Coal spontaneous combustion temperature prediction method based on ACO-BP neural network model

Abstract

The invention relates to the technical field of coal mine safety monitoring, in particular to a coal spontaneous combustion temperature prediction method based on an ACO-BP neural network model. The method comprises the following steps: carrying out gas collection on a space in which coal is located through gas monitoring equipment at preset time intervals to obtain gas corresponding to the coal; component and concentration analysis is conducted on the gas through a gas analyzer, a gas concentration data set is obtained, and the gas concentration data set comprises concentration data corresponding to at least one gas; and inputting the gas concentration data set into a target ACO-BP neural network model for identification processing, predicting a first temperature corresponding to the coal, and determining spontaneous combustion information of the coal according to the first temperature. According to the invention, the accuracy and convenience of coal spontaneous combustion temperature prediction can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of coal mine safety monitoring, and particularly to a method for predicting coal spontaneous combustion temperature based on an ACO-BP neural network model. Background Art

[0002] Coal spontaneous combustion is a major hidden danger in coal mine safety production. The resulting fire accidents not only cause serious waste of coal resources, but also generate a large amount of toxic and harmful gases, seriously threatening the lives of miners and the ecological environment of the mine. With the development of science and technology, for example, coal spontaneous combustion can be monitored. Therefore, how to accurately predict the coal spontaneous combustion temperature has become the focus of coal mine safety production research. Summary of the Invention

[0003] The present disclosure provides a method for predicting coal spontaneous combustion temperature based on an ant colony optimization (ACO)-back propagation (BP) neural network model, which can improve the accuracy and convenience of coal spontaneous combustion temperature prediction. The technical solution of the present disclosure is as follows: According to the first aspect of the embodiments of the present disclosure, a method for predicting coal spontaneous combustion temperature based on an ACO-BP neural network model is provided, including: Collect gas from the space where the coal is located through a gas monitoring device at preset intervals to obtain the gas corresponding to the coal; Analyze the composition and concentration of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; Input the gas concentration data set into a target ACO-BP neural network model for recognition processing, predict the first temperature corresponding to the coal, and determine the spontaneous combustion information of the coal according to the first temperature.

[0004] According to some embodiments, the method further includes: Obtain the preprocessing method corresponding to each gas concentration data in the gas concentration data set; Process each gas concentration data by using the preprocessing method to obtain a processed gas concentration data set.

[0005] According to some embodiments, the determining the spontaneous combustion information of the coal according to the first temperature includes: Obtain the attribute information corresponding to the coal; Determine the temperature change information of the coal according to the attribute information; Obtain the stage at which the coal spontaneous combustion occurs according to the temperature change information and the first temperature.

[0006] According to some embodiments, the method further includes: Obtain a training data set, perform normalization processing on the training data set, and obtain a processed training data set; Obtain the topological structure of the initial BP neural network model, initialize the first weight matrix and the first threshold vector of the BP neural network model, and obtain the initial BP neural network model; Perform iterative operations on the ant colony algorithm using the processed training data set to obtain the pheromone content of the ants; According to the highest pheromone content, obtain the displacement information corresponding to the ants, and update the first position of the ants according to the displacement information and the path selection probability to obtain the second position; When the iterative information of the ant colony algorithm meets the information requirements, use the target ant position corresponding to the ant colony algorithm as the second weight matrix and the second threshold vector; Update the initial BP neural network model using the second weight matrix and the second threshold vector to obtain an updated BP neural network model; Train the updated BP neural network model using the processed training data set, and when the updated BP neural network model meets the model training requirements, obtain a target ACO-BP neural network model, and use the target ACO-BP neural network model as the target ACO-BP neural network model.

[0007] According to some embodiments, the method further includes: Obtain the prediction accuracy corresponding to the coal; Determine the number of hidden layer nodes of the target ACO-BP neural network model according to the prediction accuracy, the number of first nodes in the input layer, and the number of second nodes in the output layer.

[0008] According to some embodiments, the method further includes: Obtain the warning level corresponding to the spontaneous combustion information of the coal; Obtain and execute the warning measures corresponding to the warning level, and display the spontaneous combustion information of the coal and the warning level.

[0009] According to some embodiments, after obtaining the gas concentration data set, it further includes: Update the first weight matrix and the first threshold vector of the initial BP neural network model using a genetic algorithm to obtain a target GA-BP neural network model; Input the gas concentration data set into a target Genetic Algorithm (GA)-BP neural network model for identification processing to predict the second temperature corresponding to the coal, and determine the spontaneous combustion information of the coal based on the second temperature.

[0010] According to a second aspect of the embodiments of the present disclosure, there is provided a device for predicting the spontaneous combustion temperature of coal based on an ACO-BP neural network model, including: A gas acquisition unit, configured to collect gas in the space where the coal is located through a gas monitoring device at preset time intervals to obtain the gas corresponding to the coal; A set acquisition unit, configured to analyze the components and concentrations of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; A temperature prediction unit, configured to input the gas concentration data set into a target ACO-BP neural network model for identification processing to predict the first temperature corresponding to the coal, and determine the spontaneous combustion information of the coal based on the first temperature.

[0011] According to a third aspect of the embodiments of the present disclosure, there is provided a network device, including: A processor; A memory for storing executable instructions of the processor; Wherein, the processor is configured to execute the instructions to implement the method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model according to any one of the foregoing aspects.

[0012] According to a fourth aspect of the embodiments of the present disclosure, there is provided a storage medium, when the instructions in the storage medium are executed by a processor of a network device, enabling the network device to execute the method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model according to any one of the foregoing aspects.

[0013] According to a fifth aspect of the embodiments of the present disclosure, there is provided a computer program product, including a computer program, where the computer program implements the method according to any one of the foregoing aspects when executed by a processor.

[0014] The technical solutions provided by the embodiments of the present disclosure at least bring the following beneficial effects: In some or related embodiments, gas is collected from the space where the coal is located through a gas monitoring device at preset time intervals to obtain the gas corresponding to the coal; the gas is analyzed for its composition and concentration by a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; the gas concentration data set is input into a target ACO-BP neural network model for identification processing to predict the first temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined based on the first temperature. Therefore, using the target ACO-BP neural network model to predict the spontaneous combustion temperature of coal can reduce the situation where the BP neural network model falls into a local optimal solution due to the random setting of the initial weights and thresholds of the BP neural network model during identification processing, resulting in poor accuracy in predicting the spontaneous combustion temperature of coal. Compared with the BP neural network model, the ACO-BP neural network model can improve the convergence speed, can improve the prediction accuracy and prediction stability of the ACO-BP neural network model, and does not require building a complex test platform underground, can improve the prediction efficiency, and can improve the accuracy and convenience of predicting the spontaneous combustion temperature of coal.

