Evaluation Method for Regional Support Quality of Hydraulic Supports at Working Face

Abstract

This invention relates to a method for evaluating the regional support quality of hydraulic supports in a working face, belonging to the field of intelligent coal mining technology. It includes: acquiring and constructing an overall spatiotemporal support pressure matrix for the working face based on the column pressure data of each hydraulic support on the working face; sliding from the overall spatiotemporal pressure matrix to obtain multiple current-region spatiotemporal support pressure sub-matrices representing the current regional support quality of the working face; inputting each current-region spatiotemporal support pressure sub-matrices into a pre-trained spatiotemporal regional support quality evaluation model for hydraulic supports in the working face; and determining the evaluation result of the current hydraulic support quality in each region of the working face based on the output results. This invention can evaluate the current support quality of hydraulic support groups in each region of the working face, thereby capturing the dynamic characteristics of regional support quality and predicting regional support quality for a future period, greatly improving the level of intelligence in working face production.

Description

Technical Field

[0001] This invention relates to the field of intelligent coal mining technology, and in particular to a method for evaluating the regional support quality of hydraulic supports at working faces. Background Technology

[0002] With the advancement of intelligent coal mining, the information requirements for production and management processes are constantly increasing. One important task is the intelligent sensing of the production environment, especially the intelligent sensing of the roof support condition. The stability of the roof support directly affects whether production can proceed safely and efficiently. Through early sensing of information, corresponding warnings can be issued to reduce accidents such as roof falls and roof collapses.

[0003] Existing research on the quality of roof support is generally limited to the parameter evaluation of the support effect of a single hydraulic support. This evaluation relies on analyzing statistical patterns of a single hydraulic support, such as initial support force, final resistance, and weighted time resistance, to assess the support quality of that support. However, this method lacks the evaluation and dynamic analysis process for the support quality of hydraulic support groups. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for evaluating the regional support quality of hydraulic supports at working faces. The technical solution is as follows:

[0005] A method for evaluating the regional support quality of a hydraulic support at a working face includes the following steps:

[0006] S1, acquire the column pressure data of each hydraulic support on the working surface. The column pressure data includes the historical column pressure data collected during the historical working process of each hydraulic support and the current column pressure data collected during the current working process of each hydraulic support.

[0007] S2, construct the overall spatiotemporal support pressure matrix of the working face based on the column pressure data of each hydraulic support on the working face;

[0008] S3, obtain multiple current area spatiotemporal support pressure sub-matrices representing the support quality of the current area of ​​the working face from the overall spatiotemporal pressure matrix of the working face;

[0009] S4, input the spatiotemporal support pressure sub-matrices of each current region into the pre-trained spatiotemporal regional support quality evaluation model of the working face hydraulic support, which is trained by the historical pressure data of the columns of each hydraulic support on the working face.

[0010] S5. Determine the evaluation results of the current hydraulic support quality of each area of ​​the working face based on the output results of the spatiotemporal regional support quality evaluation model of the working face hydraulic support.

[0011] Optionally, when S2 constructs the overall spatiotemporal support pressure matrix of the working face based on the column pressure data of each hydraulic support on the working face, it includes:

[0012] S21, perform data cleaning on the column pressure data of each hydraulic support on the working surface to obtain the preliminary column pressure data of each hydraulic support on the working surface;

[0013] S22, select several time points with equal spacing to construct the overall spatiotemporal support pressure matrix of the working face;

[0014] S23, the preliminary column pressure data of each hydraulic support on the working surface are filtered and sorted according to the hydraulic support number and time to obtain the column pressure spatial sequence and column pressure time sequence of each hydraulic support on the working surface.

[0015] S24, combining the column pressure spatial sequence and the column pressure time sequence, yields the overall spatiotemporal support pressure matrix of the working face.

[0016] Optionally, when S21 cleans the column pressure data of each hydraulic support on the working surface to obtain the preliminary column pressure data of each hydraulic support on the working surface, it includes: processing data loss and data anomaly of the column pressure data of each hydraulic support on the working surface.

