Coal sample gas content detection method and system based on digitization and storage medium
Through the digital sensor array and gas content prediction model, the accuracy of gas content detection and environmental condition monitoring of coal sample are solved, and higher detection accuracy and real-time performance are achieved, reducing detection losses.
Patent Information
- Application Number
- CN202411193058.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-19
AI Technical Summary
The prior art fails to accurately predict the gas content and detection environmental conditions in coal samples, resulting in inaccuracy and reliability of detection data.
The digital-based gas content detection method of coal sample is adopted, and data is collected by setting up a digital sensor array, a gas content prediction model is established, and key data and predictive analysis results are transmitted through dynamic transmission mode to monitor and evaluate operation risks in real time.
It improves the accuracy and reliability of gas content detection, reduces detection losses, and ensures that the real-time transmission of data is not suppressed by non-real-time signals, improving the reliability and real-timeness of digital functions.
Smart Images

Figure CN118980608A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of data detection, and in particular relates to a coal sample gas content detection method, system and storage medium based on digitization. Background Art
[0002] In the production process, after the coal samples are collected, a part of them is used in the desorber to measure the gas desorption amount in the coal samples, and at the same time calculate the gas loss during the sampling process. The remaining coal samples are sent to the laboratory to determine the residual gas content, and finally the gas content in the coal seam is calculated by summing the desorbed gas amount and the residual gas amount. With the development of technology, some automated and digital equipment can improve the efficiency and accuracy of the measurement. In order to accurately analyze the detection of gas content, it is necessary to ensure that various data in the desorber are accurately transmitted in real time.
[0003] Similar prior art includes a Chinese patent application with publication number CN117606976A, which discloses a system for direct underground measurement of coal seam gas content, including a ball mill coal sample tank, a desorption automatic metering device, a data processing mobile terminal and a pneumatic crushing device. The desorption automatic metering device is assembled on the outer end of the ball mill coal sample tank, and a method for direct underground measurement of coal seam gas content is disclosed. The invention determines the mass of the coal sample by a weighing sensor, determines the exposure time of the coal sample by an image recognition module and a micro-flow sensor, and determines the volume of lost gas under standard conditions, the volume of gas before crushing, the volume of gas after crushing and the mass of the coal sample by parameters detected by the micro-flow sensor, the gas pressure sensor and the temperature sensor to obtain the desorbable gas content of the coal seam by the ratio. The desorbable gas content of the coal seam and the residual gas content of the coal are added to obtain the gas content of the coal seam. The entire test process is determined by sensors and data processing mobile terminals to collect, determine and process key data, thereby avoiding low gas content due to human errors. There is also a Chinese patent application with publication number CN108444856A, which proposes a coal mine underground coal seam gas content and pressure rapid measuring instrument and measuring method. The coal mine underground coal seam gas content and pressure rapid measuring instrument includes a host and a coal sample tank. The host includes a measuring pipeline, a flow sensor, a display screen, a keyboard, a battery, a charging data shared serial port and a controller. The flow sensor is arranged on the measuring pipeline, one end of the measuring pipeline is connected to the coal sample tank via a transmission pipeline, the controller signal is connected to the flow sensor, the display screen, the keyboard and the charging data shared serial port, and the battery is electrically connected to the charging data shared serial port. The invention has the beneficial effects of being easy to operate and highly practical. Only 50g of 1 to 3mm coal sample is required to be screened, and the coal seam gas pressure P and the coal seam real gas content W can be accurately measured within 10 minutes at the coal mine underground site, which greatly shortens the measuring cycle of the coal seam gas pressure P and the coal seam real gas content W, and innovates the traditional coal seam gas pressure and content measuring principle and method.
[0004] However, none of the above-mentioned existing technologies accurately predicts the detection process of the gas content in the coal sample, and does not monitor the detection environmental conditions in real time. In actual situations, it is necessary to record the necessary detection data and environmental conditions in real time, and accurately predict the gas content based on the environmental conditions to improve the accuracy and reliability of the data. Summary of the invention
[0005] In order to solve the above problems, the present invention provides a coal sample gas content detection method, system and storage medium based on digitization to solve the problems in the prior art.