[0015] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present disclosure. Brief Description of the Drawings

[0016] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure and do not constitute an improper limitation of the present disclosure.

[0017] Figure 1 is a flowchart of the first method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model provided by an embodiment of the present disclosure; Figure 2 is a flowchart of the second method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model provided by an embodiment of the present disclosure; Figure 3 is an example schematic diagram of a BP neural network model provided by an embodiment of the present disclosure; Figure 4 is an example schematic diagram of an ant colony algorithm provided by an embodiment of the present disclosure; Figure 5 is an example schematic diagram of the structure of a BP neural network model provided by an embodiment of the present disclosure; Figure 6 is a process schematic diagram of a BP neural network model provided by an embodiment of the present disclosure; Figure 7 is an example schematic diagram of a prediction network model provided by an embodiment of the present disclosure; Figure 8 It is a flowchart of the third coal spontaneous combustion temperature prediction method based on the ACO-BP neural network model provided by the embodiments of the present disclosure; Figure 9 It is a schematic diagram of a coal spontaneous combustion temperature prediction device based on the ACO-BP neural network model shown according to an exemplary embodiment; Figure 10 It is a block diagram of a network device shown according to an exemplary embodiment. Detailed implementation manners

[0018] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0019] The embodiments of the present disclosure propose a coal spontaneous combustion temperature prediction method, device, network device, and storage medium based on the ACO-BP neural network model. In some embodiments, the coal spontaneous combustion temperature prediction method based on the ACO-BP neural network model can be mutually replaced with terms such as information processing methods and communication methods, the coal spontaneous combustion temperature prediction device based on the ACO-BP neural network model can be mutually replaced with terms such as information processing devices and communication devices, and terms such as information processing systems and communication systems can be mutually replaced.

[0020] The embodiments of the present disclosure are not exhaustive, but only for the illustration of some embodiments, and do not constitute a specific limitation on the protection scope of the present disclosure. Without contradiction, each step in a certain embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, the solution after removing some steps in a certain embodiment can also be implemented as an independent embodiment, and the order of the steps in a certain embodiment can be arbitrarily exchanged. Additionally, the optional implementation manners in a certain embodiment can be arbitrarily combined; furthermore, the embodiments can be arbitrarily combined. For example, parts or all of the steps of different embodiments can be arbitrarily combined, and a certain embodiment can be arbitrarily combined with the optional implementation manners of other embodiments.

[0021] In each embodiment of the present disclosure, if there is no special description and logical conflict, the terms and / or descriptions among the embodiments are consistent and can be mutually referred to, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0022] The terms used in the embodiments of the present disclosure are only for the purpose of describing specific embodiments, and are not intended as a limitation on the present disclosure.

[0023] In the embodiments of the present disclosure, unless otherwise specified, elements expressed in the singular form, such as "a", "an", "the", "above-mentioned", "said", "aforementioned", "this", etc., may mean "one and only one", or may also mean "one or more", "at least one", etc. For example, in the case of using articles such as "a", "an", "the" in English translation, the noun after the article can be understood as a singular expression or a plural expression.

[0024] In the embodiments of the present disclosure, "a plurality of" means two or more.

[0025] In some embodiments, terms such as "at least one of", "one or more", "a plurality of", "multiple", etc. can be replaced with each other.

[0026] The prefix words such as "first", "second", etc. in the embodiments of the present disclosure are only used to distinguish different described objects, and do not constitute a limitation on the position, order, priority, quantity or content of the described objects. The statements of the described objects refer to the description in the context of the claims or embodiments, and should not constitute an unnecessary limitation due to the use of prefix words. For example, if the described object is "field", the ordinal numbers before "field" in "the first field" and "the second field" do not limit the position or order between the "fields", and "first" and "second" do not limit whether the "fields" they modify are in the same message, nor do they limit the order of "the first field" and "the second field". For another example, if the described object is "level", the ordinal numbers before "level" in "the first level" and "the second level" do not limit the priority between the "levels". For another example, the quantity of the described object is not limited by the ordinal number, and can be one or more. Taking "the first device" as an example, the quantity of "device" therein can be one or more. In addition, the objects modified by different prefix words can be the same or different. For example, if the described object is "device", "the first device" and "the second device" can be the same device or different devices, and their types can be the same or different; for another example, if the described object is "information", "the first information" and "the second information" can be the same information or different information, and their contents can be the same or different.

[0027] In some embodiments, a "terminal" or "terminal device" may be referred to as a "user equipment (UE)", "user terminal", "mobile station (MS)", "mobile terminal (MT)", subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, etc.

[0028] In some embodiments, data, information, etc. may be obtained after obtaining the consent of the user.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described here can be implemented in an order other than those illustrated or described here. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0030] According to some embodiments, the coal resources in certain regions are rich, and the main coal seams in most mines have characteristics such as shallow burial, occurrence of (extremely) close coal seam groups, and occurrence of (extraordinarily) thick and (easily) self-igniting coal seams. During the coal mining process, as the mining scale expands and the production level extends, the underground environment becomes increasingly complex, creating conditions for coal spontaneous combustion and thus triggering a series of serious problems.

[0031] Among them, the occurrence of coal spontaneous combustion is the result of the combined action of multiple factors. There is a large amount of residual coal in the gob area and it is widely distributed. This residual coal comes into contact with air and continuously accumulates heat during the slow oxidation process. At the same time, there is a good heat storage environment in the gob area, and the ventilation conditions are not good, making it difficult for the heat to dissipate, resulting in a continuous increase in the temperature of the coal body. In addition, the characteristics of the coal itself are also crucial. In some areas with (extra) thick and (easy) spontaneous combustion coal seams, the coal quality has a lower ignition point and stronger oxidation activity, further accelerating the process of spontaneous combustion. When these factors are superimposed and reach certain conditions, the phenomenon of coal spontaneous combustion will inevitably occur.