[0017] Optionally, when S3 slides to obtain multiple current-area spatiotemporal support pressure sub-matrices representing the support quality of the current area of ​​the working face from the overall spatiotemporal pressure matrix of the working face, it includes:

[0018] S31, Select The sliding window of the size is used as a current area spatiotemporal support pressure submatrix, where n represents the number of hydraulic supports included in a current area spatiotemporal support pressure submatrix.

[0019] S32, starting from the upper right corner of the overall spatiotemporal pressure matrix of the working face, slide from top to bottom to obtain multiple spatiotemporal support pressure sub-matrices of the current area. The step distance of the sliding window is n / 2, and if the step distance of the last slide is less than n / 2, the remaining distance is used as the step distance of the last slide.

[0020] Optionally, when S5 determines the evaluation result of the current hydraulic support quality of each region of the working face based on the output result of the spatiotemporal regional support quality evaluation model of the working face hydraulic support, it includes:

[0021] Based on the output results of the spatiotemporal support quality evaluation model of the hydraulic support in the working face, the evaluation results of the current hydraulic support quality in each area of ​​the working face are determined to be one of the following: initial deterioration of support quality, continuous deterioration of support quality, deep deterioration of support quality, generally maintained support quality, initial optimization of support quality, continuous optimization of support quality, and good maintenance of support quality.

[0022] Optionally, after determining the evaluation results of the current hydraulic support quality of each region of the working face based on the output results of the spatiotemporal regional support quality evaluation model of the working face hydraulic support, step S5 further includes:

[0023] Determine the corresponding early warning information based on the evaluation results of the current hydraulic support quality in each area of ​​the working face;

[0024] The system displays the evaluation results of the current hydraulic support quality in each area of ​​the working face, as well as its corresponding early warning information, including normal maintenance, alarm prompts, and abnormal handling.

[0025] Optionally, before inputting the spatiotemporal support pressure sub-matrices of each current region into the pre-trained spatiotemporal regional support quality evaluation model of the working face hydraulic support, step S4 further includes:

[0026] S41, slide to obtain multiple historical area spatiotemporal support pressure sub-matrices representing the historical area support quality of the working face from the overall spatiotemporal pressure matrix of the working face;

[0027] S42, obtain the evaluation result labels for the spatiotemporal support pressure sub-matrix annotation of each historical region;

[0028] S43, a spatiotemporal support quality evaluation model for hydraulic supports at the working face is trained by using multiple historical spatiotemporal support pressure sub-matrices and their corresponding evaluation result labels.

[0029] Optionally, the spatiotemporal support quality evaluation model of the hydraulic support at the working face is a convolutional neural network model.

[0030] Optionally, after constructing the overall spatiotemporal support pressure matrix of the working face based on the column pressure data of each hydraulic support on the working face, step S2 further includes:

[0031] The overall spatiotemporal support pressure matrix of the working face and the evaluation results of the current hydraulic support quality of each area of ​​the working face are updated every preset time interval.

[0032] All of the above-mentioned optional technical solutions can be combined arbitrarily, and the present invention will not provide a detailed description of the structure after each combination.

[0033] By means of the above solution, the beneficial effects of the present invention are as follows:

[0034] By acquiring and constructing an overall spatiotemporal support pressure matrix for the working face based on the column pressure data of each hydraulic support on the working face, and then sliding out multiple current area spatiotemporal support pressure sub-matrices representing the current area support quality from the overall spatiotemporal pressure matrix, these sub-matrices are input into a pre-trained spatiotemporal area support quality evaluation model for the hydraulic supports on the working face. The evaluation results for the current hydraulic support quality in each area of ​​the working face are determined based on the output results. This provides a method for evaluating the support quality of hydraulic supports in each area of ​​the working face. This method can evaluate the support quality of hydraulic support groups in each area of ​​the working face, and the evaluation results are achieved by combining the current column pressure data collected during the current operation of each hydraulic support. This allows for real-time evaluation results of the support quality of hydraulic support groups in each area, making the evaluation results more reliable. It can capture the dynamic characteristics of the regional support quality of the working face, predict the regional support quality for a future period, prevent accidents such as roof falls and support collapses, and greatly improve the intelligence level of working face production.