[0006] In order to achieve the above-mentioned purpose of the invention, the present invention proposes a coal sample gas content detection method based on digitization, comprising:
[0007] S1: Collect coal samples, set a digital sensor array based on preset tag information, the digital sensor array detects the coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all the collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset tag information;
[0008] S2: the processing unit obtains a desorption rate sequence of gas in the coal sample based on the detection data sequence, establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence, and the gas content prediction model outputs a prediction analysis result of gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than a first preset value, the first time point within the prediction time that is less than the first preset value is set as an observation time point, and the detection is terminated based on the observation time point;
[0009] S3: extracting a standard sensor from the digital sensor array based on the prediction analysis result, adjusting the transmission mode based on the standard sensor, and generating a dynamic transmission mode;
[0010] S4: setting the collected data corresponding to the standard sensor as key data, transmitting the key data and the prediction analysis results to the monitoring platform based on the dynamic transmission mode, and outputting the gas content based on the prediction analysis results to evaluate the operation risk.
[0011] Furthermore, the establishment of the gas content prediction model includes the following steps:
[0012] Based on the preset label information, all the detection data sequences are classified into independent variable sequences and dependent variable sequences, the independent variable sequences are set as training data, and multiple first prediction models are constructed, reference indicators of the first prediction models are calculated, and the first prediction model whose reference indicator is greater than or equal to a second preset value is set as a reference model. The reference model regenerates a second prediction model based on the corresponding reference indicator and sets it as the gas content prediction model.
[0013] Furthermore, generating the first prediction model comprises the following steps:
[0014] Extract the characteristic value of the independent variable sequence and generate the first prediction equation y of the first prediction model. The first prediction equation is: Where I is the total number of feature quantities in the independent variable sequence, x i is the i-th characteristic value of the independent variable sequence, α i is the weight value corresponding to the i-th feature quantity, and c is the reference coefficient.
[0015] Furthermore, the weight value corresponding to the i-th feature quantity is calculated based on the following steps:
[0016] Calculate the weight value α corresponding to the i-th feature quantity based on the first formula i , the first formula is: Where A is the estimated vector of weight values, Y is the transposed column vector of the independent variable sequence, j is a positive integer from 1 to J, J is the total number of time points included in the independent variable sequence, and t j is the j-th time point, B is the maximum likelihood estimate of the independent variable sequence, β is the preset acquisition interval, T is the transposition sign, and e is the exponent.
[0017] Further, the relearning of the reference model to generate a second prediction model based on the reference indicator comprises the following steps:
[0018] Calculate the intra-class variance of the training data in the first prediction model and set it as the reference indicator, construct a hierarchical structure, the hierarchical structure includes multiple layers of hierarchical nodes, set hierarchical conditions in the hierarchical nodes based on the preset label information, set conditional prediction equations in the hierarchical nodes at the bottom layer, obtain a first prediction result based on the first prediction equation, select the corresponding hierarchical conditions based on all the first prediction results, generate a structural path, obtain the corresponding conditional prediction equation in the hierarchical structure based on the structural path, and set it as a standard prediction equation, combine all the standard prediction equations based on the dependent variable sequence, and set them as the second prediction equation of the second prediction model.
[0019] Further, generating the second prediction equation includes:
[0020] Generate a second prediction equation z of the second prediction model, wherein the second prediction equation is: Among them, μ 0 is the value corresponding to the last time point in the dependent variable sequence, n is a positive integer from 1 to N, N is the total number of hierarchical nodes contained in the structural path, y n is the first prediction result of the first prediction model having the same preset label information as the hierarchical condition in the nth hierarchical node, μ n is a reference index of the first prediction model having the same preset label information as the stratification condition in the nth stratification node, and b is an error coefficient;
[0021] Setting the result corresponding to the second prediction equation as the prediction analysis result based on the prediction time point;
[0022] The sensor in the digital sensor array having the same preset label information as the hierarchical node of the structural path is set as the standard sensor.
[0023] Further, the generating of the dynamic transmission mode comprises the following steps:
[0024] A serial transmission path is constructed, wherein the dynamic transmission mode includes an upper control unit, a master unit and a slave unit, wherein the upper control unit sends a command to the master unit based on a communication signal, and the master unit receives and identifies the command. If the command is sent to the slave unit, the communication signal is converted into a compatible signal. The serial transmission path selects a bus transmission path based on the compatible signal and transmits data based on the bus transmission path. If the command is sent to the master unit, the data is transmitted based on the communication signal.