[0032] According to some embodiments, the hazards brought about by coal spontaneous combustion are multifaceted and extremely serious. In terms of personnel safety, a large amount of toxic and harmful gases, such as carbon monoxide, hydrogen sulfide, etc., will be released during the process of coal spontaneous combustion. Once these gases accumulate underground, it is extremely easy to cause personnel poisoning and asphyxiation, seriously threatening the lives of miners. At the same time, the fire caused by coal spontaneous combustion may also lead to secondary disasters such as the collapse of underground roadways and the caving of roof, further increasing the risk of casualties. In terms of environmental impact, the high temperature generated by coal spontaneous combustion will damage the underground rock formation structure, resulting in surface subsidence and collapse, forming a large area of subsidence areas, destroying surface vegetation and cultivated land, and affecting the local ecological balance. Moreover, the harmful gases generated by combustion are discharged into the atmosphere, causing serious air pollution and exacerbating environmental problems such as haze, posing a long-term harm to the physical health of surrounding residents. From the perspective of coal production, coal spontaneous combustion will cause a large amount of waste of coal resources, reduce the coal mining efficiency and recovery rate. The fire may also damage the underground production equipment and facilities, interrupt the normal production operation, bring huge economic losses to coal enterprises, and seriously restrict the sustainable development of the coal industry. Therefore, the accurate prediction of coal spontaneous combustion temperature is an effective method for preventing and controlling coal spontaneous combustion accidents in the gob area and ensuring the safe progress of the production process.

[0033] According to some embodiments, for example, a machine learning prediction model using an optimization algorithm can be used for predicting the coal spontaneous combustion temperature. However, due to problems such as the random setting of the initial weights and thresholds of the BP neural network, it is easy to fall into a local optimal solution, and the slow convergence speed, the prediction accuracy is relatively low.

[0034] Figure 1 It is a flowchart of the first method for predicting the coal spontaneous combustion temperature based on the ACO-BP neural network model provided by the embodiments of the present disclosure. As Figure 1 shown, the method for predicting the coal spontaneous combustion temperature based on the ACO-BP neural network model can be used in the scenario of determining whether coal spontaneous combustion occurs, and includes the following steps: In step S11, the gas in the space where the coal is located is collected once every preset time period through a gas monitoring device to obtain the gas corresponding to the coal; According to some embodiments, the execution entity of the embodiments of the present disclosure may be, for example, a server. This server does not specifically refer to a certain fixed server. For example, when the server identifier changes, the server can also change accordingly. For example, when the structure or components of the server change, the server can also change accordingly. Among them, the server in the embodiments of the present disclosure may be, for example, a single server or a server cluster, and the embodiments of the present disclosure do not limit this.

[0035] In some embodiments, the preset duration may be, for example, the duration between two adjacent gas samplings. This preset duration does not specifically refer to a certain fixed duration. The preset duration may be, for example, a certain value between 5 minutes and 10 minutes. Among them, the preset duration may be, for example, variable. For example, when the self-ignition stage of the coal changes, the preset duration can also change accordingly. For example, the higher the coal temperature, the shorter the preset duration. For example, different coals may correspond to different preset durations. The preset duration may also be determined according to a duration setting instruction. The duration setting instruction does not specifically refer to a certain fixed instruction. The duration setting instruction includes but is not limited to a click setting instruction, a voice setting instruction, or a timing setting instruction, etc.

[0036] In some embodiments, the gas monitoring device may be, for example, a device for collecting the gas around the coal. This gas monitoring device does not specifically refer to a certain fixed device. For example, when the modules included in the gas monitoring device change, the gas monitoring device can also change accordingly. For example, when the device identifier of the gas monitoring device changes, the gas monitoring device can also change accordingly.

[0037] In some embodiments, the gas corresponding to the coal may be, for example, the gas around the coal. This gas may include, for example, air and the gas generated by the coal during the heating process. The gas corresponding to the coal does not specifically refer to a certain fixed gas. For example, when the gas type corresponding to the gas changes, the gas corresponding to the coal can also change accordingly. For example, when the concentration of each gas type corresponding to the gas changes, the gas corresponding to the coal can also change accordingly. Among them, the coal can also be referred to as a coal sample, a coal block, etc., and the embodiments of the present disclosure do not limit this.

[0038] According to some embodiments, the gas in the space where the coal is located is sampled once every preset duration through the gas monitoring device to obtain the gas corresponding to the coal.

[0039] In step S12, the gas is analyzed for its composition and concentration by a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes the concentration data corresponding to at least one gas; According to some embodiments, a gas analyzer can be, for example, a device for analyzing the composition and concentration of gases. The gas analyzer does not specifically refer to a certain fixed device. For example, when the types of gases that the gas analyzer can analyze change, the gas analyzer can also change accordingly. For example, when the instrument identification of the gas analyzer changes, the gas analyzer can also change accordingly.

[0040] In some examples, a gas concentration data set can be, for example, a collective formed by aggregating at least one gas concentration data. The gas concentration data set does not specifically refer to a certain fixed set. For example, when the amount of data corresponding to the gas concentration data set changes, the gas concentration data set can also change accordingly. For example, when a certain gas concentration data in the gas concentration data set changes, the gas concentration data set can also change accordingly.

[0041] In some embodiments, the gas concentration data set can include concentration data corresponding to at least one gas. Among them, the at least one gas can include, for example, one or more of oxygen O2, carbon monoxide CO, carbon dioxide CO2, methane CH4, ethylene C2H4, ethane C2H6, nitrogen N2.

[0042] In some embodiments, the gas can be analyzed for its composition and concentration by a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas.

[0043] In step S13, the gas concentration data set is input into the target ACO - BP neural network model for identification processing to predict the first temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined according to the first temperature.

[0044] In some embodiments, the target ACO - BP neural network model can be, for example, a model that has been trained and can be used for predicting the spontaneous combustion temperature of coal. The target ACO - BP neural network model does not specifically refer to a certain fixed model. For example, when the model parameters corresponding to the target ACO - BP neural network model change, the target ACO - BP neural network model can also change accordingly. For example, when the topological structure corresponding to the target ACO - BP neural network model changes, the target ACO - BP neural network model can also change accordingly.

[0045] In some embodiments, the first temperature can be, for example, the temperature obtained by predicting the temperature using the target ACO-BP neural network model based on the currently collected gas. The first temperature does not specifically refer to a certain fixed temperature. For example, when the target ACO-BP neural network model or the gas concentration data set changes, the first temperature can also change accordingly. When the specific value corresponding to the first temperature changes, the first temperature can also change accordingly.

[0046] In some embodiments, the gas concentration data set can be input into the target ACO-BP neural network model for identification processing to predict the first temperature corresponding to the coal, and the spontaneous combustion information of the coal can be determined based on the first temperature.

