Coal mining machine roller temperature monitoring control system based on hydroelectric interlocking mechanism and coal mining machine

Abstract

The invention provides a coal mining machine roller monitoring control system based on a hydroelectric interlocking mechanism, and the system comprises a sensing layer which comprises a waterway pressure sensor for monitoring the waterway pressure of a spraying system, a waterway flow sensor, a motor current transformer for monitoring the working current of a cutting motor, and an infrared temperature sensor for monitoring the temperature of a roller and a coal mining working surface; the control layer is internally provided with an abnormal temperature diagnosis data model so as to predict the working temperature under the current working condition and judge whether the working temperature is abnormal or not, and therefore a water pressure flow interlocking control strategy and a temperature gradient interlocking control strategy are generated; comprising an anti-explosion audible and visual alarm, a contactor for controlling start and stop of a cutting motor, a flow regulating valve for regulating the flow of a water spraying path of a spraying system, and a power-off protection device, and an execution layer executes matching actions based on a control strategy. And the coal mining machine can be automatically stopped and forcibly cooled when the cooling spray fails or the temperature suddenly rises.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of coal mining machine, and particularly relates to a coal mining machine drum temperature monitoring control system based on a water-electric interlocking mechanism and a coal mining machine comprising the control system. BACKGROUND

[0002] As the core equipment of the fully-mechanized coal mining face in a coal mine, the coal mining machine mainly undertakes the tasks of cutting, loading and unloading and transporting the coal seam in the working face. The rocker arm cutting unit is the most important component of the coal mining machine, which has a relatively complex structure and comprises electrical, oil and mechanical parts, and is responsible for the tasks of cutting the coal seam and transmission. Since the cutting unit of the coal mining machine is operated for a long time under the high dust environment in the coal mine, the friction temperature between the drum and the coal rock body can reach more than 200 DEG C. Statistics show that about 67% of the cutting unit failures are caused by bearing burnout, coal dust explosion and other accidents caused by cooling spray failure or temperature out of control. The traditional single threshold alarm system has problems such as high false alarm rate and control lag. SUMMARY

[0003] The present application aims at the deficiencies of the prior art, and provides a coal mining machine drum temperature monitoring control system based on a water-electric interlocking mechanism, which collects the temperature data of the cutting unit of the coal mining machine, monitors the state of the spray system and actively controls the safety to ensure that the coal mining machine automatically stops and forcibly cools when the cooling spray fails or the temperature rises suddenly, solves the safety hidden trouble problem caused by cooling failure or friction temperature rise of the cutting drum of the coal mining machine in the coal mine in the prior art, and improves the intelligence of the control system of the coal mining machine.

[0004] The present application provides a coal mining machine drum monitoring control system based on a water-electric interlocking mechanism, comprising: a perception layer comprising a water pressure sensor for monitoring the water pressure of the spray system of the coal mining machine, a water flow sensor for monitoring the water flow of the spray system, a motor current transformer for monitoring the working current of the cutting motor of the coal mining machine, and an infrared temperature sensor for monitoring the temperature of the drum of the coal mining machine and the coal mining face; a control layer electrically connected to each sensor of the perception layer to receive the signals of each sensor, the control layer comprising a distributed module control system, wherein an abnormal temperature diagnosis data model is established in the distributed module control system to predict the working temperature under the current working condition and judge whether the working temperature is abnormal, so as to generate a water pressure flow interlocking control strategy and a temperature gradient interlocking control strategy, an execution layer comprising an explosion-proof sound and light alarm, a contactor for controlling the start and stop of the cutting motor, a flow regulating valve for adjusting the water flow of the spray system, and a power-off protection device, the execution layer being electrically connected to the control layer to perform matching actions based on the corresponding control strategies.

[0005] Further, the water pressure flow interlock control strategy is: When the pressure of the spray waterway of the spray system is monitored to be lower than a first preset threshold or the flow is lower than a second preset threshold, the control layer issues a control instruction for controlling the explosion-proof sound-light alarm to issue a voice alarm, and after the abnormality continues for a first preset delay, issues a control instruction for controlling the contactor to stop the cutting motor.

[0006] Further, the temperature gradient interlock control strategy includes: Based on the abnormal temperature diagnosis data model, the normal working temperature under the current working condition is predicted, and whether the currently monitored temperature is abnormal is judged based on the 3σ criterion; when the judgment result is abnormal, the control layer issues a control instruction for controlling the flow regulating valve of the spray system to adjust to the maximum flow.