[0025] Furthermore, the monitoring platform is set as the upper-level control unit. If the collected data corresponding to the standard sensor belongs to the independent variable category, the standard sensor is set as the slave unit. If the collected data corresponding to the standard sensor belongs to the dependent variable category, the standard sensor is set as the master unit.
[0026] The present invention also provides a coal sample gas content detection system based on digitization, which is used to implement the above-mentioned coal sample gas content detection method based on digitization. The system mainly includes:
[0027] A sample collection module collects coal samples, sets a digital sensor array based on preset label information, detects the coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all the collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset label information;
[0028] A gas prediction module, wherein the processing unit obtains a desorption rate sequence of gas in the coal sample based on the detection data sequence, establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence, and the gas content prediction model outputs a prediction analysis result of gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than a first preset value, the first time point within the prediction time that is less than the first preset value is set as an observation time point, and the detection is terminated based on the observation time point;
[0029] A dynamic transmission module, which extracts standard sensors from the digital sensor array based on the prediction analysis result, adjusts the transmission mode based on the standard sensors, and generates a dynamic transmission mode;
[0030] The digital detection module sets the collected data corresponding to the standard sensor as key data, transmits the key data and the prediction and analysis results to the monitoring platform based on the dynamic transmission mode, and outputs the gas content based on the prediction and analysis results to evaluate the operation risk.
[0031] The present invention also provides a computer storage medium, which stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the above-mentioned coal sample gas content detection method based on digitization.
[0032] Compared with the prior art, the beneficial effects of the present invention are at least as follows:
[0033] The present invention first sets a digital sensor array to generate collected data, and based on the transmission mode, a detection data sequence corresponding to preset tag information can be obtained. Then, a gas content prediction model is constructed under different stratification conditions, and the prediction and analysis results can be accurately output, thereby improving the accuracy of gas content prediction and reducing detection losses. Finally, key data and prediction and analysis results are transmitted through a dynamic transmission mode, so that signals that need to transmit data in real time will not be suppressed by signals of non-real-time transmission numbers, thereby improving the reliability and real-time performance of the digital function. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 A flow chart of the steps of a coal sample gas content detection method based on digitization of the present invention;
[0035] Figure 2 A schematic diagram of combining and generating a second prediction model in the present invention;
[0036] Figure 3 This is a structural diagram of a coal sample gas content detection system based on digitization according to the present invention. DETAILED DESCRIPTION
[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0038] It is understood that the terms "first", "second", etc. used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script.
[0039] like Figure 1 As shown, the coal sample gas content detection method based on digitization includes:
[0040] S1: Collect coal samples, set up a digital sensor array based on preset label information, the digital sensor array detects coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset label information.
[0041] Specifically, during the detection of coal samples, it is necessary to transmit various sensor data in the gas content meter in real time to realize the digital analysis function and improve the accuracy of gas content detection. At the same time, various sensor data can be predicted and analyzed, and the change process of gas content can be predicted and analyzed to improve the reliability and accuracy of gas content detection. In this embodiment, coal sample collection refers to collecting representative coal samples from the coal seam and immediately putting them into a sealed processing tank after collection to reduce gas loss. The preset label information refers to various sensor types used for measurement, including but not limited to gas, pressure, temperature, image, moisture, flow and weight, etc. The digital sensor array refers to a sensor set corresponding to the preset label information, for example, a gas sensor, a pressure sensor, a temperature sensor, a weighing system, a timer, a mass flow sensor and a moisture determination system, etc. The preset collection interval refers to the measurement interval of each sensor in the digital sensor array, and the collection data refers to the measurement data of each sensor in the digital sensor array. The transmission mode means that the sensors corresponding to different preset label information have different transmission modes, which can make the sensors in the digital sensor array work together so that each collection data can be transmitted quickly and accurately. The digital sensor array can transmit the collected data to the processing unit in real time through the transmission mode, where the processing unit refers to the unit used for data integration and display, which can realize the digital function. The monitoring data sequence refers to the time series data generated by combining and arranging the collected data in sequence based on the preset collection interval.
[0042] S2: The processing unit obtains a desorption rate sequence of gas in the coal sample based on the detection data sequence, and establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence. The gas content prediction model outputs a prediction analysis result of the gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than a first preset value, the first time point in the prediction time that is less than the first preset value is set as the observation time point, and the detection is terminated based on the observation time point.