[0035] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0036] Figure 1 This is a flowchart of the present invention.

[0037] Figure 2 This is the overall flowchart of the present invention. Detailed Implementation

[0038] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0039] like Figure 1 As shown, the method for evaluating the regional support quality of hydraulic supports at working faces provided by the present invention includes the following steps:

[0040] S1, acquire the column pressure data of each hydraulic support on the working surface. The column pressure data includes the historical column pressure data collected during the historical working process of each hydraulic support and the current column pressure data collected during the current working process of each hydraulic support.

[0041] Among these, the column pressure data of the hydraulic supports is a very important data point for the working face, as it is a key parameter reflecting the quality of the working face support. Therefore, this embodiment of the invention is based on the column pressure data of each hydraulic support.

[0042] During production, the controllers and sensors installed on the equipment group on the working surface continuously collect data. By accumulating the data collected each time, the column pressure data of each hydraulic support on the working surface can be obtained. Therefore, this step is achieved by acquiring the data collected by the controllers and sensors installed on the equipment group on the working surface.

[0043] S2, construct the overall spatiotemporal support pressure matrix of the working face based on the column pressure data of each hydraulic support on the working face.

[0044] The overall spatiotemporal support pressure matrix of the working face is used to describe the temporal and spatial distribution of the column pressure of each hydraulic support on the working face. Step S2, which constructs the overall spatiotemporal support pressure matrix of the working face based on the column pressure data of each hydraulic support on the working face, can be achieved through the following steps S21 to S24:

[0045] S21, perform data cleaning on the column pressure data of each hydraulic support on the working surface to obtain the preliminary column pressure data of each hydraulic support on the working surface.

[0046] Specifically, due to the complex production environment, various sensors frequently malfunction, leading to errors in the collected data. Common errors include data loss and data anomalies. Data loss refers to a break in the data transmission path, commonly caused by damage to the sensor itself or its signal lines. Data anomalies refer to an incorrect value transmitted from the sensor, typically an outlier, often due to damage to internal sensor components. Therefore, the data cleaning in this embodiment of the invention includes both data loss handling and data anomaly handling.

[0047] Furthermore, to address the issue of data loss and ensure data integrity, this embodiment of the invention employs a method of filling adjacent values ​​in the time domain. For example, if the column pressure value of a hydraulic support is missing at a certain moment, it is filled with the pressure value of the hydraulic support at the previous moment. In the case of data anomalies, this embodiment of the invention uses the average pressure of the five adjacent hydraulic supports at that moment to replace the abnormal value.

[0048] S22, select several time points with equal spacing to construct the overall spatiotemporal support pressure matrix of the working face.

[0049] The interval between time points can be set as needed. These time points are arranged in chronological order in the overall spatiotemporal support pressure matrix of the working face.

[0050] S23, the preliminary column pressure data of each hydraulic support on the working surface are filtered and sorted according to the hydraulic support number and time to obtain the column pressure spatial sequence and column pressure time sequence of each hydraulic support on the working surface.

[0051] Since the initial column pressure data at each specific time point may not be representative for any given hydraulic support, in order to obtain representative initial column pressure data at each time point, in this embodiment of the invention, the initial column pressure data at any time point for any hydraulic support is the average value of the column pressure data over a specified period of time preceding that time point. This specified period can be set as needed, such as 5 minutes.

[0052] S24, combining the column pressure spatial sequence and the column pressure time sequence, yields the overall spatiotemporal support pressure matrix of the working face.

[0053]

[0054] The matrix shown in the table above is a schematic diagram of the overall spatiotemporal support pressure matrix of a working face. M represents the number of hydraulic supports on the working face, and N represents the number of time points. This represents the initial column pressure value of hydraulic support i at time point j.

[0055] S3, obtain multiple current area spatiotemporal support pressure sub-matrices representing the current area support quality of the working face from the overall spatiotemporal pressure matrix of the working face.

