target pressure is lower than or equal to the safety threshold, use the target pressure as the actual control value directly; if the target pressure exceeds the safety threshold, use the safety threshold as the actual control value. After determining the actual control value, set the main circuit pressure value through the electro-hydraulic proportional valve controller. The control of the main circuit pressure usually adopts a closed-loop control mode, and the controller dynamically adjusts the opening of the proportional valve according to the actual pressure value fed back by the pressure sensor, so that the system pressure is stabilized near the set value. Accurate control of the main circuit pressure is crucial for improving mining efficiency and protecting equipment.

[0122] For the unloading valve opening, similarly compare the target opening corresponding to the current period with the safety threshold of the unloading valve opening. If the target opening is lower than or equal to the safety threshold, use the target opening as the actual control value directly; if the target opening exceeds the safety threshold, use the safety threshold as the actual control value. After determining the actual control value, adjust the opening of the unloading valve through the electro-hydraulic proportional control module. As a key element for system pressure regulation, the opening of the unloading valve directly affects the pressure distribution and energy utilization efficiency of the system. By accurately controlling the opening of the unloading valve, energy loss can be minimized while ensuring the working pressure of the system.

[0123] Through the model prediction-based segmented adaptive control method, the electro-hydraulic mining equipment can adjust the control parameters in real time according to the ore body characteristics and the state of the equipment, and realize intelligent mining operation. This can not only improve the mining efficiency and resource utilization rate, but also prolong the service life of the equipment and reduce the maintenance cost, thereby creating significant economic benefits for the mining enterprises. In practical application, with the accumulation of mining experience and the enrichment of data, the equipment parameter optimization model can be further updated and optimized periodically, so that the intelligent level and adaptive ability of the system can be further improved.

[0124] In several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are merely schematic, for example, the division of the units is only one, and actual implementation can have another division mode, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0125] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0126] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method of controlling an intelligent electro-hydraulic production equipment, characterized in that, The method comprises the following steps: obtaining historical ore body mining data and screening to obtain an ore body mining learning data set, wherein the ore body mining data comprises ore body characteristics, equipment state data and a control parameter sequence, the control parameter sequence is composed of control parameters collected at a preset frequency during the mining process, and the control parameters comprise a hydraulic pump rotating speed, a main circuit pressure set value and an unloading valve opening degree; the step of obtaining historical ore body mining data and screening to obtain an ore body mining learning data set comprises: for any historical ore body mining data, obtaining equipment state data in the ore body mining data; determining whether the ore body mining data can be used as reference data according to the equipment state data; screening out the ore body mining data with a determination result of yes to form the ore body mining learning data set; training a device parameter optimization model using the ore body mining learning data set, wherein the device parameter optimization model is used to obtain a control parameter sequence according to ore body characteristics and equipment state data; the step of training a device parameter optimization model using the ore body mining learning data set comprises: obtaining the ore body mining learning data set, and pre-processing the data, wherein the pre-processing comprises anomaly value detection and processing, feature normalization and standardization; randomly dividing the data in the data set into a training set and a test set; calculating a hardness comprehensive index according to the ore body characteristics, and a stability index according to the equipment state data; combining the hardness comprehensive index and the stability index as main features with the ore body characteristics and the equipment state data features to form a feature matrix; building a gradient boosting decision tree model, and setting learning rate, maximum depth of tree and minimum number of split samples and other hyperparameters; using the feature matrix as an input variable, using the control parameter sequence as an output variable, training the model using the training set data, taking root mean square error as a loss function, and optimizing the hyperparameters using a cross-validation method; evaluating the model performance using the test set, saving the model as a device parameter optimization model when the model accuracy reaches a preset threshold, and adjusting the hyperparameters and retraining until the model accuracy reaches the preset threshold if the accuracy does not reach the threshold; obtaining real-time equipment state data and ore body characteristics, and inputting the real-time equipment state data and ore body characteristics into the device parameter optimization model to obtain a control parameter sequence; using the hydraulic pump rotating speed in the obtained control parameter sequence as a target rotating speed, the main circuit pressure set value as a target pressure, and the unloading valve opening degree as a target opening degree to obtain a target control sequence; and controlling the electro-hydraulic mining equipment to perform mining operations according to the target control sequence.