[0043] Specifically, the gas desorption rate refers to the amount of gas released from the coal sample per unit time under the current conditions. The corresponding desorption rate can be obtained from the gas desorption amount in the detection data sequence, and the analysis rate sequence is generated in sequence, which can clearly describe the change in gas. The gas content prediction model is used to predict the change process of the gas content of the coal sample under different environmental conditions within the prediction time. The prediction analysis result refers to the detection result of the gas content in the coal sample at the prediction time output by the gas content prediction model. The first preset value is used to limit the minimum value of the prediction analysis result. If the prediction analysis results within the prediction time are all less than the first preset value, it means that the detection of the gas content in the coal sample has ended. Therefore, the starting time point in the prediction analysis results that is less than the first preset value is marked, and the detection can be terminated at the starting time point. The predicted starting time point can effectively control the detection process and reduce losses.
[0044] S3: extract standard sensors from the digital sensor array based on the prediction analysis results, adjust the transmission mode based on the standard sensors, and generate a dynamic transmission mode.
[0045] Specifically, standard sensors refer to a set of sensors that must transmit data in real-time detection. Simply put, when detecting gas content under different environmental conditions, there are necessary transmission data and unnecessary transmission data. For example, for a pressure sensor in a digital sensor array, if the collected data corresponding to five consecutive preset collection intervals are all within the same preset interval, the pressure sensor can be set as one of the standard sensors in the first collection of data, and the pressure sensor is not set as one of the standard sensors in the remaining four collections of data. The standard sensor can reduce interference during data transmission and increase the transmission rate of necessary transmission data. Since the standard sensor changes with the changes in the collected data, the transmission mode can be automatically adjusted to form a dynamic transmission mode corresponding to the standard sensor, which can ensure that the signal that needs to transmit data in real time will not be suppressed by the signal of non-real-time transmission, thereby improving the stability and real-time performance of the digital function.
[0046] S4: Set the collected data corresponding to the standard sensor as key data, transmit the key data and prediction analysis results to the monitoring platform based on the dynamic transmission mode, and output the gas content based on the prediction analysis results to evaluate the operation risk.
[0047] Specifically, in this embodiment, the key data and the prediction analysis results are transmitted to the monitoring platform by the dynamic transmission mode, wherein the necessary transmission data includes the gas content and other collected data, and the monitoring platform refers to the mobile monitoring terminal. In this way, the digital coal sample gas content detection process can be realized, and the accuracy and reliability of gas content detection can be improved. At the same time, by adjusting the standard sensor to generate a dynamic transmission mode, the suppression of the transmission signal can be reduced and the real-time performance of data transmission can be improved. Based on the gas content and the prediction analysis results, the gas decomposition rate in the coal sample can be detected in real time. If the decomposition rate is greater than the preset threshold and the gas content is too high, there will be danger during coal seam mining. Therefore, a level assessment of the operation risk is required.
[0048] As a preferred technical solution of the present invention, establishing a gas content prediction model includes the following steps:
[0049] Based on the preset label information, all detection data sequences are classified into independent variable sequences and dependent variable sequences, the independent variable sequences are set as training data, and multiple first prediction models are constructed. The reference indicators of the first prediction models are calculated, and the first prediction model whose reference indicator is greater than or equal to the second preset value is set as the reference model. The reference model regenerates the second prediction model based on the corresponding reference indicator and sets it as the gas content prediction model.
[0050] Specifically, when detecting the gas content in a coal sample, the role of the sensor can be analogized to the variables in the detection experiment. The independent variable is a variable that is actively changed to observe its effect on the result, and the dependent variable is the amount that changes with the change of the independent variable. Therefore, in this embodiment, the independent variable sequence is a detection data sequence composed of the collected data generated in sequence by the sensors corresponding to the independent variable type, for example, the detection data sequence corresponding to the temperature sensor and the pressure sensor is used to simulate the conditions of the underground coal seam, wherein the pressure sensor can measure the pressure in the experimental environment, and its reading may affect the desorption behavior of the gas. Similarly, the dependent variable sequence is a detection data sequence composed of all the collected data generated in sequence by the sensors corresponding to the dependent variable type, for example, the detection data sequence corresponding to the gas sensor and the mass flow sensor, wherein the gas sensor can measure the gas concentration desorbed from the coal sample, which is the direct result of the gas content detection, and the mass flow sensor can measure the gas desorption rate, that is, the amount of gas released per unit time, which is the key measurement indicator of the experiment. The training data is set by the independent variable sequence so that each independent variable sequence has a corresponding first prediction model. The reference index is used to measure the referenceability of the first prediction model. Therefore, the reference model can be screened out from all the first prediction models through the reference index, and then the corresponding reference index is re-learned to generate the second prediction model, namely the gas content prediction model.