[0056] Since the overall spatiotemporal pressure matrix of the working face is arranged according to time, the current pressure data of the column is located in the last few columns of the overall spatiotemporal support pressure matrix of the working face. In addition, in order to obtain the transition relationship between the current hydraulic support quality and the historical hydraulic support quality of each area of ​​the working face, in this embodiment of the invention, the multiple current area spatiotemporal support pressure sub-matrices representing the current area support quality of the working face include the preliminary column pressure data determined by the current column pressure data in the last few columns of the overall spatiotemporal support pressure matrix of the working face, and several columns of preliminary column pressure data determined by the historical column pressure data adjacent to these columns.

[0057] The number of columns covered by the preliminary column pressure data determined from the current column pressure data in the current regional spatiotemporal support pressure sub-matrix is ​​related to the time interval between the current column pressure data and the previously acquired column pressure data, as well as the time interval between adjacent time points. For example, if the time interval between the current column pressure data and the previously acquired column pressure data is 15 minutes, and the time interval between adjacent time points is 5 minutes, then the number of columns covered by the preliminary column pressure data determined from the current column pressure data in the current regional spatiotemporal support pressure sub-matrix will be three.

[0058] Furthermore, the number of columns in the current regional spatiotemporal support pressure sub-matrix, which is determined by the historical pressure data of the columns, can be selected as two, three, or other columns as needed. This embodiment does not limit this.

[0059] Furthermore, the number of rows covered by the current region's spatiotemporal support pressure submatrix is ​​related to the number of hydraulic supports included in a region. For example, if a region is preset to include three hydraulic supports, then the current region's spatiotemporal support pressure submatrix will cover three rows. As shown in the table below, the boxed area represents an example of a current region's spatiotemporal support pressure submatrix.

[0060]

[0061] Optionally, when S3 slides to obtain multiple current-area spatiotemporal support pressure sub-matrices representing the support quality of the current area of ​​the working face from the overall spatiotemporal pressure matrix of the working face, it includes:

[0062] S31, Select The sliding window of size serves as a current region spatiotemporal support pressure submatrix, where n represents the number of hydraulic supports included in a current region spatiotemporal support pressure submatrix.

[0063] S32, starting from the upper right corner of the overall spatiotemporal pressure matrix of the working face, slide from top to bottom to obtain multiple spatiotemporal support pressure sub-matrices of the current area. The step distance of the sliding window is n / 2, and if the step distance of the last slide is less than n / 2, the remaining distance is used as the step distance of the last slide.

[0064] The step size of the sliding window, n / 2, is chosen by taking into account the problems of losing important information if the step size is too large, and the slow calculation speed and high information repetition rate if the step size is too small.

[0065] S4. Input the spatiotemporal support pressure sub-matrices of each current region into the pre-trained spatiotemporal regional support quality evaluation model of the working face hydraulic support. The spatiotemporal regional support quality evaluation model of the working face hydraulic support is trained by the historical pressure data of the columns of each hydraulic support on the working face.

[0066] Specifically, before inputting the spatiotemporal support pressure sub-matrices of each current region into the pre-trained spatiotemporal regional support quality evaluation model of the working face hydraulic support in step S4, it is also necessary to train the spatiotemporal regional support quality evaluation model of the working face hydraulic support first. The specific method for training the spatiotemporal regional support quality evaluation model of the working face hydraulic support includes, but is not limited to, the following steps S41 to S43:

[0067] S41, obtain multiple historical area spatiotemporal support pressure sub-matrices representing the historical area support quality of the working face from the overall spatiotemporal pressure matrix of the working face.

[0068] To ensure the trained spatiotemporal support quality evaluation model for the hydraulic support at the working face is applicable to subsequently collected data, the number of rows and columns covered by the historical spatiotemporal support pressure sub-matrix is ​​the same as that covered by the current spatiotemporal support pressure sub-matrix. For example, the size of the historical spatiotemporal support pressure sub-matrix is ​​also... .

[0069] When acquiring multiple historical spatiotemporal support pressure sub-matrices representing the support quality of historical areas of the working face, the process starts from the top left corner of the overall spatiotemporal pressure matrix of the working face and slides down. The step size of the sliding window from top to bottom is n / 2. If the step size of the last slide is less than n / 2, the remaining distance is used as the step size of the last slide. After acquiring n columns, the process slides horizontally to the next n columns and then slides down again.