[0007] Further, the temperature gradient interlock control strategy further includes: Judge whether the current temperature is less than a third preset threshold and whether the spray system is normal, and when the judgment result is no, the control layer issues a control instruction for prohibiting the restart of the cutting motor, and further judges whether the temperature rising rate of the working face temperature monitored by the infrared temperature sensor is greater than a fourth preset threshold, and when the judgment is yes, the control layer issues a control instruction for controlling the coal mining machine to be powered off and stopped.

[0008] Further, the abnormal temperature diagnosis data model is a neural network model trained based on a machine learning algorithm; specifically: A four-layer feedback neural network is constructed, including an input layer, a CNN spatial feature extraction layer, a two-stage time series progressive GRU layer, and an output and decision layer, The input layer takes the historical data corresponding to the temperature of the cutting motor and the sensor data collected by the perception layer as input data, and standardizes the original input data to obtain four kinds of time series information; The CNN spatial feature extraction layer includes a convolution layer and a pooling layer, which respectively capture more specific local time pattern features and perform dimension reduction processing to retain important features in the input data; The two-stage time series progressive GRU layer extracts local time features through the first layer and outputs complete sequences for the second layer, and the second layer outputs final time features and further captures long-term time dependencies based on the features of the first layer; The output and decision layer is composed of three fully connected layer networks, which gradually compress and integrate high-level features to enhance the non-linear modeling capability and prevent overfitting; further, by controlling the model error rate and after the error converges, the abnormal temperature diagnosis data model is obtained, so as to predict the drum and coal mining working face temperature.

[0009] Further, the temperature rising rate is obtained based on the following formula:

[0010] wherein n=5 is the number of sliding window sampling points, Δt is the sampling interval, and is an adaptive adjustment parameter. Ti and Ti respectively represent the temperature values obtained by the infrared temperature sensor monitoring the drum and the working surface at the i th and the i th sampling points.

[0011] Further, the control layer is further configured with a safety power recovery mechanism, which prohibits the cutting motor from restarting before the temperature monitored by the infrared temperature sensor drops below the safety temperature and the parameters of the spraying system return to normal.

[0012] Further, the electrical control loop structure of the safety power recovery mechanism comprises a spraying pump manual control switch, two connection points connected to a spraying pump pilot loop of the spraying system, two connection points connected to a spraying pump control component, and a pressure-sensitive component for circuit protection. Only when the spraying pump manual control switch is closed and the circuit voltage is stable, the spraying pump control component can obtain normal power supply, thereby enabling the spraying pump to normally operate according to the control logic.

[0013] Further, the control layer is further integrated with an analog input and output module for receiving the sensor signals of the perception layer and for outputting matched control instructions to the execution layer.

[0014] The application also provides a coal mining machine comprising the coal mining machine drum temperature monitoring control system based on the water-electric interlocking mechanism according to any of the above-mentioned schemes.

[0015] On the basis of common sense in the art, the above-mentioned preferred conditions can be combined arbitrarily, i.e., to obtain each preferred example of the application.

[0016] The positive progress effect of the application is that: The coal winning machine drum temperature monitoring control system based on the water-electric interlocking mechanism according to the present application, through the waterway pressure sensor and the waterway flow sensor, the state of the spraying system of the cutting part of the coal winning machine is monitored in real time, the infrared temperature sensor is combined to dynamically analyze the temperature of the drum and the cutting working face, and a triple interlocking protection mechanism of'spraying-temperature-current load' is constructed; when the waterway flow of the spraying is insufficient or the temperature rise rate is abnormal, the system triggers the voice alarm, the cutting motor stops, the water spraying amount is maximized, and the forced power-off protection is controlled, and through the power-on locking logic, the safety restart of the coal winning machine is ensured; the coal winning machine is ensured to automatically stop and forcedly cool down when the cooling spraying fails or the temperature rises suddenly, and the intelligent degree of the coal winning machine control is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is the structural block diagram of the coal winning machine drum temperature monitoring control system based on the water-electric interlocking mechanism in the embodiment of the present application.

[0018] Figure 2 is the component structure schematic diagram of the coal winning machine drum temperature monitoring control system based on the water-electric interlocking mechanism in the embodiment of the present application.