[0051] As a preferred technical solution of the present invention, generating the first prediction model includes the following steps:
[0052] The characteristic quantity of the independent variable sequence is extracted to generate the first prediction equation y of the first prediction model. The first prediction equation is: Among them, I is the total number of features in the independent variable sequence, x i is the i-th characteristic value of the independent variable sequence, α i is the weight value corresponding to the i-th feature quantity, and c is the reference coefficient.
[0053] Specifically, in this embodiment, the feature quantity refers to a specific attribute or measurement of the data, which can be a scalar or a vector. In this embodiment, the feature quantity in the form of a vector is directly used, and the feature quantity of the independent variable sequence can be generated through the machine learning model. By constructing the first prediction equation of the first prediction model, the first prediction result can be directly output, that is, the value of the independent variable sequence at the next prediction time point. In the first prediction equation, the weight value of each feature quantity is a key factor in determining the accuracy of the first prediction model. Therefore, the weight value of each feature quantity needs to be calculated later.
[0054] As a preferred technical solution of the present invention, the weight value corresponding to the i-th feature quantity is calculated based on the following steps:
[0055] Calculate the weight value α corresponding to the i-th feature value based on the first formula i , the first formula is: Where A is the estimated vector of weight values, Y is the transposed column vector of the independent variable sequence, j is a positive integer from 1 to J, J is the total number of time points included in the independent variable sequence, and t j is the j-th time point, B is the maximum likelihood estimate of the independent variable sequence, β is the preset collection interval, T is the transposed sign, and e is the exponent.
[0056] Specifically, in this embodiment, the first formula is to calculate the weight value of each feature value in the independent variable sequence by the least square method, where B is a design matrix, expressed as the maximum likelihood estimate, and the dependent variable sequence is an upward edge, which means that the slope of the dependent variable sequence corresponding to the jth time point is greater than 0. On the contrary, the dependent variable sequence is a downward edge, which means that the slope of the dependent variable sequence corresponding to the jth time point is less than or equal to 0. Through the estimated vector A of the weight value, the weight value α of the i-th feature value can be calculated in turn. i .
[0057] As a preferred technical solution of the present invention, relearning the reference model to generate a second prediction model based on the reference indicator includes the following steps:
[0058] The intra-class variance of the training data in the first prediction model is calculated and set as a reference indicator, a hierarchical structure is constructed, the hierarchical structure includes multiple layers of hierarchical nodes, hierarchical conditions are set in the hierarchical nodes based on preset label information, a conditional prediction equation is set in the bottom-level hierarchical nodes, a first prediction result is obtained based on the first prediction equation, a corresponding hierarchical condition is selected based on all the first prediction results, a structural path is generated, a corresponding conditional prediction equation is obtained in the hierarchical structure based on the structural path, and is set as a standard prediction equation, all standard prediction equations are combined based on the dependent variable sequence, and are set as the second prediction equation of the second prediction model.
[0059] Specifically, the intra-class variance usually refers to the variance of data within a specific category or group of the time series. In time series analysis, this may involve grouping the time series according to a certain standard and then calculating the variance within each group. In this embodiment, the training data can be grouped according to the rate of change of the training data, and the intra-class variance of the training data is statistically calculated. First, for each group, the square of the difference between each observation and the mean of the group is calculated, and then the square differences are summed. Finally, the sum is divided by the number of observations in the group (for sample variance) or the number of observations minus one (for unbiased estimation), which is a reference indicator, wherein the observation is the value corresponding to the independent variable sequence in the training data. The hierarchical structure refers to a screening structure for generating a second prediction model, which includes multiple layers of hierarchical nodes, and each hierarchical node is set with a hierarchical condition, wherein the hierarchical condition is a judgment condition for generating a structural path, and each hierarchical node at the bottom layer is set with a conditional prediction equation, and the first prediction result can be calculated by the above first formula, and all the first prediction results contain the values of each independent variable sequence at the next time point. The corresponding hierarchical condition can be selected by each first prediction result, and the structural path is generated in sequence. The conditional prediction equation can be obtained at the hierarchical node corresponding to the bottom layer of the structural path, which is the standard prediction equation. Combining all standard prediction equations can generate the second prediction equation corresponding to the second prediction model. Figure 2 As shown, the stratification condition R11 is set to whether the temperature is greater than T1. If the value of the last time point in the independent variable sequence of the preset label information temperature sensor is greater than T1, the stratification condition R21 is entered. The stratification condition R21 is set to whether the pressure is less than P1. If the value of the last time point in the independent variable sequence of the preset label information pressure sensor is less than P1, the stratification condition R31 is entered. Among them, the setting of each stratification condition can be set according to the value corresponding to the time point with the highest correlation between all detection data sequences. For example, when the independent variable sequence of the preset label information temperature sensor is at temperature T1, the dependent variable sequence of the preset label information mass flow sensor has a mutation, that is, the data change rate is greater than the change threshold, then the temperature T1 is set as a stratification condition.