[0047] In some or related embodiments, the gas in the space where the coal is located is collected once every preset time period through a gas monitoring device to obtain the gas corresponding to the coal; the gas is analyzed for its composition and concentration by a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes the concentration data corresponding to at least one gas; the gas concentration data set is input into the target ACO-BP neural network model for identification processing to predict the first temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined based on the first temperature. Therefore, using the target ACO-BP neural network model to predict the spontaneous combustion temperature of coal can reduce the situation where the BP neural network model falls into a local optimal solution due to the random setting of the initial weights and thresholds of the BP neural network model during the identification process, resulting in poor accuracy of predicting the spontaneous combustion temperature of coal. Compared with the BP neural network model, the ACO-BP neural network model can improve the convergence speed, can improve the prediction accuracy and prediction stability of the ACO-BP neural network model, and does not require building a complex test platform underground, which can improve the accuracy and convenience of predicting the spontaneous combustion temperature of coal.

[0048] Figure 2 is a flowchart of the second method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model provided by the embodiments of the present disclosure. As Figure 2 shown, the method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model can be used in the scenario of determining whether the coal is spontaneously combusted, and includes the following steps: In step S21, the gas in the space where the coal is located is collected once every preset time period through a gas monitoring device to obtain the gas corresponding to the coal; Among them, the relevant descriptions can be as described above, and will not be repeated here.

[0049] In step S22, the gas is analyzed for its composition and concentration by a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes the concentration data corresponding to at least one gas; Among them, the related descriptions can be as described above and will not be elaborated here.

[0050] According to some embodiments, the selection of gas types can be, for example: Coal spontaneous combustion is a complex physical and chemical process, and different gases are generated at different stages. The decrease in O2 concentration can intuitively reflect the degree of oxidation reaction between coal and oxygen; A large amount of CO is generated at the early stage of low-temperature oxidation of coal and is an important sign of early coal spontaneous combustion; CO2 is one of the main products of coal oxidation, and the change in its concentration can reflect the intensity of the oxidation reaction; Hydrocarbon gases such as CH4, C2H4, and C2H6 appear only at a certain stage of the development of coal spontaneous combustion, and the increase in their concentration indicates that the degree of spontaneous combustion is intensifying. By monitoring the concentrations of these gases, the stage of coal spontaneous combustion can be comprehensively understood, and then the change in coal temperature can be predicted.

[0051] Therefore, there is an inherent relationship between the concentrations of these gases and the coal temperature. As the coal temperature rises, the oxidation reaction of coal intensifies, leading to a faster consumption of O2 and an increase in oxidation products such as CO. When the coal temperature reaches a certain threshold, a certain gas begins to be generated in large quantities. By analyzing the change rules of the concentrations of these gases, a mathematical relationship model with the coal temperature is established, and the coal temperature is predicted using the gas concentration data to improve the accuracy of coal temperature prediction.

[0052] According to some embodiments, the method further includes: Obtain the preprocessing methods corresponding to the gas concentration data in the gas concentration data set; Process each gas concentration data using the preprocessing method to obtain a processed gas concentration data set.

[0053] In some embodiments, the gas monitoring device and the gas analyzer can be arranged, for example, at the preset positions of the coal with the temperature to be measured. The gas analyzer can wirelessly transmit the data to the server. Among them, the gas monitoring device can be, for example, gas concentration sensors (O2, CO, CO2, etc.) and temperature sensors arranged on both sides of the gob roadway and the transverse working face. Among them, different gas concentration data can correspond to different formats.

[0054] In step S23, input the gas concentration data set into the target ACO-BP neural network model for recognition processing to predict the first temperature corresponding to the coal; Among them, the related descriptions can be as described above and will not be elaborated here.

[0055] According to some embodiments, the target ACO-BP neural network model can be deployed to the server. The server collects the gas concentration data of each monitoring point once every 5 - 10 minutes (can be set), forms a feature vector and inputs it into the model. The target ACO-BP neural network model outputs the predicted coal temperature value, and combines the results of multiple monitoring points in the area (such as taking the maximum value) to determine the current predicted temperature of the area, that is, the first temperature.

[0056] According to some embodiments, the method further includes: Obtain a training data set, and perform normalization processing on the training data set to obtain a processed training data set; Obtain the topological structure of the initial BP neural network model, and initialize the first weight matrix and the first threshold vector of the BP neural network model to obtain the initial BP neural network model; Perform iterative operations on the ant colony algorithm using the processed training data set to obtain the pheromone content of the ants; According to the highest pheromone content, obtain the displacement information corresponding to the ants, and update the first position of the ants according to the displacement information and the path selection probability to obtain the second position; When the iterative information of the ant colony algorithm meets the information requirements, use the target ant position corresponding to the ant colony algorithm as the second weight matrix and the second threshold vector; Use the second weight matrix and the second threshold vector to update the initial BP neural network model to obtain an updated BP neural network model; Train the updated BP neural network model using the processed training data set, and when the updated BP neural network model meets the model training requirements, obtain the target ACO-BP neural network model, and use the target ACO-BP neural network model as the target ACO-BP neural network model. Therefore, the ant colony optimization algorithm can be used to optimize the BP neural network model, which can improve the accuracy of temperature prediction, avoid defects such as low prediction accuracy caused by randomly obtaining the initial weights and thresholds of the model, and its directional search ability can reduce the number of model iterations, improve the convergence speed, greatly shorten the training cycle, and improve the speed and accuracy of model prediction.

[0057] According to some embodiments, before performing iterative operations on the ant colony algorithm using the processed training data set, for example, the parameters of the ant colony algorithm can be initialized.

[0058] According to some embodiments, Figure 3 is an example schematic diagram of a BP neural network model provided by an embodiment of the present disclosure, such as Figure 3As shown, the BP neural network is a multi-layer feedforward network based on error backpropagation training, consisting of non-linear transformation units. By simulating the way the human brain understands and processes information, it continuously adjusts the network connection weights and thresholds to obtain a non-linear mapping between the input layer and the output layer. The BP neural network model consists of an input layer, an output layer, and several hidden layers. Among them, increasing the number of hidden layers can reduce the network error and improve the accuracy, but it also complicates the network, increases the network training time and the risk of overfitting. Therefore, only one hidden layer can be set, and by changing the number of hidden layer nodes, the model parameters can be made close to their optimal network parameters.