[0019] Figure 3 is the neural network schematic diagram of the TTP-GRU abnormal temperature diagnosis data model in the embodiment of the present application.

[0020] Figure 4 is the two-stage time sequence progressive TTP-GRU layer structure schematic diagram of the abnormal temperature diagnosis data model in the embodiment of the present application.

[0021] Figure 5 is the electrical design diagram of the safety power-on mechanism of the coal winning machine drum temperature monitoring control system based on the water-electric interlocking mechanism in the embodiment of the present application.

[0022] Figure 6 is the flow chart of the coal winning machine drum temperature monitoring control system based on the water-electric interlocking mechanism in the embodiment of the present application. DETAILED DESCRIPTION

[0023] The present application will be further described below in combination with the embodiments shown in the drawings. EMBODIMENT

[0024] As Figure 1 and Figure 2As shown, the embodiment discloses a shearer drum monitoring control system 100 based on a water-electric interlocking mechanism, applied to a shearer, aiming at the safety hidden danger caused by cooling failure or friction temperature rise of the shearer cutting drum underground, a multi-parameter cooperative control system based on a water-electric interlocking mechanism is proposed, which is used for temperature monitoring and control of the shearer cutting drum, can realize automatic shutdown and forced cooling when the spray system for cooling of the shearer fails or the temperature rises suddenly, and can avoid accidents, etc.

[0025] Specifically, the shearer drum monitoring control system 100 includes a three-level architecture composed of a perception layer 10, a control layer 20 and an execution layer 30, and each layer has different function modules for information perception, information processing and instruction execution operation, so as to achieve intelligent control of the shearer.

[0026] As shown in Figure 1 and Figure 2 , the perception layer 10 includes a waterway pressure sensor 11 for monitoring the waterway pressure of the spray system of the shearer, a waterway flow sensor 12 for monitoring the waterway flow of the spray system, a motor current transformer 13 for monitoring the working current of the cutting motor of the shearer, and an infrared temperature sensor 14 for monitoring the temperature of the drum and the coal mining face of the shearer, so as to realize multiple data acquisition, so as to accurately judge whether the cutting part of the shearer is abnormal.

[0027] The control layer 20 is electrically connected to each sensor of the perception layer 10 to receive each sensor signal. In addition, the control layer 20 includes a distributed module control system 21, which has an abnormal temperature diagnosis data model established therein, which is used to predict the working temperature under the current working condition and judge whether the working temperature is abnormal, so as to generate a water pressure flow interlocking control strategy and a temperature gradient interlocking control strategy.

[0028] The execution layer 30 is electrically connected to the control layer 20 to perform matching actions based on the corresponding control strategy. Specifically, the execution layer 30 includes an explosion-proof sound and light alarm 31, a contactor 32 for controlling the start and stop of the cutting motor, a flow regulating valve 33 for adjusting the water flow of the spray system, and a power-off protection device 34, to respectively perform matching operations such as alarm, shutdown, etc.

[0029] In another embodiment, the water pressure flow interlocking control strategy is: when it is monitored that the pressure of the spray waterway of the spray system is lower than the first preset threshold or the flow is lower than the second preset threshold, the control layer 20 issues a control instruction for controlling the explosion-proof sound and light alarm 31 to issue a voice alarm, and after the continuous abnormality reaches the first preset delay, issues a control instruction for controlling the contactor 32 to stop the cutting motor.

[0030] In another embodiment, the temperature gradient interlocking control strategy comprises: predicting a normal working temperature under a current working condition based on an abnormal temperature diagnosis data model, and determining whether the current monitored temperature is abnormal based on a 3σ criterion; when the determination result is abnormal, the control layer 20 issues a control instruction to control the flow regulating valve 33 of the spraying system to adjust to the maximum flow (such as 180 L / min).

[0031] In a specific embodiment, the abnormal temperature diagnosis data model is a neural network model trained based on a machine learning algorithm; specifically: a four-layer feedback neural network is constructed: an input layer, a CNN spatial feature extraction layer, a two-stage temporal progressive GRU layer, and an output and decision layer. The following is combined with Table 1 and Figure 4 、 Figure 5 The data processing process of each layer is specifically described.