[0060] As a preferred technical solution of the present invention, generating the second prediction equation includes:
[0061] Generate a second prediction equation z of the second prediction model, the second prediction equation is: Among them, μ 0 is the value corresponding to the last time point in the dependent variable sequence, n is a positive integer from 1 to N, N is the total number of hierarchical nodes contained in the structural path, y n is the first prediction result of the first prediction model with the same preset label information as the hierarchical condition in the nth hierarchical node, μ n is the reference index of the first prediction model having the same preset label information as the stratification condition in the nth stratification node, and b is the error coefficient.
[0062] The result corresponding to the second prediction equation is set as the prediction analysis result based on the prediction time point.
[0063] The sensors in the digital sensor array having the same preset label information as the hierarchical nodes of the structural path are set as standard sensors.
[0064] Specifically, after combining all standard prediction equations, a second prediction equation is generated. In this embodiment, the detection data sequence corresponding to the mass flow sensor is used as the dependent variable sequence. The last time point in the dependent variable sequence refers to the collected data at the current time point, such as Figure 2 As shown, the structural path is (R11→R21→R31→Equation 1), in which the total number of hierarchical nodes contained therein is 3, and the preset label information contained therein is temperature, pressure and granularity. Then, the first prediction model with the same preset label information is extracted from all the first prediction models, and the corresponding first prediction results are calculated. The second prediction equation is used for cumulative combination to calculate the prediction analysis results of the dependent variable sequence at the prediction time point.
[0065] The setting of standard sensors can extract the sensors that need to transmit collected data from the preset tag information, reducing the interference of data communication transmission.
[0066] As a preferred technical solution of the present invention, generating a dynamic transmission mode includes the following steps:
[0067] Construct a serial transmission path. The dynamic transmission mode includes an upper control unit, a master unit and a slave unit. The upper control unit sends a command to the master unit based on a communication signal. The master unit receives and identifies the command. If the command is sent to the slave unit, the communication signal is converted into a compatible signal. The serial transmission path selects a bus transmission path based on the compatible signal and transmits data based on the bus transmission path. If the command is sent to the master unit, the data is transmitted based on the communication signal.
[0068] Specifically, in this embodiment, the upper control unit is a high-level control unit for issuing instructions and receiving data. It usually acts as a central control unit, responsible for sending control commands to other components in the system, such as the master unit and the slave unit, and receiving their responses. The master unit is used to communicate with the upper control unit and manage and control the function of one or more slave units. The slave unit is connected to the master unit through a serial transmission path and is controlled and operated by the master unit. The serial transmission path is a multi-point serial connection method for data transmission between the master unit and the slave unit. This communication method allows multiple devices to communicate through the same physical line, which includes multiple types of bus transmission paths. For signals that require real-time performance, a high-speed bus transmission path will be selected in the serial transmission path for transmission, and for signals that do not require real-time performance, a low-speed bus transmission path will be selected. There is a transmission path determination function in the master unit, and the corresponding bus transmission path can be selected according to whether the compatible signal has real-time performance. If the command is sent to the master unit, it responds to the received communication signal and transmits data, wherein the transmitted data refers to the collected data of the sensor.
[0069] As a preferred technical solution of the present invention, the monitoring platform is set as an upper-level control unit. If the collected data corresponding to the standard sensor belongs to the independent variable category, the standard sensor is set as a slave unit. If the collected data corresponding to the standard sensor belongs to the dependent variable category, the standard sensor is set as a master unit.