[0059] In some embodiments, Figure 4 is an example schematic diagram of an ant colony algorithm provided by an embodiment of the present disclosure. As Figure 4 shown, the ant colony optimization algorithm is a bionic intelligent optimization algorithm that simulates the foraging behavior of ants. In the foraging actions of natural ant populations, the law of their path selection is generally random. They perceive the pheromone concentration on the ground in front of them, analyze the information left by other ants in the population, and are more likely to forage along the path with a higher pheromone concentration. Each ant has the ability to release pheromones to exchange information with other ants. Because the number of round trips of ants on the foraging route that is closer is often more, and the round-trip efficiency is higher, the concentration of the pheromones they leave behind is higher, guiding the ant population to forage along the shorter route. So when an ant encounters a fork in the road during foraging, the way it chooses a route is no longer random, and it is more likely to move along the shortest foraging route. Pheromones act as positive feedback, making more ants in the population have higher foraging efficiency and being able to continuously optimize and update the shortest route information. And as time goes by, the concentration of the pheromones left on the remaining non-optimal routes will gradually decrease. Eventually, all the ants in the population choose to move along the shortest foraging route.

[0060] In some embodiments, Figure 5 is an example schematic diagram of the structure of a BP neural network model provided by an embodiment of the present disclosure. Figure 6 is a schematic flow diagram of a BP neural network model provided by an embodiment of the present disclosure. As Figure 5 and Figure 6 shown, the ant colony optimization algorithm can be used to optimize the BP neural network. Specifically, for example, the weight matrix and the threshold vector can be regarded as the route coordinates of the ant population. According to the situation that the closer the route distance of the ant from the ant nest to the target food is, the larger the value of the pheromone concentration left on its foraging route is, the mean square error of the optimization result is regarded as the fitness value of the ant population.

[0061] (1) In the formula, MSE TrainingSetRefers to the mean squared error (MSE) trained using the training set samples TestingSet Refers to the mean squared error trained using the test set samples.

[0062] Therefore, the shortest foraging route selected by the last ant population is regarded as the optimal initial weights and thresholds of the prediction model. The optimal initial weights and thresholds obtained by optimizing the ant colony optimization algorithm are substituted into the BP neural network prediction model to obtain the ACO-BP neural network model.

[0063] According to some embodiments, such as Figure 6 As shown, it may specifically include: (1) Read the data and perform an initialization process on the parameters of the BP neural network structure and the ant colony optimization algorithm, including the maximum number of evolutionary generations, the number and value range of independent variables, the evaporation coefficient and total release amount of pheromone, the transfer probability constant, the number of ants, etc.

[0064] (2) Analyze and calculate the dimension of the solution space, and initialize the highest pheromone value and the initial ant coordinates. Let the ant colony algorithm perform iterative optimization. After the initial ant position, the probability of the next movement position of the ant is shown in formula (2): (2) In the formula, is the pheromone at (i, j) for the ant; refers to the heuristic factor for the ant to transfer from i to j; a k refers to the set of nodes to be visited by ant k; α refers to the pheromone importance factor; β refers to the heuristic function importance degree factor.

[0065] (3) Calculate the pheromone content according to the position of the ant. The way to update the pheromone is shown in formula (3) (3) In the formula, refers to the pheromone content released by the k-th ant between i and j.

[0066] (4) Calculate the highest pheromone and update the optimal individual position.

[0067] (5) Transfer according to the probability sum and update the ant position.

[0068] (6) Execute the loop from (3) to (5) until the termination generation is reached.

[0069] (7) Use the optimized best ant position coordinates to obtain the optimal initial weight matrix and threshold vector, and substitute the obtained optimal weight and threshold parameters into the established BP neural network.

[0070] (8) The optimized BP neural network is trained and tested, and the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are used to determine whether the optimized BP neural network model meets the model requirements.

[0071] According to some embodiments, when obtaining the gas concentration data set, the gas can be collected by a gas monitoring device and a gas analyzer to obtain the gas concentration data set. Among them, the concentration data of O2, CO, CO2, CH4, C2H4, C2H6, and N2 can be collected and used as input values to facilitate the construction of a prediction model based on the ACO-BP neural network, while ensuring data reliability and improving the practicability and operability of the model.

[0072] According to some embodiments, as the main component in the air, the change in the concentration of N2 can reflect the stability of the gas environment in the goaf and the air flow situation. For example, when a large amount of air rushes in, the concentration of N2 may be relatively stable or change little, while the concentrations of other gases participating in the reaction will change. Combining the concentration data of other gases helps to more accurately analyze the coal spontaneous combustion environment, thus providing a more comprehensive basis for coal temperature prediction. Therefore, the concentration of 7 gases, namely O2, CO, CO2, CH4, C2H4, C2H6, and N2, can be selected as the input parameters of the prediction model, and the coal temperature can be used as the model output parameter.

[0073] According to some embodiments, for example, the optimized BP neural network model can be trained under experimental conditions. Specifically, it can include: ① Coal sample collection and processing: Representative coal samples are collected from different areas and different coal seam depths in the coal mine to avoid oxidation, pollution, etc. of the coal samples during collection and transportation. The collected coal samples are crushed to a certain particle size, generally 0.1 - 0.25 mm, to ensure sufficient contact between the coal samples and air and uniform properties of each part, which is convenient for subsequent model training.

[0074] ② Preparation of the coal sample programmed temperature rise experimental device: Among them, the experimental device mainly includes a heating furnace, a sample tank, a gas collection system, a gas analysis instrument, etc. The heating furnace needs to have an accurate temperature control function and can heat the coal sample at a set heating rate; the sample tank is used to hold the coal sample and requires good sealing to prevent gas leakage; the gas collection system is connected to the sample tank through a pipeline and can accurately collect the gas released by the coal sample during the temperature rise process; a gas analyzer such as a gas chromatograph can analyze the composition and concentration of the collected gas.

[0075] ③ Experimental procedure: It can be determined that the prepared coal sample is in the sample tank, and the filling quantity and filling compactness are ensured to be consistent to guarantee the repeatability and comparability of the experiment. It can be determined that the filling is completed, and the sample tank is sealed to meet the requirements and connected to the gas collection system. In this case, the heating rate in the heating furnace control system can be set to 0.5 - 5 °C / min, and the heating range is from room temperature (about 17 °C) to 300 °C, which can simulate the process of coal spontaneous combustion from slow low-temperature oxidation to intense high-temperature reaction. Start the heating furnace and begin the programmed heating of the coal sample. During the heating process, gas is collected at certain time intervals (such as every 5 - 10 min). The gas collection system transports the gas extracted from the sample tank to the gas analyzer for detection. When collecting gas, it is necessary to ensure that there is no residual gas interference in the collection pipeline. The pipeline is purged before each collection to ensure that the collected gas is the gas generated by the coal sample in real time at the current temperature.

[0076] ④ Data collection and processing: The collected gas can be analyzed using a gas chromatograph. The gas chromatograph separates different gas components and determines the concentrations of gases such as O2, CO, CO2, CH4, C2H4, C2H6, N2, etc. according to the response signals of each component on the detector. Different gases have different retention times in the chromatographic column. By comparing with the retention times and peak areas of the standard gases, the concentrations of the gases to be measured are calculated. The gas concentration data measured each time, as well as the corresponding temperature, time, and other information, are recorded in detail.