[0070] S42, obtain the evaluation result labels for the spatiotemporal support pressure sub-matrix annotation of each historical region.

[0071] Among them, the evaluation result labels of each historical region's spatiotemporal support pressure sub-matrix are obtained by manual labeling based on mining technology, combined with on-site observations.

[0072] S43, a spatiotemporal support quality evaluation model for hydraulic supports at the working face is trained by using multiple historical spatiotemporal support pressure sub-matrices and their corresponding evaluation result labels.

[0073] The spatiotemporal support quality evaluation model for the hydraulic support at the working face is a convolutional neural network model. In this embodiment of the invention, the convolutional neural network model is an improved version of LeNet-5, containing 3 convolutional kernels. It has 3 convolutional layers, 3 pooling layers, and 3 fully connected layers, and incorporates Dropout and Earlystopping to prevent overfitting. It also uses the ReLU activation function and the Adam optimizer.

[0074] The specific training process involves continuously optimizing the parameters of the convolutional neural network model. Training is completed when all parameters reach their optimal values. This embodiment of the invention will not elaborate on this part in detail.

[0075] S5. Determine the evaluation results of the current hydraulic support quality of each area of ​​the working face based on the output results of the spatiotemporal regional support quality evaluation model of the working face hydraulic support.

[0076] Specifically, the output of the spatiotemporal regional support quality evaluation model for the hydraulic support at the working face is the probability of various evaluation results. For a certain region, the evaluation result corresponding to the highest probability value among these probabilities is the evaluation result of the current hydraulic support quality for that region.

[0077] The evaluation results for the current hydraulic support quality in each area of ​​the working face are categorized as follows: initial deterioration of support quality, continuous deterioration of support quality, deep deterioration of support quality, generally good support quality, initial optimization of support quality, continuous optimization of support quality, and good support quality. These seven evaluation results do not overlap with each other and are determined by combining the experience of on-site technicians, on-site working condition observations, and knowledge of hydraulic support process requirements.

[0078] The initial deterioration of support quality refers to the following: the regional support quality gradually deteriorates from good to poor, with some hydraulic supports experiencing significant working resistance, and this trend is expected to continue.

[0079] The term "continuous deterioration of support quality" refers to the following: the regional support quality gradually deteriorates from "average" to "poor", resulting in small areas with high resistance or insufficient working resistance, and there is a trend of continued deterioration.

[0080] The aforementioned deep deterioration of support quality refers to the following: the regional support quality changes from being poor to remaining unchanged, but then changes to large areas with high working resistance, large areas with insufficient working resistance, or alternating high and low pressures between adjacent hydraulic supports, which damages the integrity of the roof.

[0081] Maintaining the quality of support generally means that the regional support quality remains unchanged from generally good to poor, while the working resistance in small areas is either too high or too low, and there is no trend of change.

[0082] The initial optimization of support quality refers to the improvement of regional support quality from poor to good, and the improvement of situations where the working resistance of large areas is high or low, or where the working resistance of large areas is insufficient or where the pressure of adjacent hydraulic supports is high or low.

[0083] The continuous optimization of support quality refers to the gradual improvement of regional support quality from generally good to poor, and the improvement of situations where small areas have high or insufficient working resistance.

[0084] The term "good support quality" means that the regional support quality remains unchanged from "good" to "good", the working resistance of individual hydraulic supports is high or insufficient, there is no trend of deterioration, and the roof is well protected.

[0085] The term "small area" refers to an area where the number of adjacent hydraulic supports is less than a preset value, and the term "large area" refers to an area where the number of adjacent hydraulic supports is greater than or equal to a preset value. This preset value is selected as needed; for example, if the preset value is 5, then an area with fewer than 5 adjacent hydraulic supports is considered a small area, and an area with at least 5 adjacent hydraulic supports is considered a large area.

[0086] "High working resistance" means the working resistance is greater than the first preset threshold, while "insufficient working resistance" means the working resistance is less than the second preset threshold. The first and second preset thresholds can be set as needed; for example, the first preset threshold is 43 MPa and the second preset threshold is 18 MPa.