[0032] Table 1 TTP-GRU (Two-Stage Temporal Progressive GRU)

[0033] 1. Input layer: The historical data corresponding to each sensor data collected by the perception layer 10 and the temperature of the cutting motor (the temperature PT100 measured inside the motor winding) are used as input data, and the original input data is standardized to obtain four kinds of time series information; specifically, The input data is composed of the waterway flow, waterway pressure, cutting motor load current, and historical temperature of the cutting motor of the spraying system cooling water four physical quantity data in time sequence. Since the original data often has missing values, inconsistent dimensions, etc., the original data needs to be standardized to ensure the data quality of the input model.

[0034] 1) Missing value filling: for missing values, the effective sampling value of the previous time point is used to avoid destroying the time sequence continuity.

[0035] 2) Construction of time sample extraction: based on the least common multiple (LCM) of the sampling frequencies of the above four physical quantities, the LCM is integrated to divide the sample segment from the continuous time data as the input sample, in order to solve the problem of inconsistent time sequence dimension of the input sample.

[0036] 3) Normalization: Min-Max normalization is used to map the input physical quantity-time sequence to the [0, 1] interval to prevent the model weight from being biased towards high-dimensional physical quantity data due to unit differences.

[0037] Through the above processing, four kinds of time series information with a sequence length of 60 are obtained.

[0038] 2. CNN spatial feature extraction layer: It includes convolutional and pooling layers, which capture more specific local temporal pattern features and reduce dimensionality, respectively, to preserve important features in the input data. Specifically, 1) Convolutional layer The convolutional layer consists of two layers.

[0039] The first layer consists of 64 convolutional kernels with a size of 3*1. The activation function is ReLU, and the padding is set to padding=same to ensure that the length of the time series remains unchanged after convolution.

[0040] Input: 60x4 → 64-channel convolutional kernel → Output: 60x64 The second layer uses the same size of convolutional kernel to capture more specific local temporal pattern features, with ReLU as the activation function, padding set to padding=same, and 32 as the number of convolutional kernels.

[0041] Input: 60x64 → 32-channel convolution → Output: 60x32 2) Pooling layer MaxPooling is used with a 2x1 pooling kernel to reduce dimensionality, preserve important features in the input information, and enhance position invariance.

[0042] Input: 60x32 → Output: 30x32 3. Two-stage time-series progressive GRU layer: The data model designs a two-stage time-series progressive GRU layer and integrates it into the network, replacing the traditional GRU layer and optimizing the processing of multi-sensor time-series data. As shown in Figure 5 The two GRU layers use a "series stacking" method.

[0043] The first layer is responsible for extracting local temporal features and outputting complete sequences for the second layer: The first layer returns the complete sequence for the second layer to process; Input: 30x32 → 128 GRU → Output: 30x128 The second layer outputs the final temporal features; Input: 30x128 → 64 GRU → Output: 64 output layer The second layer further captures long-term temporal dependencies based on the features from the first layer.

[0044] 4. Output and decision layer: Composed of fully connected layer networks, a total of three layers. These layers gradually compress and integrate high-level features, enhance nonlinear modeling capabilities, and prevent overfitting. ReLU activation function is used. Further by controlling the model error rate and after error convergence, an abnormal temperature diagnosis data model is obtained to predict the temperature of the coal mining face. Specifically, Input: 64→32 dense→16 dense→output: predicted value The system will detect and store the temperature of the cutting part of the coal mining machine Within an hour, there are a total of n time series data points, as shown in equation (1).

[0045] (1) In the data sequence, σ is used to measure the deviation (dispersion) between data points, as shown in equation (2).

[0046] (2) is the temperature value predicted by the model, is the temperature value measured by the temperature sensor at the current time. If the following equation (3) is satisfied, an abnormal temperature alarm is output.

[0047] (3) In summary, the distributed module control system 21 establishes an abnormal temperature diagnosis data model by establishing a relationship between the water flow, water pressure of the spray system cooling water, motor load current of the cutting motor, and historical temperature of the cutting motor and the drum and coal mining face temperature monitored by the infrared temperature sensor. After training, testing, and controlling the model error rate to be 0.95 or less, the error converges, and an abnormal temperature diagnosis data model is obtained. In this way, when the coal mining machine is working, the same sensor data (current temperature of the drum and coal mining face, cutting motor current, etc.) is collected and input into the model to predict the normal working temperature under the current working condition. When the temperature monitored by the infrared temperature sensor 14 exceeds the 3σ criterion, it is determined that the temperature is abnormal.