[0077] When the data is collected, preprocessing can be performed on the collected data, which can specifically include: a. Data standardization: To ensure more efficient gradient descent, more stable training, and higher accuracy of the prediction model, the collected data can be standardized to eliminate the influence of different feature dimensions on model training. In the collected data sample set, for example, O2 and N2 can be in percentage units, and CO, CO2, CH4, C2H4, C2H6 can be in ppm units to ensure the convergence speed and accuracy of the ACO - BP neural network model.

[0078] b. To ensure the reliability and generalization ability of the prediction model, for example, the data samples can be divided into a training sample set and a test sample set according to a ratio of 8:2. For example, 56 groups of data can be obtained from the programmed heating experiments of two coal samples. Among them, 45 groups of sample data are used as the training sample set of the model, and 11 groups of sample data are used as the test sample set. Among them, all the data of the first coal sample are used as training data, and the first 18 groups of data of the second coal sample are used as training data, with a total of 24 groups of sample data as the training data of the prediction model; the remaining 11 groups of sample data in the second group are all used as the test data of the prediction model.

[0079] According to some embodiments, the collected data can be sorted and analyzed to establish a training data sample library for the prediction model of the collected 7 gas concentration and coal temperature data. According to some embodiments, for example, 7 index characteristic data including 6 characteristic gas concentrations (O2, CO, CO2, CH4, C2H4, C2H6) and N2 concentration and coal sample temperature data can be collected. Among them, the division ratio of the sample set is not limited.

[0080] According to some embodiments, the number of training times and the learning rate of the ACO - BP neural network model have a great influence on the accuracy of the model prediction results. Therefore, the optimal learning rate and the number of training times can be gradually approached by observing the results through repeated iterations. Among them, the number of training times of the ACO - BP neural network model can be set to 1000, for example, the learning rate is set to 0.01, and the minimum training error is set to 0.0001. When the specified number of training times is reached or the training error reaches the specified range, the target ACO - BP neural network model is output. Among them, other parameters of the ACO - BP neural network model are determined according to the pre - set information. For example, the display frequency can be set to 25, and it is displayed once every 25 training times. The minimum performance gradient is set to 10 - 6, the maximum number of failures is set to 6, the initial population size in the ACO algorithm is set to 10, and the maximum number of evolutions is set to 50, and the pheromone evaporation coefficient is set to 0.9.

[0081] According to some embodiments, the value range of the number of hidden layer nodes can be determined according to formula (4): (4) In the formula, l is the number of hidden layer nodes, m and n are the number of nodes in the input layer and the output layer respectively, and a is an adjustment constant between 1 and 10. Because there are 7 input parameters and 1 output parameter, the value range of the number of hidden layer nodes is 3 - 12.

[0082] Among them, the learning and prediction functions of the blasting fragmentation prediction model based on the ACO - BP neural network model can be implemented by writing programs on the MATLAB platform.

[0083] According to some embodiments, for example, it can be determined whether the target ACO - BP neural network model is obtained according to the root mean square error (RMSE) and the mean absolute percentage error (MAPE). Among them, RMSE represents the deviation of the predicted data from the actual value, and MAPE represents the error magnitude between the predicted value and the actual value. Their formulas can be as shown in formulas (5) and (6) for example.

[0084] (5) (6) In the formula, n is the number of samples; is the predicted lump size value; is the actual lump size of blasting.

[0085] According to some embodiments, the method further includes: obtaining the prediction accuracy corresponding to the coal; According to the prediction accuracy, the number of first nodes in the input layer, and the number of second nodes in the output layer, determine the number of hidden layer nodes of the target ACO-BP neural network model. Therefore, the matching between the prediction efficiency and the prediction accuracy can be improved, and the convenience and accuracy of temperature prediction can be enhanced. Among them, different numbers of hidden layers can correspond to different prediction accuracies, for example. The prediction accuracy can be, for example, the RMSE value. Among them, the corresponding relationship between different numbers of hidden layer nodes and the root mean square error can be as shown in Table 1, for example.

[0086] Table 1

[0087] According to some embodiments, the number of hidden layer nodes can be 3, for example, and the prediction network model can be as Figure 7 shown.

[0088] Among some embodiments, for example, the collected gas concentration data set can be regularly used to retrain the model and update the model parameters, so as to adapt to factors such as coal quality changes and mining environment changes. An incremental learning strategy can also be adopted to avoid repeating the training of historical data and improve the efficiency.

[0089] In step S24, obtain the attribute information corresponding to the coal; Among them, the related descriptions can be as described above and will not be elaborated here.

[0090] Among some embodiments, the attribute information can be used to indicate the information possessed by the coal itself, for example. The attribute information does not specifically refer to a certain fixed information. The attribute information can include, for example, the volume of the coal, the composition of the coal, etc.

[0091] According to some embodiments, the execution order of step S24 is not limited. It can be before step S25.

[0092] In step S25, determine the temperature change information of the coal according to the attribute information; Among them, the related descriptions can be as described above and will not be elaborated here.

[0093] According to some embodiments, the temperature change information can be used to indicate the temperature change information of the current coal during the spontaneous combustion process, for example. The temperature change information does not specifically refer to a certain fixed information. For example, when the coal changes, the temperature change information can also change accordingly.

[0094] According to some embodiments, the temperature change information of the coal can be determined according to the attribute information.

[0095] In step S26, according to the temperature change information and the first temperature, obtain the stage of coal spontaneous combustion.

[0096] Among them, the relevant descriptions can be as described above and will not be elaborated here.

[0097] In some embodiments, for example, according to the temperature change information and the first temperature, obtain the stage of coal spontaneous combustion. That is, the stage of coal spontaneous combustion can be determined according to the temperature range where the first temperature is located.

[0098] According to some embodiments, the method further includes: Obtain the warning level corresponding to the spontaneous combustion information of the coal; Obtain and execute the warning measures corresponding to the warning level, and display the spontaneous combustion information and warning level of the coal.

[0099] In some embodiments, for example, a temperature change curve and a gas concentration trend chart can also be drawn to visually present the development trend of coal spontaneous combustion.

[0100] According to some embodiments, the attribute information corresponding to the coal can be obtained; the temperature change information of the coal can be determined according to the attribute information; the stage of coal spontaneous combustion can be obtained according to the temperature change information and the first temperature. Therefore, the accuracy of determining the stage of coal spontaneous combustion can be improved.