[0070] Specifically, in this embodiment, since the standard sensor includes sensors corresponding to the independent variable and the dependent variable, the independent variable category refers to the sensor corresponding to the independent variable, including but not limited to temperature sensors, particle size sensors and pressure sensors, etc., and the dependent variable category refers to the sensor corresponding to the dependent variable, including but not limited to mass flow sensors. The standard sensor corresponding to the dependent variable category is set as the main unit, which can be used to adjust and control the standard sensor corresponding to the independent variable category, and the dependent variable sequence can be transmitted in real time. When the monitoring platform sends a command to the standard sensor corresponding to the slave unit, the above-mentioned bus transmission path can be used to transmit data, ensuring that key real-time signals are processed first, while allowing non-real-time signals to be transmitted without affecting system performance, so as to achieve digital precision control and rapid response.
[0071] like Figure 3 As shown, the present invention also provides a coal sample gas content detection system based on digitization, which is used to implement the above-mentioned coal sample gas content detection method based on digitization, and the system mainly includes:
[0072] The sample collection module collects coal samples, sets a digital sensor array based on preset label information, detects coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset label information.
[0073] Gas prediction module, the processing unit obtains the desorption rate sequence of gas in the coal sample based on the detection data sequence, and establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence. The gas content prediction model outputs the prediction analysis results of the gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than the first preset value, the first time point in the prediction time that is less than the first preset value is set as the observation time point, and the detection is terminated based on the observation time point.
[0074] The dynamic transmission module extracts standard sensors from the digital sensor array based on the prediction analysis results, adjusts the transmission mode based on the standard sensors, and generates a dynamic transmission mode.
[0075] The digital detection module sets the collected data corresponding to the standard sensor as key data, transmits the key data and predictive analysis results to the monitoring platform based on the dynamic transmission mode, and outputs the gas content based on the predictive analysis results to evaluate the operation risk.
[0076] The present invention also provides a computer storage medium, which stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the above-mentioned coal sample gas content detection method based on digitization.
[0077] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.
[0078] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0079] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0080] The above embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.
[0081] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A coal sample gas content detection method based on digitalization, characterized in that: The method comprises the following steps: S1: Collect coal samples, set a digital sensor array based on preset tag information, the digital sensor array detects the coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all the collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset tag information; S2: the processing unit obtains a desorption rate sequence of gas in the coal sample based on the detection data sequence, establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence, and the gas content prediction model outputs a prediction analysis result of gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than a first preset value, the first time point within the prediction time that is less than the first preset value is set as an observation time point, and the detection is terminated based on the observation time point; S3: extracting a standard sensor from the digital sensor array based on the prediction analysis result, adjusting the transmission mode based on the standard sensor, and generating a dynamic transmission mode; S4: setting the collected data corresponding to the standard sensor as key data, transmitting the key data and the prediction analysis results to the monitoring platform based on the dynamic transmission mode, and outputting the gas content based on the prediction analysis results to evaluate the operation risk.
2. The method according to claim 1, characterized in that The gas content prediction model is established by the following steps: Based on the preset label information, all the detection data sequences are classified into independent variable sequences and dependent variable sequences, the independent variable sequences are set as training data, and multiple first prediction models are constructed, reference indicators of the first prediction models are calculated, and the first prediction model whose reference indicator is greater than or equal to a second preset value is set as a reference model. The reference model regenerates a second prediction model based on the corresponding reference indicator and sets it as the gas content prediction model.
3. The method according to claim 2, characterized in that Generating the first prediction model comprises the following steps: Extract the characteristic value of the independent variable sequence and generate the first prediction equation y of the first prediction model. The first prediction equation is: Where I is the total number of feature quantities in the independent variable sequence, x i is the i-th characteristic value of the independent variable sequence, α i is the weight value corresponding to the i-th feature quantity, and c is the reference coefficient.
4. The method according to claim 3, characterized in that The weight value corresponding to the i-th feature quantity is calculated based on the following steps: Calculate the weight value α corresponding to the i-th feature quantity based on the first formula i , the first formula is: Where A is the estimated vector of weight values, Y is the transposed column vector of the independent variable sequence, j is a positive integer from 1 to J, J is the total number of time points included in the independent variable sequence, and t j is the j-th time point, B is the maximum likelihood estimate of the independent variable sequence, β is the preset acquisition interval, T is the transposition sign, and e is the exponent.