[0048] In another embodiment, regarding the temperature gradient interlocking control strategy, based on the temperature abnormality judgment, it can also include: judging whether the current temperature is less than a third preset threshold and whether the spray system has returned to normal, and when the judgment result is no, the control layer 20 issues a control instruction to prohibit the restart of the cutting motor, and further judges whether the temperature of the working face monitored by the infrared temperature sensor 14 is greater than a fourth preset threshold, and when the judgment is yes, the control layer 20 issues a control instruction to shut down the coal mining machine.

[0049] In one embodiment, the temperature rise rate is obtained based on the following formula: (4) where n=5 is the number of sampling points in the sliding window, Ti and Ti -1 represent the temperature values obtained by the infrared temperature sensor monitoring the drum and the working surface at the first and the last sampling point, respectively; Δt is the sampling interval, and is an adaptive adjustment parameter, which can be set to 0.2 s, for example. i i To prevent the sampling frequency from being too high or too low, and to avoid frequent adjustment of the frequency, a minimum and maximum sampling frequency limit is set. At the same time, instead of adjusting the frequency frequently, the evaluation and adjustment are performed after a certain amount of time or a certain number of sampling points. The specific algorithm can be set as follows: Basic frequency (

[0050] ): Temperature: 0.5 Hz fbase Dynamic adjustment rule: If Δt Δt-1: 2× new = min( current × 2, f max) f If Δt f Δt-1: 0.5× new = max( current ÷ 2, f min) f Frequency boundary: f max: temperature 10 Hz f min: temperature 0.2 Hz f Emergency mode: If the temperature suddenly changes by more than 5% per second, it is immediately raised to max. f

[0051] In another embodiment, as shown in Figure 1 , the control layer 20 is also configured with a safety power recovery mechanism 22 for prohibiting the restart of the cutting motor before the temperature monitored by the infrared temperature sensor 14 drops below the safety temperature and the parameters of the spraying system return to normal.

[0052] Specifically, as Figure 3 ​​As shown, the electrical control loop structure of the safe power recovery mechanism 22 includes: a spray pump manual control switch PWBC12, two connection points connected to the spray pump pilot circuit of the spray system (XQA:5 and XQA:11 in the figure), two connection points connected to the spray pump control component V8 (XQA:10 and XQA:6 in the figure), and a pressure-sensitive component VSR5 for circuit protection.

[0053] In Figure 3 the electrical design diagram, the spray pump control component V8 is the core control unit of the spray pump, responsible for power control and logic execution of the spray pump. The PWBC12 control switch belongs to the "spray system control" switch, which is a hardware trigger element for starting the spray pump. Manual operation of the switch can trigger the control logic of the spray pump. The pressure-sensitive component VSR5 is a protective element of the circuit, which will perform a protective short circuit when the circuit voltage fluctuates abnormally (too high or unstable), preventing fluctuating voltage from damaging electrical elements such as spray pump control components and switches.

[0054] The electrical control loop structure of the safe power recovery mechanism 22, as part of the spray pump pilot circuit, combines the "hardware-software double locking of the coal mining machine power recovery" requirement, and its control logic is designed as follows: Only when the spray pump manual control switch PWBC12 control switch is closed (manually triggering the spray system control) and the circuit voltage is stable (the pressure-sensitive component VSR5 is not triggered for protection), the spray pump control component V8 can obtain normal power supply, and then the spray pump operates normally according to the control logic.

[0055] That is, if the voltage is abnormal, the pressure-sensitive component VSR5 will short circuit to protect the circuit; at the same time, with the software logic, the double locking of "the coal mining machine cannot automatically recover power when the temperature does not reach the safe level, and manual reset is required" is realized, preventing misoperation and electrical damage from the hardware level, and ensuring system safety.

[0056] In short, the configuration of the safe power recovery mechanism 22 realizes the safety and anti-misoperation function of the spray pump control through the cooperation of hardware switch triggering, voltage protection elements and software logic.

[0057] In another embodiment, as shown in Figure 1 , the control layer 20 also integrates an analog input and output module 23 for receiving sensor signals from the sensing layer and outputting matching control instructions to the execution layer 30.