2. The intelligent electro-hydraulic mining equipment control method of claim 1, wherein, The ore body characteristics comprise target ore hardness and content, main impurity hardness and content, and hardest impurity hardness and content, and the equipment state data comprises a hydraulic oil temperature change rate, a maximum hydraulic oil temperature, a system pressure fluctuation rate and a device vibration intensity. The hydraulic oil temperature change rate is calculated by linear regression method using all the hydraulic oil temperature data collected at a preset frequency during the mining process, and the maximum hydraulic oil temperature is the maximum value of all the collected hydraulic oil temperature data. The system pressure fluctuation rate is calculated by the ratio of the standard deviation to the average value of the system pressure data collected at a preset frequency during the mining process. The device vibration intensity is obtained by calculating the root mean square value of the vibration frequency data collected at a preset frequency during the mining process.

3. The intelligent electro-hydraulic mining equipment control method of claim 2, wherein, The device state data includes: determining whether the hydraulic oil temperature change rate is less than a preset hydraulic oil temperature change rate threshold value; determining whether the maximum hydraulic oil temperature is less than a preset maximum hydraulic oil temperature threshold value; determining whether the system pressure fluctuation rate is less than a preset system pressure fluctuation rate threshold value; determining whether the device vibration intensity is less than a preset device vibration intensity threshold value; If the hydraulic oil temperature change rate, the maximum hydraulic oil temperature, the system pressure fluctuation rate and the device vibration intensity are all less than the corresponding threshold values, it is determined that the ore body mining data can be used as reference data; otherwise, it is determined that the ore body mining data cannot be used as reference data.

4. The intelligent electro-hydraulic mining equipment control method of claim 1, wherein, The real-time device state data and ore body characteristics include: Before the mining operation, collect samples of the ore body to be mined, and obtain the hardness and content of the target ore through mineral analysis technology; identify the main impurities in the sample through mineral composition analysis technology, and measure their hardness and content; identify the impurities with the highest hardness in the sample through mineral hardness testing technology, and measure their hardness and content; combine to obtain the ore body characteristics; Start the device for low-load trial operation, collect hydraulic oil temperature data, system pressure data and device vibration data at a preset frequency; calculate the current device state data according to the collected data.

5. The intelligent electro-hydraulic mining equipment control method of claim 4, wherein, The real-time device state data and ore body characteristics include: preprocess the real-time ore body characteristics and device state data; calculate the hardness comprehensive index according to the ore body characteristic data, and calculate the stability index according to the device state data; combine the hardness comprehensive index and the stability index as the main features with the ore body characteristics and device state data features to construct a feature matrix; input the feature matrix into the device parameter optimization model.

6. The intelligent electro-hydraulic mining equipment control method of claim 5, wherein, The target control sequence is divided into multiple control periods in time sequence, and each control period corresponds to a group of control parameter values. Set the hydraulic pump speed safety threshold, the main circuit pressure safety threshold and the unloading valve opening safety threshold; For each control period, compare the corresponding target speed with the hydraulic pump speed safety threshold, if the target speed exceeds the safety threshold, use the safety threshold as the actual control value, pass it to the hydraulic pump motor through the frequency converter, and adjust the actual speed of the hydraulic pump; compare the corresponding target pressure with the main circuit pressure safety threshold, if the target pressure exceeds the safety threshold, use the safety threshold as the actual control value, set the main circuit pressure value through the electro-hydraulic proportional valve controller; compare the corresponding target opening with the unloading valve opening safety threshold, if the target opening exceeds the safety threshold, use the safety threshold as the actual control value, adjust the opening of the unloading valve through the electro-hydraulic proportional control module. It includes:

7. An intelligent electro-hydraulic mining equipment control system for implementing an intelligent electro-hydraulic mining equipment control method according to any one of claims 1-6, characterized by, ​ The historical data screening module is configured to obtain historical ore body mining data and screen the ore body mining data to obtain an ore body mining learning data set, the ore body mining data including ore body characteristics, equipment state data and a control parameter sequence, the control parameter sequence being composed of control parameters collected at a preset frequency during the mining process, the control parameters including a hydraulic pump rotating speed, a main circuit pressure set value and an unloading valve opening degree; The model training module is configured to train an equipment parameter optimization model using the ore body mining learning data set, the equipment parameter optimization model being configured to obtain a control parameter sequence according to ore body characteristics and equipment state data; The control parameter obtaining module is configured to obtain real-time equipment state data and ore body characteristics, and input the real-time equipment state data and ore body characteristics into the equipment parameter optimization model to obtain a control parameter sequence; The control execution module is configured to obtain a target control sequence by taking the hydraulic pump rotating speed in the obtained control parameter sequence as a target rotating speed, the main circuit pressure set value as a target pressure and the unloading valve opening degree as a target opening degree; and control the electro-hydraulic mining equipment to perform a mining operation according to the target control sequence.

Images