[0087] Furthermore, based on the above S1 to S5, the method provided by the embodiments of the present invention may further include the following S6 and S7:

[0088] S6 determines the early warning information corresponding to the evaluation results of the current hydraulic support quality in each area of ​​the working face.

[0089] In this embodiment of the invention, a pre-defined correspondence between the hydraulic support quality and early warning information can be established. Based on this, in step S6, when determining the early warning information corresponding to the evaluation results of the current hydraulic support quality in each area of ​​the working face, the pre-defined correspondence can be used.

[0090] S7 displays the evaluation results of the current hydraulic support quality in each area of ​​the working face, as well as its early warning information.

[0091] Specifically, an early warning interface can be created on the working face visualization platform to display the evaluation results of the current hydraulic support quality of each area and the corresponding early warning information on the early warning interface.

[0092] Furthermore, the early warning information may include normal operation status, alarm prompts, and abnormal handling.

[0093] The term "normal maintenance" refers to situations where the support quality remains average, the support quality continues to improve, or the support quality remains good, indicating that the support effect is relatively good and within a reasonable range, and is relatively safe.

[0094] The alarm message indicates that the support quality is either initially deteriorating or initially improving, suggesting that the support effect is average and in a changing phase, requiring continuous monitoring.

[0095] The aforementioned abnormal handling refers to situations where the support quality continues to deteriorate or deteriorates significantly, indicating that the support effect is poor and is continuously worsening, requiring manual adjustment.

[0096] Furthermore, during the production process, in order to obtain timely evaluation results of the current hydraulic support quality in each area of ​​the working face, it is necessary to evaluate the current hydraulic support quality in each area of ​​the working face in real time. Therefore, it is necessary to update the overall spatiotemporal support pressure matrix of the working face in real time. Thus, the method provided in this embodiment of the invention further includes updating the overall spatiotemporal support pressure matrix of the working face every preset time interval, and further updating the evaluation results of the current hydraulic support quality in each area of ​​the working face every preset time interval based on the update results. The preset time interval can be selected as needed, for example, set to 15 minutes. In addition, the preset time interval can also be determined based on the number of columns covered by the preliminary column pressure data determined by the current column pressure data in the current area spatiotemporal support pressure submatrix. For example, if the current area spatiotemporal support pressure submatrix includes five columns, and the number of columns covered by the preliminary column pressure data determined by the current column pressure data is three columns, and the interval between adjacent time points is 5 minutes, then the preset time interval is 15 minutes.

[0097] In summary, as Figure 2 As shown, it is the overall flowchart of the present invention.

[0098] The method provided by the embodiments of the present invention can classify the regional support quality of the working face in real time and issue different early warning information for different support types. It can not only replace the repetitive physical labor of workers and save a lot of manpower, but also capture the dynamic characteristics of the regional support quality of the working face and predict the regional support quality for a period of time in the future, so as to prevent accidents such as roof collapse and frame crush, and greatly improve the intelligence level of working face production.

[0099] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for evaluating regional support quality of a hydraulic support of a working face, characterized in that, The method comprises the following steps: S1, obtaining column pressure data of each hydraulic support on the working face, the column pressure data comprising column historical pressure data collected in a historical working process of each hydraulic support and column current pressure data collected in a current working process of each hydraulic support; S2, constructing a working face overall time-space support pressure matrix according to the column pressure data of each hydraulic support on the working face; S3, slidingly obtaining a plurality of current regional time-space support pressure sub-matrices representing current regional support quality of the working face from the working face overall time-space pressure matrix; S4, inputting each current regional time-space support pressure sub-matrix into a working face hydraulic support time-space regional support quality evaluation model pre-trained, the working face hydraulic support time-space regional support quality evaluation model being trained by column historical pressure data of each hydraulic support on the working face; S5, determining an evaluation result of current hydraulic support quality of each region of the working face according to an output result of the working face hydraulic support time-space regional support quality evaluation model; In the step S2, the step of constructing the working face overall time-space support pressure matrix according to the column pressure data of each hydraulic support on the working face comprises: S21, performing data cleaning on the column pressure data of each hydraulic support on the working face to obtain preliminary column pressure data of each hydraulic support on the working face; S22, selecting a plurality of equidistantly distributed time points for constructing the working face overall time-space support pressure matrix; S23, sorting and sequencing the preliminary column pressure data of each hydraulic support on the working face according to the support number and time of the hydraulic support to obtain column pressure space sequences and column pressure time sequences of each hydraulic support on the working face; S24, combining the column pressure space sequences and the column pressure time sequences to obtain the working face overall time-space support pressure matrix.