[0058] According to the coal winning machine drum temperature monitoring control system based on the water interlocking mechanism related to the application, the state of the spraying system of the cutting part of the coal winning machine is monitored in real time through the waterway pressure sensor and the waterway flow sensor, dynamic analysis is performed on the drum and the cutting working face temperature by combining the infrared temperature sensor, a "spraying-temperature-current load" triple interlocking protection mechanism is constructed, when the waterway flow of the spraying is insufficient or the temperature rise rate is abnormal, the system triggers voice alarm, cutting motor shutdown, maximum water spraying control and forced power-off protection, and the safe restart of the coal winning machine is ensured through the power-on locking logic; the coal winning machine is ensured to automatically stop and forcedly cool down when the cooling spraying fails or the temperature rises sharply.

[0059] Compared with the problem of response lag existing in the alarm of the traditional monitoring means depending on a single temperature threshold, the temperature rise rate prediction algorithm based on time series analysis is proposed in the embodiment. Meanwhile, the historical data of the cutting part and the temperature gradient are cooperatively interlocked for control, and the intelligence degree of the coal winning machine control system is improved.

[0060] According to the structure of the coal winning machine drum monitoring control system 100, in specific application, in view of the complex environment of the coal winning face, which has the characteristics of high humidity, much dust and strong electromagnetic interference, the accuracy and explosion-proof performance of the sensor are required to be harsh, and the selection and installation of the sensor are a key point of the system. The selection of the pressure sensor, the flow sensor and the infrared temperature sensor needs to meet the explosion-proof standard, while the accuracy and reliability are considered. The selected sensor must be durable and good in anti-interference, and the related technical parameters are shown in Table 2.

[0061] Table 2 Selection of each sensor

[0062] Based on the structure of the above embodiment, the function is tested through simulation test, and the simulation content and results are as follows: A 1:1 test system is constructed in the simulation control center, in the temperature of 37℃ and the humidity of 85%, two typical fault scenes are set: Scene A: artificially closing 50% of the nozzles to simulate pipeline blockage (flow rate reduced to 10 L / min) Scene B: loading abnormal friction load to simulate bearing failure (temperature rise rate of 12℃ / min) During the simulation process, as shown in the flowchart, Figure 6 the control system reads the sensor data of pressure, flow, temperature and temperature change rate continuously, judges the state of the cutting part of the coal winning machine according to the set threshold condition, triggers alarm, stops the motor, cuts off the power and the like when the abnormality occurs, so as to ensure the safe operation of the system.

[0063] Wherein the first preset threshold is set to 1.5 MPa, the second preset threshold is set to 102 L / min, the first preset delay is set to 5 seconds, the third preset threshold is set to 40 DEG C, the fourth preset threshold is set to 5 DEG C / min, and the maximum flow of the flow regulating valve is 180 L / min.

[0064] Through the simulation test, the performance test results are as follows: Scenario A experimental results

[0065] Scenario B experimental results

[0066] Based on the above simulation test structure, it can be directly concluded that the shearer drum monitoring control system 100 based on the water-electric interlocking mechanism of the embodiment can ensure that the shearer automatically stops and forcibly cools down when the cooling spray fails or the temperature rises sharply through the multi-sensor collection of the temperature data of the shearer cutting part, the spray system state monitoring and the active safety control, and improve the safety of the shearer.

[0067] Based on the structure of the above embodiment, the embodiment further provides a shearer comprising the shearer drum temperature monitoring control system based on the water-electric interlocking mechanism described in the above embodiment.

[0068] The scope of protection of the present application is not limited to the above-mentioned embodiments, and obviously, those skilled in the art can make various modifications and changes to the present application without departing from the scope and spirit of the present application. If these modifications and changes belong to the scope of the claims of the present application and its equivalent technologies, the intention of the present application also includes these modifications and changes.

Claims

1. A monitoring and control system for a coal mining machine drum based on a hydropower interlocking mechanism, characterized in that, include: The sensing layer includes a water pressure sensor for monitoring the water pressure of the spray system of the coal mining machine, a water flow sensor for monitoring the water flow of the spray system, a motor current transformer for monitoring the operating current of the cutting motor of the coal mining machine, and an infrared temperature sensor for monitoring the temperature of the drum and the coal mining face of the coal mining machine. The control layer, electrically connected to each sensor in the sensing layer to receive signals from each sensor, includes a distributed module control system. This is used to predict the operating temperature under current conditions and determine whether the operating temperature is abnormal, thereby generating water pressure and flow rate interlocking control strategies and temperature gradient interlocking control strategies. The execution layer includes an explosion-proof audible and visual alarm, a contactor for controlling the start and stop of the cutting motor, a flow regulating valve for adjusting the water flow of the spray system, and a power failure protection device. The execution layer is electrically connected to the control layer to perform matched actions based on the corresponding control strategy.