5. The method according to claim 2, characterized in that: The relearning of the reference model to generate a second prediction model based on the reference indicator comprises the following steps: Calculate the intra-class variance of the training data in the first prediction model and set it as the reference indicator, construct a hierarchical structure, the hierarchical structure includes multiple layers of hierarchical nodes, set hierarchical conditions in the hierarchical nodes based on the preset label information, set conditional prediction equations in the hierarchical nodes at the bottom layer, obtain a first prediction result based on the first prediction equation, select the corresponding hierarchical conditions based on all the first prediction results, generate a structural path, obtain the corresponding conditional prediction equation in the hierarchical structure based on the structural path, and set it as a standard prediction equation, combine all the standard prediction equations based on the dependent variable sequence, and set them as the second prediction equation of the second prediction model.
6. The method according to claim 5, characterized in that Generating the second prediction equation includes: Generate a second prediction equation z of the second prediction model, wherein the second prediction equation is: Wherein, μ0 is the value corresponding to the last time point in the dependent variable sequence, n is a positive integer from 1 to N, N is the total number of hierarchical nodes contained in the structural path, and y n is the first prediction result of the first prediction model having the same preset label information as the hierarchical condition in the nth hierarchical node, μ n is a reference index of the first prediction model having the same preset label information as the stratification condition in the nth stratification node, and b is an error coefficient; Setting the result corresponding to the second prediction equation as the prediction analysis result based on the prediction time point; The sensor in the digital sensor array having the same preset label information as the hierarchical node of the structural path is set as the standard sensor.
7. The method according to claim 1, characterized in that Generating a dynamic transmission mode comprises the following steps: A serial transmission path is constructed, wherein the dynamic transmission mode includes an upper control unit, a master unit and a slave unit, wherein the upper control unit sends a command to the master unit based on a communication signal, and the master unit receives and identifies the command. If the command is sent to the slave unit, the communication signal is converted into a compatible signal. The serial transmission path selects a bus transmission path based on the compatible signal and transmits data based on the bus transmission path. If the command is sent to the master unit, the data is transmitted based on the communication signal.
8. The method according to claim 7, characterized in that The monitoring platform is set as the upper control unit. If the collected data corresponding to the standard sensor belongs to the independent variable category, the standard sensor is set as the slave unit. If the collected data corresponding to the standard sensor belongs to the dependent variable category, the standard sensor is set as the master unit.
9. A coal sample gas content detection system based on digitization, used to implement the method according to any one of claims 1 to 8, characterized in that: The system includes the following modules: A sample collection module collects coal samples, sets a digital sensor array based on preset label information, detects the coal samples based on preset collection intervals, generates collection data, sets a transmission mode, transmits all the collection data to a processing unit based on the transmission mode, and sequentially combines to generate a detection data sequence corresponding to the preset label information; A gas prediction module, wherein the processing unit obtains a desorption rate sequence of gas in the coal sample based on the detection data sequence, establishes a gas content prediction model based on the detection data sequence and the desorption rate sequence, and the gas content prediction model outputs a prediction analysis result of gas content based on the desorption rate sequence of gas in the coal sample. If the prediction analysis results within the prediction time are all less than a first preset value, the first time point within the prediction time that is less than the first preset value is set as an observation time point, and the detection is terminated based on the observation time point; A dynamic transmission module, which extracts standard sensors from the digital sensor array based on the prediction analysis result, adjusts the transmission mode based on the standard sensors, and generates a dynamic transmission mode; The digital detection module sets the collected data corresponding to the standard sensor as key data, transmits the key data and the prediction and analysis results to the monitoring platform based on the dynamic transmission mode, and outputs the gas content based on the prediction and analysis results to evaluate the operation risk.
10. A computer storage medium, characterized in that: The computer storage medium stores program instructions, wherein when the program instructions are executed, the device where the computer storage medium is located is controlled to execute the method according to any one of claims 1 to 8.
Citation Information
Patent Citations
Underground coal mine coal seam gas-bearing capacity and pressure rapid measuring instrument and measuring method
CN108444856A
Underground direct measuring system and method for coal seam gas content
CN117606976A
Cited By
Method and system for monitoring change of coal reserve of coal storage yard based on three-dimensional digital twinning
CN121147781A
A coal storage change monitoring method and system based on three-dimensional digital twinning
CN121147781B