2. The method of evaluating the quality of regional support of the hydraulic support of the coal face according to claim 1, characterized in that, In the step S21, the step of performing data cleaning on the column pressure data of each hydraulic support on the working face to obtain preliminary column pressure data of each hydraulic support on the working face comprises data loss processing and data anomaly processing on the column pressure data of each hydraulic support on the working face.

3. The method of evaluating the quality of regional support of the hydraulic support of the coal face according to claim 1, characterized in that, In the step S3, the step of slidingly obtaining a plurality of current regional time-space support pressure sub-matrices representing current regional support quality of the working face from the working face overall time-space pressure matrix comprises: S31, selecting The size of the sliding window is taken as a current region and space support pressure sub-matrix, and n represents the number of hydraulic supports included in the current region and space support pressure sub-matrix. S32, slidingly obtaining the plurality of current regional time-space support pressure sub-matrices from top to bottom starting from the upper right corner of the working face overall time-space pressure matrix, wherein the step distance of the sliding window is n / 2, and if the step distance of the last sliding is less than n / 2, the remaining distance is used as the step distance of the last sliding.

4. The method of evaluating the quality of regional support of a hydraulic support of a coal face according to claim 1, characterized in that, In the step S5, the step of determining an evaluation result of current hydraulic support quality of each region of the working face according to an output result of the working face hydraulic support time-space regional support quality evaluation model comprises: According to the output result of the working face hydraulic support space-time area support quality evaluation model, the evaluation result of the current hydraulic support quality of each area of the working face is determined as one of support quality preliminary deterioration, support quality continuous deterioration, support quality deep deterioration, support quality general, support quality preliminary optimization, support quality continuous optimization and support quality good.

5. The method of evaluating the quality of regional support of the hydraulic support of the coal face according to claim 1 or 4, characterized in that, The S5 further includes, after determining the evaluation result of the current hydraulic support quality of each area of the working face according to the output result of the working face hydraulic support space-time area support quality evaluation model: determining the corresponding early warning information of the evaluation result of the current hydraulic support quality of each area of the working face; displaying the evaluation result of the current hydraulic support quality of each area of the working face and the corresponding early warning information, wherein the early warning information includes normal maintenance, alarm prompt and abnormal disposal.

6. The method of evaluating the quality of regional support of a hydraulic support of a coal face according to claim 1, characterised in that, The S4 further includes, before inputting each current area space-time support pressure sub-matrix into the working face hydraulic support space-time area support quality evaluation model trained in advance: S41, slidingly obtaining a plurality of historical area space-time support pressure sub-matrices representing the historical area support quality of the working face from the working face overall space-time pressure matrix; S42, obtaining the evaluation result label labeled for each historical area space-time support pressure sub-matrix; S43, training the working face hydraulic support space-time area support quality evaluation model through the plurality of historical area space-time support pressure sub-matrices and the corresponding evaluation result labels.

7. The method of evaluating the quality of regional support of the hydraulic support of the coal face according to claim 1 or 6, characterized in that, The working face hydraulic support space-time area support quality evaluation model is a convolutional neural network model.

8. The method of evaluating the quality of regional support of a hydraulic support of a coal face according to claim 1, characterized in that, The S2 further includes, after constructing the working face overall space-time support pressure matrix according to the column pressure data of each hydraulic support on the working face: updating the working face overall space-time support pressure matrix and the evaluation result of the current hydraulic support quality of each area of the working face every preset time length.

Images