2. The system according to claim 1, characterized in that, The water pressure and flow rate interlock control strategy is as follows: When the pressure of the spray water path of the spray system is detected to be lower than the first preset threshold or the flow rate is lower than the second preset threshold, the control layer issues a control command to control the explosion-proof audible and visual alarm to issue a voice alarm, and after the continuous abnormality reaches the first preset delay, issues a control command to control the contactor to stop the cutting motor.

3. The system according to claim 2, characterized in that, The temperature gradient interlocking control strategy includes: Based on the abnormal temperature diagnostic data model, the normal operating temperature under the current working conditions is predicted, and the 3σ criterion is used to determine whether the currently monitored temperature is abnormal. When the determination result is abnormal, the control layer issues a control command to adjust the flow regulating valve of the spray system to the maximum flow rate.

4. The system according to claim 3, characterized in that, The temperature gradient interlocking control strategy also includes: The control layer determines whether the current temperature is less than a third preset threshold and whether the spray system has returned to normal. If the determination result is no, the control layer issues a control command to prohibit the cutting motor from restarting. It further determines whether the heating rate of the working face temperature monitored by the infrared temperature sensor is greater than a fourth preset threshold. If the determination result is yes, the control layer issues a control command to control the coal mining machine to shut down.

5. The system according to claim 1, characterized in that, The abnormal temperature diagnostic data model is a neural network model trained based on machine learning algorithms; specifically: Construct a four-layer feedback neural network: an input layer, a CNN spatial feature extraction layer, a two-stage temporal progressive GRU layer, and an output and decision layer. The input layer takes the sensor data collected by the sensing layer and the historical data corresponding to the temperature of the cutting motor as input data, and performs standardization processing on the original input data to obtain four types of time series information. The CNN spatial feature extraction layer includes convolutional layers and pooling layers, which respectively capture more specific local temporal pattern features and perform dimensionality reduction processing to retain important features in the input data; The two-stage temporal progressive GRU layer extracts local temporal features through the first layer and outputs the complete sequence for the second layer. The second layer outputs the final temporal features and further captures long-term temporal dependencies based on the features of the first layer. The output and decision layer consists of a three-layer fully connected network. By gradually compressing and integrating high-level features, it enhances nonlinear modeling capabilities and prevents overfitting. Furthermore, by controlling the model error rate and after the error converges, the abnormal temperature diagnostic data model is obtained, which facilitates the prediction of the drum and coal face temperatures.

6. The system according to claim 4, characterized in that, The heating rate is obtained based on the following formula: 。 7. Among them, n=5 is the number of sampling points in the sliding window, and Δt is the sampling interval, which is an adaptive adjustment parameter; Ti and Ti -1 represents the first... i The and the first i The infrared temperature sensor at -1 sampling points monitors the temperature values ​​of the drum and working surface.

8. The system according to claim 4, characterized in that, The control layer is also equipped with a safety power-off mechanism, which prevents the cutting motor from restarting until the temperature detected by the infrared temperature sensor drops below the safe temperature and the parameters of the spray system return to normal.

9. The system according to claim 7, characterized in that, The electrical control circuit structure of the safety power restoration mechanism includes: a manual control switch for the spray pump, two connection points connected to the pilot circuit of the spray pump in the spray system, two connection points connected to the spray pump control component, and a pressure-sensitive component for circuit protection. The spray pump control component can only receive normal power supply when the manual control switch of the spray pump is closed and the circuit voltage is stable, thereby enabling the spray pump to operate normally according to the control logic.

10. The system according to claim 1, characterized in that, The control layer also integrates an analog input / output module, which is used to receive signals from each sensor in the sensing layer and to output matching control commands to the execution layer.

11. A coal mining machine, characterized in that, The system includes a coal mining machine drum temperature monitoring and control system based on a hydropower interlocking mechanism as described in any one of claims 1 to 9.

Images