[0101] Figure 8 It is a flowchart of the third method for predicting the temperature of coal spontaneous combustion based on the GA-BP neural network model provided by the embodiments of the present disclosure. As Figure 8 shown, the method for predicting the temperature of coal spontaneous combustion based on the GA-BP neural network model can be used in the scenario of determining whether the coal is spontaneously combusted, and includes the following steps: In step S31, gas collection is performed on the space where the coal is located through a gas monitoring device at preset time intervals to obtain the gas corresponding to the coal; Among them, the relevant descriptions can be as described above and will not be elaborated here.

[0102] In step S32, the gas is analyzed for its composition and concentration through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; Among them, the relevant descriptions can be as described above and will not be elaborated here.

[0103] In step S33, use the genetic algorithm to update the first weight matrix and the first threshold vector of the initial BP neural network model to obtain the target GA-BP neural network model; According to some embodiments, a genetic algorithm can optimize the initial weight matrix and threshold vector randomly obtained by a BP neural network model. The genetic algorithm uses binary coding to represent the weight matrix and threshold vector of the BP network model, searches for the optimal solution through selection, crossover, and mutation operations, and obtains the optimal weight matrix and optimal threshold vector of the BP neural network model. The genetic algorithm has the core advantages of strong parallel search ability and suitability for large-scale parameter optimization, and its global exploration ability is better than that of the traditional BP algorithm.

[0104] According to some embodiments, a genetic algorithm can be used to update the first weight matrix and the first threshold vector of the initial BP neural network model to obtain a target GA-BP neural network model.

[0105] In step S34, the gas concentration data set is input into the target GA-BP neural network model for recognition processing to predict the second temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined according to the second temperature.

[0106] Among them, the second temperature is the temperature obtained by performing recognition processing using the target GA-BP neural network model. The gas concentration data set can be input into the target GA-BP neural network model for recognition processing to predict the second temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined according to the second temperature.

[0107] In some embodiments, for example, a long short-term memory network (LSTM) can also be used to predict the spontaneous combustion temperature of coal. The long short-term memory network can predict the spontaneous combustion temperature of coal by constructing a three-layer LSTM network to process the time-series temperature data (the input layer includes monitoring values such as coal temperature and oxygen concentration, and the output layer predicts the future temperature). Capture the time-series characteristics in the given text data of the spontaneous combustion temperature of coal from the time dimension, strengthen the model's ability to model long-distance dependencies in time-series data, and use the LSTM context time series as the input of a fully connected neural network to predict the spontaneous combustion temperature of coal at the next moment. This method has the core advantages of being good at dealing with long-term dependencies in time-series data and having strong modeling ability for nonlinear dynamic systems.

[0108] According to some embodiments, a genetic algorithm can be used to update the first weight matrix and the first threshold vector of the initial BP neural network model to obtain a target GA-BP neural network model; the gas concentration data set is input into the target GA-BP neural network model for recognition processing to predict the second temperature corresponding to the coal, and the spontaneous combustion information of the coal is determined according to the second temperature. Therefore, using the target GA-BP neural network model to predict the spontaneous combustion temperature of coal can reduce the situation that when using the BP neural network model for recognition processing, due to the random setting of the initial weights and thresholds of the BP neural network model, it falls into a local optimal solution, resulting in poor accuracy of predicting the spontaneous combustion temperature of coal. Compared with the BP neural network model, the GA-BP neural network model can improve the convergence speed, can improve the prediction accuracy and prediction stability of the GA-BP neural network model, and does not require building a complex test platform underground, which can improve the accuracy and convenience of predicting the spontaneous combustion temperature of coal.

[0109] The block diagram of a device for predicting the spontaneous combustion temperature of coal based on an ACO-BP neural network model shown according to an exemplary embodiment. Refer to Figure 9 , the device 900 includes: A gas acquisition unit 901, configured to collect gas corresponding to the coal in the space where the coal is located through a gas monitoring device at preset time intervals, and obtain the gas corresponding to the coal; A set acquisition unit 902, configured to analyze the components and concentrations of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; A temperature prediction unit 903, configured to input the gas concentration data set into the target ACO-BP neural network model for recognition processing, predict the first temperature corresponding to the coal, and determine the spontaneous combustion information of the coal according to the first temperature.

[0110] According to some embodiments, the set acquisition unit 902 is further specifically configured to: Obtain the preprocessing methods corresponding to the gas concentration data in the gas concentration data set; Process the gas concentration data by using the preprocessing methods to obtain a processed gas concentration data set.

[0111] According to some embodiments, when the set acquisition unit 902 is used to determine the spontaneous combustion information of the coal according to the first temperature, it is specifically configured to: Obtain the attribute information corresponding to the coal; Determine the temperature change information of the coal according to the attribute information; Obtain the stage at which the coal spontaneous combustion is located according to the temperature change information and the first temperature.

[0112] According to some embodiments, the set acquisition unit 902 is further specifically configured to: Obtain the training data set, and perform normalization processing on the training data set to obtain the processed training data set; Obtain the topological structure of the initial BP neural network model, and initialize the first weight matrix and the first threshold vector of the BP neural network model to obtain the initial BP neural network model; Perform iterative operations on the ant colony algorithm using the processed training data set to obtain the pheromone content of the ants; According to the highest pheromone content, obtain the displacement information corresponding to the ants, and update the first position of the ants according to the displacement information and the path selection probability to obtain the second position; When the iterative information of the ant colony algorithm meets the information requirements, use the target ant position corresponding to the ant colony algorithm as the second weight matrix and the second threshold vector; Use the second weight matrix and the second threshold vector to update the initial BP neural network model to obtain the updated BP neural network model; Use the processed training data set to train the updated BP neural network model, and when the updated BP neural network model meets the model training requirements, obtain the target ACO-BP neural network model, and use the target ACO-BP neural network model as the target ACO-BP neural network model.

[0113] According to some embodiments, the set acquisition unit 902 is further specifically configured to: Obtain the prediction accuracy corresponding to the coal; Determine the number of hidden layer nodes of the target ACO-BP neural network model according to the prediction accuracy, the number of first nodes in the input layer, and the number of second nodes in the output layer.

[0114] According to some embodiments, the temperature prediction unit 903 is further specifically configured to: Obtain the warning level corresponding to the spontaneous combustion information of the coal; Obtain and execute the warning measures corresponding to the warning level, and display the spontaneous combustion information and the warning level of the coal.

[0115] According to some embodiments, after the temperature prediction unit 903 is used to obtain the gas concentration data set, it is further specifically configured to: Use the genetic algorithm to update the first weight matrix and the first threshold vector of the initial BP neural network model to obtain the target GA-BP neural network model; Input the gas concentration data set into the target GA-BP neural network model for recognition processing, predict the second temperature corresponding to the coal, and determine the spontaneous combustion information of the coal according to the second temperature.

[0116] Regarding the device in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0117] In some or related embodiments, there is a gas acquisition unit for collecting gas in the space where the coal is located through a gas monitoring device at preset time intervals to obtain the gas corresponding to the coal; a set acquisition unit for analyzing the composition and concentration of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; and a temperature prediction unit for inputting the gas concentration data set into a target ACO-BP neural network model for recognition processing, predicting the first temperature corresponding to the coal, and determining the spontaneous combustion information of the coal based on the first temperature. Therefore, using the target ACO-BP neural network model for predicting the spontaneous combustion temperature of coal can reduce the situation where the BP neural network model falls into a local optimal solution due to the random setting of the initial weights and thresholds of the BP neural network model during recognition processing, resulting in poor prediction accuracy of the spontaneous combustion temperature of coal. Compared with the BP neural network model, the ACO-BP neural network model can improve the convergence speed, can improve the prediction accuracy and prediction stability of the ACO-BP neural network model, and does not require building a complex test platform underground, which can improve the accuracy and convenience of predicting the spontaneous combustion temperature of coal.

[0118] Figure 10 It is a block diagram of a network device 1000 provided by an embodiment of the present disclosure. For example, the network device 1000 may be provided as a network device. Referring to Figure 10 , the network device 1000 includes a processing component 1022, which further includes at least one processor, and memory resources represented by a memory 1032 for storing instructions executable by the processing component 1022, such as application programs. The application programs stored in the memory 1032 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 1022 is configured to execute instructions to perform any of the methods described above applied to the network device.

[0119] The network device 1000 may further include a power supply component 1027 configured to perform power management of the network device 1000, a wired or wireless network interface 1050 configured to connect the network device 1000 to a network, and an input / output (I / O) interface 1058. The network device 1000 may operate based on an operating system stored in the memory 1032, such as Windows Server TM, Mac OS XTM, Unix TM, Linux TM, Free BSDTM or the like.

[0120] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0121] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0122] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0123] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0124] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), the Internet, and blockchain networks.

[0125] A computer system can include a client and a server. The client and the server are generally far from each other and usually interact with each other through a communication network. The relationship between the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services ("Virtual Private Server", or simply "VPS"). The server can also be a server of a distributed system, or a server combined with blockchain.

[0126] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and no limitation is made herein.

[0127] The above specific embodiments do not constitute a limitation to the scope of protection of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present disclosure shall be included within the scope of protection of the present disclosure.

Claims

1. A method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model, characterized in that Including: Performing gas collection on the space where the coal is located through a gas monitoring device at preset time intervals to obtain the gas corresponding to the coal; Analyzing the composition and concentration of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas; Inputting the gas concentration data set into a target ACO-BP neural network model for recognition processing, predicting the first temperature corresponding to the coal, and determining the spontaneous combustion information of the coal according to the first temperature.

2. The method according to claim 1, wherein The method further includes: Obtaining the preprocessing method corresponding to each gas concentration data in the gas concentration data set; Processing each gas concentration data by using the preprocessing method to obtain a processed gas concentration data set.

3. The method according to claim 1 or 2, characterized in that, The determining the spontaneous combustion information of the coal according to the first temperature includes: Obtaining the attribute information corresponding to the coal; Determining the temperature change information of the coal according to the attribute information; Obtaining the stage at which the coal spontaneous combustion occurs according to the temperature change information and the first temperature.

4. The method according to claim 1, wherein The method further includes: Obtaining a training data set and performing normalization processing on the training data set to obtain a processed training data set; Obtaining the topological structure of an initial BP neural network model, initializing the first weight matrix and the first threshold vector of the BP neural network model, and obtaining an initial BP neural network model; Performing iterative operations on the ant colony algorithm by using the processed training data set to obtain the pheromone content of the ants; Obtaining the displacement information corresponding to the ants according to the highest pheromone content, and updating the first position of the ants according to the displacement information and the path selection probability to obtain a second position; When the iterative information of the ant colony algorithm meets the information requirements, using the target ant position corresponding to the ant colony algorithm as the second weight matrix and the second threshold vector; Updating the initial BP neural network model by using the second weight matrix and the second threshold vector to obtain an updated BP neural network model; Training the updated BP neural network model by using the processed training data set, and when the updated BP neural network model meets the model training requirements, obtaining a target ACO-BP neural network model, and using the target ACO-BP neural network model as the target ACO-BP neural network model.

5. The method according to claim 1, wherein The method further includes: Obtaining the prediction accuracy corresponding to the coal; Determining the number of hidden layer nodes of the target ACO-BP neural network model according to the prediction accuracy, the number of first nodes in the input layer, and the number of second nodes in the output layer.

6. The method according to claim 1, wherein The method further includes: Obtaining the warning level corresponding to the spontaneous combustion information of the coal; Obtaining and executing the warning measures corresponding to the warning level, and displaying the spontaneous combustion information of the coal and the warning level.

7. The method according to claim 1, wherein After obtaining the gas concentration data set, it further includes: Updating the first weight matrix and the first threshold vector of the initial BP neural network model by using a genetic algorithm to obtain a target GA-BP neural network model; Input the gas concentration data set into the target GA-BP neural network model for recognition processing, predict the second temperature corresponding to the coal, and determine the spontaneous combustion information of the coal according to the second temperature.

8. A coal spontaneous combustion temperature prediction device based on an ACO-BP neural network model, characterized in that, It includes: A gas acquisition unit, which is used to collect the gas corresponding to the coal in the space where the coal is located through a gas monitoring device every preset time interval. A set acquisition unit, which is used to analyze the components and concentrations of the gas through a gas analyzer to obtain a gas concentration data set, where the gas concentration data set includes concentration data corresponding to at least one gas. A temperature prediction unit, which is used to input the gas concentration data set into the target ACO-BP neural network model for recognition processing, predict the first temperature corresponding to the coal, and determine the spontaneous combustion information of the coal according to the first temperature.

9. A network device, characterized in that, It includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the instructions to implement the method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model according to any one of claims 1 to 7.

10. A storage medium, the storage medium stores instructions, characterized in that, When the instructions are running on the network device, the network device is caused to execute the method for predicting the spontaneous combustion temperature of coal based on the ACO-BP neural network model according to any one of claims 1 to 7.

Images